复杂场景下基于热图的车牌检测

doi:10.19678/j.issn.1000-3428.0062085

摘要/Abstract

摘要： 随着车牌识别的应用场景不断扩展，处理的图像复杂性也随之提高，车牌检测面临车牌定位困难、检测速度慢和精度低等挑战。为提高光照不均衡、透视变形、雨雾天气、低分辨率等复杂场景下车牌检测的准确率，提出一种基于车牌角点热图的检测网络LPHD-Net。不同于传统模板匹配和目标检测中矩形先验框的方式，该网络通过车牌角点热图和车牌边界向量场的方法对车牌进行检测。在中国城市停车数据集中进行训练和测试，使用目标检测任务中常用的平均精度和召回率对模型的整体性能进行评价。实验结果表明，LPHD-Net模型对多种复杂场境下的车牌检测精确率和速度分别达到99.2%和78 frame/s，较LMAFLPD模型提升1.15个百分点和14 frame/s。同时，其对场景中的多车牌检测也具有较好的检测效果。

关键词: 目标检测, 热图, 复杂场景, 车牌检测定位, 深度学习

Abstract: As the application scenarios of license plate recognition continue to expand, the complexity of the processed images is also increasing.Thereforelicense plate detection also faces challenges such as difficulty in locating license platesslow detection speedand low accuracy.For license plate detection in complex scenes such as uneven illuminationperspective distortionrainy and foggy weatherand low resolutiona heat map-based license plate detection NetworkLPHD-Netbased on the license plate corner heatmap is proposed.This network is different from the rectangular prior frame method in template matching and target detection in the past. Insteadthe license plate is detected using the license plate corner heat map and license plate boundary vector field methods.The network is trained and tested using the China Urban Parking Data Set(CDPP) and the overall performance of the model is evaluated using the average accuracy and recall rate commonly used in target detection tasks.The experimental results show that the algorithm's license plate detection accuracy and speed in complex scenes reaches 99.2% and 78 frame/s, respectively, which are increased by 1.15 percentage points and 14 frame/s compared with LMAFLPD model.At the same time, it also has a good detection effect for multiple license plate detections in a scene.

Key words: target detection, heat map, complex scene, license plate detection and positioning, deep learning

中图分类号:

TP18

郝超杰, 贾振堂. 复杂场景下基于热图的车牌检测[J]. 计算机工程, 2022, 48(7): 315-320.

HAO Chaojie, JIA Zhengtang. Heat Map-Based License Plate Detection in Complex Scenes[J]. Computer Engineering, 2022, 48(7): 315-320.

http://www.ecice06.com/CN/Y2022/V48/I7/315

图/表 8

20221029180237

20221029180240

20221029180244

20221029180247

20221029180251

20221029180254

20221029180257

20221029180300

参考文献

[1] 杨鼎鼎, 陈世强, 刘静漪.基于车牌背景和字符颜色特征的车牌定位算法[J].计算机应用与软件, 2018, 35(12):216-221. YANG D D, CHEN S Q, LIU J Y.License plate location algorithm based on background and character color feature[J].Computer Applications and Software, 2018, 35(12):216-221.(in Chinese)
[2] 王永杰, 裴明涛, 贾云得.多信息融合的快速车牌定位[J].中国图象图形学报, 2014, 19(3):471-475. WANG Y J, PEI M T, JIA Y D.License plate detection based on multiple features[J].Journal of Image and Graphics, 2014, 19(3):471-475.(in Chinese)
[3] 魏亭, 邱实, 李晨, 等.计算机多尺度辅助定位车牌算法[J].电子学报, 2018, 46(9):2188-2193. WEI T, QIU S, LI C, et al.License plate detection algorithm based on computer multi scale assist[J].Acta Electronica Sinica, 2018, 46(9):2188-2193.(in Chinese)
[4] 张小瑞, 陈旋, 孙伟, 等.基于深度学习的车辆再识别研究进展[J].计算机工程, 2020, 46(11):1-11. ZHANG X R, CHEN X, SUN W, et al.Progress of vehicle re-identification research based on deep learning[J]. Computer Engineering, 2020, 46(11):1-11.(in Chinese)
[5] 单美静, 秦龙飞, 张会兵.L-YOLO:适用于车载边缘计算的实时交通标识检测模型[J].计算机科学, 2021, 48(1):89-95. L-YOLO:real time traffic sign detection model for vehicle edge computing[J].Computer Science, 2021, 48(1):89-95.(in Chinese)
[6] LIU W, ANGUELOV D, ERHAN D, et al.SSD:single shot multibox detector[C]//Proceedings of ECCV'16.Berlin, Germany:Springer, 2016:21-37.
[7] 张彩丽, 刘广文, 詹旭, 等.基于Mask R-CNN的多目标跟踪算法[J].吉林大学学报(理学版), 2021, 59(3):609-618. ZHANG C L, LIU G W, ZHAN X, et al.Multiple object tracking algorithm based on Mask R-CNN[J].Journal of Jilin University(Science Edition), 2021, 59(3):609-618.(in Chinese)
[8] LAROCA R, SEVERO E, ZANLORENSI L A, et al.A robust real-time automatic license plate recognition based on the YOLO detector[C]//Proceedings of International Joint Conference on Neural Networks.Washington D.C., USA:IEEE Press, 2018:1-10.
[9] 胡逸龙, 金立左.基于深度学习方法的中文车牌识别算法[J].工业控制计算机, 2021, 34(5):63-65. HU Y L, JIN L Z.Chinese license plate recognition algorithm based on deep learning method[J].Industrial Control Computer, 2021, 34(5):63-65.(in Chinese)
[10] HSU G S, AMBIKAPATHI A, CHUNG S L, et al.Robust license plate detection in the wild[C]//Proceedings of the 14th IEEE International Conference on Advanced Video and Signal Based Surveillance.Washington D.C., USA:IEEE Press, 2017:1-6.
[11] TIAN Y, LU X, LI W X.License plate detection and localization in complex scenes based on deep learning[C]//Proceedings of 2018 Chinese Control And Decision Conference.Washington D.C., USA:IEEE Press, 2018:6569-6574.
[12] SILVA S M, JUNG C R.License plate detection and recognition in unconstrained scenarios[C]//Proceedings of ECCV'18.Berlin, Germany:Springer, 2018:593-609.
[13] 余烨, 付源梓, 陈维笑, 等.自然场景下变形车牌检测模型DLPD-Net[J].中国图象图形学报, 2021, 26(3):556-567. YU Y, FU Y Z, CHEN W X, et al.DLPD-Net:distorted license plate detection model in natural scenarios[J].Journal of Image and Graphics, 2021, 26(3):556-567.(in Chinese)
[14] CAO Z, SIMON T, WEI S H, et al.Realtime multi-person 2D pose estimation using part affinity fields[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:1302-1310.
[15] OSOKIN D.Real-time 2d multi-person pose estimation on CPU:lightweight OpenPose[EB/OL].[2021-06-10].https://arxiv.org/pdf/1811.12004.pdf.
[16] HUANG G, LIU Z, VAN DER MAATEN L, et al.Densely connected convolutional networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:2261-2269.
[17] WANG S Z, LEE H J.A cascade framework for a real-time statistical plate recognition system[J].IEEE Transactions on Information Forensics and Security, 2007, 2(2):267-282.
[18] REN S Q, HE K M, GIRSHICK R, et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.
[19] REDMON J, FARHADI A.YOLOv3:an incremental improvement[EB/OL].[2021-06-10].https://arxiv.org/abs/1804.02767.
[20] LI H, WANG P, SHEN C H.Toward end-to-end car license plate detection and recognition with deep neural networks[J].IEEE Transactions on Intelligent Transportation Systems, 2019, 20(3):1126-1136.
[21] XU Z B, YANG W, MENG A J, et al.Towards end-to-end license plate detection and recognition:a large dataset and baseline[C]//Proceedings of ECCV'18.Berlin, Germany:Springer, 2018:261-277.
[22] WANG W W, YANG J, CHEN M, et al.A light CNN for end-to-end car license plates detection and recognition[J].IEEE Access, 2019, 7:173875-173883.

选择文件类型/文献管理软件名称

选择包含的内容