Improved YOLOv8n Model for Detecting Helmets and License Plates on Electric Bicycles

Wearing helmetswhile riding electric bicycles can significantly reduce head injuries resulting fromtraffic accidents.To effectively monitor compliance,the utilization of target detection algorithms through traffic cameras plays a vital role in identifying helmet usage by electric bicycle riders and recognizing license plates on electric bicycles.However,manual enforcement by traffic police is time-consuming and labor-intensive.Traditional methods face challenges in accurately identifying small targets such as helmets and license plates using deep learning techniques.This paper proposes an enhanced model for detecting helmets and license plates on electric bicycles,addressing these challenges.The proposedmodel improves uponYOLOv8n by deepening the network structure,incorporating weighted connections,and introducing lightweight convolutional modules.These modifications aim to enhance the precision of small target recognition while reducing the model’s parameters,making it suitable for deployment on low-performance devices in real traffic scenarios.Experimental results demonstrate that the model achieves an mAP@0.5 of 91.8%,showing an 11.5%improvement over the baselinemodel,with a 16.2%reduction in parameters.Additionally,themodel achieves a frames per second(FPS)rate of 58,meeting the accuracy and speed requirements for detection in actual traffic scenarios.

An Efficient Method for Vehicle License Plate Detection in Complex Scenes

In this paper, we propose an efficient method for license plate localization in the images with various situations and complex background. At the first, in order to reduce problems such as low quality and low contrast in the vehicle images, image contrast is enhanced by the two different methods and the best for following is selected. At the second part, vertical edges of the enhanced image are extracted by sobel mask. Then the most of the noise and background edges are removed by an effective algorithm. The output of this stage is given to a morphological filtering to extract the candidate regions and finally we use several geometrical features such as area of the regions, aspect ratio and edge density to eliminate the non-plate regions and segment the plate from the input car image. This method is performed on some real images that have been captured at the different imaging conditions. The appropriate experimental results show that our proposed method is nearly independent to environmental conditions such as lightening, camera angles and camera distance from the automobile, and license plate rotation.

Multi-Object Detection of Chinese License Plate in Complex Scenes

Multi-license plate detection in complex scenes is still a challenging task because of multiple vehicle license plates with different sizes and classes in the images having complex background.The edge features of high-density distribution and the high curvature features of stroke turning of Chinese character are important signs to distinguish Chinese license plate from other objects.To accurately detect multiple vehicle license plates with different sizes and classes in complex scenes,a multi-object detection of Chinese license plate method based on improved YOLOv3 network was proposed in this research.The improvements include replacing the residual block of the YOLOv3 backbone network with the Inception-ResNet-A block,imbedding the SPP block into the detection network,cutting the redundant Inception-ResNet-A block to suit for the multi-license plate detection task,and clustering the ground truth boxes of license plates to obtain a new set of anchor boxes.A Chinese vehicle license plate image dataset was built for training and testing the improved network,and the location and class of the license plates in each image were accurately labeled.The dataset has 62,153 pieces of images and 4 classes of China vehicle license plates,almost images have multiple license plates with different sizes.Experiments demonstrated that the multilicense plate detection method obtained 83.4%mAP,98.88%precision,98.17%recall,98.52 F1 score,89.196 BFLOPS and 22 FPS on the test dataset,and whole performance was better than the other five compared networks including YOLOv3,SSD,Faster-RCNN,EfficientDet and RetinaNet.

License Plate Recognition via Attention Mechanism

License plate recognition technology use widely in intelligent trafficmanagement and control. Researchers have been committed to improving thespeed and accuracy of license plate recognition for nearly 30 years. This paperis the first to propose combining the attention mechanism with YOLO-v5and LPRnet to construct a new license plate recognition model (LPR-CBAMNet).Through the attention mechanism CBAM(Convolutional Block AttentionModule), the importance of different feature channels in license platerecognition can be re-calibrated to obtain proper attention to features. Forceinformation to achieve the purpose of improving recognition speed andaccuracy. Experimental results show that the model construction methodis superior in speed and accuracy to traditional license plate recognitionalgorithms. The accuracy of the recognition model of the CBAM model isincreased by two percentage points to 97.2%, and the size of the constructedmodel is only 1.8 M, which can meet the requirements of real-time executionof embedded low-power devices. The codes for training and evaluating LPRCBAM-Net are available under the open-source MIT License at: https://github.com/To2rk/LPR-CBAM-Net.

Vehicle Plate Number Localization Using Memetic Algorithms and Convolutional Neural Networks

This paper introduces the third enhanced version of a genetic algorithm-based technique to allow fast and accurate detection of vehicle plate numbers(VPLN)in challenging image datasets.Since binarization of the input image is the most important and difficult step in the detection of VPLN,a hybrid technique is introduced that fuses the outputs of three fast techniques into a pool of connected components objects(CCO)and hence enriches the solution space with more solution candidates.Due to the combination of the outputs of the three binarization techniques,many CCOs are produced into the output pool from which one or more sequences are to be selected as candidate solutions.The pool is filtered and submitted to a new memetic algorithm to select the best fit sequence of CCOs based on an objective distance between the tested sequence and the defined geometrical relationship matrix that represents the layout of the VPLN symbols inside the concerned plate prototype.Using any of the previous versions will give moderate results but with very low speed.Hence,a new local search is added as a memetic operator to increase the fitness of the best chromosomes based on the linear arrangement of the license plate symbols.The memetic operator speeds up the convergence to the best solution and hence compensates for the overhead of the used hybrid binarization techniques and allows for real-time detection especially after using GPUs in implementing most of the used techniques.Also,a deep convolutional network is used to detect false positives to prevent fake detection of non-plate text or similar patterns.Various image samples with a wide range of scale,orientation,and illumination conditions have been experimented with to verify the effect of the new improvements.Encouraging results with 97.55%detection precision have been reported using the recent challenging public Chinese City Parking Dataset(CCPD)outperforming the author of the dataset by 3.05%and the state-of-the-art technique by 1.45%.

基于顶点与主体区域同步检测的精准车牌定位

为应对非约束环境下的车牌精定位问题,提出一种基于顶点局部区域与主体区域同步检测策略的非约束性车牌定位算法。通过删减YOLOv5网络的输出结构,训练得到可同步检测车牌及顶点区域的车牌检测网络,在兼顾精度与计算速度的前提下,实现车牌顶点和主体区域的同步定位。针对一幅图中存在多个车牌区域及顶点区域存在少量漏检和误检的情况,分别设计了车牌顶点归类和单一缺失顶点预测后处理算法,借助顶点间的空间位置关系进行漏检目标预测和误检目标排查,有效改善了因场景复杂导致的个别顶点目标检测效果差的问题。所提算法在中国城市停车场数据集(CCPD)上的测试结果显示,平均精准率达99.25%,平均召回率达98.70%。所提算法不仅能够准确预测出车牌的4个顶点坐标,而且在中端GPU硬件平台上处理速度可达121帧/s,具有较好的应用价值。

Attention U-Net with Multilevel Fusion for License Plate Detection

In recent years,license plate recognition system(LPRS)is widely used in various places.Fast and accurate license plate detection is the first and critical step in LPRS.In order to improve the performance of license plate detection in complex environment,we propose a novel attention U-net with multilevel fusion(AUMF).At first,input images are fed to the network.Then,the feature maps of each level are generated by convolution operations of the original images.Before the feature connection,there are multi-layer splicing and convolution to detect more features.The attention mechanisms are used to retain the information of important regions.In order to ensure that the size of the input and output images are the same,down-sampling and up-sampling are employed to resize the feature mappings between the upper and lower levels.In the complex environment,the AUMF can accurately detect the license plate.To validate the effectiveness of the proposed method,we conducted a series of experiments on the AOLP dataset.The experimental results show that our approach effectively improves the performance of license plate detection under the three different license plate environments of AOLP dataset.

雾霾环境下基于PLATE-YOLO的车牌检测方法

针对目前车牌识别领域中,雾霾环境下车牌检测准确率低的问题,本文提出一种基于深度学习的抗雾霾车牌检测方法,该方法能够检测民用车牌和机场民航车辆车牌。该方法首先利用一种基于卷积神经网络的去雾算法对车牌图片进行去雾预处理,然后将处理过的无雾霾图片送入PLATE-YOLO网络中检测车牌的位置。该PLATE-YOLO网络是本文针对车牌检测的特点,对YOLOv3网络做了修改后得到的适用于车牌检测的网络。主要改进点有两处:第一,提出了一种基于层次聚类算法的锚盒(Anchor Box)个数和初始簇中心的计算方法;第二,针对车牌目标较大的特点,对网络的多尺度特征融合做了优化。优化后的PLATE-YOLO网络更适合于车牌检测,且提高了检测速度。实验证明,PLATE-YOLO网络检测车牌的速度较YOLOv3提高了5 FPS;在雾霾环境下,经去雾预处理的PLATE-YOLO车牌检测方法比未经去雾处理的车牌检测方法准确率提高了9.2%。

基于改进YOLOv5m的电动车骑行者头盔与车牌检测方法

电动车上路必须佩戴安全头盔已成为交管部门的强制性规定.为了能自动检测出电动车骑行者的头盔佩戴情况,提出一种基于改进的YOLOv5m模型的头盔与车牌检测方法,在检测出骑行者未佩戴头盔的同时还能检测出电动车车牌.模型使用自建电动车骑行者头盔与车牌检测数据集进行训练,用DIOU损失函数代替GIOU损失函数,DIOU_NMS代替加权NMS,增强模型对密集骑行场景的识别能力.在Backone部位与预测中小目标的Neck部位加入ECA注意力机制,使得模型对中小目标的识别率有所提高;用K-means算法对锚框尺寸重新进行聚类.最后,改进Mosaic数据增强方式.实验结果表明:改进的YOLOv5m电动车骑行者头盔与车牌检测模型的mAP为92.7%,较原YOLOv5m模型提高2.15个百分点,较YOLOv4-tiny、Faster RCNN模型分别提高5.7个百分点与6.9个百分点.改进后的YOLOv5m模型能有效提高对头盔与车牌的识别率.

基于卷积神经网络的车牌识别研究

为了满足交通管理和安全监控需求,针对车牌识别技术在智能交通系统中的重要性,本文旨在运用卷积神经网络实现汽车牌照的识别,首先通过对采集到的图像进行滤波降噪等预处理,运用Sobel算子进行边缘检测;其次对图像进行形态学运算,利用轮廓检测确定出车牌的位置;然后采取边缘检测的方法进行字符分割;最后运用卷积神经网络对车牌字符实现识别。模型测试结果表明,车牌识别系统汉字识别的准确率为91.4%,数字和字母共同识别的准确率为95.5%。

低能见度下GYOLOv5-SPD算法的车牌检测

车牌自动检测技术是构建智慧城市、加强交通管理等方面的重要内容。目前车牌检测技术正在逐步完善,但对于光照条件过低、雨天、雪天等能见度极低的气候环境下所拍摄的车牌照片,车牌检测技术还处于一个相对落后的水平。本文在中国城市停车数据集(CCPD)中能见度极低的车牌图像的基础上利用添加了伽玛(Gamma)变换的你只需看一次(YOLO)v5s模型与空间到深度层及无步长卷积层(SPD-Conv)进行融合用于车牌检测,将低能见度条件下的车牌视为小物体进行检测,目的是最大限度地识别其特征以提高精度。实验结果表明,本文用到的方法在车牌定位检测阶段达到了99.68%的召回率,在低能见度条件下对车牌的定位检测相较其他算法确实有一定的优势。

基于深度学习的轻量化车牌检测算法研究

针对现有车牌检测算法存在的模型参数量过大、实时性差和检测效果不佳等问题,提出一种基于深度学习的轻量化车牌检测网络(Lightweight License Plate Detection Networks,LW-LPDNet)模型。该模型以PP-LCNet作为骨干网络,大幅减少模型参数量,同时融入压缩-激励网络(Squeeze and Excitation Networks,SE-Net)注意力模块,增加车牌信息的通道权重。最后,引入SimSPPF和GSConv,对多尺度特征进行融合,增大感受野,进一步提高检测准确率。通过对模型进行训练和测试,LW-LPDNet在中国城市停车数据集(Chinese City Parking Dataset,CCPD)上获得98.9%的平均精确率,优于其他车牌检测方法,且模型参数量仅有0.13 MB,检测速度达到243 f·s^(-1),具备较高的实时性。

复杂道路环境下的车辆牌照检测与识别

针对复杂道路环境中车牌因倾斜、模糊、遮挡导致图像定位检测效果不佳和识别精度低等问题,提出一种基于全局阈值的灰度二值化图像预处理方法,采用YoloV5l算法对后处理阶段的数据集进行定位检测和检测结果评估,并通过R-CNN模型识别定位检测后的车牌图像字符。结果表明:当训练过程持续到100轮次时,相比于Faster R-CNN算法,该模型检测的平均精度均值(mAP)提升9.2%,识别准确率提升17.33%,验证该方法检测和识别车牌的有效性与优越性。

冶金废渣物流管控系统的应用

为了解决冶金废渣物流的有效控制,通过远程访问和控制技术、卡口监控摄像机车牌号自动识别技术、手机App开发,实现从合同、计划、清洁化运输、称重、料场装卸货确认、自动出门岗等于一体的运销流程智能化管控系统,解决了冶金废渣销售、运输、计量过程中的漏洞,杜绝作弊行为,提高企业计量管控、信息化管理水平。

结合双金字塔特征融合与级联定位的车牌检测

为了解决复杂环境中不同因素干扰车牌检测精确度的问题,提出了一种基于双金字塔特征融合的复杂环境下车牌检测算法。通过采用Mish激活函数的残差网络(ResNet101-M)对输入图像进行初级特征提取;在传统特征金字塔网络(feature pyramid network,FPN)的基础上,提出了一种改进的双金字塔特征融合网络(siamese feature pyramid network,SFPN)。被提取的初级特征被送入该网络进行多层特征融合。融合后的特征被送入基于形状先验的锚点设置网络来确定感兴趣区域。将所生成的感兴趣区域送入级联定位网络从而得到准确的车牌检测结果。实验结果表明,该算法在AOLP与CCPD车牌数据集上均能够有效提升检测性能。

基于深度学习的美式车牌检测与识别技术

针对美式车牌存在背景图案多样、文本信息复杂,传统车牌识别方法难以同时满足对不同样式车牌进行识别的问题,通过对文本建议网络和卷积递归神经网络研究,本文提出了一种美式车牌检测与识别方法。针对美式车牌检测时产生多个文本框,且有些车牌号由于被图案隔断不在同一文本框的情况,设计了一套用于筛选车牌号的锚点机制。使用Adam优化算法训练卷积递归神经网络。本文构建了美式车牌数据集,进行验证实验,州名的识别率达到了92%,车牌号的识别率达到了84%,共同识别率达到了82%。且本文提出的网络模型大小不到60 MB,实时性强。提出的SE-MobileNetV2快速特征提取模型,大幅提高特征提取速度。

基于并行CNN和识别策略优化的车牌识别方法研究

为改善车牌自动识别系统的通用性,在一般卷积神经网络(CNN)的基础上,提出一种具有两个浅层独立子网络的CNN,且具有并行卷积层计算的功能.一个用于推理车牌的概率;另一个利用线性激活对仿射参数进行回归.支持对汽车(包括公交车和卡车等)、摩托车等不同类型的交通工具牌照的检测识别.此外,使用基于YOLO v3的车牌字符识别模块,并施加了一系列的优化策略,实现对车牌中汉字的准确读取.实验结果表明所提方法的识别精度优于一些同类优秀方法,在AOLP数据集上的车牌检测准确率达到98.9%,在CLPD数据集上的字符识别准确率达到96.2%.所提方法有助于促进智能交通系统的进一步发展.

面向激光光源的光纤智能结构应变监测研究

光纤智能结构在应变监测环境中,受到多种因素干扰,导致应变监测误差大。为了获得理想的光纤智能结构应变监测结果,提出面向激光光源的光纤智能结构应变监测研究。采用激振信号产生的应力波回馈力和反射波延时,根据两者设计光纤智能结构应变监测目标函数。通过目标函数解实现应力脉冲信号约束,使得监测点的接收和反馈信号的同步性,提取应变监测冲击和应力波特征。考虑外界影响对监测结果干扰,采用周期性方式获取监测信号序列,查找序列畸变点,实现光纤智能结构应变精准监测。测试结果表明,本方法的光纤智能结构应变监测精准度高,应力波回馈延时短,具有较好的光纤智能结构应变监测实时性。

基于YOLO的无约束场景中文车牌检测与识别

传统中文车牌识别方法对场景约束有要求,算法实时性差,且无法被部署在边缘设备上。针对上述问题,文中提出一种基于YOLO(You Only Look Once)的无约束场景中文车牌检测与识别方法。该方法分为车牌检测和车牌字符识别两个模块。在车牌检测部分,使用改进的YOLOv5模型,在预测目标候选区域的基础上多预测4组关键点用于车牌矫正,并使用在COCO数据集上训练的预训练模型进行训练,减少了由环境复杂引起的误检问题,具有高实时性。在车牌字符识别部分,改进了CRNN(Convolutional Recurrent Neural Network)模型,减少了算法的参数量和计算量,使其能成功部署于各类边缘设备。实验结果表明所提出的车牌识别方法能在复杂环境中高效检测并识别车牌。文中提出的车牌检测模型在车牌检测数据集上的map值相较Retina-face提升了3.0%,车牌字符识别模型在车牌识别数据集上精确度相比LPR-Net提升了4.2%。