基于SAM与MobileNetV4的茶叶叶部病害识别方法研究

It is challenging to achieve accurate tea bud detection in optical images with complex backgrounds since distinguishing between the foregrounds and backgrounds of these images remains difficult. Although several studies have been proposed to implicitly distinguish foregrounds and backgrounds via various attention mechanisms, explicit distinction between foregrounds and backgrounds has been seldom explored. Inspired by recent successful applications of the Segment Anything Model (SAM) in computer vision, this study proposes a SAM-assisted dual-branch YOLOv8 model named SD-YOLOv8 for tea bud detection to address the challenges of explicit distinction between foregrounds and backgrounds. The SD-YOLOv8 model mainly consists of two key components: (1) the SAM-based foreground segmenter (SFS) to generate foreground masks of tea bud images without any training, and (2) the heterogeneous feature extractor to parallelly capture both color features in optical images and edge features in foreground masks. The experimental results show that the proposed SD-YOLOv8 significantly improves the performance of tea bud detection based on the explicit distinction between foregrounds and backgrounds. The mean Average Precision of the SD-YOLOv8 model reaches 86.0%, surpassing the YOLOv8 (mAP 81.6%) by 5 percentage points and outperforming recent object detection models, including Faster R-CNN (mAP 60.7%), DETR (mAP 64.6%), YOLOv5 (mAP 72.4%), and YOLOv7 (mAP 80.6%). This demonstrates its superior capability in efficiently detecting tea buds against complex backgrounds. Additionally, this study proposes a self-built tea bud dataset with three seasons to address the data shortages in tea bud detection.

Plant diseases threaten global food security, yet traditional visual inspection often fails to detect early-stage symptoms critical for timely intervention. While deep learning models have shown promise for automated disease detection, their performance often degrades in realistic field conditions. This study investigates whether data-centric preprocessing improves apple leaf disease detection. We present the first systematic evaluation of the Segment Anything Model (SAM) combined with morphological quality assessment for leaf segmentation, compared against whole-image classification using the Plant-Pathology FGVC7 dataset (3,642 apple orchard images). To ensure segmentation reliability, we introduce a five-metric morphological framework (area ratio, aspect ratio, spatial coverage, centroid proximity, border penalty). Experiments with ResNet-18 under 3-fold cross-validation reveal class-specific effects: SAM improves F1 by 3.0% for the minority multiple diseases class, but decreases by $1. 4 {\%}$ for healthy leaves where contextual cues aid detection. Rust and scab remain stable above 95% F1, reflecting their distinctive visual signatures. GradCAM ++ confirms that preprocessing redirects attention toward diseaserelevant regions, particularly in complex multiple-disease cases. Overall, these findings show that adaptive preprocessing, rather than universal background removal, offers practical benefits for precision agriculture.

Accurate segmentation of leaf diseases is crucial for crop health management and disease prevention. However, existing studies fall short in addressing issues such as blurred disease spot boundaries and complex feature distributions in disease images. Although the vision foundation model, Segment Anything Model (SAM), performs well in general segmentation tasks within natural scenes, it does not exhibit good performance in plant disease segmentation. To achieve fine-grained segmentation of leaf disease images, this study proposes an advanced model: Enhanced Multi-Scale SAM (EMSAM). EMSAM employs the Local Feature Extraction Module (LFEM) and the Global Feature Extraction Module (GFEM) to extract local and global features from images respectively. The LFEM utilizes multiple convolutional layers to capture lesion boundaries and detailed characteristics, while the GFEM fine-tunes ViT blocks using a Multi-Scale Adaptive Adapter (MAA) to obtain multi-scale global information. Both outputs of LFEM and GFEM are then effectively fused in the Feature Fusion Module (FFM), which is optimized with cross-branch and channel attention mechanisms, significantly enhancing the model’s ability to handle blurred boundaries and complex shapes. EMSAM integrates lightweight linear layers as classification heads and employs a joint loss function for both classification and segmentation tasks. Experimental results on the PlantVillage dataset demonstrate that EMSAM outperforms the second-best state-of-the-art semantic segmentation model by 2.45% in Dice Coefficient and 6.91% in IoU score, and surpasses the baseline method by 21.40% and 22.57%, respectively. Particularly, for images with moderate and severe disease levels, EMSAM achieved Dice Coefficients of 0.8354 and 0.8178, respectively, significantly outperforming other semantic segmentation algorithms. Additionally, the model achieved a classification accuracy of 87.86% across the entire dataset, highlighting EMSAM’s effectiveness and superiority in plant disease segmentation and classification tasks.

Plant disease detection and segmentation are essential to maintaining healthy crops and improving agricultural productivity. Using Artificial Intelligence (AI) for this task enables farmers and researchers to identify diseases early and take preventive action. However, accurately segmenting diseased regions remains challenging, as most methods require a large number of labeled images or manual guidance to train the model. This limits their scalability and practical use in real-world agricultural settings. To address this issue, we propose the YOLO_SAM model for plant disease detection and segmentation, using self-prompted mask generation. Experiments were conducted on a self-generated dataset containing samples of different leaf diseases. The YOLO11 model was first used to detect infected regions, and its bounding boxes were automatically passed to the Segment Anything Model (SAM), which generated detailed segmentation masks without manual input. Experimental results showed that YOLOv11s achieved the best performance with a mean Average Precision (mAP) of 0.845, outperforming YOLOv11n (0.825), YOLOv11l (0.807), and YOLOv11m (0.764). Based on this superior performance, YOLOv11s was chosen as the prompt generator for SAM2, enabling more accurate and reliable segmentation of disease regions. This combined approach enabled the model to generate precise lesion masks without manual prompting, allowing clear boundary extraction even for small or irregular disease spots.

Recently, the Segment Anything Model (SAM) has demonstrated superior performance across numerous visual benchmarks. It is a highly influential foundational vision model capable of achieving excellent segmentation results through zero-shot approaches in various domains. However, studies have indicated that SAM does not always perform exceptionally in all areas. Notably, its performance in segmenting crop diseases and pests in agriculture has been found wanting. To address this shortfall, we introduce a specialized fine-tuning method for agriculture, termed SAM-Adapter. This approach enhances the SAM model's performance in agricultural detection by embedding domain-specific knowledge in conjunction with SAM's universal framework. Our experiments conducted on the BRACOT and PEST datasets demonstrate that the SAM-Adapter effectively improves SAM's performance in agricultural applications.

Agricultural production is a critical sector that directly impacts the economy and social life of any society. The identification of plant disease in a real-time environment is a significant challenge for agriculture production. For accurate plant disease detection, precise detection of plant leaves is a meaningful and challenging task for developing smart agricultural systems. Most researchers train and test models on synthetic images. So, when using that model in a real-time scenario, it does not give a satisfactory result because when a model trained on images of leaves is fed with the image of the plant, then its accuracy is affected. In this research work, we have integrated two models, the Segment Anything Model (SAM) with YOLOv8, to detect the tomato leaf of a tomato plant, mask the leaf, and extract the leaf in a real-time environment. To improve the performance of leaf disease detection in plant leaves in a real-time environment, we need to detect leaves accurately. We developed a system that will detect the leaf, mask the leaf, extract the leaf, and then detect the disease in that specific leaf. For leaf detection, the modified YOLOv8 is used, and for masking and extraction of the leaf images from the tomato plant, the Segment Anything Model (SAM) is used. Then, for that specific leaf, an image is provided to the deep neural network to detect the disease.

The detection of plant leaf diseases is essential for ensuring crop health and productivity. This study uses a comprehensive merged dataset, including the Mendeley Plant Leaf Dataset, which includes 22 classes, while the Jute and Mulberry datasets provide 2 classes each, representing healthy and diseased categories from various species. The final dataset, consisting of 26 classes, was augmented to ensure 500 training samples per class. In this study, the Segment Anything Model (SAM) was employed for zero-shot segmentation, enabling the automatic extraction of precise regions of interest (ROIs) from leaf images without the need for task-specific training for region-focused analysis. An EfficientNetB7- Multi-Head Attention (MHA) model that combines EfficientNetB7 with MHA was used to improve plant disease classification across 26 distinct classes. The proposed model is designed to handle the variety and diversity of leaf diseases, achieving a high classification accuracy of 98.01%, with precision, recall, and F1-scores all exceeding 97.9%. The integration of MHA allows the model to focus on disease-specific features, significantly enhancing its ability to generalize across diverse agricultural settings while maintaining scalability. Experimental results show that the EfficientNetB7-MHA model consistently outperforms several state-of-the-art (SOTA) models in terms of accuracy and robustness, making it a promising tool for precision agriculture. Explainable AI (XAI) including the use of Grad-CAM was utilized to detect the precise regions of interest for each class providing interpretability and insights into the model’s decision-making process. Finally, to demonstrate the model’s performance, a web application was developed that displays the predicted outputs when given different images of healthy and diseased leaves from the dataset. This comprehensive approach demonstrates significant potential for improving agricultural disease management through accurate, scalable, interpretable, and efficient disease detection.

The cultivation of date palm (Phoenix dactylifera L.) is acutely impacted by numerous fungal, bacterial, and pest-related diseases that diminish yield, spoil fruit quality, and undermine long-term agricultural sustainability. The traditional methods of monitoring diseases, which rely heavily on expert knowledge, are not scalable and depend heavily on classical models that do not generalize readily to real-world conditions. Recent improvements in deep learning over the last two decades, particularly with Convolutional Neural Networks (CNNs), have led to significantly greater automation. However, CNNs still require relatively large labeled datasets and struggle with ambiguous or complex background features, small lesions, and overlapping symptoms when diagnosing plant diseases. To address these difficulties, we introduce an innovative hybrid Transformer deep learning framework based on four sophisticated modules: 1) Swin Transformer for hierarchical image classification, 2) YOLOv12 for real-time detection, 3) Grounding DINO with SAM2.1 for zero-shot segmentation, and 4) Vision Transformer (ViT) and a regression head for predicting disease severity. This collective architecture can deliver accurate detection, the most accurate segmentation, and quantification of disease severity in real-world, low-annotation-based scenarios and adverse environmental context situations. Experimental findings on a curated dataset of 13,459 palm leaf images show that the proposed model outperforms all previous CNN-based models with a classification accuracy of 98.91%, a precision of 98.85%, a recall of 96.8%, and an F1-score of 96.4%. These results represent a new standard for automated, scalable, and interpretable disease diagnosis in precision agriculture.

The Segment Anything Model (SAM) is a versatile image segmentation model that enables zero-shot segmentation of various objects in any image using prompts, including bounding boxes, points, texts, and more. However, studies have shown that the SAM performs poorly in agricultural tasks like crop disease segmentation and pest segmentation. To address this issue, the agricultural SAM adapter (ASA) is proposed, which incorporates agricultural domain expertise into the segmentation model through a simple but effective adapter technique. By leveraging the distinctive characteristics of agricultural image segmentation and suitable user prompts, the model enables zero-shot segmentation, providing a new approach for zero-sample image segmentation in the agricultural domain. Comprehensive experiments are conducted to assess the efficacy of the ASA compared to the default SAM. The results show that the proposed model achieves significant improvements on all 12 agricultural segmentation tasks. Notably, the average Dice score improved by 41.48% on two coffee-leaf-disease segmentation tasks.

Accurate identification of tomato leaf diseases is essential for improving crop health management and reducing agricultural losses. However, the wide variety of tomato diseases, the limited availability of real-world datasets, and the variability of agricultural environments make developing reliable and explainable tomato disease identification systems challenging. Traditional deep learning models trained on controlled datasets often fail to generalize in real-world farm environments due to complex backgrounds and variations in illumination and image conditions. This paper presents a hierarchical, multi-stage framework that integrates leaf detection, segmentation, and classification with interpretability to achieve robust and explainable disease identification. The proposed pipeline employs YOLO11 for leaf detection, the Segment Anything Model (SAM) for segmentation, a ResNet-50 classifier, and LIME interpretability. To evaluate the impact of the pipeline on robustness and generalization, we conducted cross-domain experiments in two settings, using PlantVillage for laboratory conditions and PlantDoc for real-world conditions. The proposed pipeline improved real-world accuracy from 22.79 % with a flat ResNet-50 to 55.33 %, cutting the accuracy drop by 34.0 %. The pipeline stages focused on leaf isolation to ensure that predictions are driven by symptomatic tissue rather than background. Experiments demonstrate that our approach significantly improves generalization and transparency, addressing major gaps in existing plant disease identification systems.

Tomato leaf diseases significantly impact crop productivity, necessitating accurate and efficient diagnostic tools. This study proposes a hybrid framework that integrates deep learning-based localization and segmentation with handcrafted feature extraction and classical machine learning for tomato leaf disease classification. Specifically, YOLOv8 is used for object detection and SAM for segmenting diseased regions. Features are then extracted using HSV color space, GLCM, and LBP descriptors. To address class imbalance, the SMOTE technique was applied, expanding the original 48,243 image dataset to 102,465 balanced samples across 11 disease categories. Multiple classifiers were evaluated, with Random Forest achieving the highest performance over 90% accuracy and a macro F1-score of 0.90. Importantly, recall for minority classes improved markedly after balancing. The proposed system demonstrates strong potential for deployment in real-world agricultural environments due to its low computational cost and robustness under varying conditions. Future work will explore multi-crop generalization, real-time inference, and eld validation under challenging conditions such as lighting variation and occlusion.

Tomato leaf diseases pose a significant threat to global food security, necessitating accurate and efficient detection methods. This paper introduces the Tomato Leaf Disease Visual Language Model (TLDVLM), a novel approach based on the BLIP-2 architecture enhanced with Low-Rank Adaptation (LoRA), for precise classification of 10 distinct tomato leaf diseases. Our methodology integrates a sophisticated image preprocessing pipeline, utilizing GroundingDINO for robust leaf detection and SAM-2 for pixel-level segmentation, ensuring that the model focuses solely on relevant plant tissue. The TLDVLM leverages the powerful multimodal understanding of BLIP-2, with LoRA applied to its Q-Former module, enabling parameter-efficient fine-tuning without compromising performance. Comparative experiments demonstrate that the TLDVLM significantly outperforms baseline models, including CLIP-LoRA and ConvNeXT-tiny, achieving an accuracy of 97.27%, a precision of 0.9587, a recall of 0.9789, and an F1-score of 0.9681. Beyond classification, the finetuned TLDVLM checkpoints are integrated into a practical application for new image inference. This application displays the raw and segmented images, the predicted disease, and offers functionalities to fetch comprehensive information on disease causes and remedies using external APIs (e.g., OpenAI), with an option to download a PDF summary for offline access on a portable device. This research highlights the potential of LoRA-adapted Vision-Language Models in developing highly accurate, efficient, and user-friendly agricultural diagnostic tools.

Ensuring sustainable and profitable agriculture is critical for addressing global food security challenges. This has resulted in the need for automation in plant health identification. However, this objective is hampered by the lack of efficient image-processing methods and the need for extensive datasets for training deep learning models for plant disease diagnosis. To overcome the need for extensive training data, the proposed Localized Normalized Difference Vegetation Index (LNDVI) uses zero-shot plant detection models such as Grounded Dino and state-of-the-art methods for image segmentation such as Segment Anything Model (SAM) are leveraged. This also expands the capabilities of the system to diagnose plant health beyond known plant species available as part of training set. The proposed system uses synthetic Normalized Difference Vegetation Index (NDVI) to estimate the chlorophyll content of the plant through RGB images alone instead of using the combination of RGB and near Infra-red (nIR) bands used in contemporary works. Since NDVI value is greatly affected by the amount of light present while the image is captured, we also present an irradiation estimation metric that uses CIE XYZ (Tristimulus values), Hue, Saturation and Value (HSV) and CIE LAB color spaces as well as correlated color temperatures, which automatically normalizes the NDVI threshold for health classification of the image, enabling a more precise analysis of plant health. Using the Grounding Dino provided an accuracy of 99.994% in terms of detecting plants from the phenotyping dataset. The segmentation of plant region in images is reported using Intersection over Union (IoU). While using the Segment Anything Model (SAM), an average accuracy of 95.884% was obtained for clustered plants while the average accuracy was even better at 97.031% for individual plants. Significant differences were observed for plant health classification while using Localized Normalized Difference Vegetation Index (LNDVI) approach when compared to NDVI.

Cotton is one of the world’s most important economic crops, and its yield and quality have a significant impact on the agricultural economy. However, Verticillium wilt of cotton, as a widely spread disease, severely affects the growth and yield of cotton. Due to the typically small and densely distributed characteristics of this disease, its identification poses considerable challenges. In this study, we introduce YOLO-MSPM, a lightweight and accurate detection framework, designed on the YOLOv11 architecture to efficiently identify cotton Verticillium wilt. In order to achieve a lightweight model, MobileNetV4 is introduced into the backbone network. Moreover, a single-head self-attention (SHSA) mechanism is integrated into the C2PSA block, allowing the network to emphasize critical areas of the feature maps and thus enhance its ability to represent features effectively. Furthermore, the PC3k2 module combines pinwheel-shaped convolution (PConv) with C3k2, and the mobile inverted bottleneck convolution (MBConv) module is incorporated into the detection head of YOLOv11. Such adjustments improve multi-scale information integration, enhance small-target recognition, and effectively reduce computation costs. According to the evaluation, YOLO-MSPM achieves precision (0.933), recall (0.920), mAP50 (0.970), and mAP50-95 (0.797), each exceeding the corresponding performance of YOLOv11n. In terms of model lightweighting, the YOLO-MSPM model has 1.773 M parameters, which is a 31.332% reduction compared to YOLOv11n. Its GFLOPs and model size are 5.4 and 4.0 MB, respectively, representing reductions of 14.286% and 27.273%. The study delivers a lightweight yet accurate solution to support the identification and monitoring of cotton Verticillium wilt in environments with limited resources.

Researchers have designed various models trained on public or private datasets for plant disease detection to help farmers remedy crop yield losses on their farms due to plant diseases. Plantvillage is the most widely used plant disease dataset with laboratory images captured under controlled conditions with a single leaf on each image and a uniform background. Models trained on such datasets have extremely low classification accuracies when running on field images captured directly from plantations with various interwoven leaves, complex backgrounds, and different lighting conditions. In this study, we propose a model ensemble solution for the accurate identification and classification of plant diseases using field images. The model uses Segment Anything Model to efficiently circumscribe all identifiable objects in the image. Image Processing techniques are then used to isolate the identified objects from the original image. Background objects are separated from actual leaf objects using Fully Convolutional Data Description, which is an explainable deep one-class classification model for anomaly detection. Finally, the selected leaves are submitted to a Plantvillage-trained classification model for inference. Our model can detect diseases appearing on individual leaves of the same image and improves classification accuracy by more than 10% on public field plant disease datasets such as PlantDoc, thus providing a reliable solution for farmers and practitioners.

Leaf segmentation plays a crucial role in plant phenotyping and precision agriculture, enabling the monitoring of growth, disease detection, and informed crop management. However, accurate segmentation in natural environments is challenging due to complex backgrounds, overlapping structures, irregular boundaries, and varying illumination. This paper proposes a hybrid six-stage framework that integrates U-Net with the Fast Segment Anything Model (FastSAM) to achieve accurate and efficient leaf segmentation. The pipeline consists of initial U-Net segmentation, largest component filtering, contour extraction with convex hull transformation, bounding box derivation via distance transform, promptable refinement with FastSAM, and final contour selection. The experiments conducted used 633 images from the Pl@ntLeaves database: 333 images for model development with a train/validation split of 266/67 (20% validation), and a held-out test set of 300 images. On the 300-image test set, the proposed framework achieved superior results (Precision = 0.966, Recall = 0.945, Intersection over Union (IoU) = 0.917, Dice = 0.953, HD95 = 27.859), outperforming DeepLabV3 and CLIPSeg. These findings confirm that combining U-Net's fine-grained feature extraction with FastSAM's efficient prompt-based refinement provides a robust and scalable solution for plant phenotyping and precision agriculture, particularly by enhancing boundary accuracy in complex natural scenes.

Introduction Precise semantic segmentation of microbial alterations is paramount for their evaluation and treatment. This study focuses on harnessing the SegFormer segmentation model for precise semantic segmentation of strawberry diseases, aiming to improve disease detection accuracy under natural acquisition conditions. Methods Three distinct Mix Transformer encoders - MiT-B0, MiT-B3, and MiT-B5 - were thoroughly analyzed to enhance disease detection, targeting diseases such as Angular leaf spot, Anthracnose rot, Blossom blight, Gray mold, Leaf spot, Powdery mildew on fruit, and Powdery mildew on leaves. The dataset consisted of 2,450 raw images, expanded to 4,574 augmented images. The Segment Anything Model integrated into the Roboflow annotation tool facilitated efficient annotation and dataset preparation. Results The results reveal that MiT-B0 demonstrates balanced but slightly overfitting behavior, MiT-B3 adapts rapidly with consistent training and validation performance, and MiT-B5 offers efficient learning with occasional fluctuations, providing robust performance. MiT-B3 and MiT-B5 consistently outperformed MiT-B0 across disease types, with MiT-B5 achieving the most precise segmentation in general. Discussion The findings provide key insights for researchers to select the most suitable encoder for disease detection applications, propelling the field forward for further investigation. The success in strawberry disease analysis suggests potential for extending this approach to other crops and diseases, paving the way for future research and interdisciplinary collaboration.

This research addresses the critical need for efficient image annotation in precision agriculture, using the wild blueberry (Vaccinium angustifolium Ait.) cropping system as a representative application to enable data-driven crop management. Tasks such as automated berry ripeness detection, plant disease identification, plant growth stage monitoring, and weed detection rely on extensive annotated datasets. However, manual annotation is labor-intensive, time-consuming, and impractical for large-scale agricultural systems. To address this challenge, this study evaluates an automated annotation pipeline that integrates zero-shot detection models from two frameworks (Grounding DINO and YOLO-World) with the Segment Anything Model version 2 (SAM2). The models were tested on detecting and segmenting ripe wild blueberries, developmental wild blueberry buds, hair fescue (Festuca filiformis Pourr.), and red leaf disease (Exobasidium vaccinii). Grounding DINO consistently outperformed YOLO-World, with its Swin-T achieving mean Intersection over Union (mIoU) scores of 0.694 ± 0.175 for fescue grass and 0.905 ± 0.114 for red leaf disease when paired with SAM2-Large. For ripe wild blueberry detection, Swin-B with SAM2-Small achieved the highest performance (mIoU of 0.738 ± 0.189). Whereas for wild blueberry buds, Swin-B with SAM2-Large yielded the highest performance (0.751 ± 0.154). Processing times were also evaluated, with SAM2-Tiny, Small, and Base demonstrating the shortest durations when paired with Swin-T (0.30–0.33 s) and Swin-B (0.35–0.38 s). SAM2-Large, despite higher segmentation accuracy, had significantly longer processing times (significance level α = 0.05), making it less practical for real-time applications. This research offers a scalable solution for rapid, accurate annotation of agricultural images, improving targeted crop management. Future research should optimize these models for different cropping systems, such as orchard-based agriculture, row crops, and greenhouse farming, and expand their application to diverse crops to validate their generalizability.

No abstract available

Branch canker caused by Macrophoma theicola is a major stem disease of tea plants (Camellia spp.). In tea plantations, this disease causes crop loss and it is one of the major limiting factor for yield stagnation. In very few instances it causes considerable damage in new clearings (about 3 or 4 years old) and large number of bushes have been killed. As there is no control measures for branch canker disease in south Indian tea plantation, this field study was conducted in naturally infected pruned tea field at UPASI Tea Research Institute (Good Agricultural Practice), Valparai, Tamil Nadu, India. The chemical fungicides, biological agents and bio products were evaluated under naturally infected field of seedling plants for two consecutive disease seasons (2014–2015) and there was 11 treatments with three applications. All the treatments were carried out in the time of February–March and October–November (2014–2015). The two set of application was conducted per year. Each set contains eight rounds during the month of February–March as well as October–November (2014–2015). The chemical fungicides, biological agents and commercial bio products were measured as per UPASI- TRF, recommendation viz., COC (50 g/ha and 0.2 g/plot), Companion (20 g/ha and 0.08 g/plot), biological agent of Bacillus amyloliquefaciens, Tichoderma harzianum, Gliocladium virens and Beauveria bassiana (5 kg/ha and 20.8 g/plot) and bio product of Tari (1 L/ha and 4.2 ml/plot) and Tricure (1 L/ha and 4.2 ml/plot). The present investigation revealed the integrated application of Companion/Bacillus amyloliquefaciens showed superior control of branch canker disease followed by the treatment with Companion alone under field condition. Copper oxychloride/Bacillus amyloliquefaciens was moderately effective followed by Copper oxychloride. The significantly reduced canker size was recorded with treatment of Bacillus amyloliquefaciens followed by commercial organic fungicides of Tari (Organic Tea Special) and Tricure (0.03% Azadirachtin). The least canker size was observed with Gliocladium virens followed by Beauveria bassiana. Branch canker disease incidence was increased in untreated control plants when compared to treated plants. Among these 11 treatments, the integrated treatment of companion at rate of 0.08 g and Bacillus amyloliquefaciens (20.8 g) showed the most significantly decreased canker size (DPL, 5.76) followed by another treatment with companion (0.08 g) (DPL, 4.11). The moderate reduction of canker size was observed by the treatment with Copper oxychloride (0.2 g)/Bacillus amyloliquefaciens (20.8 g) (DPL, 3.05) followed by the treatment of copper oxychloride alone (DPL, 1.74). Therefore, the integrated application of Companion/Bacillus amyloliquefaciens proved significantly effective in the management of branch canker disease under the field conditions.

No abstract available

Climate change triggered by global warming poses a major threat to agricultural systems globally. This phenomenon is characterized by emergence of pests and diseases, extreme weather events, such as prolonged drought, high intensity rains, hailstones and frosts, which are becoming more frequent ultimately impacting negatively to agricultural production including rain-fed tea cultivation. Kenya is predominantly an agricultural based economy, with the tea sector generating about 26% of the total export earnings and about 4% gross domestic product (GDP). In the recent years, however, the country has witnessed unstable trends in tea production associated with climate driven stresses. Toward mitigation and adaptation of climate change, multiple approaches for impact assessment, intensity prediction and adaptation have been advanced in the Kenyan tea sub-sector. Further, pressure on tea breeders to release improved climate-compatible cultivars for the rapidly deteriorating environment has resulted in the adoption of a multi-targeted approach seeking to understand the complex molecular regulatory networks associated with biotic and abiotic stresses adaptation and tolerance in tea. Genetic modeling, a powerful tool that assists in breeding process, has also been adopted for selection of tea cultivars for optimal performance under varying climatic conditions. A range of physiological and biochemical responses known to counteract the effects of environmental stresses in most plants that include lowering the rates of cellular growth and net photosynthesis, stomatal closure, and the accumulation of organic solutes such as sugar alcohols, or osmolytes have been used to support breeding programs through screening of new tea cultivars suitable for changing environment. This review describes simulation models combined with high resolution climate change scenarios required to quantify the relative importance of climate change on tea production. In addition, both biodiversity and ecosystem based approaches are described as a part of an overall adaptation strategy to mitigate adverse effects of climate change on tea in Kenya and gaps highlighted for urgent investigations.

Epigenetic regulation includes a set of regulatory processes, such as histone modification, DNA methylation and small noncoding RNA regulation, which modify gene expression without changing its original DNA sequence (1). Among all the epigenetic modifications, DNA methylation has been extensively studied in the past decades. DNA methylation mainly refers to the covalent addition of methyl groups to the 5’ position of a cytosine. Such process is catalysed by DNA methyl transferases (DNMTs) and mainly occurs at cytosine-phosphate-guanine (CpG) sites in the human genome (2). The function of DNA methylation varies with the location it takes place within the genome. Usually, gene expression is repressed with elevated DNA methylation levels in promoters while gene body methylation is linked with upregulated transcription (3). The removal of methyl groups (i.e., DNA demethylation) can happen either passively by failing to maintain the methylation patterns after replication or actively through oxidizing methylated cytosine into 5-hydroxymethylcytosine (5-hmC), a process mediated by ten-eleven translocation (TET) enzymes (4). DNA methylation is a dynamic process that progressively diverges across lifetime. Heyn et al. compared peripheral blood DNA methylation patterns of newborns and centenarians, and reported a globally decreased methylation level in the centenarian while in gene promoters of the aged group, the methylation level was higher than the newborn (5). The study of Kananen et al. on blood methylation patterns of a middle-aged population (40-49 years) identified 1,202 ageingassociated CpG sites (6), the majority (987 sites) of which were also found in a methylation comparison between young adults and nonagenarians (7). Besides blood, age-related CpG sites have been uncovered from other tissues such as saliva (8), skeletal muscle (9) and brain (10). The discovery of agerelated CpG sites provides a novel approach for age prediction. In the year 2011, Bocklandt et al. pioneered such prediction by building a regression model, which explained 73% of the variance in age, based on two CpG sites identified from saliva (8). Two years later, Horvath introduced an epigenetic clock, also known as DNA methylation (DNAm) age, based on 353 CpG sites identified across 51 tissues with a median absolute prediction error of 3.6 years (11). Despite that other epigenetic clocks have been continuously created (10, 12–14), Horvath’s clock gains large popularity for its ability to predict age across multiple tissues. Although the epigenetic clock was initially developed for age prediction, the discovery of its connection with lifespan has made it a hallmark of biological age with an accelerated tick of the clock indicating a faster degeneration speed (15). Meanwhile, the concept of epigenetic age acceleration is proposed as the difference between DNAm age and chronological age (16). Positive epigenetic age acceleration (i.e., epigenetic age greater than chronological age) indicates that the tissue ages faster than would be expected, and has been linked with multiple age-related conditions such as obesity (17), frailty (18), osteoarthritis (19), cognitive decline (20), Alzheimer’s disease (21), Parkinson’s Disease (22) and allcause mortality (23). DNA methylation can be regulated by nutrients. Many B-vitamins are directly linked with DNA methylation by serving as substrates or cofactors in relevant pathways. With folate as the main substrate and related B-vitamins (e.g., vitamin B-2, vitamin B-6 and vitamin B-12) as cofactors, onecarbon metabolism plays a critical role in generating methyl groups that are later used for DNA methylation. Other nutrients such as choline and betaine are involved in the methionine cycle for the regeneration of methionine, a precursor for the universal methyl donor S-adenosylmethionine (SAM) (24). Multiple cohort studies have shown increased DNA methylation associated with higher intakes or blood levels of folate, vitamin B-2 and vitamin B-6 (25–27). Some bioactive food components, such as green tea polyphenols and soybean genistein, have inhibitory effect on DNMTs and can reduce cancer activities by reversing hypermethylation status of key tumour suppression genes (28). Besides DNA methylation pathways, a study of Yin et al. have found that vitamin C can promote DNA demethylation by enhancing activities of TET enzymes (29). The close connection between nutrients and DNA methylation makes epigenetic clocks as ideal indices in ageingrelated nutrition studies. Firstly, the emerge of epigenetic clocks provides an additional insight to verify the importance of diet in health management. The study of Quach et al. found that lower epigenetic age acceleration was associated with higher fish and poultry intake, moderate alcohol consumption and higher fruit and vegetable consumption, supporting the benefits of a plant-based diet with lean meats (30). Secondly, the epigenetic clock can be used to test DNA methylationrelated hypothesis. Vitamin D deficiency has been associated with chronic pain, the leading cause of which is osteoarthritis (31). While underlying mechanisms of the association remain poorly understood, a possible explanation might be epigenetic alterations that have been related to both vitamin D (32) and © Serdi and Springer-Verlag International SAS, part of Springer Nature

Kombucha is a potential probiotic tea-based drink with increasing worldwide consumption. Studies on this probiotic beverage are growing rapidly, especially about micronutrients and microbial population. As such, the present study performed the molecular identification of the microorganism and evaluated 5-methyltetrahydrofolate content by HPLC-DAD, phenolic compounds, flavonoids, carotenoids, antioxidant activity by spectrophotometric methods, and physicochemical composition of green tea kombucha on fermentation days 1, 3, 7, 14, and 21. DNA sequencing identified the Microbacterium genus as predominant. However, was unable to safely determine the species level because of the rRNA 16S gene sequence similarity between four species M. ureisolvens, M. yannicii, M. chocolatum e M. atlanticum. The concentration of 5-methyltetrahydrofolate found on the third day was 39.12 ± 1.32 μg/mL (liquid) and 45.78 ± 8.42 μg/mL (polymeric biofilm); On the twenty-first day it was 50.87 ± 3.56 μg/mL (liquid) and 54.88 ± 3.89 μg/mL (polymeric biofilm). Total phenolic compounds increased with fermentation; however, flavonoids and carotenoids were degraded by the process. The information on 5-methyltetrahydrofolate is unprecedented and highly relevant for food guidelines, since related deficiencies can lead to fetal malformation in the first three months of pregnancy. Lastly, the best fermentation time to obtain 5-methyltetrahydrofolate and others bioactive compounds is between days 7–14. Further analyses are also encouraged to understand the bioavailability of the vitamin.

Tea production and quality are largely determined by the many genetic and biochemical characteristics that occur in tea plant cultivars. Worldwide, tea is consumed for its pleasing and refreshing effects due to its caffeine content. The present study performed transcriptomics analyses of two tea species (Camellia sinensis var. Shuchazao (SCZ) and Camellia ptilophylla (CAF)) and identified diversity in the gene expression levels and major regulatory transcription factors (TFs) for the characterization of purine alkaloids and phenylpropanoid biosynthesis pathways. The RNA-seq analysis of two species (SCZ and CAF) revealed the differences in caffeine and catechins synthesis. In the purine alkaloid biosynthesis pathway, the S-adenosyl methionine (SAM) and adenosine monophosphate (AMP) pathway genes were significantly related to xanthosine synthesis in contrasting purine alkaloids among (Camellia sinensis var. Shuchazao (SCZ) and Camellia ptilophylla (CAF)). The significant expression of SAMS-5, PPAT-2, IMPDH-2, TCS-2, TCS-3, XMT-1, XMT-13, and XDH-4 in the xanthosine degradation pathway in CAF is attributed to higher theobromine content as compared to SCZ. Moreover, the transcription factors (TFs) AP2/ERF (20%), WRKY (12%), NAC (11%), and MYB (8%) were significantly correlated. The upregulated expression of caffeine synthesis genes in SCZ was correlated with MYB and AP2/ERF transcription factors. This study provides the basis for differences in the genetic mechanism in purine alkaloids, phenylpropanoid, and flavonoid biosynthesis pathways, which would be helpful in the development and selection of tea plant species with high or low caffeine concentrations. This study also provides a road map for future genetic improvement in tea species and cultivars.

No abstract available

Tea producers are in demand of new high yielding cultivars, which produce high quality tea liquors. To breed for these phenotypic traits is challenging due to their polygenic disposition and influence by environment. Two C. sinensis populations, namely Comm cultivars from open pollinated field selections, and NComm cultivars from the reciprocal cross of two parents were used. These cultivars were employed to identify the metabolites responsible for distinguishing Comm cultivars, with high yield, high quality and DT from NComm cultivars that did not show these traits. PCA and PLS-DA models were constructed on UPLC/DAD data, which showed clear separation between the Comm and NComm cultivars. CHAID decision trees constructed aimed to classify the 303 genotypes as either Comm or NComm cultivars using subset of compounds. Breeders can predict the quality of new selections from mature seedling fields by employing CHAID decision trees, or the CAF/EC ratio, as predictors.

No abstract available

The two original plants of the oolong tea cultivar (‘Tieguanyin’) are “Wei shuo” ‘Tieguanyin’—TGY (Wei) and “Wang shuo” ‘Tieguanyin’—TGY (Wang). Another cultivar, ‘Benshan’ (BS), is similar to TGY in its aroma, taste, and genetic make-up, but it lacks the “Yin Rhyme” flavor. We aimed to identify differences in biochemical characteristics and gene expression among these tea plants. The results of spectrophotometric, high performance liquid chromatography (HPLC), and gas chromatography-mass spectrometry (GC-MS) analyses revealed that TGY (Wei) and TGY (Wang) had deeper purple-colored leaves and higher contents of anthocyanin, catechins, caffeine, and limonene compared with BS. Analyses of transcriptome data revealed 12,420 differentially expressed genes (DEGs) among the cultivars. According to a Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, the flavonoid, caffeine, and limonene metabolic pathways were highly enriched. The transcript levels of the genes involved in these three metabolic pathways were not significantly different between TGY (Wei) and TGY (Wang), except for two unigenes encoding IMPDH and SAMS, which are involved in caffeine metabolism. The comparison of TGY vs. BS revealed eight up-regulated genes (PAL, C4H, CHS, F3’H, F3H, DFR, ANS, and ANR) and two down-regulated genes (FLS and CCR) in flavonoid metabolism, four up-regulated genes (AMPD, IMPDH, SAMS, and 5′-Nase) and one down-regulated XDH gene in caffeine metabolism; and two down-regulated genes (ALDH and HIBADH) in limonene degradation. In addition, the expression levels of the transcription factor (TF) PAP1 were significantly higher in TGY than in BS. Therefore, high accumulation of flavonoids, caffeine, and limonene metabolites and the expression patterns of their related genes in TGY might be beneficial for the formation of the “Yin Rhyme” flavor. Transcriptomic, HPLC, and GC-MS analyses of TGY (Wei), TGY (Wang), and BS indicated that the expression levels of genes related to secondary metabolism and high contents of catechins, anthocyanin, caffeine, and limonene may contribute to the formation of the “Yin Rhyme” flavor in TGY. These findings provide new insights into the relationship between the accumulation of secondary metabolites and sensory quality, and the molecular mechanisms underlying the formation of the unique flavor “Yin Rhyme” in TGY.

Tea quality and yield is influenced by various factors including developmental tissue, seasonal variation and cultivar type. Here, the molecular basis of these factors was investigated in three tea cultivars namely, Him Sphurti (H), TV23 (T), and UPASI-9 (U) using RNA-seq. Seasonal variation in these cultivars was studied during active (A), mid-dormant (MD), dormant (D) and mid-active (MA) stages in two developmental tissues viz. young and old leaf. Development appears to affect gene expression more than the seasonal variation and cultivar types. Further, detailed transcript and metabolite profiling has identified genes such as F3′H, F3′5′H, FLS, DFR, LAR, ANR and ANS of catechin biosynthesis, while MXMT, SAMS, TCS and XDH of caffeine biosynthesis/catabolism as key regulators during development and seasonal variation among three different tea cultivars. In addition, expression analysis of genes related to phytohormones such as ABA, GA, ethylene and auxin has suggested their role in developmental tissues during seasonal variation in tea cultivars. Moreover, differential expression of genes involved in histone and DNA modification further suggests role of epigenetic mechanism in coordinating global gene expression during developmental and seasonal variation in tea. Our findings provide insights into global transcriptional reprogramming associated with development and seasonal variation in tea.

Tea is one of the world's most popular beverages, known for its cultural significance and numerous health benefits. A clear understanding of the origin and history of domestication of the tea species is a fundamental pre-requisite for effective germplasm conservation and improvement. Though there is a general consensus about the center of origin of the tea plant, the evolutionary origin and expansion history of the species remain shrouded in controversy, with studies often reporting conflicting findings. This mini review provides a concise summary of the current state of knowledge regarding the origin, domestication, and dissemination of the species around the world. We note that tea was domesticated around 3000 B.C. either from non-tea wild relatives (probably Camellia grandibracteata and/or C. leptophylla) or intra-specifically from the wild Camellia sinensis var. assamica trees, and that the genetic origins of the various tea varieties may need further inquiry. Moreover, we found that lineage divergence within the tea family was apparently largely driven by a combination of orogenic, climatic, and human-related forces, a fact that could have important implications for conservation of the contemporary tea germplasm. Finally, we demonstrate the robustness of an integrative approach involving linguistics, historical records, and genetics to identify the center of origin of the tea species, and to infer its history of expansion. Throughout the review, we identify areas of debate, and highlight potential research gaps, which lay a foundation for future explorations of the topic.

Africa is one of the key centers of global tea production. Understanding the genetic diversity and relationships of cultivars of African tea is important for future targeted breeding efforts for new crop cultivars, specialty tea processing, and to guide germplasm conservation efforts. Despite the economic importance of tea in Africa, no research work has been done so far on its genetic diversity at a continental scale. Twenty-three nSSRs and three plastid DNA regions were used to investigate the genetic diversity, relationships, and breeding patterns of tea accessions collected from eight countries of Africa. A total of 280 African tea accessions generated 297 alleles with a mean of 12.91 alleles per locus and a genetic diversity (HS) estimate of 0.652. A STRUCTURE analysis suggested two main genetic groups of African tea accessions which corresponded well with the two tea types Camellia sinensis var. sinensis and C. sinensis var. assamica, respectively, as well as an admixed “mosaic” group whose individuals were defined as hybrids of F2 and BC generation with a high proportion of C. sinensis var. assamica being maternal parents. Accessions known to be C. sinensis var. assamica further separated into two groups representing the two major tea breeding centers corresponding to southern Africa (Tea Research Foundation of Central Africa, TRFCA), and East Africa (Tea Research Foundation of Kenya, TRFK). Tea accessions were shared among countries. African tea has relatively lower genetic diversity. C. sinensis var. assamica is the main tea type under cultivation and contributes more in tea breeding improvements in Africa. International germplasm exchange and movement among countries within Africa was confirmed. The clustering into two main breeding centers, TRFCA, and TRFK, suggested that some traits of C. sinensis var. assamica and their associated genes possibly underwent selection during geographic differentiation or local breeding preferences. This study represents the first step toward effective utilization of differently inherited molecular markers for exploring the breeding status of African tea. The findings here will be important for planning the exploration, utilization, and conservation of tea germplasm for future breeding efforts in Africa.

The understanding of black tea quality and percent relative water content (%RWC) traits in tea (Camellia sinensis) by a quantitative trait loci (QTL) approach can be useful in elucidation and identification of candidate genes underlying the QTL which has remained to be difficult. The objective of the study was to identify putative QTL controlling black tea quality and percent relative water traits in two tea populations and their F1 progeny. A total of 1,421 DArTseq markers derived from the linkage map identified 53 DArTseq markers to be linked to black tea quality and %RWC. All 53 DArTseq markers with unique best hits were identified in the tea genome. A total of 5,592 unigenes were assigned gene ontology (GO) terms, 56% comprised biological processes, cellular component (29%) and molecular functions (15%), respectively. A total of 84 unigenes in 15 LGs were assigned to 25 different Kyoto Encyclopedia of Genes and Genomes (KEGG) database pathways based on categories of secondary metabolite biosynthesis. The three major enzymes identified were transferases (38.9%), hydrolases (29%) and oxidoreductases (18.3%). The putative candidate proteins identified were involved in flavonoid biosynthesis, alkaloid biosynthesis, ATPase family proteins related to abiotic/biotic stress response. The functional annotation of putative QTL identified in this current study will shed more light on the proteins associated with caffeine and catechins biosynthesis and % RWC. This study may help breeders in selection of parents with desirable DArTseq markers for development of new tea cultivars with desirable traits.

Tea (Camellia sinensis) is a major cash crop and leading foreign exchange earner, contributing to poverty alleviation by providing employment and livelihood to many stakeholders in the producing countries. Production has increased faster than consumption causing price stagnation, especially for CTC black teas. Kenya is the third-largest tea producer and Lake Victoria Basin produces over 60% of her tea. Selection of tea cultivars in Kenya has been cantered in one location before the selected clones are introduced to other growth environments. This study evaluated if tea clones maintain their yield and plain black tea quality attributes when grown at different locations within Lake Victoria Basin. The basin produces mainly plain black teas whose quality is due to levels of polyphenolic compounds, especially green leaf flavan-3-ols that are oxidized to theaflavins and thearubigins during black tea processing. The theaflavins and thearubigins contribute to the color and brightness of black teas. The trials were done in two sites Timbilil and Kipkebe using twenty clones. All the plain tea quality parameters including individual theaflavins and yields varied (p less than 0.05) with clones, demonstrating diversity in the cultivars used. The levels of the parameters and yields also changed (p less than 0.05) with the location of production. These results demonstrated that clonal tea quality and yields vary depending on the geographical location of production. There were also significant interactions effects between the clones and location of production in the quality parameters and clones showing the extent of the changes varied from clone to clone. Indeed the relative ranking of the clones varied with location. No clone retained its relative superiority ranking at the two locations. Both the Spearman correlation coefficients (rs) and the Pearson correlation coefficients (r) between the individual parameters were positive but low and insignificant, except for theaflavin-3-gallate. These results demonstrate the need for location-specific evaluation of both new and old clones to establish clonal yield and quality potentials in new locations of cultivation.

India's diverse cuisine and lack of nutritional awareness contribute significantly to lifestyle diseases. The rising prevalence of such diseases has increased the demand for personalized dietary advice, highlighting the need for effective nutrition monitoring. Current solutions often rely on time-consuming manual food data entry or costly consultations with dietitians, making them inaccessible to many. The proposed Android application personalizes nutrition management by first gathering user data (age, weight, height, physical activity level, health conditions) to calculate TDEE and individual macronutrient requirements. Users then capture meal images, which are analyzed using YOLOv11 for food detection and SAM for precise segmentation. Based on nutritional analysis and the user's health condition, the app offers safe, healthier food recommendations and updates a dashboard that tracks daily nutrient intake and meal history. The research focuses on integrating YOLO, SAM, and recommendation-based meal tracking to enhance dietary management. In NutriVision, the object detection model reliably identifies and classifies food items, while SAM ensures precise segmentation with clearly defined object masks. The outputs from these models are used for nutritional analysis, which evaluates the macronutrient content of each item. This analysis then informs tailored food recommendations, delivering personalized dietary suggestions and enhancing the overall user experience.

Rice is an essential crop in the world food supply. As the consumption of rice grows, its production has fallen due to diseases. These diseases are able to spread quickly, leading to outbreaks that cause major yield loss. It is important to identify the disease quickly before it spreads out. However, most rice disease classification model is that the training uses datasets containing images with a single leaves. This is unrealistic as it is cumbersome to take a picture of each leaves one by one. This paper proposes the use of segmentation using Mask R-CNN, this will allowed the model to focus on identifying key features of three types of rice disease, namely rice blast, brown spot, and leaf blight. DenseNet classification model is trained using the cropped dataset and uncropped dataset. The result shows an increased in F1 score, which is counted as the frequency of a model making correct predictions. The F1 score when using cropped dataset compared to uncropped dataset increased by 8,22%, from 77,91% to 86,13%. This proves the idea of image segmentation using Mask-RCNN is able to improve the accuracy and performance of image classification models.

The date palm (Phoenix dactylifera L.) is a vital crop in arid and semi-arid regions, contributing over $13 billion annually to the global economy. However, it faces significant yield losses due to pests, such as the red palm weevil, and diseases, including Bayoud and Black Scorch. Currently, expert visual inspection is the primary method of management, but it is time-consuming, subjective, and unsuitable for detecting large-scale or early-stage damage. Automated approaches based on classical machine learning offer limited improvements due to their lack of generalizability and environmental sensitivity. Recent deep learning methods, such as CNNs and Vision Transformers, have improved classification accuracy, but treat tasks like classification, detection, segmentation, and severity estimation as separate. This paper proposes an integrated Reveal-Aware Hybrid Vision-Language and Transformer-based AI framework that combines GAN-based augmentations for feature generation, CLIP for multimodal classification, PaliGemma2 for text-based detection, Grounding DINO + SAM 2.1 for zero-shot segmentation, and a Vision Transformer regression model for severity prediction. This end-to-end explainable diagnostic pipeline achieved 98% classification accuracy, 95.8% precision, 91.3% recall, and 94.2% F1-score across two datasets: nine classes of infected date palm leaves and three classes of date palm diseases. The proposed framework demonstrated detection accuracy of 94-98%, high-quality segmentations, and reliable severity estimates. This integrated approach highlights the potential of combining AI, vision-language models, and transformers for scalable, accurate, and sustainable plant disease management.

Poultry farming is an indispensable part of global agriculture, playing a crucial role in food safety and economic development. Managing and preventing diseases is a vital task in the poultry industry, where semantic segmentation technology can significantly enhance the efficiency of traditional manual monitoring methods. Furthermore, traditional semantic segmentation has achieved excellent results on extensively manually annotated datasets, facilitating real-time monitoring of poultry. Nonetheless, the model encounters limitations when exposed to new environments, diverse breeding varieties, or varying growth stages within the same species, necessitating extensive data retraining. Overreliance on large datasets results in higher costs for manual annotations and deployment delays, thus hindering practical applicability. To address this issue, our study introduces HSDNet, an innovative semantic segmentation model based on few-shot learning, for monitoring poultry farms. The HSDNet model adeptly adjusts to new settings or species with a single image input while maintaining substantial accuracy. In the specific context of poultry breeding, characterized by small congregating animals and the inherent complexities of agricultural environments, issues of non-smooth losses arise, potentially compromising accuracy. HSDNet incorporates a Sharpness-Aware Minimization (SAM) strategy to counteract these challenges. Furthermore, by considering the effects of imbalanced loss on convergence, HSDNet mitigates the overfitting issue induced by few-shot learning. Empirical findings underscore HSDNet’s proficiency in poultry breeding settings, exhibiting a significant 72.89% semantic segmentation accuracy on single images, which is higher than SOTA’s 68.85%.

Alfalfa is the most widely cultivated forage legume, with approximately 30 million hectares planted worldwide. Genetic improvements in alfalfa have been highly successful in developing cultivars with exceptional winter hardiness and disease resistance traits. However, genetic improvements have been limited for complex economically important traits such as biomass. One of the major bottlenecks is the labor-intensive phenotyping burden for biomass selection. In this study, we employed two alfalfa fields to pave a path to overcome the challenge by using UAV images with fully automatic field plot segmentation for high-throughput phenotyping. The first field was used to develop the prediction model and the second field to validate the predictions. The first and second fields had 808 and 1025 plots, respectively. The first field had three harvests with biomass measured in May, July, and September of 2019. The second had one harvest with biomass measured in September of 2019. These two fields were imaged one day before harvesting with a DJI Phantom 4 pro UAV carrying an additional Sentera multispectral camera. Alfalfa plot images were extracted by GRID software to quantify vegetative area based on the Normalized Difference Vegetation Index. The prediction model developed from the first field explained 50–70% (R Square) of biomass variation in the second field by incorporating four features from UAV images: vegetative area, plant height, Normalized Green–Red Difference Index, and Normalized Difference Red Edge Index. This result suggests that UAV-based, high-throughput phenotyping could be used to improve the efficiency of the biomass selection process in alfalfa breeding programs.

Powdery mildew disease significantly impacts crop productivity and financial losses in agriculture. Detecting and diagnosing the disease is challenging and requires expertise. Bell pepper production is particularly vulnerable, affecting both quality and quantity. To ensure effective treatment and control measures, the disease must be quickly identified and investigated. A comprehensive system has been developed to help farmers make informed decisions about crop management, improving overall quality and production. A smartphone app is developed to help farmers detect powdery mildew infection in bell pepper plants. Utilizing deep learning algorithms, the app processes photos through classification, feature extraction, and image segmentation. The program uses a powerful algorithm to identify illnesses, utilizing object detection, transfer learning, and Mask R-CNN to improve disease detection early. The smart polytunnel system, developed by this research, can identify and examine plant powdery mildew disease early on. This technology can improve agricultural yields and crop quality, enhancing sustainability and profitability. By enabling proactive and informed crop management decisions, the system contributes to a more robust and successful future for bell pepper growers and the agricultural sector.

Grape cluster architecture and compactness are complex traits influencing disease susceptibility, fruit quality, and yield. Evaluation methods for these traits include visual scoring, manual methodologies, and computer vision, with the latter being the most scalable approach. Most of the existing computer vision approaches for processing cluster images often rely on conventional segmentation or machine learning with extensive training and limited generalization. The Segment Anything Model (SAM), a novel foundation model trained on a massive image dataset, enables automated object segmentation without additional training. This study demonstrates out-of-the-box SAM’s high accuracy in identifying individual berries in 2-dimensional (2D) cluster images. Using this model, we managed to segment approximately 3,500 cluster images, generating over 150,000 berry masks, each linked with spatial coordinates within their clusters. The correlation between human-identified berries and SAM predictions was very strong (Pearson’s r2 = 0.96). Although the visible berry count in images typically underestimates the actual cluster berry count due to visibility issues, we demonstrated that this discrepancy could be adjusted using a linear regression model (adjusted R2 = 0.87). We emphasized the critical importance of the angle at which the cluster is imaged, noting its substantial effect on berry counts and architecture. We proposed different approaches in which berry location information facilitated the calculation of complex features related to cluster architecture and compactness. Finally, we discussed SAM’s potential integration into currently available pipelines for image generation and processing in vineyard conditions.

Imaging sensors (e.g., multispectral cameras) mounted on unmanned aerial systems (UAS) have emerged as a powerful tool for deriving insights about agricultural fields, from plant morphology phenotyping to plant disease monitoring. Advances in computer vision-based image analysis have enabled researchers to rapidly and accurately isolate crop spectra in UAS images. Specialty crops often employ unique production styles, such as trellising or inter-cropping. This presents a barrier to using existing image processing methodologies developed for broad-acre, row cropped systems (i.e. corn, wheat, soybean). Here, we present MAUI, a customizable image processing workflow built for specialty crops. Using a pathology research vineyard and hemp breeding trial as test cases, MAUI streamlines the generation of multispectral orthomosaic time-series, the segmentation of crops at the unit of research interest, and the extraction of crop spectra for downstream analysis. We successfully used MAUI to collect and analyze UAS data at two field sites over two growing seasons. Of the five canopy segmentation methods we tested, a supervised deep convolutional neural network (DeepLabv3) and a vision foundation model (SAM) produced the most accurate crop masks for the vineyard and hemp images, with mean intersection over union (mIoU) values of 0.85 and 0.95, respectively. Segmentation accuracy decreased when we applied each method to the other dataset, highlighting the importance of modular, flexible segmentation workflows for UAS imaging analysis in specialty crops. We present a modular framework to efficiently extract spectral data for specialty crops from UAS imagery. We highlight two kinds of segmentation applied to trellised and row cropping systems to demonstrate the modularity and versatility of the proposed methodology. MAUI improved spectral discrimination between individual plants and treatment groups for hemp and grapevine, respectively. With the containerized deployment package and open-source codebase, MAUI can be widely adopted by specialty crop researchers to facilitate the integration of UAS imagery analysis into routine research.

Grape cluster compactness is a key trait that influence fruit quality, yield, and disease susceptibility. Understanding the genetic basis of this trait is essential for optimizing vineyard management and improving grapevine cultivars. In this study, we performed quantitative trait locus (QTL) mapping to identify genomic regions associated with cluster architecture and yield components in a bi-parental population derived from Vitis vinifera cv. Riesling × Cabernet Sauvignon. A total of 138 full-sibling progeny were evaluated over two growing seasons at Oakville, Napa Valley, California. Traditional yield-related traits were measured, including cluster number, total cluster weight, and average cluster weight. Additionally, an image-based phenotyping pipeline leveraging the foundation model Segment Anything Model (SAM) was employed to segment individual berries, measure their size and shape, and compute cluster compactness with minimal manual intervention. Trait correlations revealed that compact clusters tended to have a higher berry count but smaller berry size, highlighting the role of compactness in modulating cluster structure. Heritability estimates varied across traits, with berry dimensions and compactness displaying moderate to high heritability, indicating strong genetic control. Two parental linkage maps were constructed using a pseudo-test cross strategy. QTL mapping identified multiple loci associated with cluster architecture and yield components, with several stable QTLs detected across both years. Notably, a QTL for cluster compactness was found in both seasons on chromosome 1 in Cabernet Sauvignon. Other stable QTLs were associated with berry size (chromosomes 6 and 17) and berry count (chromosome 5 in Cabernet Sauvignon and chromosome 7 in Riesling). Additional QTLs were detected in a single year, reflecting the influence of environmental variation. Our findings provide valuable insights into the application of foundation models requiring no prior training and minimal intervention for high-quality segmentation and enhance our understanding of the genetic architecture of cluster compactness and yield traits. The genomic regions identified in this study offer promising targets for breeding programs aimed at improving grape quality and disease resistance.

ABSTRACT In Asian countries, rice is the major agricultural product. Hence, the initial identification of plant disorder via the IoT (Internet of Things) offers to ignore the rice from critical disorders. To increase crop production, the measurements should be considered to completely destroy the rice plant disorders by an effective system. This paper proposed an Intelligent IoT-aided deep learning model for detecting rice blast fungal along with a hybrid heuristic algorithm. The proposed work encompasses with multiple stages that are explained as follows. At first, the required paddy images are gathered from online data resources. Next, the pre-processing of the collected images is made by adaptive mean filtering and contrast enhancement. Further, the adaptive thresholding and morphological operation are adopted for leaf segmentation purposes, where the threshold value is tuned by a Fitness-based Billiards-inspired Rat Swarm Optimizer (FBRSO). Consequently, from the segmented image, the Region of Interest (ROI) is cropped. Finally, the cropped ROI is subjected to the Optimized MobileNetv2 and Multiscale Residual Attention Network (OMMRAN), where it includes MobileNetV2 and Multiscale Residual Attention Network, in which some of the hyperparameters are tuned by FBRSO approach. The performance is validated and compared with other existing approaches for detecting rice diseases effectively.

Flavescence dorée (FD) is a grapevine disease caused by phytoplasmas and transmitted by leafhoppers that has been spreading in European vineyards despite significant efforts to control it. In this study, we aim to develop a model for the automatic detection of FD-like symptoms (which encompass other grapevine yellows symptoms). The concept is to detect likely FD-affected grapevines so that samples can be removed for FD laboratory identification, followed by uprooting if they test positive, all to be conducted quickly and without omission, thus avoiding further contamination in the fields. Developing FD-like symptoms detection models is not simple, as it requires dealing with the complexity of field conditions and FD symptoms’ expression. To address these challenges, we use deep learning, which has already been proven effective in similar contexts. More specifically, we train a Convolutional Neural Network on image patches, and convert it into a Fully Convolutional Network to perform inference. As a result, we obtain a coarse segmentation of the likely FD-affected areas while having only trained a classifier, which is less demanding in terms of annotations. We evaluate the performance of our model trained on a white grape variety, Chardonnay, across five other grape varieties with varying FD symptoms expressions. Of the two largest test datasets, the true positive rate for Chardonnay reaches 98.48% whereas for Ugni-Blanc it drops to 8.3%, underlining the need for a multi-varietal training dataset to capture the diversity of FD symptoms. To obtain more transparent results and to better understand the model’s sensitivity, we investigate its behavior using two visualization techniques, Guided Gradient-weighted Class Activation Mapping and the Uniform Manifold Approximation and Projection. Such techniques lead to a more comprehensive analysis with greater reliability, which is essential for in-field applications, and more broadly, for all applications impacting humans and the environment.

Large Language Models (LLM) are revolutionizing robotics. Vision-Language-Models (VLMs) and Vision-Language-Action (VLA) have gained wide acceptance in general robot applications. Recently, robots are built with the ability to convert text prompts to actions. In this study the efficient deep-learning model to build a foundational backbone for a farmland-based VLM for generalist robot was engaged, we evaluated the performance of Convolutional Neural Networks (CNN) models including ResNet34, ResNet50, ResNet101, and Vision Transformers (ViT) model. Three plants were selected (Soybean, Groundnut, and Tomato) and a plant disease Tomato Tuta Absoluta was considered for experimentation. The models can be further extended and fine-tuned using transfer learning. The pre-trained versions of these models were passed raw images of these plants for validation. The accuracy was 0% in all cases because ImageNet or CIFAR-10 were used for the pretraining instead of datasets oriented to the concerned disease detection. Transfer learning was then done on the pre-trained models to accommodate our plant dataset. We found that the training from pretrained models improved the performance. Each of the four models in consideration was evaluated to select the best performer on our plant dataset. Among the considered methods, ViT model performed best with 82.1% training accuracy and 72.9% validation accuracy. ViT model had a higher generalization capacity, which is a very important characteristic that is needed while building a VLM backbone that will be extended into a generalist robot VLA. The research drew motivation from OpenVLA, and future research is to extend OpenVLA with farmland scenarios.

Grape disease image recognition is an important part of agricultural disease detection. Accurately identifying diseases allows for timely prevention and control at an early stage, which plays a crucial role in reducing yield losses. This study addresses the problems in grape leaf disease recognition under small-sample conditions, such as the difficulty in capturing multi-scale features, the minuteness of features, and the weak adaptability of traditional data augmentation methods. It proposes a solution that combines a multi-scale feature hybrid fusion architecture with data augmentation. The innovation of this study lies in the following four dimensions: (1) Utilize generative models to enhance the cross-category data balancing ability under small-sample conditions and enrich the sample information in the dataset. (2) Innovatively propose the LVT Block, a multi-scale information perception hybrid module based on the Ghost and Transformer structures. This module can effectively acquire and fuse multi-scale information and global information in the feature map. (3) Use the dense connection method to combine the LVT Block and the MARI Block to propose a new architecture, the DLVT Block. By fusing multi-scale information and global information, it improves the richness of feature information. It also uses the MARI to enhance the model’s perception of disease areas and constructs an end-to-end lightweight model, DLVTNet, using the DLVT Block. Experiments show that this method achieves an average recognition rate of 98.48% on the New Plant Diseases Dataset. The number of parameters is reduced to 42.7% of that of MobileNetV4, and it maintains an accuracy of 96.12% in the tomato leaf disease test. This paper embeds pathological features into the generative adversarial process, which can effectively alleviate the problem of insufficient samples in intelligent agricultural detection. It provides a new method system with strong interpretability and excellent generalization performance for disease detection.

To address the challenges of high computational complexity and difficult deployment of existing deep learning models on mobile devices for grape leaf disease diagnosis, this paper proposes a lightweight image classification algorithm named MobileNet-GDR (Grape Disease Recognition), built upon the MobileNetV4-small architecture. The algorithm constructs an efficient feature extraction module based on depthwise separable convolutions and grouped convolutions to optimize the feature fusion process, while incorporating PReLU activation functions to enhance nonlinear representation capability. Experimental results on a grape leaf disease dataset demonstrate that MobileNet-GDR achieves high accuracy while significantly reducing computational overhead: with only 1.75M parameters and 0.18G FLOPs, it attains real-time inference speed of 184.89 FPS and a classification accuracy of 99.625%. Ablation studies validate the effectiveness of each module, and comparative experiments show that its computational efficiency surpasses mainstream lightweight models such as FasterNet and GhostNet. MobileNet-GDR provides a practical lightweight solution for real-time disease diagnosis in field conditions, demonstrating significant value for agricultural applications.

The latest improvements in computer vision formulated through deep learning have paved the method for how to detect and diagnose diseases in plants by using a camera to capture images as basis for recognizing several types of plant diseases. This study provides an efficient solution for detecting multiple diseases in several plant varieties. The system was designed to detect and recognize several plant varieties specifically apple, corn, grapes, potato, sugarcane, and tomato. The system can also detect several diseases of plants. Comprised of 35,000 images of healthy plant leaves and infected with the diseases, the researchers were able to train deep learning models to detect and recognize plant diseases and the absence these of diseases. The trained model has achieved an accuracy rate of 96.5% and the system was able to register up to 100% accuracy in detecting and recognizing the plant variety and the type of diseases the plant was infected.

Sugarcane is a vital crop worldwide and the main source of sugar and ethanol. One problem in the sugar industry is sugarcane diseases that leads in eradicating growing crops infested with the disease resulting in the financial loss of small-scale farmers if these diseases are not treated and detected early. With the fast-growing classes of diseases and inadequate know-how of farmers in identification and recognition of diseases was the motivation in conducting this study. Machine learning through computer vision using deep learning techniques provides a solution to solve this problem. This study trained and test a deep learning model consisting of 13,842 sugarcane image dataset of disease infected leaves and healthy leaves achieving an accuracy of 95%. The trained model achieved its purpose by detecting and classifying sugarcane images into healthy and unhealthy or diseased class of sugarcane leaves. Therefore, this paper provides an idea of helping farmers with the aid of deep learning algorithm in detecting and classifying sugarcane diseases.

This research explores the potential of leveraging computer vision and artificial intelligence (AI) to revolutionise crop disease detection and prevention specifically within smallholder farming systems. Smallholder farmers, who are critical to global food security, face significant challenges due to crop diseases that can lead to substantial yield losses. Traditional disease management methods are often inadequate, highlighting the urgent need for scalable, accurate, and timely solutions. This paper presents a conceptual framework for integrating AI-driven image recognition and data analytics to enable real-time disease detection and facilitate proactive prevention strategies tailored to the constraints of resource-limited smallholder farms. By examining existing methodologies, the applications of computer vision in agriculture, and current research gaps, this work outlines a system design, compares suitable AI models, and discusses crucial implementation considerations such as scalability, accessibility, and ethical implications. Ultimately, this paper envisions the transformative impact of AI in bolstering resilience against disease outbreaks, promoting sustainable farming practices, and ensuring global food security by empowering smallholder farmers with advanced technological tools.

Efficient livestock management is vital for enhancing agricultural productivity and ensuring animal welfare. Traditional cattle management methods are often time-consuming, error-prone, and dependent on manual supervision, leading to delayed disease detection and inconsistent feeding or vaccination schedules. This paper proposes an intelligent and automated cattle monitoring framework that integrates Convolutional Neural Networks (CNNs) with Inertial Measurement Unit (IMU) sensors for accurate activity recognition. The objective is to enhance behavioral analysis and health monitoring by leveraging data augmentation techniques to overcome the limitations of small and imbalanced datasets. The proposed methodology involves realtime data collection from accelerometers and gyroscopes, preprocessing and augmentation of motion signals, and CNNbased classification of cattle activities such as feeding, walking, and resting. Experimental results demonstrate a classification accuracy of 96.8 percent, outperforming conventional machine learning approaches. The findings highlight the system’s ability to operate efficiently in low-resource environments while maintaining high reliability and accuracy. This research aims to advance precision livestock farming by providing a scalable, low-cost, and data-driven solution that reduces manual intervention and supports proactive cattle health management. Keywords-CNN, Inertial Sensors, Data Augmentation, Cattle Management, IoT, Precision Livestock Farming.

This study, involving 120 poultry farmers (82.00% male, 18.00% female) raising layers (63.00%), broilers (35.00%), and mixed flocks (2.00%) with sizes from 200 to 100,000 birds, investigated the interplay between knowledge, attitudes, and practices (KAP) and the occurrence of Infectious Bronchitis (IB) in poultry. A critical observation was that vaccine administration, a key preventive measure, was often entrusted to farm attendants (45.00%) who may lack adequate understanding of biological material handling, potentially leading to vaccine failure and disease spread. While 45.00% of respondents limited their IB vaccination knowledge to laying birds, 55.00% understood its relevance to all bird types. Despite 95.00% acknowledging the economic impact of IB, 48.30% did not vaccinate against IB after 3-in-1 administration, and 23.30% had no vaccination history. Although all farmers kept medication records, only 25.00% consistently screened for maternal derived antibodies, a practice crucial for effective vaccination scheduling. Furthermore, a notable portion of farmers (15.00% strongly disagreed, 5.00% disagreed) lacked knowledge regarding the importance of priming birds before 3-in-1 vaccination. The findings show a critical need for increased awareness among poultry farmers, particularly concerning comprehensive vaccination protocols for all bird types, accurate disease recognition beyond clinical signs, and the significance of practices like maternal derived antibody screening to effectively mitigate IB and its economic consequences.

The poultry farming industry encounters considerable obstacles stemming from viral diseases, resulting in elevated mortality rates and substantial economic losses. Current research highlights the significant involvement of long non-coding RNAs (lncRNAs) in the interactions between hosts and pathogens by enhancing antiviral responses at different levels, such as the activation of pathogen recognition receptors, as well as through epigenetic, transcriptional, and post-transcriptional modifications. Specific long non-coding RNAs (lncRNAs), including ERL lncRNA, linc-GALMD3, and loc107051710, have been recognized as significant contributors to the antiviral immune response to multiple avian viral pathogens. Understanding the mechanisms by which long non-coding RNAs (lncRNAs) act offers valuable insights into prospective diagnostic and therapeutic approaches aimed at improving disease resistance in poultry. Differentially expressed lncRNAs may also be utilized as biomarkers for both prognosis and diagnosis of avian viral diseases. This review delves into the various roles of long non-coding RNAs (lncRNAs) in the context of viral diseases in chickens, such as avian leukosis, Marek’s disease, infectious bursal disease, avian influenza, infectious bronchitis, and Newcastle disease. It highlights the pivotal role of lncRNAs in the complex dynamics between the host and viral pathogens, particularly their interactions with specific viral proteins. Understanding these interactions may provide valuable insights into the spatial and temporal regulation of lncRNAs, aid in the identification of potential drug targets, and reveal the expression patterns of lncRNA and coding gene transcripts in response to different viral infections in avian species.

Bacterial Leaf Streak (BLS), caused by Xanthomonas oryzae pv. oryzicola, has emerged as a major threat to rice cultivation in Senegal, leading to yield losses of up to 30 %. Traditional agricultural extension methods remain limited in reach and interactivity, particularly in remote rural areas. This study proposes SEN-Agri, an innovative awareness platform designed to educate Senegalese rice farmers about BLS using Hybrid Broadcast Broadband TV (HbbTV) technology. The primary objective is to enhance farmers' capacity for early disease recognition and management through interactive and easily accessible television-based content. The proposed methodology involves the design and implementation of an HbbTV-based architecture integrating MySQL for persistent data management, Redis for real-time interactivity, and SRS for video streaming. The user interface, accessible via the red button on standard remote controls, provides farmers with localized educational videos, symptom illustrations, management recommendations, and diagnostic quizzes in local languages. Experimental validation confirms the system's technical feasibility, with an average interaction latency below 2 seconds and a high acceptance rate (82 %) among surveyed farmers. Qualitative feedback indicates improved understanding of disease symptoms and increased adoption of recommended phytosanitary practices. This research demonstrates the potential of interactive television as a scalable medium for agricultural knowledge transfer in low-connectivity regions. By leveraging existing broadcast infrastructures and open HbbTV standards, SEN-Agri contributes to reducing the digital divide in rural Africa, offering a replicable model for other crops and plant health challenges.

This study was conducted to address the demand for interpretable intelligent recognition of fruit tree diseases in smart horticultural environments. A KAD-Former framework integrating an agricultural knowledge graph with a visual Transformer was proposed and systematically validated through extensive cross-regional, multi-variety, and multi-disease experiments. The primary objective of this work was to overcome the limitations of conventional deep models, including insufficient interpretability, unstable recognition of weak disease features, and poor cross-regional generalization. In the experimental evaluation, the model achieved significant advantages across multiple representative tasks: in the overall performance comparison, KAD-Former reached an accuracy of 0.946, an F1-score of 0.933, and a mAP of 0.938, outperforming classical models such as ResNet50, EfficientNet, and Swin-T. In the cross-regional generalization assessment, a DGS of 0.933 was obtained, notably surpassing competing models. In terms of explainability consistency, a Consistency@5 score of 0.826 indicated strong alignment between the model’s attention regions and expert annotations. The ablation experiments further demonstrated that the three core modules—AKG (agricultural knowledge graph), SAM (semantic alignment module), and KGA (knowledge-guided attention)—each contributed substantially to final performance, with the complete model exhibiting the best results. These findings collectively demonstrate the comprehensive advantages of KAD-Former in disease classification, symptom localization, model interpretability, and cross-domain transfer. The proposed method not only achieved state-of-the-art performance in pure visual tasks but also advanced knowledge-enhanced and interpretable reasoning by emulating the diagnostic logic employed by agricultural experts in real orchard scenarios. Through the integration of the agricultural knowledge graph, semantic alignment, and knowledge-guided attention, the model maintained stable performance under challenging conditions such as complex illumination, background noise, and weak lesion features, while exhibiting strong robustness in cross-region and cross-variety transfer tests. Furthermore, the experimental results indicated that the approach enhanced fine-grained recognition capabilities for various fruit tree diseases, including apple ring rot, brown spot, powdery mildew, and downy mildew.

Stevia rebaudiana Bertoni, a herbaceous perennial known for its natural sweetness, has garnered global recognition and is found in various regions of India. A study was carried out to investigate the leaf spot disease in Stevia, caused by Alternaria alternata (FR.) Keissler to find out the most suitable design of dual culture technique. Since the leaves are the primary site for synthesizing sweet glycosides, in Stevia this disease leads to significant losses and ultimately reduces the yield which leads to a serious concern. Due to the harmful effects of chemical fungicides, finding a safer alternative to control the pathogen became a priority. This prompted experiments with bioagents for pathogen control. Bio fungicides derived from Trichoderma are increasingly being recognized as successful agricultural applications, with over 50 registered products available worldwide. The present study was conducted in the Laboratory, Department of Plant Pathology, Naini Agricultural Institute, Sam Higginbottom University of Agriculture, Technology and Sciences, Prayagraj (Uttar Pradesh). The dual culture technique was carried out on Completely Randomized Design (CRD) with three replications and six treatments. Examination of fungal colony characteristics was done through microscopic examination. At 7 DAI, maximum mycelial inhibition of 96.70% was recorded in the treatment T6 (six discs of Trichoderma harzianum against one disc of Alternaria alternata).

Agriculture in Nigeria is a branch of its economy providing employment for over 70% of its population and contributing about 41% to it gross domestic production (GDP). Nigeria’s wide range of climate variations allows it to produce a variety of food and cash crops. Cucumber/watermelon is among the few varieties of fruits and vegetables produced in the country. The need for steady monitoring of the plants is necessary to detect and control the spread of its diseases. Common practices for detecting these plants diseases are mostly based on direct observation of the affected plant, which are sometimes erroneous. Laboratory analysis can also be used for plant diseases detection but mostly costly and time consuming. Digital image processing can identify and grade the diseases in cucumber/watermelon. This will aid in describing and predicting the performance of the said cultivated crops, hence increasing the production yield. Although, official disease recognition is a responsibility of professional agriculturists, low-cost observation and computational assisted diagnosis can effectively help in the recognition of a plant disease in its early stages. The application software was designed based on Object Oriented System Analysis and Design Methodology while UML was used to model the system; MATLAB was used in designing the front-end and the database is MYSQL Server. The developed system works efficiently and can successfully detect and classify the specified diseases in cucumber/watermelon with a precision between 92% and 98%.

Background and Aim Ticks and tick-borne diseases (TBDs) remain a major constraint to cattle production, responsible for substantial economic losses through reduced productivity, increased treatment costs, and high mortality. Beyond livestock impacts, ticks transmit a range of zoonotic pathogens, posing significant health risks to communities in close contact with cattle. Despite Borno State having the largest cattle population in Nigeria, there is no prior documentation of cattle farmers' knowledge, attitudes, and practices (KAP) regarding ticks, TBDs, and their zoonotic implications. This study aimed to assess farmers' awareness, preventive behaviors, and tick-control strategies, while evaluating the influence of formal and informal education on these variables. Materials and Methods A cross-sectional descriptive KAP survey was conducted among 492 cattle farmers across Maiduguri Metropolitan Council and Jere Local Government Area between November 2024 and February 2025. Data were collected using a semi-structured and pre-validated questionnaire translated into local languages. Descriptive statistics summarized KAP, while Chi-square tests assessed associations between education and key outcome variables (significance level: p ≤ 0.05). Results Most farmers (77.2%) reported observing ticks on their cattle, and 82.9% recognized their role in livestock disease transmission. Tick occurrence was highest during the rainy season (83.7%). Although awareness of livestock TBDs was high, more than half (54.4%) were unaware that ticks transmit diseases to humans. A large proportion (59.8%) reported previous tick bites, but only 10.2% sought medical care afterward. Combined control through acaricides and handpicking was the predominant practice (78.9%). Significant differences between formally and informally educated farmers were observed for lesion recognition after tick bites (χ² = 128.04; p ≤ 0.001), tick-control method (χ² = 26.30; p ≤ 0.001), frequency of handpicking (χ² = 44.27; p ≤ 0.001), and acaricide application methods (χ² = 57.45; p ≤ 0.001). Conclusion Farmers demonstrated good knowledge of ticks and livestock TBDs but exhibited low awareness of zoonotic risks and poor health-seeking behavior following tick bites. Strengthening public health education, promoting protective practices, and integrating zoonotic TBDs into One Health policies are essential to reducing risks among high-exposure populations.

Xanthomonas species, Pseudomonas syringae and Ralstonia species are bacterial plant pathogens that cause significant yield loss in many crop species. Generating disease-resistant crop varieties can provide a more sustainable solution to control yield loss compared to chemical methods. Plant immune receptors encoded by nucleotide−binding, leucine−rich repeat (NLR) genes typically confer resistance to pathogens that produce a cognate elicitor, often an effector protein secreted by the pathogen to promote virulence. The diverse sequence and presence/absence variation of pathogen effector proteins within and between pathogen species usually limits the utility of a single NLR gene to protecting a plant from a single pathogen species or particular strains. The NLR protein Recognition of XopQ 1 (Roq1) was recently identified from the plant Nicotiana benthamiana and mediates perception of the effector proteins XopQ and HopQ1 from Xanthomonas and P. syringae respectively. Unlike most recognized effectors, alleles of XopQ/HopQ1 are highly conserved and present in most plant pathogenic strains of Xanthomonas and P. syringae. A homolog of XopQ/HopQ1, named RipB, is present in most Ralstonia strains. We found that Roq1 confers immunity to Xanthomonas, P. syringae, and Ralstonia when expressed in tomato. Strong resistance to Xanthomonas perforans was observed in three seasons of field trials with both natural and artificial inoculation. The Roq1 gene can therefore be used to provide safe, economical, and effective control of these pathogens in tomato and other crop species and reduce or eliminate the need for traditional chemical controls.

Sugarcane disease causes a threat in the sugarcane industry that leads to massive demolition of infected crops, decreases cultivation and financial loss of farmers. Early detection of the diseases using machine learning technology can avoid such loss and disaster. Subsequently, deep learning is an interesting technology to solve this problem through machine learning. CNN has proven to be a popular technique in recognizing and detecting plant diseases through deep learning. This paper aims to integrate various CNN architectures of deep learning models to achieve the highest accuracy rate in detecting and recognizing sugarcane diseases. The models were trained using 14,725 images of a healthy sugarcane leaves and infested sugarcane diseases and achieves a maximum 95.40% accuracy rate during training. Three architectures of CNN; StridedNet, LeNet, and VGGNet were used in the conduct of this study. VGGNet model tops the other two models achieving an accuracy rate of 95.40%, LeNet model achieves the rate of 93.65%, while StridedNet model achieves 90.10% accuracy rate during training. The three models trained serve its purpose by classifying images of a healthy sugarcane leaves and diseased infested sugarcane leaves based on the leaf features. This paper also bids a successful result for the researchers who are in the development of recognition system for sugarcane diseases.

No abstract available

The agricultural industry plays a pivotal role in supporting global food security and sustaining economies. Within this industry, cattle farming represent a significant sector, providing essential resources like meat, milk, and other dairy products. However, traditional cattle farming practices often face challenges related to monitoring, managing, and predicting various aspects of cattle health and productivity. In recent years, the advent of technology has opened up new possibilities for transforming traditional farming practices into more efficient, data-driven systems. The integration of smart devices, IoT sensors, machine learning algorithms, and real-time data analytics has paved the way for innovative solutions that can address the limitations of conventional cattle farming. The integrated embedded system proposed in this paper aims to revolutionize cattle farming practices by providing a comprehensive solution to enhance cattle well-being and optimize farming efficiency. It encompasses four crucial areas: real-time monitoring and health management, milk production prediction, artificial insemination scheduling, and disease analysis with first aid recommendations. By utilizing cutting-edge IoT, sensors, machine learning algorithms, and image recognition techniques the system enables farmers to monitor cattle health in real-time, predict milk production accurately, schedule artificial insemination effectively, and promptly identify and manage cattle skin diseases. Overall, the system has archived 91% high accuracy through these advancements, also empowers farmers to make data-driven decisions, ensuring proactive measures to prevent health issues, enhance productivity, and fosters a sustainable and profitable agriculture industry.

The extravasation of T cells at sites of inflammation is critically dependent on the activity of homing receptors (HR) involved in endothelial cell recognition and binding. Two such HR (the cutaneous lymphocyte antigen [CLA] and L-selectin) have been shown to be selectively involved in T cell migration to skin and peripheral lymph nodes, respectively. This study was designed to assess the relationship between the organ specificity of an allergic reaction to food and the expression of HR on T cells activated in vitro by the relevant food allergen. Peripheral blood mononuclear cells were isolated from seven milk allergic children with a history of eczema when exposed to milk. All patients had a positive prick skin test and double-blind placebo-controlled food challenge to milk. 10 children with either allergic eosinophilic gastroenteritis or milk-induced enterocolitis and 8 nonatopic adults served as controls. Five-parameter flow cytometry using monoclonal antibodies was used for detection of the specific HR on freshly isolated T cells versus T cell blasts induced by a 6-d incubation with casein, as compared with Candida albicans. After in vitro stimulation with casein, but not C. albicans, patients with milk allergy and atopic dermatitis had a significantly greater percentage of CLA+ T cells (P < 0.01) than controls with milk-induced enterocolitis, allergic eosinophilic gastroenteritis, or nonatopic healthy controls. In contrast, the percentage of L-selectin-expressing T cells did not differ significantly between these groups. These data suggest that after casein stimulation allergic patients with milk-induced skin disease have an expanded population of CLA+ T cells, as compared with nonatopics or allergic patients without skin involvement. We postulate that heterogeneity in the regulation of HR expression on antigen-specific T cells may play a role in determining sites of involvement in tissue-directed allergic responses.

A key step in characterization of germplasm is the identification of phenotypic variation present in a given population. A study was carried out to determine the effect of different dosages of gamma rays (50 and 100Gy) on phenotypic variation using 21 standardized morphological descriptors of the UPOV Tea Test Guidelines. The trial comprised of open-pollinated seed stocks from six commercial tea cultivars namely TRFCA SFS150, TRFK 303/1199, EPK C12, GW Ejulu-L, TRFK 301/1 and TRFK 301/4 along with untreated controls. Data was collected for three seasons (dry, warm wet and cold wet) using five randomly selected plants from each treatment. Principle Component Analysis using 17 informative descriptors showed the first eight principal components accounted for 78% of the total variance, with 15 being highly informative. Cluster analysis further identified characters such as young shoot anthocyanin colouration at base of the petiole, leaf blade shape/color/length, shoot color/length, density of pubescence, plant vigour and density of branches as most discriminating descriptors resulting in four phenotypically well-defined groups. Most traits showed significant correlation, an indication that the traits could be used for indirect selection. The study provides a basis for rapid and early screening of base populations for identification of elite cultivars.