城市绿地智能提取与时空演变分析

高分辨率遥感影像是城市绿地信息快速提取的主要数据源 文中以多尺度影像分割与面向对象影像分析方法为主要技术 利用样本多边形对象的成员函数建立训练区 自动提取大庆市城市绿地覆盖信息 达到清查城市绿地的目的。该方法信息获取周期短、精度高、成本低 实现了城市绿地信息精确获取与快速更新。

探讨应用高分辨率遥感影像提取城市地区绿地信息。利用自相关函数计算30 个绿地样区, 结果表明在影像位移2 像素时, 自相关系数还能达到0. 95, 从而确定纹理窗口大小为53 5。纹理值计算是在全色影像灰度共生矩阵的基础上, 方向取45°、135°、225°、315°4 个方向的平均值, 计算5 个纹理参数:Mean、variance、homogeneity、contrast、secondmomen t。对全色影像和5 个纹理影像进行多分辨率分割, 对分割所形成的目标根据绿地5 个纹理特征设定阈值, 提取出绿地信息, 通过精度评定正确率达92. 8%。结果表明所采取的方法在高分辨率遥感影像的城市绿地信息的提取上具有很好的应用性。

应用TM影像对南京城市景观生态格局进行研究并对研究方法进行了探讨。为了提取正确的生态绿地专题信息,对提取的方法进行了探讨。研究中尝试用NDVI波段及其它波段组合对南京城市植被进行解译,效果较好。研究结果表明,南京生态绿地分布很不平衡。紫金山区的林地占主体,而繁华的中心城区,生态绿地相对总体绿地来说占据的面积很小。提出了南京生态绿地建设的合理化建议。

以德国Vaihingen城区的高分辨率遥感影像为数据源，提出一种结合多尺度引导滤波特征与核主成分分析特征的提取方法，利用多尺度引导滤波提取不同尺度的绿地特征，通过具有非线性映射能力的核主成分分析算法，对多尺度特征进行降维，最后将降维后的特征输入支持向量机分类器，得到城市绿地的分类结果，并与现有的绿地提取方法进行对比分析。结果表明：该方法能充分利用空间邻域信息，获得比现有单尺度分析方法更高的分类精度，且明显减少传统像素级分类方法产生的结果细碎问题。

本文通过无人机获取四川省北川县高分辨率影像数据，经过空三加密，正射校正等一系列预处理，利用ENVI软件分别计算研究区可见光波段差异植被指数VDVI （visible-band difference vegetation index），归一化绿红差值指数NGRDI （normalized green-red difference index）、归一化绿蓝差值指数NGBDI （normalized green-blue difference index）。采用面向对象的影像分类方式，分别提取城市绿地并进行精度评价。结果表明：3种植被指数均能较好地提取城市绿地，总体提取精度均在83%以上，其中VDVI提取效果最优，总体精度达到89.5%。因此，利用无人机遥感技术进行城市绿地的提取统计是可行的。基于VDVI统计结果，通过去除小斑块以及目视解译校正城市绿地分类结果，统计得到北川县建成区绿地面积为2.3948 km 2 ，城市绿化覆盖率为40.04%。

利用GS变换、主成分分析、Ehlers变换、Wavelet分析、HIS变换5种方法对城区WorldView-2和PL-1A影像进行融合，并从影像融合质量和绿地信息提取精度两方面对融合方法的有效性进行了评价。结果表明：① 5种融合方法中，GS变换融合的效果最好；主成分分析和Ehlers变换融合WorldView-2质量较好，但融合PL-1A影像质量较差；Wavelet变换、HIS变换融合两种影像质量都较差；② 用于绿地信息提取时，GS、PCA融合影像获取的精度最高，其次为Ehlers、Wavelet融合影像，均明显高于多光谱影像的提取精度；Ehlers、Wavelet变换精度最低，绿地信息提取精度低于多光谱影像的提取精度。可以得出，影像融合可以明显地提高绿地信息提取精度，5种影像融合方法中，GS变换普适性较好，影像融合质量最好，提高分类精度效果最明显。

随着城市化不断推进，北京市中心城社会经济发展和环境破坏严重威胁着绿色空间发展，理清绿色空间的演变机制为绿地系统规划方案制定提供重要的理论依据。研究以北京市中心城为对象，选择1992年、2000年、2008年和2016年4个重要节点，对其遥感影像进行解译，探究北京中心城绿色空间的时空变化并分析其转变影响因素。研究表明，研究期北京市中心城耕地、林地和湿地及水域面积减少，草地面积增加，总绿色空间大面积减少；中心城用地间的转换主要集中在耕地向建设用地、林地的转换，林地和草地向建设用地的转换上；社会经济发展对北京市中心城绿色空间面积演变影响显著，自然因素对绿色空间演变起到一定限制作用，政策因素对于结构性大型绿色空间的建设具有积极的推动作用。

绿色空间是城市的基础要素之一,对城市环境问题起到一定缓解作用。对福建省整体绿色空间的时空演变及驱动力研究有助于福建省生态文明示范区建设。以2000年、2010年和2020年福建省地表覆盖数据和社会经济及自然数据为基础,运用动态度计算、转移矩阵,结合PLSR模型分析福建省不同尺度的绿色空间演变规律及驱动机制。结果表明① 2000—2020年间,福建省整体绿色空间面积呈下降走向,非绿色空间面积增加,耕地、林地、草地3类绿色空间分别与建设用地之间存在较大面积的转换,转出的位置主要位于福建省沿海地区;② 9个地区的绿色空间面积变化趋势与福建省域总体绿色空间面积的变化趋势基本一致,即绿色空间面积减少,非绿色空间面积增加,绿色空间面积减少的区域大部分位于福建省沿海地区,但各地存在差异;③回归模型结果显示,省域的国民经济、人口增长对各类绿色空间面积均有重要的影响(VIP值均大于1),而其他影响因素作用力大小存在一定差异。社会经济因素对于沿海地区的绿色空间面积具有较重要的影响;自然因素对不同类型的绿色空间面积有不同程度的影响,且对不同地区绿色空间面积的影响有差异。分析揭示了经济发展、产业结构比重、人口增长促使绿色空间转向非绿色空间,且不同区域的社会经济发展差异也导致绿色空间演变存在空间差异。

探究绿色空间演变与驱动因素对于城镇群跨区域协同发展具有重要意义。以南太行城镇群为例,基于2000、2010、2020年土地利用和社会经济统计数据,采用动态度、转移矩阵、景观格局指数分析各类绿色空间的演变规律,通过皮尔森相关性和参数最优地理探测器分析驱动机制。结果表明:1)2000—2020年,南太行城镇群绿色空间面积呈下降趋势,耕地和草地面积持续减少,耕地大量转为建设用地;2)多数绿色空间的景观格局指数呈下降趋势; 3)绿色空间面积指标与社会经济和自然环境指标相关性显著,地形因子解释力最高。综上可知,近20年各类绿地斑块数量降低、破碎度增加,景观连接性降低;城乡建设发展和自然环境条件是影响绿色空间演变的重要因素,地形地貌是山地城镇群绿色空间演变的直接驱动力。

目的 绿色空间是城乡发展的生态基础，良好的绿色空间体系是协调城市发展和自然保护的重要保障。全面测度黄河下游城市区域绿色空间时空演变规律并探究其演变机制，有助于下游绿色生态走廊建设和滩区生态综合整治中的规划应对。 方法 以黄河下游4个城市济南、菏泽、郑州、新乡为研究对象，以1990、2000、2010、2020年为时间节点，基于土地利用二级分类和植被覆盖密度划分区域绿色空间类型，进而运用转移矩阵、景观格局指数、城乡梯度、地理探测等方法，对绿色空间时空演变的规律与机制开展定量研究。 结果 （1）30年间，绿色空间合计向裸露地表转出3 223 km 2 ，裸露地表向各类绿色空间转出1 181 km 2 ；绿色空间之间互相转换735 km 2 ，其中较高密度向较低密度绿色空间转出466 km 2 ，较低密度向较高密度绿色空间转出269 km 2 。（2）各类绿色空间斑块数量减少，斑块平均面积、功能连通度增加，低密度绿色空间斑块面积比、面积加权平均形状指数减少是景观格局指数变化的普遍规律。（3）城乡梯度上，低密度绿色空间的波峰、波谷移动明显，高、中密度绿色空间相对固定。（4）自然环境因素对济南、郑州、新乡区域绿色空间分布产生主导影响，社会经济因素解释力的累积值增加了3.1% ~ 8.4%，因子间的非线性增强效果逐年提升。 结论 30年间，区域绿色空间总量损失，绿色空间内部转移以中、高密度向低密度的转出为主，各级城镇建成区边缘、宽滩区沿线转移较多；景观格局的总体变化趋势为由明显波动到趋近平稳、由破碎分散向整合连通，中、高密度绿色空间指标的改善态势在区域性中心城市更为明显；城乡梯度特征在城市间、绿色空间类型间存在较大差异；影响因子间对绿色空间地理分布的协同作用渐增，绿色空间逐渐成为自然−社会互构的结果。未来应“因城制宜”地促进城市区域绿色空间由“屏障”转为“枢纽”。

在区域生态环境退化与快速城市化叠加压力下，生态韧性成为区域可持续发展的重要标尺，而绿色空间作为城市生态系统的核心载体，其韧性演变机制亟待系统性解析。从生态韧性视角，探讨了2000－2023年陕西省绿色空间生态韧性的时空演变特征及其驱动机制。基于多源遥感与统计数据，构建“抵抗力-适应力-恢复力”三维生态韧性评估框架，结合地理探测器与GWR方法，解析陕西省绿色空间生态韧性的时空演变特征及驱动机制。结果表明，1）2000－2023年陕西省绿色空间规模整体缩减1877 km 2 ，但结构显著优化。其中耕地持续减少5096 km 2 （集中于关中平原），林地因退耕还林政策驱动稳步增加2206 km 2 ，水域受气候与人为调控波动增长。2）生态韧性水平整体提升，但空间分异显著。高值区（0.46－0.86）集中于陕南秦巴山地（林地覆盖率高、景观连通性强）及延安南部退耕还林区，低值区（0－0.16）分布于关中平原（耕地主导型结构脆弱）和榆林北部（干旱与能源开发叠加压力）。3）驱动机制呈现显著空间分异。降水提升陕北生态韧性，陕南因水资源过剩存在风险；温度主要加剧汉中、安康、延安北部压力；坡度在榆林与延安交界处支撑韧性，而在延安中部削弱韧性。人口密度在榆林北部城区促进韧性，但在延安北部和关中（如西安、咸阳）加剧脆弱性；GDP与绿色空间生态韧性在榆林北部呈负相关，在延安以及部分陕南区域呈正向效应；土地利用强度在全省普遍削弱韧性，尤其在关中地区表现突出。

防尘防护绿网（防尘绿网）作为抑尘的主要措施被建筑工地广泛使用。快速获取防尘绿网覆盖及时空变化信息，对防尘抑尘、生态环境保护措施的制定具有重要指导意义。本文基于Sentinel-2时间序列遥感影像，使用DeepLabv3+深度学习语义分割模型生成了济南市中心城区2016年—2020年逐年防尘绿网数据，随后利用景观格局、重心-标准差椭圆等方法分析了其空间分布特征和时空扩张趋势。研究结果表明：（1）本文方法的分割精准度、召回率、F1值、IoU分别为84.05%、80.09%、0.82、69.72%，可快速准确地提取城市防尘绿网进而实现大范围、长时序的防尘绿网动态监测。（2）相较于传统的遥感分类方法和其他语义分割模型，本方法的精度最优，防尘绿网的提取结果更精细；另外对北京和天津市的提取实验也证实了本模型的迁移能力。（3）2016年—2020年济南市防尘绿网铺设范围明显扩张，斑块面积和数量不断增多，破碎程度和形状复杂度不断增强，平均斑块面积增大，斑块间凝聚度和聚集度呈波动状态，景观格局复杂且不稳定；绿网铺设范围的扩张呈现出明显的方向性。防尘绿网的分布形态与动态扩张变化受到城市规划、项目进程等人为因素影响，城市规划决定了防尘绿网的分布，而防尘绿网的使用现状在一定程度上又反映了城市建设进程。因此，利用遥感手段对城市防尘绿网进行动态监测与管理，可为城市规划、生态环境建设和城市精细管理提供数据和技术支持，对城市扩张模式与城市重建工作管理具有重要意义。

Land Use Land Cover (LULC) mapping is essential for urban and resource planning, and is one of the key elements in developing smart and sustainable cities.This study evaluates advanced LULC mapping techniques, focusing on Look-Up Table (LUT)-based Atmospheric Correction applied to Cartosat Multispectral (MX) sensor images, followed by supervised and semi-supervised learning models for LULC prediction. We explore DeeplabV3+ and Cross-Pseudo Supervision (CPS). The CPS model is further refined with dynamic weighting, enhancing pseudo-label reliability during training. This comprehensive approach analyses the accuracy and utility of LULC mapping techniques for various urban planning applications. A case study of Hyderabad, India, illustrates significant land use changes due to rapid urbanization. By analyzing Cartosat MX images over time, we highlight shifts such as urban sprawl, shrinking green spaces, and expanding industrial areas. This demonstrates the practical utility of these techniques for urban planners and policymakers.

Urban forests play a key role in enhancing environmental quality and supporting biodiversity in cities. Mapping and monitoring these green spaces are crucial for urban planning and conservation, yet accurately detecting trees is challenging due to complex landscapes and the variability in image resolution caused by different satellite sensors or UAV flight altitudes. While deep learning architectures have shown promise in addressing these challenges, their effectiveness remains strongly dependent on the availability of large and manually labeled datasets, which are often expensive and difficult to obtain in sufficient quantity. In this work, we propose a novel pipeline that integrates domain adaptation with GANs and Diffusion models to enhance the quality of low-resolution aerial images. Our proposed pipeline enhances low-resolution imagery while preserving semantic content, enabling effective tree segmentation without requiring large volumes of manually annotated data. Leveraging models such as pix2pix, Real-ESRGAN, Latent Diffusion, and Stable Diffusion, we generate realistic and structurally consistent synthetic samples that expand the training dataset and unify scale across domains. This approach not only improves the robustness of segmentation models across different acquisition conditions but also provides a scalable and replicable solution for remote sensing scenarios with scarce annotation resources. Experimental results demonstrated an improvement of over 50% in IoU for low-resolution images, highlighting the effectiveness of our method compared to traditional pipelines.

The modernization of offi cial statistics involves the use of new data sources, such as data collected through remote sensing. The document contains a description of how an urban green index, derived from the SDG 11.7 objective, was obtained for Romania's 41 county seat cities based on free data sets collected by remote sensing from the European and North American space agencies. The main result is represented by an estimate of the areas of surfaces covered with vegetation for the 40 county seat towns and the municipality of Bucharest, relative to the total surface. To estimate the area covered with vegetation, we used two data sets obtained by remote sensing, namely data provided by the MODIS mission, the TERRA satellite, and data provided by the Sentinel 2 mission from the Copernicus space program. Based on the results obtained, namely the surface area covered with vegetation, estimated in square kilometers, and the percentage of the total surface area or urban green index, we have created a national top of the county seat cities

ABSTRACT The recent global urban expansion has seen a massive decline and loss in the connectivity of urban vegetation. In urban areas, some vegetation patches have increasingly become isolated and less connected by a matrix composed of impervious surfaces and transportation networks like roads. Vegetation fragmentation is a global threat to the remaining urban green spaces and has an impact on biodiversity conservation, environmental quality and urban microclimates. So far, a lot of work has been done on mapping and monitoring urban green spaces, some using conventional methods and of late using remotely sensed data. However, not much is known or well documented on the new developments in the remote sensing of vegetation fragmentation in urban areas over the last two decades. Thus, the objective of this research work is to present a detailed and comprehensive synthesis of the progress of remote sensing in assessing and monitoring landscape structure of urban green spaces and vegetation fragmentation. Specifically, scientific literature from the year 2000 to 2020 was reviewed to provide a state-of-the-art progress on the remote sensing of vegetation fragmentation in urban areas. Results indicate that between 2000 and 2020, there was a considerable increase in the number of scientific publications on vegetation fragmentation in urban landscapes. The discrete landscape pattern indices are the most widely used method. Comparatively, Land Remote Sensing Satellite System (Landsat) data was widely used due to its suitable spatial and temporal resolution, free availability and the presence of historical archival data that spans over a period of 40 years. The most commonly used scale was local (a city and/or municipality) followed by regional (more than one municipality in one continent) and then global (several selected cities and urban areas across continents). Only two studies were conducted at the global level. Further, geographic bias was observed in most of the accessed vegetation fragmentation studies. The review showed that vegetation fragmentation studies were done mostly for cities in China, North America and Europe while cities in most parts of Africa, Asia, Eurasia, Oceania and South America have not been comprehensively studied. The study also showed that the majority of cities across the globe have experienced severe vegetation fragmentation over the years. This review underscores the relevance of scientific findings in urban and spatial planning to minimize the loss of urban green spaces and to conserve and restore affected areas.

Abstract Urban green space (UGS) is an indispensable component of urban environmental systems and is important to urban residents. Both physical features (e.g., shrubs, trees) and social functions (e.g., public parks, green buffers) are important in UGS mapping. Most UGS studies rely solely on remote sensing data to conduct UGS mapping of physical features, and few studies have focused on UGS mapping from a social function perspective. Due to the limitations of remote sensing in identifying social features; social sensing, which can reflect socioeconomic characteristics, is needed. As a result, a novel methodological framework for integrating these two different data sources to conduct the social functional mapping of UGS has been required. Consequently, we first extracted vegetation patches from an area in Beijing, via the Hyperplanes for Plant Extraction Methodology (HPEM) and considered the parcels segmented by the OpenStreetMap (OSM) road networks as the basic analytical units. Then, near-convex-hull analysis (NCHA) and text-concave-hull analysis (TCHA) were performed to integrate the multi-source data. The results show that the Level I and Level II (refer to Table 3) social function types of UGS had overall accuracies of 92.48% and 88.76%, respectively. Our study provides an improved understanding of UGS and can assist government departments in urban planning. It can also help researchers broaden their research scope by acting as a freely available data source for their work.

Abstract Taking the main city of Fuzhou as the study area, the relationship between the spatiotemporal evolution of urban green space (UGS) and the urban thermal environment from 1993 to 2013 was investigated using a set of remote sensing images. The evolution of UGS is obvious in the study area, where UGS loss (42.83 km2) > UGS extension (4.99 km2) > UGS exchange (2.61 km2). UGS loss affects forest/grass > water > wetland. Furthermore, the area defined as high temperature zones increased by 23.11 km2 in 2013, twice as much as that in 1993. However, the influence of UGS on the urban thermal environment differs by type and evolution: water has the greatest cooling effect, followed by wetland and forest/grass, and UGS loss (8.67 ℃) > UGS exchange (4.00 ℃) > UGS extension (2.90 ℃) > UGS unchanged (2.45 ℃). Finally, the vegetation and cooling index classified the mechanism of temperature response induced by different types of UGS evolution. The evolution of UGS loss usually simulated the movement of the corresponding pixel from the low land surface temperature and high vegetation coverage to the opposite situation. Regression analyses demonstrated that the effect of elevated land surface temperature generated from the reduction of water and forest/grass reached 0.81 ℃ and 0.72 ℃, respectively, in 20 years, indicating that the loss of a significant amount of UGS during urbanization was the primary influence on the urban thermal environment. This study may provide more useful information for researchers and decision-makers engaged in urban planning, urban regeneration, and sustainable land development, especially focusing on the issues of climate adaption and the urban heat island (UHI) effect mitigation.

The real-time, accurate, and refined monitoring of urban green space status information is of great significance in the construction of urban ecological environment and the improvement of urban ecological benefits. The high-resolution technology can provide abundant information of ground objects, which makes the information of urban green surface more complicated. The existing classification methods are challenging to meet the classification accuracy and automation requirements of high-resolution images. This paper proposed a deep learning classification method for urban green space based on phenological features constraints in order to make full use of the spectral and spatial information of green space provided by high-resolution remote sensing images (GaoFen-2) in different periods. The vegetation phenological features were added as auxiliary bands to the deep learning network for training and classification. We used the HRNet (High-Resolution Network) as our model and introduced the Focal Tversky Loss function to solve the sample imbalance problem. The experimental results show that the introduction of phenological features into HRNet model training can effectively improve urban green space classification accuracy by solving the problem of misclassification of evergreen and deciduous trees. The improvement rate of F1-Score of deciduous trees, evergreen trees, and grassland were 0.48%, 4.77%, and 3.93%, respectively, which proved that the combination of vegetation phenology and high-resolution remote sensing image can improve the results of deep learning urban green space classification.

… for UGS area correction based on landscape pattern indices. The results of this study will facilitate the application of remote sensing data at different spatial resolutions in urban areas. …

A recently conducted study by the Centers for Disease Control and Prevention encouraged access to urban green space for the public over the prevalence of COVID-19 in that exposure to urban green space can positively affect the physical and mental health, including the reduction rate of heart disease, obesity, stress, stroke, and depression. COVID-19 has foregrounded the inadequacy of green space in populated cities. It has also highlighted the extant inequities so as to unequal access to urban green space both quantitatively and qualitatively. In this regard, it seems that one of the problems related to Malatya is the uncoordinated distribution of green space in different parts of the city. Therefore, knowing the quantity and quality of these spaces in each region can play an effective role in urban planning. The aim of the present study has been to evaluate urban green space per capita and to investigate its distribution based on the population of the districts of Battalgazi county in Malatya city through developing an integrated methodology (remote sensing and geographic information system). Accordingly, in Google Earth Engine by images of Sentinel-1 and PlanetScope satellites, it was calculated different indexes (NDVI, EVI, PSSR, GNDVI, and NDWI). The data set was prepared and then by combining different data, classification was performed according to support vector machine algorithm. From the landscaping maps obtained, the map was selected with the highest accuracy (overall accuracy: 94.43; and kappa coefficient: 90.5). Finally, by the obtained last map, the distribution of urban green space per capita and their functions in Battalgazi county and its districts were evaluated. The results of the study showed that the existing urban green spaces in the Battalgazi/Malatya were not distributed evenly on the basis of the districts. The per capita of urban green space is twenty-four regions which is more than 9m2 and in twenty-three ones is less than 9m2. The recommendation of this study was that Türkiye city planners and landscape designers should replan and redesign the quality and equal distribution of urban green spaces, especially during and following COVID-19 pandemic. Additionally, drawing on the Google Earth Engine cloud system, which has revolutionized GIS and remote sensing, is recommended to be used in land use land cover modeling. It is straightforward to access information and analyze them quickly in Google Earth Engine. The published codes in this study makes it possible to conduct further relevant studies.

… urban green space distribution by focusing on the landscape fragmentation in city of Osmaniye using remote sensing … vegetation index (NDVI) and urban landscape ratio (ULR) were …

Urban green space (UGS) plays a pivotal role in improving urban ecosystem services and building a livable environment for urban dwellers. However, remotely sensed investigation of UGS at city scale is facing a challenge due to the pixels’ mosaics of buildings, squares, roads and green spaces in cities. Here we developed a new algorithm to unmix the fraction of UGS derived from Landsat TM/ETM/8 OLI using a big-data platform. The spatiotemporal patterns and dynamics of UGSs were examined for 70 major cities in China between 2000 and 2018. The results showed that the total area of UGS in these cities grew from 2780.66 km2 in 2000 to 6764.75 km2 in 2018, which more than doubled its area. As a result, the UGS area per inhabitant rose from 15.01 m2 in 2000 to 18.09 m2 in 2018. However, an uneven layout of UGS occurred among the coastal, western, northeastern and central zones. For example, the UGS percentage in newly expanded urban areas in the coastal zone rose significantly in 2000–2018, with an increase of 2.51%, compared to the decline in UGS in cities in the western zone. Therefore, the effective strategies we have developed should be adopted to show disparities and promote green infrastructure capacity building in those cities with less green space, especially in western China.

… In such urban development endeavour, green spaces planning … quality in Colombo city based on green spaces and to provide … In this study, green space areas were extracted from …

… green space for proper urban management. This study analyzed green spaces of Dhaka city between 1989 and 2020 (30 years) using GIS and Remote Sensing. 30-meter resolution …

… Furthermore, the remotely sensed maps allow the derivation of certain social indicators, such as the accessibility of urban green space. Finally, remote sensing is used to generate maps …

ABSTRACT The water consumption of green space in a large region is difficult to attain through traditional methods. In this article, a practical method is developed using different sources of remote-sensing data. The green space was first derived from a high spatial resolution RapidEye image using the stratified classification method. Then the primary vegetation types of green space were identified using the object-oriented classification method. Afterwards regional green space evapotranspiration was inversed based on multi-temporal Landsat 8 images using the Surface Energy Balance Algorithm for Land model. Finally, water consumption patterns for different types of vegetation were analysed, and regional water consumption was estimated. The method was applied to the northwest region of Beijing City with an area of 147.5 km2 where the green space area was 56.87 km2, and the deciduous broadleaf forest area was the largest among six vegetation types. The total quantity of water consumption for green space in the growing period in the study region was 41.52 × 106 m3 (Mm3). The quantity of water consumed by different types of vegetation in an order from high to low were deciduous broadleaf forest, mixed green space, grassland, evergreen needleleaf forest, golf course, and aquatic vegetation, ranging from 17.43 to 0.79 Mm3. The results are helpful for identifying vegetation types, monitoring vegetation growth status, managing green space, and optimizing green space ecological functions in the Beijing region. The method presented in this article, having higher accuracy and more convenience, has great potential to be applied to other areas across the world.

Urban Green Spaces (UGS) offer social and environmental benefits that enhance quality of life of the residents. However, due to the underlying social and economic disparities, different sections of urban population have disproportionate level of access to UGS. The environmental inequity owing to the varied UGS distribution poses a challenge to urban planners in efficient resource allocation. This study attempts to counter this challenge using a novel remote sensing-based approach. The variations in UGS distribution (in terms of quantity, quality and accessibility) across the neighbourhoods in Mumbai vis-à-vis the socio-economic status (SES) of neighbourhood residents are assessed using remote sensing-based indicators. Further, as these indicators are susceptible to the effect of changing scales, a multi-scale approach is adopted to study the potential variations in the relationship between SES and spatial metrics of UGS with spatial resolution. The neighbourhood SES was assessed using the newly developed Socio-Economic Status Index (SESI) and the neighbourhoods were classified into multiple SES categories. The UGS were extracted from remotely sensed data using Normalized Difference Vegetation Index (NDVI), and their spatial distribution aspects were characterized using indicators at neighbourhood level. The variations in indicators of UGS distribution in the neighbourhoods belonging to different SES categories were analysed using a logistic regression model. The results showed that, while quantity of UGS is not statistically associated with neighbourhoods SES, the quality and accessibility aspects of UGS share a statistically significant relation with SES. Also, this relation was found to vary significantly with spatial resolutions. Further, it was found that the neighbourhoods with higher SES in Mumbai have a better access to green spaces, indicating spatial inequities in UGS distribution in Mumbai. This study has important implications for planning equitable green spaces in cities that are currently in urbanization transition.

Urban green space, discovered by optical remote sensors, is the area covered by terrestrial vegetation in urban areas, and is considered an important factor in urban sustainability. Two sensors ALOS/AVNIR-2 and Landsat/OLI&TIR were used in this study to determine green space by Maximum Likelihood Classification method. The investigated area was Nha Trang city, located in the central Vietnam. This was found that the impervious surfaces were rapidly increased leading to significantly reduce urban green space within 10 years from 2007–2017. In urban areas, the green index was very low compared to the TCXDVN 9257: 2012. Based on the Markov chain, it is projected that over the next 10 years, the total vegetation cover of the city will continue to decline compared to that of today. This is likely to lead to increase catastrophe and environmental risks, especially floods and erosion in the coastal city of Nha Trang. The process could be very useful in mapping urban green space as indicator serving city sustainable development.

… by urban expansion. This study aims to investigate the spatial-temporal fragmentation trends of urban green spaces … to understand trends in urban green spaces from 1986 to 2021. Five …

Urban green spaces are known to provide ample benefits to human society and hence play a vital role in safeguarding the quality of life in our cities. In order to optimize the design and management of green spaces with regard to the provisioning of these ecosystem services, there is a clear need for uniform and spatially explicit datasets on the existing urban green infrastructure. Current mapping approaches, however, largely focus on large land use units (e.g., park, garden), or broad land cover classes (e.g., tree, grass), not providing sufficient thematic detail to model urban ecosystem service supply. We therefore proposed a functional urban green typology and explored the potential of both passive (2 m-hyperspectral and 0.5 m-multispectral optical imagery) and active (airborne LiDAR) remote sensing technology for mapping the proposed types using object-based image analysis and machine learning. Airborne LiDAR data was found to be the most valuable dataset overall, while fusion with hyperspectral data was essential for mapping the most detailed classes. High spectral similarities, along with adjacency and shadow effects still caused severe confusion, resulting in class-wise accuracies <50% for some detailed functional types. Further research should focus on the use of multi-temporal image analysis to fully unlock the potential of remote sensing data for detailed urban green mapping.

With advancements in urbanization, natural lands are constantly being encroached upon by artificial impervious surfaces, leading to serious ecosystem damage. Calls for Green Infrastructure to address urban environmental issues and resource reallocation are growing. How to optimize Green Infrastructure networks are becoming increasingly important under rapid urbanization. In this study, we used the main city zone in Hangzhou as the study area, and we extracted 2000, 2010 and 2020 land-use data. We used morphological spatial pattern analysis to identify Green Infrastructure landscape types and further extract Green Infrastructure elements. We identified the spatial priority of Green Infrastructure network elements through landscape connectivity evaluation according to ecological importance and development vulnerability. After the construction of a Green Infrastructure network, we analyzed its spatio-temporal characteristics to determine the Green Infrastructure network’s spatial priority. Through spatial prioritization, the gradual construction and optimization of Green Infrastructure networks will help to improve urban green spaces in stages. Smartly coordinating urban growth and ecological protection based on Green Infrastructure spatial prioritization may help improve urban living environments and enhance sustainable urban development capabilities. In conclusion, sources dominate corridors and codes are changing. If sources are fragmented, the integration degree decreases and the first-level source advantage is weakened. The corridor morphology continuously develops, and the corridor structure stabilizes. Second-level corridors gradually replace third-level corridors to guide Green Infrastructure network structure development. Codes present a scatter distribution and tend to average, closely following corridor change.

… spatiotemporal characteristics of surface urban heat island (SUHI) effect from the perspective of urban expansion and urban green infrastructure. … The core area of green infrastructure in …

BackgroundThe spatiotemporal analysis of urban land use/land cover change (LULCC) helps to understand the dynamics of the changing environment of green infrastructure (GI) on the basis of sustainable city development. There are important links between spatiotemporal land use/land cover and GI change in urban areas. Therefore, the main objective of this study was to examine the spatiotemporal trends of urban land use/land cover and GI changes in Bahir Dar and Hawassa cities for the last four decades (1973–2015). Three different sets of Landsat satellite data were procured from EMA for Bahir Dar and Hawassa from 1973, 2000 and 2015 using Landsat 4 MSS, 7 TM and 8 OLI respectively. Based on this, using ERDAS Imagine (ver. 9.2) and Arc GIS (Ver.10.3) five LULCC classes were identified for analysis purpose.ResultThe results show that vegetation decreased by 30 and 14% in Bahir Dar and Hawassa respectively for the period 1973–2015, while built-up areas expanded by 10 and 24% respectively in the two cities. These land use changes have significant impacts on spatiotemporal trends of GI in urban areas. GI has increased in Bahir Dar and Hawassa in association with built-up area expansion and deliberate activity of city administrations with effective implementation of spatial plans of corresponding cities.ConclusionsThere is a growing concern about GI in cities. Policy makers and stakeholders should also decide on how to use the land at present and in the future. LULCC policymaking processes should aim to balance GI and other types of land use/land cover for sustainable urban development. Urban LULCC has important effects on the urban GI system.

The reasonable layout of green infrastructure is conducive to the low-carbon, livable and high-quality sustainable development of cities. The framework of spatio-temporal evolution characteristics and prediction analysis of Urban Green Infrastructure (UGI) was constructed by integrating morphological spatial pattern analysis (MSPA) and CA-Markov in the study. We analyzed the spatio-temporal evolution characteristics of UGI in Beijing from 1990 to 2019, predicted its future change trend in 2030, and put forward the optimization scheme for the ecological network of UGI. The area change of UGI presented a "V" shape from 1990 to 2019 in Beijing, and the turning point was around 2009. Its spatial distribution revealed a significant heterogeneity. The comprehensive change rate index showed a "rising and then falling" trend from 1990 to 2019. Core with an area of over 1000 km2 had inclined "C" shape, connecting the north, west and south of the study area. Among the three prediction scenarios for 2030, the area of UGI under the ecological conservation priority scenario is the largest, accounting for 86.35% of the total area. The area of UGI under the economic development priority scenario is the smallest, accounting for 76.85%. The optimization of zoning and road network are effective measures to improve the connectivity of UGI in Beijing. This study is beneficial to extend the research ideas of UGI and promote sustainable urban development.

Abstract The ‘green infrastructure typology’ (GIT) scheme is a standardised framework to map and classify urban landscapes into 34 standard classes, each defined by a specific land cover composition and spatial configuration of vegetation. Previous studies have confirmed that GIT classifications can be successfully derived from airborne remote sensing data; nonetheless, the promotion of the GIT scheme as a framework for the assessment of ecosystem services such as ‘climate moderation’ requires further validations using a range of study areas with different vegetation conditions, and datasets from different seasons and times of the day. This study expands on previous research and evaluates the quality of thermal delineations by examining the spatio-temporal patterns and intra-/inter-typology differences of land surface temperatures (LSTs) using Sydney as case study. Further, this paper discusses the advantages and disadvantages of the classification framework and methods for mapping and assessing the thermal conditions of green infrastructure (GI). Evidence indicates a strong spatial dependency of LSTs that may have significant implications in the interpretation and precision of numerical or predictive models. Results for spatial clustering demonstrate that the GIT scheme can be implemented for a rapid identification of hotspots to prioritise urban areas for heat mitigation. Statistical results confirm that LST differences among GIT classes are statistically significant for different times of the day and seasons. Significant thermal contrasts were found for most GITs at daytime (86.9% in summer and 85.5% in winter) and night-time (80.9% in summer and 73.8% in winter). Temperature differences are more distinguishable in summer and daytime, due to longer solar exposure of surfaces. It was found that the cooling effects of pervious surfaces, water and trees are significantly disturbed by transmission of heat from surrounding impervious materials. Despite good thermal differentiations among GITs, a considerable intra-variability of LSTs was detected in classes with a large proportion of impervious materials with contrasting radiative properties. This causes numerous complexities and challenges that should be explored by future studies.

Abstract The morphology of urban green infrastructure (UGI) will affect the quality of urban environment and the way people perceive. The three-dimensional morphological features of UGI have been proven to be the key factors to effect urban ecological environment, which have rarely been incorporated into the UGI morphology in the previous researches. In this paper, a systematic approach to develop a 3D spatiotemporal morphological database for UGI is proposed. The database is built on a complete set of information describing the form of UGI in the plane, the vertical direction and the temporal changes. In addition, three categories of morphological parameters of UGI are calculated and integrated in the database. User operation and visualization and morphological parameter operation is achievable in the database. The database can be further integrated with simulation programs and analytical models so that it can be used in the design and research of various urban sustainable subjects. In a case study, we further create a complete 3D spatiotemporal morphological database of UGI for an urban district of 4km² in Nanjing, China.

Developing green infrastructure (GI) has drawn increasing attention as a strategic planning approach for advancing urban sustainability. The connectivity of green spaces, a central principle of GI, has been considered in planning studies regarding its structure and functions for biodiversity conservation and ecosystem services delivery; however, aspects of GI connectivity across temporal and spatial scales are rarely addressed. This paper aims to develop and apply a method for the GI connectivity analysis at multiple spatiotemporal scales. A transferable and multi-scale workable approach is presented to reveal the changes of structural and spatial heterogeneity of urban GI. Our method includes i) morphological spatial patterns analysis for central and green corridors recognition, ii) a graph-based quantification of GI connectivity based on the Conefor model, and iii) least-cost path analysis for identifying potential green corridors. We apply the GI connectivity analysis method in the Ruhr Metropolitan Area (RMA), one of Europe's largest agglomerations. We use spatial Urban Atlas data from 2006 to 2018. At the metropolitan scale, we find that GI connectivity in the RMA decreases 3.9% from 2006 to 2018, even though the general distributions of GI changes only slightly. With reference to the municipal scale from 2006 to 2018, four major types of GI connectivity changes were discovered in RMA's 15 cities, namely consistent decreasing, consistent increasing, increase followed by decrease, and vice-versa. Our findings provide new evidence on GI connectivity changes across a twelve-year difference and at metropolitan and municipal scales, as well as the identification of priority areas for increasing GI connectivity. It provides insights on the evolving and heterogenous nature of GI connectivity in support of decision-making for more sustainable metropolitan development for people and nature.

Quantifying the dynamics of green infrastructure (GI) in agricultural peri-urban areas is of great significance to the regional ecological security, food security, and the sustainable development of urban integration. Based on remote sensing images, this study aims to provide a spatiotemporal dynamic assessment of the GI in Baisha District from 2007 to 2018 to improve the layout of GI and planning policies from the perspective of ecological security and food security. Research methods include landscape pattern indices, spatial autocorrelations, and grid analyses in this case study. The results suggest that ensuring the dominant position of farmland is critical to maintaining the composition and connectivity of the overall GI. The recreation, inheritance of farming culture, and ecosystem service functions of farmland should be improved to meet the growing needs of urban residents. GI includes the farmland, greenspace, and wetland on both sides of the Jialu River that should be retained and restored as much as possible to protect natural ecological processes. Simultaneously, construction of important urban facilities and residential areas in flooded areas should be banned. A part of the evenly distributed large greenspace patches should be moved to both sides of the Jialu River to increase the agglomeration effect of GI. Optimization measures in this case study also offer a perspective for other agricultural peri-urban areas that have experienced similar urbanization.

Integrating green infrastructure in urban planning for urban sustainability to stay environmentally equitable, ecologically resilient, and climate adaptive is gradually becoming significant. Using remote sensing data, GIS analytical methods, and urban forestry indicators, this study analyses the spatiotemporal changes in the urban green space and forest coverage of the Sichuan Province of China during 2002–2022. The results show a 20% to 40% addition to urban green space and a 24% to 38% extension in forest coverage resulting from urban greening programmes and reforestation schemes. Urban sprawl has contributed to biodiversity loss, the fragmentation of habitats, and a reduced carbon sequestration potential, notably in peri-urban areas. To address these issues, we propose sustainable green infrastructure by introducing nature-based solutions, carbon offset strategies, and ecological connectivity corridors. Specific proposed policies encompass enhancing the urban forestry legal framework, establishing ecological red lines, and optimising land use policies by coordinating urban development with ecological conservation. This work provides a scientific foundation for urban planners and policymakers to enhance climate resilience, carbon neutrality, and sustainable urban ecosystems.

The contemporary globalized world characterizes the rapid population growth, its significant concentration in cities, and an increase in the urban population. Currently, many socio-cultural, economic, environmental, and other challenges are arising in modern cities, negatively affecting the state of the urban environment, health, and quality of life. There is a need to study these problems in order to solve them. Urban Green Areas (UGAs) are a part of the social space and a vital part of the urban landscape. They act as an environmental framework of the territory and a factor ensuring a more comfortable environment of human life. This study aims at substantiating the importance of the UGAs, identifying the spatiotemporal dynamics of their functioning, and transforming changes in their infrastructure given the expansion of their functions. This research was carried out as a case study of the second city in Ukraine, Kharkiv. The authors developed and used an original integrated approach using urban remote sensing (URS) and GIS for changes detection to evaluate the current state and monitor spatial transformations of the UGAs. We used several GIS platforms and online resources to overcome the lack of digital cadastre of the thematic municipal area of Kharkiv. This resulted in the present original study. The study analyses the dynamics of the spatial and functional organization of the UGAs according to the Master Plans, plans, maps, and functional zoning of the city for the period from 1867 to 2019. The peripheral green areas became important after the large-scale development of the extensive residential areas during the rapid industrial development in remote districts of the city. They provide opportunities for population recreation near living places. Central UGAs are modern, comprehensively developed clusters with multidisciplinary infrastructure, while the peripheral UGAs are currently being developed. The use of URS/GIS tools in the analysis of the satellite images covering 2000–2020 allowed identifying the factors of the UGAs losses in Kharkiv and finding that UGAs were not expanding and partially shrinking during the study period. It is caused by the intensive construction of the residential neighborhoods, primarily peripheral areas, infrastructure development, and expansion of the city transport network. Nonetheless, some sustainable trends of UGA functioning without more or less significant decrease could be proved as existing in a long-term perspective. The authors analyzed and evaluated changes and expansion of the UGAs functions according to modern social demand. The research value of this is the usage of different approaches, scientific sources, URS/GIS tools to determine the UGAs transformation in the second-largest city in Ukraine (Kharkiv), to expand and update the main functions of UGAs and their role in the population’s recreation. The obtained scientific results can be used to update the following strategies, programs, and development plans of Kharkiv.

Under the background of rapid urbanization and frequent global climate extreme events, giving full play to the ecological services and functions of blue-green infrastructure has become an important way to ensure regional sustainable development. However, current studies on blue-green infrastructure resilience are mostly limited to single-dimension assessments, overlooking the systematic disclosure of the evolution and driving mechanism of comprehensive ecological resilience under challenges like flood risk, fragile ecosystems, and biodiversity decline, causing fragmented cognition. Taking the Dawen River Basin as an example, this study innovatively integrated the “pattern - process - function” cascading coupling framework of resilience, combined with Fragstats 4.2, InVEST and other models, analyzed the spatial and temporal changes of watershed pattern, process and functional resilience from 1990 to 2020, and explored the driving factors affecting ecological resilience with the help of the optimal parameter geographic detector. Furthermore, the geographically weighted regression model is used to reveal the spatio-temporal non-stationarity of the main driving factors. The results show that: (1) From 1990 to 2020, the proportion of the total area of blue-green infrastructure in the Dawen River Basin has always remained above 80%, which is dominated by green infrastructure, but it shows a downward trend from 90.17% to 82.73%. (2) The ecological resilience of the blue-green infrastructure in the basin shows a step-change trend of “descending, ascending, ascending”, in which the pattern and process resilience show a fluctuating downward trend, decreasing by 2.12% and 3.37% respectively, while the functional resilience shows an increasing trend, increasing by 6.78%. (3) The absolute values of regression coefficients of slope length factor, runoff velocity, blue-green infrastructure connectivity, night light and policy factors continue to increase, and the intensity of their influence on ecological resilience is increasing year by year. To improve ecological resilience in the future, on the basis of protecting and expanding high-resilience regions, differentiated strategies should be implemented for low-resilience regions, so as to enhance the spatial and functional connections among various infrastructures, and systematically build an interconnected regional ecological resilience network.

… a good spatio-temporal correspondence with … spatio-temporal processes offer many opportunities for identifying, protecting and restoring key elements in an urban green infrastructure …

In recent decades, the City of Stockholm, Sweden, has grown substantially and is now the largest city in Scandinavia. Recent urban growth is placing pressure on green areas within and around the city. In order to protect biodiversity and ecosystem services, green infrastructure is part of Stockholm municipal planning. This research quantifies land-cover change in the City of Stockholm between 2003 and 2018 and examines what impact urban growth has had on its green infrastructure. Two 2018 WorldView-2 images and three 2003 QuickBird-2 images were used to produce classifications of 11 land-cover types using object-based image analysis and a support vector machine algorithm with spectral, geometric and texture features. The classification accuracies reached over 90% and the results were used in calculations and comparisons to determine the impact of urban growth in Stockholm between 2003 and 2018, including the generation of land-cover change statistics in relation to administrative boundaries and green infrastructure. For components of the green infrastructure, i.e., habitat networks for selected sensitive species, habitat network analysis for the European crested tit (Lophophanes cristatus) and common toad (Bufo bufo) was performed. Between 2003 and 2018, urban areas increased by approximately 4% while green areas decreased by 2% in comparison with their 2003 areal amounts. The most significant urban growth occurred through expansion of the transport network, paved surfaces and construction areas which increased by 12%, mainly at the expense of grassland and coniferous forest. Examination of urban growth within the green infrastructure indicated that most land area was lost in dispersal zones (28 ha) while the highest percent change was within habitat for species of conservation concern (14%). The habitat network analysis revealed that overall connectivity decreased slightly through patch fragmentation and areal loss mainly caused by road expansion on the outskirts of the city. The habitat network analysis also revealed which habitat areas are well-connected and which are most vulnerable. These results can assist policymakers and planners in their efforts to ensure sustainable urban development including sustaining biodiversity in the City of Stockholm.

Loss of natural forests by forest clearcutting has been identified as a critical conservation challenge worldwide. This study addressed forest fragmentation and loss in the context of the establishment of a functional green infrastructure as a spatiotemporally connected landscape‐scale network of habitats enhancing biodiversity, favorable conservation status, and ecosystem services. Through retrospective analysis of satellite images, we assessed a 50‐ to 60‐year spatiotemporal clearcutting impact trajectory on natural and near‐natural boreal forests across a sizable and representative region from the Gulf of Bothnia to the Scandinavian Mountain Range in northern Fennoscandia. This period broadly covers the whole forest clearcutting period; thus, our approach and results can be applied to comprehensive impact assessment of industrial forest management. The entire study region covers close to 46,000 km2 of forest‐dominated landscape in a late phase of transition from a natural or near‐natural to a land‐use modified state. We found a substantial loss of intact forest, in particular of large, contiguous areas, a spatial polarization of remaining forest on regional scale where the inland has been more severely affected than the mountain and coastal zones, and a pronounced impact on interior forest core areas. Salient results were a decrease in area of the largest intact forest patch from 225,853 to 68,714 ha in the mountain zone and from 257,715 to 38,668 ha in the foothills zone, a decrease from 75% to 38% intact forest in the inland zones, a decrease in largest patch core area (assessed by considering 100‐m patch edge disturbance) from 6114 to 351 ha in the coastal zone, and a geographic imbalance in protected forest with an evident predominance in the mountain zone. These results demonstrate profound disturbance of configuration of the natural forest landscape and disrupted connectivity, which challenges the establishment of functional green infrastructure. Our approach supports the identification of forests for expanded protection and conservation‐oriented forest landscape restoration.

Mapping of green vegetation in urban areas using remote sensing techniques can be used as a tool for integrated spatial planning to deal with urban challenges. In this context, multitemporal (MT) synthetic aperture radar (SAR) data have not been equally investigated, as compared to optical satellite data. This research compared various machine learning methods using single-date and MT Sentinel-1 (S1) imagery. The research was focused on vegetation mapping in urban areas across Europe. Urban vegetation was classified using six classifiers—random forests (RF), support vector machine (SVM), extreme gradient boosting (XGB), multi-layer perceptron (MLP), AdaBoost.M1 (AB), and extreme learning machine (ELM). Whereas, SVM showed the best performance in the single-date image analysis, the MLP classifier yielded the highest overall accuracy in the MT classification scenario. Mean overall accuracy (OA) values for all machine learning methods increased from 57% to 77% with speckle filtering. Using MT SAR data, i.e., three and five S1 imagery, an additional increase in the OA of 8.59% and 13.66% occurred, respectively. Additionally, using three and five S1 imagery for classification, the F1 measure for forest and low vegetation land-cover class exceeded 90%. This research allowed us to confirm the possibility of MT C-band SAR imagery for urban vegetation mapping.

Trees are the key components of urban vegetation in cities. The timely and accurate identification of existing urban tree species with their location is the most important task for improving air, water, and land quality; reducing carbon accumulation; mitigating urban heat island effects; and protecting soil and water balance. Light detection and ranging (LiDAR) is frequently used for extracting high-resolution structural information regarding tree objects. LiDAR systems are a cost-effective alternative to the traditional ways of identifying tree species, such as field surveys and aerial photograph interpretation. The aim of this work was to assess the usage of machine learning algorithms for classifying the deciduous (broadleaf) and coniferous tree species from 3D raw LiDAR data on the Davutpasa Campus of Yildiz Technical University, Istanbul, Turkey. First, ground, building, and low, medium, and high vegetation classes were acquired from raw LiDAR data using a hierarchical-rule-based classification method. Next, individual tree crowns were segmented using a mean shift clustering algorithm from high vegetation points. A total of 25 spatial- and intensity-based features were utilized for support vector machine (SVM), random forest (RF), and multi-layer perceptron (MLP) classifiers to discriminate deciduous and coniferous tree species in the urban area. The machine learning-based classification’s overall accuracies were 80%, 83.75%, and 73.75% for the SVM, RF, and MLP classifiers, respectively, in split 70/30 (training/testing). The SVM and RF algorithms generally gave better classification results than the MLP algorithm for identifying the urban tree species.

Green space is increasingly recognized as an important component of the urban environment. Adequate management and planning of urban green space is crucial to maximize its benefits for urban inhabitants and for the urban ecosystem in general. Inventorying urban vegetation is a costly and time-consuming process. The development of new remote sensing techniques to map and monitor vegetation has therefore become an important topic of interest to many scholars. Based on a comprehensive survey of the literature, this review article provides an overview of the main approaches proposed to map urban vegetation from high-resolution remotely sensed data. Studies are reviewed from three perspectives: (a) the vegetation typology, (b) the remote sensing data used and (c) the mapping approach applied. With regard to vegetation typology, a distinction is made between studies focusing on the mapping of functional vegetation types and studies performing mapping of lower-level taxonomic ranks, with the latter mainly focusing on urban trees. A wide variety of high-resolution imagery has been used by researchers for both types of mapping. The fusion of various types of remote sensing data, as well as the inclusion of phenological information through the use of multi-temporal imagery, prove to be the most promising avenues to improve mapping accuracy. With regard to mapping approaches, the use of deep learning is becoming more established, mostly for the mapping of tree species. Through this survey, several research gaps could be identified. Interest in the mapping of non-tree species in urban environments is still limited. The same holds for the mapping of understory species. Most studies focus on the mapping of public green spaces, while interest in the mapping of private green space is less common. The use of imagery with a high spatial and temporal resolution, enabling the retrieval of phenological information for mapping and monitoring vegetation at the species level, still proves to be limited in urban contexts. Hence, mapping approaches specifically tailored towards time-series analysis and the use of new data sources seem to hold great promise for advancing the field. Finally, unsupervised learning techniques and active learning, so far rarely applied in urban vegetation mapping, are also areas where significant progress can be expected.

Addressing the problems of misclassification and omissions in urban vegetation fine classification from current remote sensing classification methods, this research proposes an intelligent urban vegetation classification method that combines feature engineering and improved DeepLabV3+ based on unmanned aerial vehicle visible spectrum images. The method constructs feature engineering under the ReliefF algorithm to increase the number of features in the samples, enabling the deep learning model to learn more detailed information about the vegetation. Moreover, the method improves the classical DeepLabV3+ network structure based on (1) replacing the backbone network using MoblieNetV2; (2) adjusting the atrous spatial pyramid pooling null rate; and (3) adding the attention mechanism and the convolutional block attention module. Experiments were conducted with self-constructed sample datasets, where the method was compared and analyzed with a fully convolutional network (FCN) and U-Net and ShuffleNetV2 networks; the migration of the method was tested as well. The results show that the method in this paper is better than FCN, U-Net, and ShuffleNetV2, and reaches 92.27%, 91.48%, and 85.63% on the accuracy evaluation indices of overall accuracy, MarcoF1, and mean intersection over union, respectively. Furthermore, the segmentation results are accurate and complete, which effectively alleviates misclassifications and omissions of urban vegetation; moreover, it has a certain migration ability that can quickly and accurately classify the vegetation.

… machine learning algorithms, assessing the impact of spatial resolution on urban vegetation classification … The classification was performed with support vector machine (SVM), random …

Rapid urbanization in cities can result in a decrease in green urban areas. Reductions in green urban infrastructure pose a threat to the sustainability of cities. Up-to-date maps are important for the effective planning of urban development and the maintenance of green urban infrastructure. There are many possible ways to map vegetation; however, the most effective way is to apply machine learning methods to satellite imagery. In this study, we analyze four machine learning methods (support vector machine, random forest, artificial neural network, and the naïve Bayes classifier) for mapping green urban areas using satellite imagery from the Sentinel-2 multispectral instrument. The methods are tested on two cities in Croatia (Varaždin and Osijek). Support vector machines outperform random forest, artificial neural networks, and the naïve Bayes classifier in terms of classification accuracy (a Kappa value of 0.87 for Varaždin and 0.89 for Osijek) and performance time.

Urban gardens are an integral part of urban agricultural systems, contributing to ecosystem services, biodiversity and human wellbeing. These systems occur at fine scales, can be highly complex and therefore offer the opportunity to test mechanisms of ecological patterns and processes. The capacity to confidently characterize urban gardens and their land uses is still lacking, while it could provide the basis for assessing ecosystem service provision. Land classifications from remote sensing platforms are common at the landscape scale, but imagery often lacks the resolution required to map differences in land use of fine-scale systems such as urban gardens. Here, we present a workflow to model and map land use in urban gardens using imagery from an unoccupied aerial vehicle (UAV) and machine learning. Due to high resolutions (<5 cm) from image acquisition at low altitudes, UAV remote sensing is better suited to characterize urban land use. We mapped six common land uses in 10 urban community gardens, exhibiting distinct spatial arrangements. Our models had good predictive performance, reaching 80% overall prediction accuracy in independent validation and up to 95% when assessing model performance per cover class. Extracting spatial metrics from these land use classifications, we found that at the garden and plot scale, plant species richness can be estimated by the total area and patchiness of crops. Land use classifications like these can offer an accessible tool to assess complex urban habitats and justify the importance of urban agriculture as a service-providing system, contributing to the sustainability and livability of cities.

ABSTRACT Urban tree species classification is a challenging task due to spectral and spatial diversity within an urban environment. Unmanned aerial vehicle (UAV) platforms and small-sensor technology are rapidly evolving, presenting the opportunity for a comprehensive multi-sensor remote sensing approach for urban tree classification. The objectives of this paper were to develop a multi-sensor data fusion technique for urban tree species classification with limited training samples. To that end, UAV-based multispectral, hyperspectral, LiDAR, and thermal infrared imagery was collected over an urban study area to test the classification of 96 individual trees from seven species using a data fusion approach. Two supervised machine learning classifiers, Random Forest (RF) and Support Vector Machine (SVM), were investigated for their capacity to incorporate highly dimensional and diverse datasets from multiple sensors. When using hyperspectral-derived spectral features with RF, the fusion of all features extracted from all sensor types (spectral, LiDAR, thermal) achieved the highest overall classification accuracy (OA) of 83.3% and kappa of 0.80. Despite multispectral reflectance bands alone producing significantly lower OA of 55.2% compared to 70.2% with minimum noise fraction (MNF) transformed hyperspectral reflectance bands, the full dataset combination (spectral, LiDAR, thermal) with multispectral-derived spectral features achieved an OA of 81.3% and kappa of 0.77 using RF. Comparison of the features extracted from individual sensors for each species highlight the ability for each sensor to identify distinguishable characteristics between species to aid classification. The results demonstrate the potential for a high-resolution multi-sensor data fusion approach for classifying individual trees by species in a complex urban environment under limited sampling requirements.

… In this study, urban vegetation classification is determined to be used by supervised ensemble ML methods such as Gradient Boosting, Categorical Boosting (CatBoost), Extreme …

Urban areas feature complex and heterogeneous land covers which create challenging issues for tree species classification. The increased availability of high spatial resolution multispectral satellite imagery and LiDAR datasets combined with the recent evolution of deep learning within remote sensing for object detection and scene classification, provide promising opportunities to map individual tree species with greater accuracy and resolution. However, there are knowledge gaps that are related to the contribution of Worldview-3 SWIR bands, very high resolution PAN band and LiDAR data in detailed tree species mapping. Additionally, contemporary deep learning methods are hampered by lack of training samples and difficulties of preparing training data. The objective of this study was to examine the potential of a novel deep learning method, Dense Convolutional Network (DenseNet), to identify dominant individual tree species in a complex urban environment within a fused image of WorldView-2 VNIR, Worldview-3 SWIR and LiDAR datasets. DenseNet results were compared against two popular machine classifiers in remote sensing image analysis, Random Forest (RF) and Support Vector Machine (SVM). Our results demonstrated that: (1) utilizing a data fusion approach beginning with VNIR and adding SWIR, LiDAR, and panchromatic (PAN) bands increased the overall accuracy of the DenseNet classifier from 75.9% to 76.8%, 81.1% and 82.6%, respectively. (2) DenseNet significantly outperformed RF and SVM for the classification of eight dominant tree species with an overall accuracy of 82.6%, compared to 51.8% and 52% for SVM and RF classifiers, respectively. (3) DenseNet maintained superior performance over RF and SVM classifiers under restricted training sample quantities which is a major limiting factor for deep learning techniques. Overall, the study reveals that DenseNet is more effective for urban tree species classification as it outperforms the popular RF and SVM techniques when working with highly complex image scenes regardless of training sample size.

Rapid technological advances in airborne hyperspectral and lidar systems paved the way for using machine learning algorithms to map urban environments. Both hyperspectral and lidar systems can discriminate among many significant urban structures and materials properties, which are not recognizable by applying conventional RGB cameras. In most recent years, the fusion of hyperspectral and lidar sensors has overcome challenges related to the limits of active and passive remote sensing systems, providing promising results in urban land cover classification. This paper presents principles and key features for airborne hyperspectral imaging, lidar, and the fusion of those, as well as applications of these for urban land cover classification. In addition, machine learning and deep learning classification algorithms suitable for classifying individual urban classes such as buildings, vegetation, and roads have been reviewed, focusing on extracted features critical for classification of urban surfaces, transferability, dimensionality, and computational expense.

… The S3PVI framework advances urban vegetation assessment by providing species-specific and seasonally dynamic visual data, supporting evidence-based urban planning for …

Deep learning (DL) models combined with mobile laser scanning (MLS) datasets have demonstrated immense potential for vegetation segmentation. However, restricted performance and inconsistent behavior across datasets by generic DL models offer notable concerns. Furthermore, to capture the characteristic distribution of vegetation points toward effective segregation, a dedicated model for vegetation segmentation is essential. In addition, with curated class-specific DL models being conceptualized, the same is indispensable for vegetation. To address this problem, we propose a novel DL architecture, green segmentation network (GreenSegNet), tailored for vegetation segmentation from MLS point cloud data. Toward a comprehensive assessment, GreenSegNet has been investigated on MLS datasets from three study sites, Chandigarh, Toronto3D, and Kerala. GreenSegNet has illustrated state of the art (SOTA) as well as consistent segmentation performance across all the datasets. GreenSegNet has achieved mean intersection over union (mIoU) as follows: Chandigarh 96.43%, Toronto3D 92.70%, and Kerala 90.16%. In addition, with less than one million parameters, the architecture is the most efficient with respect to the number of parameters among the representative DL models. The associated ablation studies conform to the effectiveness of GreenSegNet. Unlike other SOTA models, GreenSegNet is found robust across different datasets and terrains.

Vegetation has become very important decision-making information in promoting tasks such as urban regeneration, urban planning, environment, and landscaping. In the past, the vegetation index was calculated by combining images of various wavelength regions mainly acquired from the Landsat satellite’s TM or ETM+ sensor. Recently, a technology using UAV-based multispectral images has been developed to obtain more rapid and precise vegetation information. NDVI is a method of calculating the vegetation index by combining the red and near-infrared bands, and is currently the most widely used. In this study, NDVI was calculated using UAV-based multispectral images to classify vegetation. However, among the areas analyzed using NDVI, there was a problem that areas coated with urethane, such as basketball courts and waterproof coating roofs, were classified as vegetation areas. In order to examine these problems, the reflectance of each land cover was investigated using the ASD FieldSpec4 spectrometer. As a result of analyzing the spectrometer measurements, the NDVI values of basketball courts and waterproof coating roofs were similar to those of grass with slightly lower vegetation. To solve this problem, the temperature characteristics of the target site were analyzed using UAV-based thermal infrared images, and vegetation area was analyzed by combining the temperature information with NDVI. To evaluate the accuracy of the vegetation classification technology, 4409 verification points were selected, and kappa coefficients were analyzed for the method using only NDVI and the method using NDVI and thermal infrared images. Compared to the kappa coefficient of 0.830, which was analyzed by applying only NDVI, the kappa coefficient, which was analyzed by combining NDVI and thermal infrared images, was 0.934, which was higher. Therefore, it is very effective to apply a technology that classifies vegetation by combining NDVI and thermal infrared images in urban areas with many urethane-coated land cover such as basketball courts or waterproof coating roofs.

… vegetation estimates are required. In this study, we make the first attempt to employ deep learning for urban vegetation … We present a multimodal deep learning (MDL) model to combine …

Urban vegetation mapping is critical in many applications, i.e., preserving biodiversity, maintaining ecological balance, and minimizing the urban heat island effect. It is still challenging to extract accurate vegetation covers from aerial imagery using traditional classification approaches, because urban vegetation categories have complex spatial structures and similar spectral properties. Deep neural networks (DNNs) have shown a significant improvement in remote sensing image classification outcomes during the last few years. These methods are promising in this domain, yet unreliable for various reasons, such as the use of irrelevant descriptor features in the building of the models and lack of quality in the labeled image. Explainable AI (XAI) can help us gain insight into these limits and, as a result, adjust the training dataset and model as needed. Thus, in this work, we explain how an explanation model called Shapley additive explanations (SHAP) can be utilized for interpreting the output of the DNN model that is designed for classifying vegetation covers. We want to not only produce high-quality vegetation maps, but also rank the input parameters and select appropriate features for classification. Therefore, we test our method on vegetation mapping from aerial imagery based on spectral and textural features. Texture features can help overcome the limitations of poor spectral resolution in aerial imagery for vegetation mapping. The model was capable of obtaining an overall accuracy (OA) of 94.44% for vegetation cover mapping. The conclusions derived from SHAP plots demonstrate the high contribution of features, such as Hue, Brightness, GLCM_Dissimilarity, GLCM_Homogeneity, and GLCM_Mean to the output of the proposed model for vegetation mapping. Therefore, the study indicates that existing vegetation mapping strategies based only on spectral characteristics are insufficient to appropriately classify vegetation covers.

… vegetation data in cities has been widely researched. However, large-scale urban vegetation … This study proposes a novel framework for 3D extraction of urban vegetation, which can be …

… of green space change. Among the green space land use types, agriculture land was largely … Forest land was also impacted but encountered a relatively moderate loss rate compared to …

… the spatio-temporal dynamics of vegetation cover in Kumasi using remote-sensing techniques. More specifically, the study analyses the vegetation change over time and space using …

(UGS) has gained increasing attention due to its environmental and social functions. However, the compound effects of climate change, population growth and economic development on UGS are largely unknown. We selected 107 medium-sized and large cities in China to investigate dynamics in the spatial pattern of UGS in relation to government policy and other drivers based on remote sensing data for the period 1990 to 2019. To explore the effect of different levels of urbanization on changes in green space, we develop a new Normalized Urban Development Index (NUDI) to classify urban-suburban-rural gradients, viz. Long-term Built-up, New Built-up and Non-Built-up. Then, we analysed changes over time in the annual peak value of fraction of vegetation cover (FVC) for 380,000 cloud-free Landsat images, and regional UGS dynamics were evaluated using the proposed Regional Greenness Dynamic Index (RGDI). Finally, to reveal the major driver(s) of changes in UGS and estimate the extent to which patterns of urban greening are due to differences in economic development, we compared the observed UGS spatio-temporal dynamics with data on several climatic, social-economic and land use related factors for the same period. The NUDI are shown to be highly effective in mapping urban development gradients, with overall accuracy in the identified classes of 89%. Annual maximum FVC analysis indicates that there was significant greening between 1990 and 2019 in both the long-term built up (10,667.52 km 2 ) and the non-built up areas (529,310.47 km 2 ), while there was a major increase in browning (25,110.43 km 2 ) in the newly built-up areas. The RGDI results indicate that 65% (71/107) of long-term built-up areas in cities trended greener over 2010 to 2019 under consideration. At the whole city scale, RGDI is negatively correlated with gross domestic product (GDP), although when considering the long-term built-up areas only, economic growth exhibits a significant positive correlation during 2010 to 2019 (R = 0.62, p < 0.01). This study offers important insights as to the patterns of change in urban greening extent over time and its underyling drivers across urban-suburban-rural gradients against the background of urban expansion, afforestation, climate change and economic development.

The purpose of this study is to reveal the spatial-temporal change and driving factors of green space in coastal cities of southeast China over the past 20 years. A supervised classification method combining support vector machines (SVMs) and visual interpretation was used to extract the green space from Landsat TM/OLI imageries from 2000–2020. The landscape pattern index was used to calculate geospatial information of green space and analyze their spatial-temporal changes. The hierarchical partitioning analysis was then used to determine the influences of anthropogenic and geographic environmental factors on the spatial-temporal changes in green space. The results indicated that the total area of green space remained constant over the past 20 years in coastal cities of southeast China (1% reduction). The spatial change of green space mainly occurred in the area near the ocean and the southern region. 41.37% of forest land was transferred from cultivated land, while 44.56%, 41.83%, 43.20%, 46.31%, 41.98% and 40.20% of shrub land, sparse woodland, other woodland, high-coverage grassland, moderate-coverage grassland and low-coverage grassland were transferred from forest land. The number of patches, patch density, edge density, landscape shape index and Shannon’s diversity index increased from 2000–2015, and then decreased to the minimum in 2020, while largest patch index continued to decline from 2000–2020. The contribution of anthropogenic factors (0.53–0.61) on the spatial-temporal changes of green space continually increased over the past 20 years, which was also higher than geographical environment factors (0.39–0.41). Our study provides a new perspective to distinguish the impact of anthropogenic activities and geographical environmental factors on the change of green space area, thereby providing a theoretical support for the construction and ecological management of green space.

Under the pressure of rapid urbanization, the spatiotemporal dynamics of urban green spaces (UGS) have enormous impacts on the local ecological system and environment at different scales. In this study, UGS in Hefei City, which has experienced rapid urbanization from 1995 to 2015, were extracted based on time-series Landsat-5 TM and Landsat-8 OLI images, and different types of parks were mapped based on GF-2 images combined with multi-source metadata. Dynamic patterns of green space were examined by drawing the spatial variations of green space at the city, inner city, and park scales. Results revealed that: (1) At the city level, UGS decreased with the rapid urbanization, especially farmland sharply lost in areas around existing urban areas and along the transportation corridors. (2) At the inner-city level, concentric analyses showed that UGS changes in different rings had unique trends, and the dynamic changes were the most significant in Rings 2 and 3. Comparative analysis of old and new districts showed that the common characteristics of UGS changes were the transfer-out of farmland and the transfer-in of built-up land, and the newer the region, the more intense the changes. (3) At the park level, the number and area of urban parks were increasing from the center to the periphery, and the type of park gradually changed from single to rich. Significant spatial heterogeneity was identified in the landscape patterns of urban parks based on fishnet cell analysis. It is fundamental to assign urban construction land for socio-economic development, while planning UGS at different scales; moreover, integrating different green space-related policies could protect the UGS and maintain its stability. Only in this way can UGS combine economic, social, and ecological benefits under the background of rapid urbanization.

Although there is extensive research demonstrating the significant loss and fragmentation of urban spaces caused by rapid urbanization, to date, no empirical research in Shanghai has investigated the spatiotemporal dynamics of urban open spaces using a comprehensive set of integrated geospatial techniques based on long-sequence time series. Based on the Google Earth Engine (GEE) platform and using the Random Forest (RF) classifier, multiple techniques, namely landscape metrics, trend analysis, open space ratio, transition matrix, Normalized Difference Vegetation Index (NDVI), and fractal dimension analysis, were applied to analyze the Landsat satellite data. Next, Geographic Detector (GeoDetector) methods were used to investigate the driving forces of such spatial variations. The results showed that (1) the RF classification algorithm, supported by the GEE, can accurately and quickly obtain a research object dataset, and that calculating the optimal spatial grain size for open space pattern was 70 m; (2) open spaces exhibited declining and contracting trends; and open spaces in the city experienced a decline from 91.83% in 1980 to 69.63% in 2020. Meanwhile, the degree of open spaces in each district increased to different extents, whilst connectivity markedly decreased. Furthermore, the open space of city center districts showed the lowest rate of decrease, with open space patterns fragmenting due to encroaching urbanization; (3) the contribution of socioeconomic factors to the spatial–temporal changes in open space continually has increased over the past 40 years, and were also higher than natural geographic factors to some extent. Apart from offering policy insights guiding the future spatial planning and development of the city, this paper has contributions from both methodological and empirical perspectives. Based on integrated remote sensing and geographic information science (GIS) techniques, this paper provides updated evidence and a clearer understanding of the spatiotemporal variations in urban spaces and their influencing mechanisms in Shanghai.

… This paper addresses this need by examining the spatio-temporal geography of UGS at the neighborhood-scale in Kumasi, the next biggest city in Ghana. Five (5) UGS distribution …

Urban green space (UGS) is a crucial physical area that supports the functioning of urban ecosystems, and its changes affect urban ecological balance. In order to accurately analyze the dynamic processes and transfer targets of UGS during urbanization, this study proposes a new method of UGS assessment based on multi-temporal Landsat remote sensing data. This method is integrated with intensity analysis and landscape pattern indices so as to explore the spatio-temporal dynamics of the evolution process, landscape pattern, and driving forces of UGS from 2000 to 2022 in the resource-based city of Taiyuan in central China. The results of the case study show that rapid urbanization brought about a continuous reduction in UGS in the study area, but the trend of decreasing gradually slowed down; UGS patches have become more dispersed and isolated, bare land has been targeted for both gains and losses of UGS, and ecological restoration of bare land mitigated the rapid reduction of UGS. The results of this study not only confirm the applicability of this methodology for monitoring and assessing the evolution of UGS, but also reveal the identification of the targeting or avoidance of other categories during the conversion of UGS. Thus, the potential factors influencing changes in UGS can be analyzed to guide and safeguard sustainable development.

… green space changes, and is essential for monitoring and assessing green space functions. This paper presents a new method for quantifying and capturing changes in green space …

In the current context of urbanization, urban agglomerations face complex challenges in maintaining an ecological balance. This study uses remote sensing images of the Central Yunnan urban agglomeration from 2000 to 2020, along with socioeconomic data, to analyze the spatiotemporal characteristics of the green space evolution. Utilizing dynamic geographically weighted regression analysis based on principal components (PCA-GWR), we identify the key socioeconomic factors influencing these changes and quantitatively analyze the driving forces in each stage. Our findings reveal a continuing trend of decreasing total green space alongside increasing individual forest types and pronounced regional disparities in green space dynamics. The results indicate that socioeconomic factors exert both positive facilitative effects and negative pressures, with evident spatial and temporal variability. Urbanization and economic development promote forest expansion in certain areas, while contributing to the reduction in farmland and shrub–grass lands. Significant variations are influenced by factors such as the urbanization rate, the agricultural population, the industrial composition, and fiscal revenue. This study enhances the in-depth understanding of the relationship between the spatiotemporal dynamics of green spaces and socially driven mechanisms, offering significant insights for sustainable urban planning and landscape management and harmonizing urban agglomeration development.

… the spatio-temporal changes in urban green spaces of Mumbai city through three significant environmental indices of land … In this study, the LST has been retrieved from Landsat data by …

… space for suburban recreation and a structural space that … of the dynamic landscape changes of green space in Nanjing… further optimize the spatial structure of green space and regional …

2020年中国明确提出2030年“碳达峰”与2060年“碳中和”目标，城市绿地是城市中唯一的、直接的自然碳汇，如何估算城市绿地碳储量对于实现“双碳”目标十分重要。以北京市海淀区五环内城市绿地为例，以高分二号遥感数据为信息源，在公园绿地、防护绿地、附属绿地、区域绿地4类绿地中分层抽取139个样地进行碳储量估算研究。研究发现各类样地碳储量值及归一化植被指数（NDVI）均存在显著差异，通过回归分析构建了4类绿地NDVI与碳储量的拟合模型，并另选40个检验样地，通过人工识别的碳储量数据检验回归模型的合理性，构建完善的城市绿地碳储量估算系统。估算结果表明，北京市海淀区五环内城市绿地固碳总量约为4.14万t，不同绿地类型碳汇能力存在一定差异，具体表现为公园绿地>附属绿地>区域绿地>防护绿地。研究对于指导全国各城市绿地碳储量的估算、实现碳中和具有重要意义。

基于"资源-能源-经济-环境"构建农业绿色全要素生产率理论分析框架，设计测算指标，改进EBM模型并结合ML指数从静态和动态视角测算并解析农业绿色全要素生产率增长源泉，采用核密度函数 and 空间马尔科夫链从时空视域分析中国农业绿色全要素生产率的动态演变规律。结论：一是农业绿色全要素生产率更真实、合理反映农业效率。二是2006-2016年间中国农业绿色全要素生产率呈微幅波动上升趋势，增长的动力源泉在于农业绿色技术进步。三是农业绿色全要素生产率区域差异明显，长期内区域差异不会缩小，区域差异的根源在于农业绿色技术效率。四是中国农业绿色全要素生产率具有空间集聚特性，空间滞后类型对区域转移的稳定程度具有显著影响；"以邻为善"与"以邻为壑"并存；与农业绿色全要素生产率水平较高区域相邻，会降低向低水平转移的概率，但跨界式增长难以实现。政策含义是制定区域差异化的AGTFP增长策略，实现中国农业区域协调、绿色增长。

针对生态环境脆弱的寒旱区开展地物要素提取以及土地覆盖变化监测研究,对农业规划、城乡建设、生态环境监测与保护等具有重要意义。借助2015—2019年新疆莫索湾垦区Landsat-8影像构建数据集,对比3种传统方法：最大似然分类（Maximum likelihood classification,MLC）、支持向量机（Support vector machine,SVM）和随机森林（Random forest,RF）及5种语义分割模型：DeepLabv3+（Xception）、DeepLabv3+（MobileNet）、SegNet（ResNet50）、U-Net（MobileNet）和PSPNet（MobileNet）,选取最优自动化地物提取模型对研究区1998—2020年农用地、建筑用地、水体和荒漠4种地物要素进行分类,并运用土地利用转移矩阵和动态度进行定量动态变化分析。结果表明：DeepLabv3+（Xception）模型可以实现更准确、更高效的地物提取,总体精确度（OA）、Kappa系数和F1值分别为96.06%、0.96和0.86,其中所选模型的平均交并比（MIoU）较其他模型提升0.03~0.39。近23 a,莫索湾垦区的荒漠、农用地和建筑用地三者的土地结构转化较为明显,荒漠总面积减少15.00%,农用地总面积增加12.68%,建筑用地总面积增加2.53%,水体面积变化较为平稳。地物类型总体转变方向为荒漠向农用地转化、农用地向建筑用地转化。该研究可为深度学习技术应用于中分辨率遥感卫星影像领域中实现土地利用及变化动态监测提供参考。

通过整合标签超分辨率（SR）和实例批量归一化网络（IBN−Net），在无本地高分辨率标签的情况下，实现了福建省光泽县的2 m分辨率土地覆盖制图，提出了一种基于深度学习的无标签土地覆盖制图方法。结果表明：利用改进的全卷积神经网络（FCN）模型能够实现标签超分辨率，将低分辨率标签提升至高分辨率，有效提高分类精度；IBN−Net网络增强了模型的泛化能力，显著提升跨域应用的效果。相比于内源低分辨率标签，使用高精度的外源标签使模型在光泽县的整体准确率提高2.55%，达到85.48%。本方法在无匹配标签条件下，显著提升土地覆盖制图的精度，可为区域生态监测和管理提供有效的技术支持。

扰动图斑是生产建设项目水土保持信息化监管基础数据。针对扰动图斑传统人机交互目视解译效率低、成果不统一等问题, 基于深度学习原理, 构建生产建设项目扰动图斑自动识别分类卷积神经网络模型, 用以提高扰动图斑解译生产效率和成果质量。然后确定深度学习模型关键超参数－优化器算法、学习速率和批大小最优值。在此基础上经过150个训练轮次得到生产建设项目扰动图斑自动识别分类卷积神经网络模型, 模型综合性能评价指标－精度值和损失值分别为0.9526和0.1670。模型在"检验样本集"应用效果表明: 模型识别分类总体精度为97.52%, 扰动样本查准率和查全率分别为72.44%和83.90%;模型识别分类结果与真实情况基本一致, 漏分类、误分类比例相对较低, 具有较强的泛化能力。这说明深度学习模型用于生产建设项目扰动图斑自动识别分类是实际可行的。研究成果为扰动图斑解译生产提供一种新方法, 可为生产建设项目水土保持信息化监管提供重要技术支撑。

The development of generative design driven by artificial intelligence algorithms is speedy. There are two research gaps in the current research: 1) Most studies only focus on the relationship between design elements and pay little attention to the external information of the site; 2) GAN and other traditional generative algorithms generate results with low resolution and insufficient details. To address these two problems, we integrate GAN, Stable diffusion multimodal large-scale image pre-training model to construct a full-process park generative design method: 1) First, construct a high-precision remote sensing object extraction system for automated extraction of urban environmental information; 2) Secondly, use GAN to construct a park design generation system based on the external environment, which can quickly infer and generate design schemes from urban environmental information; 3) Finally, introduce Stable Diffusion to optimize the design plan, fill in details, and expand the resolution of the plan by 64 times. This method can achieve a fully unmanned design automation workflow. The research results show that: 1) The relationship between the inside and outside of the site will affect the algorithm generation results. 2) Compared with traditional GAN algorithms, Stable diffusion significantly improve the information richness of the generated results.

The analysis of climate regions is very important for designers and architects, because the increase in density and built up spaces and reduction in open spaces and green lands induce the increase of heat, especially in an urban area, deteriorating the environment and causing health problems. This study analyzes the Land Surface Temperature (LST) differences in the region of Dobrogea, Romania, and compares with the land use and land cover types using TM and ETM+ data of 1989 and 2000. As the analysis is performed on large data sets, we used Grid Computing to implement a service for using on Computational Grids with a Web-based client interface, which will be greatly useful and convenient for those who are studying the ground thermal environment and heat island effects by using Landsat TM/ETM+ bands, and have typical workstations, with no special computing and storing resources for computationally intensive satellite image processing and no license for a commercial image processing tool. Based on the satellite imagery, the paper also addresses a Supervised Classification algorithm and the computation of two indices of great value in water resources management, Normalized Difference Vegetation Index (NDVI), respectively Land Surface Emissivity (LSE).

In post-disaster space-air-ground integrated networks (SAGINs), terrestrial infrastructure is often impaired, and unmanned aerial vehicles (UAVs) must rapidly restore connectivity for mission-critical ground terminals in cluttered non-line-of-sight (NLoS) urban environments. To enhance coverage, UAVs employ movable antennas (MAs), while reconfigurable intelligent surfaces (RISs) on surviving high-rises redirect signals. The key challenge is communication-limited partial observability, leaving each UAV with a narrow, fast-changing neighborhood view that destabilizes value estimation. Existing multi-agent reinforcement learning (MARL) approaches are inadequate--non-communication methods rely on unavailable global critics, heuristic sharing is brittle and redundant, and learnable protocols (e.g., CommNet, DIAL) lose per-neighbor structure and aggravate non-stationarity under tight bandwidth. To address partial observability, we propose a spatiotemporal A2C where each UAV transmits prior-decision messages with local state, a compact policy fingerprint, and a recurrent belief, encoded per neighbor and concatenated. A spatial discount shapes value targets to emphasize local interactions, while analysis under one-hop-per-slot latency explains stable training with delayed views. Experimental results show our policy outperforms IA2C, ConseNet, FPrint, DIAL, and CommNet--achieving faster convergence, higher asymptotic reward, reduced Temporal-Difference(TD)/advantage errors, and a better communication throughput-energy trade-off.

Multi-spectral imagery plays a crucial role in diverse Remote Sensing applications including land-use classification, environmental monitoring and urban planning. These images are widely adopted because their additional spectral bands correlate strongly with physical materials on the ground, such as ice, water, and vegetation. This allows for more accurate identification, and their public availability from missions, such as Sentinel-2 and Landsat, only adds to their value. Currently, the automatic analysis of such data is predominantly managed through machine learning models specifically trained for multi-spectral input, which are costly to train and support. Furthermore, although providing a lot of utility for Remote Sensing, such additional inputs cannot be used with powerful generalist large multimodal models, which are capable of solving many visual problems, but are not able to understand specialized multi-spectral signals. To address this, we propose a training-free approach which introduces new multi-spectral data in a Zero-Shot-only mode, as inputs to generalist multimodal models, trained on RGB-only inputs. Our approach leverages the multimodal models'understanding of the visual space, and proposes to adapt to inputs to that space, and to inject domain-specific information as instructions into the model. We exemplify this idea with the Gemini2.5 model and observe strong Zero-Shot performance gains of the approach on popular Remote Sensing benchmarks for land cover and land use classification and demonstrate the easy adaptability of Gemini2.5 to new inputs. These results highlight the potential for geospatial professionals, working with non-standard specialized inputs, to easily leverage powerful multimodal models, such as Gemini2.5, to accelerate their work, benefiting from their rich reasoning and contextual capabilities, grounded in the specialized sensor data.

Urban construction land expansion damages natural ecological patches, changing the relationship between residents and ecological land. This is widespread due to global urbanization. Considering nature and society in urban planning, we have established an evaluation system for urban green space construction to ensure urban development residents’ needs while considering natural resource distribution. This is to alleviate the contradiction of urban land use and realize the city’s sustainable development. Taking the Fengdong New City, Xixian New Area as an example, the study used seven indicators to construct an ecological source evaluation system, four types of factors to identify ecological corridors and ecological nodes using the minimum cumulative resistance model, and a Back Propagation neural network to determine the weight of the evaluation system, constructing an urban green space ecological network. We comprehensively analyzed and retained 11 ecological source areas, identified 18 ecological corridors, and integrated and selected 13 ecological nodes. We found that the area under the influence of ecosystem functions is 12.56 km2, under the influence of ecological demands is 1.40 km2, and after comprehensive consideration is 22.88 km2. Based on the results, this paper concludes that protecting, excavating, and developing various urban greening factors do not conflict with meeting the residents’ ecological needs. With consideration of urban greening factors, cities can achieve green and sustainable development. We also found that the BP neural network objectively calculates and analyzes the evaluation factors, corrects the distribution value of each factor, and ensures the validity and practicability of the weights. The main innovation of this study lies in the quantitative analysis and spatial expression of residents’ demand for ecological land and the positive and negative aspects of disturbance. The research results improve the credibility and scientificity of green space construction so that urban planning can adapt and serve the city and its residents.

… mitigate UHI effects, however, it is still unclear how the green space characteristics and its … the green space cool island (GCI). In this study, land surface temperature (LST) and land …

Urban green spaces are considered an appropriate way to reduce urban heat island effects and provide comfort to the nearby occupants. In addition to cooling the actual space, urban green spaces are also able to influence the surrounding area, and this phenomenon is called the urban green space cooling effect. The most important issues with regard to the cooling effects of urban green spaces are the intensity and density of the cooling, which can play a major role for urban designers and planners in dealing with urban heat island. This article reviews the latest studies that have examined the cooling effects of urban green spaces in recent years. Based on the method of evaluation of their samples, the studies are divided into three groups. The first category consists of research into a set of urban green spaces in one part of or in an entire city, mainly conducted through remote sensing and satellite maps. The second category investigates city parks or several urban parks with recognizable shapes and locations. In this section, information was mainly gathered by virtue of field observations. The third category relates to studies in which a part of urban space according to different scenarios of green space placement was modeled by simulation. The results of the present study illustrate that the highest cooling effect distance and cooling effect intensity are for large urban parks with an area of more than 10 ha; however, in addition to the area, the natural elements and qualities of the urban green spaces, as well as climate characteristics, highly inform the urban green space cooling effect.

Urbanization is a rapid global trend, leading to consequences such as urban heat islands and local flooding. Imminent climate change is predicted to intensify these consequences, forcing cities to rethink common infrastructure practices. One popular method of adaptation is green infrastructure implementation, which has been found to reduce local temperatures and alleviate excess runoff when installed effectively. As cities continue to change and adapt, land use/landcover modeling becomes an important tool for city officials in planning future land usage. This study uses a combination of cellular automata, machine learning, and Markov chain analysis to predict high resolution land use/landcover changes in Philadelphia, PA, USA for the year 2036. The 2036 landcover model assumes full implementation of Philadelphia’s green infrastructure program and past temporal trends of urbanization. The methodology used to create the 2036 model was validated by creating an intermediate prediction of a 2015 landcover that was then compared to an existing 2015 landcover. The accuracy of the validation was determined using Kappa statistics and disagreement scores. The 2036 model successfully met Philadelphia’s green infrastructure goals. A variety of landscape metrics demonstrated an overall decrease in fragmentation throughout the landscape due to increases in urban landcover.

… infrastructure construction and urban green development in 280 cities above the prefecture level from 2005 to 2019,and analyzes the spatio-temporal … of new infrastructure construction …

… Alongside the growing interest in Green Infrastructure (GI), the rapid rise of Spatio-temporal Big Data (STBD) presents potentially transformative tools for understanding and managing …

Urban green infrastructure (UGI) is considered to be an effective tool for mitigating PM 2.5 (particulate matter with an aerodynamic diameter of less than 2.5 μm) pollution in urban areas. …

… The goal of this paper is to explore the sensitivity of the simulated spatiotemporal emergence of GI to the rules used to describe GI adoption by residential property owners. This is …

Plant classification requires the eye of an expert in botanics when the subtle differences in stem or petals differentiate between different species. Hence, an accurate automatic plant classification might be of great assistance to a person who studies agriculture, travels, or explores rare species. This paper focuses on a specific task of urban plants classification. The possible practical application of this work is a tool which assists people, growing plants at home, to recognize new species and to provide the relevant caring instructions. Because urban species are barely covered by the benchmark datasets, these species cannot be accurately recognized by the state-of-the-art pre-trained classification models. This paper introduces a new dataset, Urban Planter, for plant species classification with 1500 images categorized into 15 categories. The dataset contains 15 urban species, which can be grown at home in any climate (mostly desert) and are barely covered by existing datasets. We performed an extensive analysis of this dataset, aimed at answering the following research questions: (1) Does the Urban Planter dataset provide enough information to train accurate deep learning models? (2) Can pre-trained classification models be successfully applied on Urban Planter, and is the pre-training on ImageNet beneficial in comparison to the pre-training on a much smaller but more relevant dataset? (3) Does two-step transfer learning further improve the classification accuracy? We report the results of experiments designed to answer these questions. In addition, we provide the link to the installation code of the alpha version and the demo video of the web app for urban plants classification based on the best evaluated model. To conclude, our contribution is three-fold: (1) We introduce a new dataset of urban plant images; (2) We report the results of an extensive case study with several state-of-the-art deep networks and different configurations for transfer learning; (3) We provide a web application based on the best evaluated model. In addition, we believe that, by extending our dataset in the future to eatable plants and assisting people to grow food at home, our research contributes to achieve the United Nations’ 2030 Agenda for Sustainable Development.

Abstract In many countries, the increasing population drives the spatiotemporal land use and land cover change (LULCC) at a higher rate, causing heterogeneous landscape. Hence, the LULCC is a dynamic and frequent process causing fragmented land cover, and therefore, extensive research on LULCC pattern is necessary at different spatial and temporal scales. Moreover, it is essential to identify appropriate algorithms to detect LULCC in such fragmented areas. Furthermore, the rate of change is different in rural and urban areas. Hence, the main goal of the study was to describe the performance of different machine learning algorithms on three different spatial and multispectral satellite image classification in rural and urban extents. We carried out atmospheric and geometric correction. To achieve this, we acquired and processed a set of moderate and finer resolution images (Landsat-8, Sentinel-2, and Planet images) having similar phenological stages to finding out the suitable algorithms that provide better performance in LULC classification. Random forest, Support Vector Machine (SVM), and their combined strength (stacked algorithms) were applied on Landsat-8, Sentinel-2, and Planet images separately to assess individual and overall class accuracy of the images. Two unique sets of training data were generated to classify the 6 (3 sensors x 2 locations) imagery. We find that the Sentinel-2 image performs best among the three images. Among the three different algorithms, SVM showed comparatively better results. In both Bhola and Dhaka, the SVM on Sentinel image performed best with an overall accuracy of 0.969, 0.983, and overall kappa of 0.948, and 0.968, respectively. This study illustrates the role of algorithms on LULCC studies for a better understanding of the changes that occurred in the rural and urban areas. These findings assist planners, remote sensing scientists and decision-makers to choose a suitable image classification algorithm in monitoring rapidly changing, fragmented, and diverse landscape in Bangladesh and elsewhere in the world.

The confluence of global warming, the urban heat island effect, and alterations in the nature of underlying surfaces has led to a continuous escalation in the frequency, scale, and intensity of fires within urban green spaces. Mitigating or eliminating the adverse effects of such fires on the service functions of urban ecosystems, while enhancing the resilience of urban greening systems in disaster prevention and risk reduction, has become a pivotal challenge in modern urban development and management. Academic focus has progressively broadened from isolated urban and forest domains to encompass the more intricate environments of the Wildland–Urban Interface (WUI) and urban–suburban forests, with a particular emphasis on the distinctive characteristics of urban greening and in-depth research. This study employs a combination of CiteSpace bibliometric analysis and a narrative literature review to comprehensively examine three critical aspects of urban fire safety as follows: (1) the evaluation of the fire-resistant performance of landscape plants in urban green spaces; (2) the mechanisms of fire behavior in urban greening systems; and (3) the assessment and prediction of urban fire risks. Our findings indicate that landscape plants play a crucial role in controlling the spread of fires in urban green spaces by providing physical barriers and inhibiting combustion processes, thereby mitigating fire propagation. However, the diversity and non-native characteristics of urban greenery species present challenges. The existing research lacks standardized experimental indicators and often focuses on single-dimensional analyses, leading to conclusions that are limited, inconsistent, or even contradictory. Furthermore, most current fire spread models are designed primarily for forests and wildland–urban interface (WUI) regions. Empirical and semi-empirical models dominate this field, yet future advancements will likely involve coupled models that integrate climate and environmental factors. Fire risk assessment and prediction represent a global research hotspot, with machine learning- and deep learning-based approaches increasingly gaining prominence. These advanced methods have demonstrated superior accuracy compared to traditional techniques in predicting urban fire risks. This synthesis aims to elucidate the current state, trends, and deficiencies within the existing research. Future research should explore methods for screening highly resistant landscape plants, with the goal of bolstering the ecological resilience of urban greening systems and providing theoretical underpinnings for the realization of sustainable urban environmental security.

… social analysis in regional scale, but few of them focus on the disparities of access to green spaces in spatial-temporal dynamics analysis in … approaches to analyze the spatial-temporal …

… Abstract: Based on the panel data of park green space and social economy of 284 … the spatio-temporal patterns and driving mechanism of supply and demand of park green space, …

… spatio-temporal data analysis, Dagum Spatial Gini coefficient, and geographical detector methods, we examine spatio-temporal dynamic … The overall level of green resilience in Chinese …

Urbanization is currently one of the most pressing environmental issues which cuts across all countries at unprecedented rates and intensities, with far reaching consequences on ecosystems, biodiversity and human wellbeing. This paper assessed urban expansion and land use/land cover changes in Sokoto metropolis, North-western Nigeria using Remote Sensing and GIS. Landsat images of 1990, 1999 and 2015 were processed for LULC classification and change detection using the Maximum Likelihood Classification, Post Classification Comparison techniques and the Land Change Modeler. The classification revealed five broad land cover classes which include Built-up Area, Farmland, Green Area, Open Space and Wetland/Water. The Built-up and Green areas continuously increased while Farmland and Open space decreased throughout the study period. The metropolis expanded radially at a faster rate between 1999 and 2015 with the highest rate of increase (1890.5ha per annum) recorded in the Built-up Area. This implies a doubling time of approximately 30 years at the expense of Farmland and Open space which may be completely exhausted in 40 and 29 years respectively. Infrastructural provision should thus align with the rate and direction of growth and where the Green Area is converted, replacement should be made to ensure continued supply and stability of the numerous ecosystem services green areas provide.