基于SSD的车辆目标检测

doi:10.3969/j.issn.1006-2475.2019.05.007

计算机与现代化

基于SSD的车辆目标检测

(河海大学计算机与信息学院，江苏南京211100)

收稿日期:2018-11-23 出版日期:2019-05-14 发布日期:2019-05-14
作者简介:吴水清(1994-),女,浙江江山人,硕士研究生,研究方向:目标检测与识别,E-mail: wsq30332@163.com；王宇(1979-),男,研究员,博士,研究方向:云计算；师岩(1993-),男,硕士研究生,研究方向:机器翻译。
基金资助:
国家自然青年科学基金资助项目(61103017); 中国科学院感知中国先导专项子课题(XDA06040504)

Vehicle Detection Method Based on SSD

(College of Computer and Information, Hohai University, Nanjing 211100, China)

Received:2018-11-23 Online:2019-05-14 Published:2019-05-14

摘要/Abstract

摘要： 传统的车辆目标检测算法需要为不同的图像场景选择合适的特征，导致泛化能力差。针对此问题，本文提出一种基于SSD（Single Shot MultiBox Detector）的图像车辆检测方法。该方法通过对多个尺度的卷积特征图进行预测来检测车辆，在一定程度上提升车辆的检测精度；找出原SSD方法在训练过程中的小缺陷，通过改进损失函数来优化训练速度。最后结合KITTI数据集进行训练。实验结果表明，该方法对车辆的检测具有较高的识别率，且比传统算法的效果更好。

关键词: 车辆, 目标检测； SSD；卷积神经网络

Abstract: The traditional vehicle target detection algorithm needs to select appropriate features for different image scenes, resulting in poor generalization ability. Therefore, an image detection method based on SSD (Single Shot MultiBox Detector) for vehicle is proposed. The method detects the vehicle by predicting the convolutional feature maps of multiple scales, and improves the detection accuracy of the vehicle to a certain extent. The small defects of the original SSD method in the training process are found, and the loss function is improved to optimize the training speed. Finally, the KITTI data set is used for training. The experimental results show that the method has a higher recognition rate for vehicles and the effect is better than traditional algorithms.

Key words: vehicle, target detection, SSD, convolutional neural network

中图分类号:

TP391.4

吴水清，王宇，师岩. 基于SSD的车辆目标检测[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2019.05.007.

WU Shui-qing, WANG Yu, SHI Yan. Vehicle Detection Method Based on SSD[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2019.05.007.

参考文献

［1］ SZEGEDY C, TOSHEV A, ERHAN D. Deep neural networks for object detection ［C］// Proceedings of 2013 Advances in Neural Information Processing Systems(NIPS). 2013:2553-2561.
［2］ SINGLA N. Motion detection based on frame difference method［J］. International Journal of Information & Computation Technology, 2014,4(15):1559-1565.
［3］ MERLIN P M, FARBER D J. A parallel mechanism for detecting curves in pictures［J］. IEEE Transactions on Computers, 1975,24(1):96-98.
［4］ HORN B K P, SCHUNCK B G. Determining optical flow［J］. Artificial Intelligence, 1981,17(1-3):185-203.
［5］ VIOLA P, JONES M. Rapid object detection using a boosted cascade of simple features［C］// Proceedings of 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2003:I-511-I-518.
［6］ LEE D S. Effective Gaussian mixture learning for video background subtraction［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005,27(5):827-832.
［7］ FELZENSZWALB P F, GIRSHICK R B, MC ALLESTER D, et al. Object detection with discriminatively trained part-based models［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010,32(9):1627-1645.
［8］ LOWE D G. Distinctive image features from scale-invariant keypoints［J］. International Journal of Computer Vision, 2004,60(2):91-110.
［9］ DALAL N, TRIGGS B. Histograms of oriented gradients for human detection［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2005:886-893.
［10］PANNING A, AL-HAMADI A K, NIESE R, et al. Facial expression recognition based on Haar-like feature detection［J］. Pattern Recognition & Image Analysis, 2008,18(3):447-452.
［11］罗海波,许凌云,惠斌,等. 基于深度学习的目标跟踪方法研究现状与展望［J］. 红外与激光工程, 2017,46(5):14-20.
［12］GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation［C］// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 2014:580-587.
［13］GIRSHICK R. Fast R-CNN［C］// Proceedings of IEEE International Conference on Computer Vision. 2015:1440-1448.
［14］REN S, HE K, 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.
［15］LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection［C］// Proceedings of IEEE International Conference on Computer Vision. 2017:2999-3007.
［16］LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single shot multibox detector［C］// European Conference on Computer Vision. 2016:21-37.
［17］REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: Unified, real-time object detection［C］// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 2016:779-788.
［18］唐聪,凌永顺,郑科栋,等. 基于深度学习的多视窗SSD目标检测方法［J］. 红外与激光工程, 2018,47(1):290-298.
［19］蔡汉明,赵振兴,韩露,等. 基于SSD网络模型的多目标检测算法［J］. 机电工程, 2017,34(6):685-688.
［20］ERHAN D, SZEGEDY C, TOSHEV A, et al. Scalable object detection using deep neural networks［C］// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 2014:2155-2162.
［21］杨洁,陈明志,吴智秦,等. 基于SSD卷积网络的视频目标检测研究［J］. 南华大学学报(自然科学版), 2018,32(1):78-86.
［22］IOFFE S, SZEGEDY C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift［EB/OL］. ［2015-03-02］. https://arxiv.org/abs/1502.03167.
［23］Matthew B, Richard S, Simon W. Multi-image matching using multi-scale oriented patches［C］// Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2005:510-517.

[1]	刘续, 查可可. 一种用于机场特种车辆作业的环境目标识别方法[J]. 计算机与现代化, 2023, 0(08): 18-24.
[2]	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.
[3]	万冬厚, 张德贤, 邓淼磊, . 基于非局部注意力和局部特征的车辆重识别算法[J]. 计算机与现代化, 2022, 0(03): 23-29.
[4]	张雄, 潘大志, . 融合邻域搜索策略蚁群算法求解带时间窗口的车辆路径问题[J]. 计算机与现代化, 2022, 0(03): 98-102.
[5]	张成龙, 张毅. 面向协作式自动泊车的场端单目视觉测距系统[J]. 计算机与现代化, 2021, 0(11): 112-117.
[6]	张志刚, 游安清. 基于CPN网络的车辆关键点检测[J]. 计算机与现代化, 2021, 0(10): 75-80.
[7]	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.
[8]	郑钦浩, 杨贞, 杨振 . 面向车辆和行人检测的KM-SSD方法[J]. 计算机与现代化, 2021, 0(03): 51-56.
[9]	吕坦悦1,2，陆小敏1,2，王健1,2. 基于决策树和层次分析的地铁车辆健康评估[J]. 计算机与现代化, 2020, 0(03): 29-.
[10]	王喜龙，张伟伟，吴训成 . 基于双相机协同的管制车型抓拍系统[J]. 计算机与现代化, 2019, 0(03): 23-.
[11]	刘银萍，马少辉. 基于多模型融合的车辆通过时间预测[J]. 计算机与现代化, 2019, 0(02): 66-.
[12]	李晓龙,陈国良,葛凯凯. 基于四旋翼无人机的车辆自主抄牌[J]. 计算机与现代化, 2018, 0(10): 63-.
[13]	王德宇1，徐友春2，李永乐2，陆峰1，郑凯文1. 基于深度学习的车辆检测方法[J]. 计算机与现代化, 2017, 0(8): 56-.
[14]	温凯歌. 信号控制下的动态路段行程时间模型[J]. 计算机与现代化, 2015, 0(8): 104-106,111.
[15]	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.

基于SSD的车辆目标检测

Vehicle Detection Method Based on SSD

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价