基于深度学习与微服务的车牌识别系统设计与实现

深度神经网络(Deep neural network, DNN)作为最常用的深度学习方法之一, 广泛应用于各个领域. 然而, DNN容易受到对抗攻击的威胁, 因此通过对抗攻击来检测应用系统中DNN的漏洞至关重要. 针对车牌识别系统进行漏洞检测, 在完全未知模型内部结构信息的前提下展开黑盒攻击, 发现商用车牌识别系统存在安全漏洞. 提出基于精英策略的非支配排序遗传算法(NSGA-II)的车牌识别黑盒攻击方法, 仅获得输出类标及对应置信度, 即可产生对环境变化较为鲁棒的对抗样本, 而且该算法将扰动控制为纯黑色块, 可用淤泥块代替, 具有较强的迷惑性. 为验证本方法在真实场景的攻击可复现性, 分别在实验室和真实环境中对车牌识别系统展开攻击, 并且将对抗样本用于开源的商业软件中进行测试, 验证了攻击的迁移性.

在车牌字符识别的某些场合中，获得的字符通常存在切割不均匀、光照对比度强烈、遮挡严重等强噪声污染。针对被强噪声污染的数字字符，本文提出基于Caffe深度学习框架的字符识别算法，在Caffe框架下搭建卷积神经网络，并对网络参数训练获得了一个鲁棒性强、识别精度高的网络结构。实验结果表明，在低噪声、中度噪声、强噪声污染情况下，文章中提出的方法相比当前典型的识别方法，在数字字符识别上均具有较好的识别能力，平均识别率高出将近5%，而在强噪声污染情况下，识别效果具有更加明显的优势。

针对综合交通枢纽的停车场车位引导难、停车拥堵、寻车难等问题，提出了一种基于三维全景建模、车辆识别、室内导航等多技术融合的智能停车系统。该系统通过车辆识别技术获取车辆品牌、车牌号码等信息，通过三维全景建模和室内导航技术实现停车场实景导航，做到快速停车、便捷寻车，提升用户体验。

为提高煤炭发运及装车系统的自动化与精准化，满足智慧矿山建设需求，针对东滩煤矿系统的人工依赖、装车精度差及数据孤岛等问题，实施了智能化升级。改造采用AI算法、激光雷达点云建模和PLC控制技术，优化门岗管理，升级地磅系统，实现装车全自动控制。通过IC卡、RFID验证、LED屏和语音引导，增强车辆定位、车型识别和车厢检测等功能，并与SAP系统对接，实时共享订单、车辆、称重数据。改造后，单车煤炭装车效率提升33%，年发运总量增长约30%，撒料率由0.3%降至0.1%，人员减少50%，数据传输准确率超过99%，粉尘排放减少约30%，噪声降低约20%。该系统实现了煤炭发运与装车的智能化、信息化与无人化升级，为煤矿企业提供了降本增效的保障。

深层卷积神经网络(deep convolutional neural networks, DCNN)因其能够自动学习图像有效特征,被广泛应用于视觉目标检测.为克服DCNN目标检测算法大多因采用矩形检测框,而无法有效地应对非约束环境下倾斜性车牌的准确定位问题.提出一种可同时输出矩形目标检测框与关键点的车牌定位解决方案,并具体以YOLOv3所用网络为对象,通过扩展其输出维度,增设车牌顶点相对于矩形检测输出框角点的偏移量损失,在保留其高效计算性能的前提下,训练使其可同时输出矩形检测框及车牌顶点,实现精准定位.在广泛使用的大型非约束性车牌数据集CCPD上的实验结果显示,所提算法不仅可以准确检测车牌顶点,而且能够在Base,Tilt和Weather子集上取得99%以上的定位精度.该方法还可扩展至其他需同时输出目标检测框及关键点的应用领域,具有较好的应用价值.

… using machine learning approach. In [6], a plate recognition system using deep learning … They developed an OCR system with a customized dataset. The dataset was made artificially by …

… Thus, vehicle number plate recognition is an … deep learning which forms the basis of this project[10].This is all about the existing system of the project“Automatic Vehicle Number Plate …

… In law and judiciary conditions relating to license plate detection and … number plate detection technology used [4]. The Python programming language is employed throughout the system…

: The latest advancements in highway research domain and increase in the number of vehicles everyday led to wider exposure and attentiontowards the development of efficient Intelligent Transportation System (ITS). One of the popular research areas i.e., Vehicle License Plate Recognition (VLPR) aims at determining the characters that exist in the license plate of the vehicles. The VLPR process is a difficult one due to the differences in viewpoint, shapes, colors, patterns, and non-uniform illuminationat the time of capturing images. The current study develops a robust Deep Learning (DL)-based VLPR model using Squirrel Search Algorithm (SSA)-based Convolutional Neural Network (CNN), called the SSA-CNN model. The presented technique has a total of four major processes namely preprocessing, License Plate (LP) localization and detection, character segmentation, and recognition. Hough Transform (HT) is applied as a feature extractor and SSA-CNN algorithm is applied for character recognition in LP. The SSA-CNN method effectively recognizes the characters that exist in the segmented image by optimal tuning of CNN parameters. The HT-SSA-CNN model was experimentally validated using the Stanford Car, FZU Car, and HumAIn 2019 Challenge datasets. The experimentation outcome verified that the presented method was better under several aspects. The projected HT-SSA-CNN model implied the best performance with optimal overall accuracy of 0.983%.

In this paper, we present an efficient and layout-independent Automatic License Plate Recognition (ALPR) system based on the state-of-the-art YOLO object detector that contains a unified approach for license plate (LP) detection and layout classification to improve the recognition results using post-processing rules. The system is conceived by evaluating and optimizing different models with various modifications, aiming at achieving the best speed/accuracy trade-off at each stage. The networks are trained using images from several datasets, with the addition of various data augmentation techniques, so that they are robust under different conditions. The proposed system achieved an average end-to-end recognition rate of 96.8% across eight public datasets (from five different regions) used in the experiments, outperforming both previous works and commercial systems in the ChineseLP, OpenALPR-EU, SSIG-SegPlate and UFPR-ALPR datasets. In the other datasets, the proposed approach achieved competitive results to those attained by the baselines. Our system also achieved impressive frames per second (FPS) rates on a high-end GPU, being able to perform in real time even when there are four vehicles in the scene. An additional contribution is that we manually labeled 38,334 bounding boxes on 6,237 images from public datasets and made the annotations publicly available to the research community.

Detection of vehicle license plates have many impor-tant applications including identification of traffic rules breaking vehicles, entry-exit restriction in secured places to many more. This paper presents a simple yet effective architecture for automatic license plate detection and recognition for secured places. The proposed architecture has three main parts - a) license plate detection, b) license plate segmentation where we segment the characters of the License Plate into constituents digits and letters, and c) character recognition, which uses Convolution Neural Network (CNN) classifier to recognize the segmented digits.

In this paper, a real-time vision pipeline for automatically identifying the corresponding license plates and detecting motorcycle helmet violations is presented. Four object classes—helmeted head, unhelmeted head, motorbike, and number plate—are localized in a single pass using a single YOLOv8seg network. The algorithm searches the spatially associated motorcycle region for each head that is detected without a helmet, crops the detected number plate, uses RealESRGAN to super-resolve the crop four times, and then applies PaddleOCR and a country-specific regular expression filter to produce a clean alphanumeric plate string. An annotated frame, raw and super-resolved plate crops, and the verified plate ID make up the final record, which is admissible evidence for automated traffic enforcement systems. After super-resolution, evaluation on a public image dataset shows high localization accuracy (mAP50 85%) and plate-reading recall above 90%, highlighting the pipeline’s suitability as an addition to the current CCTV infrastructure.

Abstract Nowadays, detection of license plate (LP) for non-helmeted motorcyclist has become mandatory to ensure the safety of the motorcyclists. This paper presents the real-time detection of LP for non-helmeted motorcyclist using the real-time object detector YOLO (You Only Look Once). In this proposed approach, a single convolutional neural network was deployed to automatically detect the LP of a non-helmeted motorcyclist from the video stream. The centroid tracking method with a horizontal reference line was used to eliminate the false positive generated by the helmeted motorcyclist as they leave the video frames. The overall LP detection rate was 98.52%.

The project aims to determine the most effective EasyOCR approach for real-time car license plate recognition, focusing on speed, accuracy, and the architecture’s capacity to accurately record license plates. The vehicle license plate recognition system employs Easy Optical Character Recognition (EasyOCR) for real-time processing. The procedure commences with the input image, thereafter undergoing image preprocessing with the FuzzyWuzzy or SymSpell approach to improve image quality. Subsequently, license plate detection is executed to identify the license plate inside the image. The subsequent phase is license plate preprocessing, which aims to prepare the license plate for character processing. Afterwards, Easy Optical Character Recognition is executed to identify the characters on the observed license plate. Text formatting is executed to enhance the readability of character recognition outcomes, culminating in the system’s real-time output, which facilitates immediate and efficient car license plate recognition. The experimental findings indicate that the five OCR optimization techniques demonstrate varying efficacy in recognizing car license plates. FuzzyWuzzy and SymSpelL achieved optimal outcomes with recall and precision rates of 87.71% and 85.71%, respectively, alongside an accuracy of 85.72%. Both techniques are highly useful for real-time applications where rapidity and precision are essential. Local Binary Patterns (LBP) performed well, matching the recall and precision rates of 85.71% and an accuracy of 85.72%, similar to FuzzyWuzzy and SymSpelL, but it had a better confidence score of 70.71%, showing it is more reliable in detection.

Vehicle tracking and license plate recognition (LPR) over video surveillance cameras are essential intelligent traffic monitoring systems. Due to the enormous amount of data collected each day, it would be difficult to track vehicles by license plate in a real-world traffic setting. Large volumes of data processing, real-time request responses, and emergency scenario response may not be possible using conventional approaches. By combining license plate recognition with the docker container-based structure of the Apache Kafka node ecosystem, the suggested solution takes a novel approach to vehicle tracking. The primary components of our suggested framework for reading license plates are the identification of license plates and text data queries. License plate localization is performed with You Only Look Once version 3 (YOLOv3) and character recognition with Optical Character Recognition (OCR). The detected vehicle images with license plate results are published on related topics with Apache Kafka. Apache Kafka is a publish-subscribe (producers-consumers) messaging system and one of the most popular architectures used for streaming data. For each license plate search, a topic will be created in the framework where producers publish and consumers receive data. Thus, the workload of the operators will be reduced and they will be able to pay attention to more important events in traffic.

… The system’s layered, microservices-based architecture ensures … The foundation of our license plate detection and tracking … for license plate recognition and vehicle identification. …

A study was carried out in the field of building a flexible and scalable architecture with support for fast vertical and horizontal scaling. System was developed using microservice architecture. A list of requirements for building an architecture has been developed. Key services and tools for architecture deployment are described. Proposed system allows to detect a car in a traffic flow, to recognize license plate on different angles, to track cars, that are in the in the field of view of cameras. System stores all received information about cars and their tracks and provides fast access to stored data for making some statistical calculations.

This paper presents the design and implementation of an automated traffic monitoring and violation reporting system integrating advanced computer vision and machine learning technologies. The system unifies vehicle detection based on convolutional neural networks, optical character recognition for license plate reading, and real-time speed estimation into a scalable, efficient, and autonomous platform. The architecture incorporates multiple image preprocessing steps to enhance recognition accuracy and employs a multi-threaded design for efficient parallel processing of video streams. A key innovation is the homography estimation of camera parameters to adapt to camera motion, supported by a top-down projection method for precise distance and speed estimation. Reliable communication and automated violation reporting are achieved through a distributed messaging protocol that ensures scalable data transfer and automatic alert delivery to authorities. The comprehensive integration of these components into a single cohesive system reduces complexity and enhances maintainability, providing an effective solution for intelligent traffic surveillance and enforcement within modern transportation infrastructure.

The UFC Smart Campus project applies IoT and computer vision technologies to enhance campus security through an advanced multi-layered image processing system. Using computer vision and parallel computing, the research develops innovative applications for license plate recognition and bus stop monitoring, achieving high accuracy rates. The system demonstrates a scalable approach to intelligent campus infrastructure, processing multiple camera streams simultaneously and highlighting the potential of cutting-edge AI technologies in academic environments.

… on real-time circumstances, enhancing intersection efficiency. … , LPR systems, when integrated with robust databases and … A microservices architecture further enhances flexibility and …

AI/ML techniques have been used to solve systems problems, but their applicability to customize solutions on-the-fly has been limited. Traditionally, any customization required manually changing the AI/ML model or modifying the code, configuration parameters, application settings, etc. This incurs too much time and effort, and is very painful. In this paper, we propose a novel technique using Generative Artificial Intelligence (GenAI) technology, wherein instructions can be provided in natural language and actual code to handle any customization is automatically generated, integrated and applied on-the-fly. Such capability is extremely powerful since it makes customization of application settings or solution techniques super easy. Specifically, we propose ECO-LLM (LLM-based Edge Cloud Optimization), which leverages Large Language Models (LLM) to dynamically adjust placement of application tasks across edge and cloud computing tiers, in response to changes in application workload, such that insights are delivered quickly with low cost of operation (systems problem). Our experiments with real-world video analytics applications i.e. face recognition, human attributes detection and license plate recognition show that ECO-LLM is able to automatically generate code on-the-fly and adapt placement of application tasks across edge and cloud computing tiers. We note that the trigger workload (to switch between edge and cloud) for ECO-LLM is exactly the same as the baseline (manual) and actual placement performed by ECO-LLM is only slightly different i.e. on average (across 2 days) only 1.45% difference in human attributes detection and face recognition, and 1.11% difference in license plate recognition. Although we tackle this specific systems problem in this paper, our proposed GenAI-based technique is applicable to solve other systems problems too.