嵌入空洞卷积模块的改进YOLOv3车辆检测算法

计算机与现代化 ›› 2021, Vol. 0 ›› Issue (04): 53-60.

嵌入空洞卷积模块的改进YOLOv3车辆检测算法

(1.中国石油大学（华东）海洋与空间信息学院，山东青岛266580;
2.齐鲁工业大学（山东省科学院）海洋仪器仪表研究所，山东青岛266061)

出版日期:2021-04-22 发布日期:2021-04-25
作者简介:胡昌冉(1994—)，男，山东济宁人，硕士研究生，研究方向：数字图像处理，目标检测，E-mail: 15689131805@163.com; 樊彦国(1965—)，男，河北望都人，教授，博士，研究方向：遥感图像处理，数字图像处理，E-mail: ygfan@upc.edu.cn; 禹定峰（1986—），男，山东青岛人，副研究员，博士，研究方向：水色遥感，E-mail: dfyucsas@163.com。
基金资助:
山东省重点研发计划项目(2019GHY112017)

Improved YOLOv3 Vehicle Detection Algorithm Embedded in Dilated Convolution Module

(1. College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China;
2. Institute of Oceanographic Instrumentation, Qilu University of Technology (Shandong Academy of Sciences), Qingdao 266061, China)

Online:2021-04-22 Published:2021-04-25

摘要/Abstract

摘要： 对图像或者视频中的车辆进行检测是计算机视觉领域研究的热点之一，同时也是智能交通系统的重要组成部分。鉴于车辆检测场景复杂多变以及现有的车辆检测算法不能同时满足高精度以及高实时性的要求，本文提出一种改进的YOLOv3车辆检测算法，并自建车辆检测数据集。首先在原有及特征提取网络Darknet-53中嵌入空洞卷积模块，以减少目标信息的丢失增强感受野；其次为减少错检漏检的情况，本文对传统的NMS算法进行改进，若预测框的IoU大于设定的阈值，使其以一定的方式衰减。该改进的方法在KITTI标准数据集上显示出优于其他算法的性能，同时在自建的数据集中进行验证，精度可达96%，检测速度达25.9帧/s。

关键词: 车辆检测, 实时检测, 空洞卷积, 非极大值抑制

Abstract: Vehicle detection on image or video data is one of the hotspots in the field of computer vision, and it is also an important part of intelligent transportation systems. In view of the complex and changeable vehicle detection scenes and the existing vehicle detection algorithms can not meet the requirements of high precision and high real-time at the same time, this paper proposes an improved YOLOv3 vehicle detection algorithm and builds its own vehicle detection data set. First, we embed the dilated convolution module in the original and feature extraction network Darknet-53 to reduce the loss of target information and enhance the receptive field. Secondly, in the NMS (non-maximum suppression) module, in order to reduce the missed detection, this article discusses the traditional NMS and makes improvements. If the IoU of the prediction frame is greater than the set threshold, it will be attenuated in a certain way. The improved method shows better performance than other algorithms on the KITTI standard data set, and the verification accuracy can reach 96% in the self-built data set, and the detection speed is 25.9 frames/s.

Key words: vehicle detection, real-time detection, dilated convolution, NMS

胡昌冉, 樊彦国, 禹定峰. 嵌入空洞卷积模块的改进YOLOv3车辆检测算法[J]. 计算机与现代化, 2021, 0(04): 53-60.

HU Chang-ran, FAN Yan-guo, YU Ding-feng. Improved YOLOv3 Vehicle Detection Algorithm Embedded in Dilated Convolution Module[J]. Computer and Modernization, 2021, 0(04): 53-60.

参考文献

［1］缪前明. 城市智能交通系统分析及新技术应用设想［J］. 中国市政工程, 2014(2):1-3.
［2］刘博艺,程杰仁,唐湘滟,等. 复杂动态环境下运动车辆的识别方法［J］. 计算机科学与探索, 2017,11(1):134-143.
［3］王伟锋,金杰,陈景明. 基于感受野的快速小目标检测算法［J］. 激光与光电子学进展, 2020,57(2):250-255.
［4］卞山峰,张庆辉. 基于改进YOLOv2的车辆实时检测算法［J］. 电子质量, 2019(10):19-22.
［5］ KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks［C］// Proceedings of the 25th International Conference on Neural Information Processing Systems. 2012:1097-1105.
［6］ GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation［C］// Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. 2014:580-587.
［7］ UIJLINGS J R R, VAN DE SANDE K E A, GEVERS T, et al. Selective search for object recognition［J］. International Journal of Computer Vision, 2013,104(2):154-171.
［8］ HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015,37(9):1904-1916.
［9］ GIRSHICK R. Fast R-CNN［C］// Proceedings of the 2015 IEEE International Conference on Computer Vision. 2015:1440-1448.
［10］REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks［C］// Proceedings of the 28th International Conference on Neural Information Processing Systems. 2015:91-99.
［11］REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: Unified, real-time object detection［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. 2016:779-788.
［12］REDMON J, FARHADI A. YOLO9000: Better, faster, stronger［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. 2017:6517-6525.
［13］IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift［J］// Proceedings of the 32nd International Conference on Machine Learning. 2015:448-456.
［14］REDMON J, FARHADI A. YOLOv3: An incremental improvement［J］. arXiv preprint arXiv:1804.02767, 2018.
［15］鞠默然,罗海波,王仲博,等. 改进的YOLOv3算法及其在小目标检测中的应用［J］. 光学学报, 2019,39(7):245-252.
［16］刘军,后士浩,张凯,等. 基于增强Tiny YOLOv3算法的车辆实时检测与跟踪［J］. 农业工程学报, 2019,35(8):118-125.
［17］SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. 2015:1-9.
［18］YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions［J］. arXiv preprint arXiv:1511.07122, 2015.
［19］刘学平,李玙乾,刘励,等. 嵌入SENet结构的改进YOLOv3目标识别算法［J］. 计算机工程, 2019,45(11):243-248.
［20］GEIGER A, LENZ P, URTASUN R. Are we ready for autonomous driving? The KITTI vision benchmark suite［C］// Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. 2012:3354-3361.
［21］WEN L Y, DU D W, CAI Z W, et al. UA-DETRAC: A new benchmark and protocol for multi-object detection and tracking［J］. arXiv preprint arXiv:1511.04136, 2015.
［22］LYU S W, CHANG M C, DU D W, et al. UA-DETRAC 2017: Report of AVSS2017 & IWT4S challenge on advanced traffic monitoring［C］// Proceedings of the 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance. 2017. DOI: 10.1109/AVSS.2017.8078560.
［23］ZHOU Y, LIU L, SHAO L, et al. Fast automatic vehicle annotation for urban traffic surveillance［J］. IEEE Transactions on Intelligent Transportation Systems, 2018,19(6):1973-1984.
［24］DOLLAR P, WOJEK C, SCHIELE B, et al. Pedestrian detection: A benchmark［C］// Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. 2009:304-311.

[1]	王涛1, 2, 黄丹1, 2, 刘禅奕1, 2, 朱桃1, 2. 基于YOLOv5s的无人机图像车辆检测[J]. 计算机与现代化, 2024, 0(08): 108-113.
[2]	高埂1, 肖风丽2, 杨飞1. 基于改进MobileNetV3-Small的色素减退性皮肤病诊断[J]. 计算机与现代化, 2024, 0(05): 120-126.
[3]	宁召阳1, 2, 申情2, 3, 郝秀兰1, 2, 赵康1, 2. 基于改进对抗学习及融合特征的短文本分类框架[J]. 计算机与现代化, 2024, 0(04): 66-76.
[4]	崔少国, 张宇楠. 融合多尺度空间特征的甲状腺结节超声图像分割[J]. 计算机与现代化, 2024, 0(03): 72-77.
[5]	彭露露, 朱媛媛, 金文倩, 王笑梅. 基于改进YOLOv4的汽车钢铁零件表面缺陷检测[J]. 计算机与现代化, 2022, 0(09): 32-39.
[6]	张文丽, 徐丽, 刘星星. 基于卷积神经网络的夜间车辆检测算法[J]. 计算机与现代化, 2022, 0(05): 108-113.
[7]	刘学虎, 欧鸥, 张伟劲, 杜雪垒. 融合注意力机制和空洞卷积的滑坡图像检测[J]. 计算机与现代化, 2022, 0(04): 45-51.
[8]	周幸, 陈立福. 基于双注意力机制的遥感图像目标检测[J]. 计算机与现代化, 2020, 0(08): 1-7.
[9]	唐晖，王庆，陈洪，郭浩. 基于深度学习的体感交互方法[J]. 计算机与现代化, 2019, 0(02): 7-.
[10]	李亚,王颖. 基于一维最大熵的视频图像运动背景减除[J]. 计算机与现代化, 2018, 0(03): 44-.
[11]	熊琰铖，孙涵. 结合全局信息的对称SURF算法在车辆检测中的应用[J]. 计算机与现代化, 2015, 0(1): 80-83.
[12]	刘雁斌1,2. 高速公路收费站防盗卡系统设计与实现[J]. 计算机与现代化, 2014, 0(4): 191-194.
[13]	杨小伟;徐贵力;王彪;郭瑞鹏;田裕鹏;何银南. 一种基于局部梯度矢量的车辆检测方法[J]. 计算机与现代化, 2013, 1(2): 9-14.
[14]	鲁胜强;刘瑞玲. 基于最大目标函数的医学图像边缘检测[J]. 计算机与现代化, 2011, 1(6): 5-3.
[15]	杨志奇. 车辆检测系统中背景提取与更新算法研究 [J]. 计算机与现代化, 2010, 1(7): 111-114.