构建多部件关系概率模型的车辆检测方法

doi:10.3969/j.issn.1001-2400.2017.03.016

西安电子科技大学学报

构建多部件关系概率模型的车辆检测方法

宋俊芳1,2;宋翔宇3;郭晓军2;王卫星1

(1. 长安大学信息工程学院，陕西西安 710064;
2. 西藏民族大学信息工程学院，陕西咸阳 712082;
3. 北京交通大学交通运输学院，北京 100044)

收稿日期:2016-05-04 出版日期:2017-06-20 发布日期:2017-07-17
作者简介:宋俊芳(1984-)，女，长安大学博士研究生，E-mail：sjf0732001@126.com
基金资助:
国家自然科学基金资助项目(61572083)；西藏科技厅自然科学基金资助项目(2015ZR-13-17)

Vehicle detection using the location relationship model between multi-components

SONG Junfang1,2;SONG Xiangyu3;GUO Xiaojun2;WANG Weixing1

(1. School of Information Engineering, Chang'an Univ., Xi'an 710064, China;
2. School of Information Engineering, Xizang Minzu Univ., Xianyang 712082, China;
3. School of Traffic and Transportation, Beijing Jiaotong Univ., Beijing 100044, China)

Received:2016-05-04 Online:2017-06-20 Published:2017-07-17

摘要/Abstract

摘要：

传统车辆检测方法在处理交通复杂场景时，由于判断依据单一而容易出现车辆被误检和漏检的情况．为此，笔者提出一种基于车尾灯对和车牌之间空间位置关系概率模型的车辆检测方法．首先对道路交通视频图像进行颜色模型转换，分别得到突显车牌区域和尾灯区域的灰度图像，然后，对车牌灰度图像，通过梯度特征提取、区域梯度平滑和局部极大梯度查找几个关键步骤完成车牌的进一步准确定位，同时，对尾灯灰度图像经过阈值分割、连通域分析和面积计算几个关键步骤完成尾灯对的进一步准确定位．最后，建立尾灯和车牌在同一空域下的位置关系概率模型，并计算出该模型的概率似然度，进而判断部件所属的目标是否是同一辆车．实验结果表明，引入部件的空间位置关系概率模型，可以明显提高车辆检测的精度，且对不同型号的车辆检测鲁棒性都较好．

关键词: 车辆检测, 多部件模型, 高斯混合建模, 尾灯检测, 车牌检测

Abstract:

In view of the complex traffic and changeable weather and illumination in a scene, traditional vehicle detection method based on the single part model may result in a bad detection. So, using the spatial location relationships existing in multi-components of the vehicle, license plate and rear lamps are selected to construct the probabilistic models, through which vehicles are detected in this paper. In the new method, first, the color image of the road video is decomposed to the rear lamp gray image and license plate gray image through a different color conversion model. After that, the further identification for the license plate is accomplished through the key steps of gradient feature extraction, regional gradient smoothing and local maximum gradient search; similarly, the further identification of rear lamps is accomplished through the key steps of threshold segmentation, connected domain analysis and area calculation. Finally, With the Gaussian Mixture Model, relationships between the parts of the probability are established, and for the relationship model, if it makes the likelihood probability greater than a preset threshold, we argue that these parts belong to the same vehicle, and take the test result as the final vehicle detection result. Experimental results indicate that the new vehicle detection method has a strong adaptability, which can perfectly deal with the bad illumination conditions and target occlusion conditions, as well as a variety of vehicle types.

Key words: vehicle detection, part-based models, Gaussian mixture model, rear lamp detection, license plate detection

宋俊芳;宋翔宇;郭晓军;王卫星. 构建多部件关系概率模型的车辆检测方法[J]. 西安电子科技大学学报, 2017, 44(3): 89-95.

SONG Junfang;SONG Xiangyu;GUO Xiaojun;WANG Weixing. Vehicle detection using the location relationship model between multi-components[J]. Journal of Xidian University, 2017, 44(3): 89-95.

参考文献

［1］ CHEN X, SHRIVASTAVA A, GUPTA A. Enriching Visual Knowledge Bases via Object Discovery and Segmentation［C］//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2014: 2035-2042.
［2］ O'MALLEY R, JONES E, GLAVIN M. Rear-lamp Vehicle Detection and Tracking in Low Exposure Color Video for Night Conditions ［J］. IEEE Transactions on Intelligent Transportation Systems, 2010, 11(2): 453-462.
［3］ CHEN G, SUI H G, DONG L, et al. Semi-automatic Extraction Method for Low Contrast Road Based on Gabor Filter and Simulated Annealing［C］//Proceedings of the 4th International Conference on Materials Science and Information Technology: 989-994. Pfaffikon:Trans Tech Publications Ltd, 2014, 989: 3644-3648.
［4］ HSIEH J W, CHEN L C, CHEN D Y. Symmetrical SURF and Its Applications to Vehicle Detection and Vehicle Make and Model Recognition ［J］. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(1): 6-20.
［5］ MOHAMED A, ISSAM A, MOHAMED B, et al. Real-time Detection of Vehicles Using the Haar-like Features and Artificial Neuron Networks［C］//Proceedings of the International Conference on Advanced Wireless Information and Communication Technologies: 73. Philadelphia: Elsevier, 2015: 24-31.
［6］ TSAI W K, WU S L, LIN L J, et al. Edge-based forward Vehicle Detection Method for Complex Scenes［C］//Proceedings of the IEEE International Conference on Consumer Electronics. Piscataway: IEEE, 2014: 173-174.
［7］ ZHU H J, FAN H H, YE F Y, et al. Improving Vehicle Detection Accuracy Based on Vehicle Shadow and Superposition Elimination ［J］. Open Mechanical Engineering Journal, 2015, 9(1): 1039-1044.
［8］ HA D M, LEE J M, KIM Y D. Neural-edge-based Vehicle Detection and Traffic Parameter Extraction［J］. Image and Vision Computing, 2004, 22(11): 899-907.
［9］ KUMARMISHRA P, BANERJEE B. Vehicle Classification Using Density based Multi-feature Approach in Support Vector Machine Classifier ［J］. International Journal of Computer Applications, 2013, 71(7): 103-109.
［10］ SIVARAMAN S, TRIVEDI M M. Active Learning for On-road Vehicle Detection: a Comparative Study ［J］. Machine Vision and Applications, 2014, 25(3): 599-611.
［11］ TEHRANI NIKNEJAD H, TAKEUCHI A, MITA S, et al. On-road Multivehicle Tracking Using Deformable Object Model and Particle Filter with Improved Likelihood Estimation ［J］. IEEE Transactions on Intelligent Transportation Systems, 2012, 13(2) 748-758.
［12］ JURI D, LONARI S. A Unified Approach for On-road Visual Night-time Vehicle Light Detection［C］//Lecture Notes in Computer Science: 8887. Berlin: Springer Verlag, 2014: 730-739.
［13］ CHANG J K, RYOO S, LIM H. Real-time Vehicle Tracking Mechanism with License Plate Recognition from Road Images［J］. Journal of Supercomputing, 2013, 65(1): 353-364.
［14］ CHENG D, WANG J J, WEI X, et al. Training Mixture of Weighted SVM for Object Detection Using EM ALGORITHM ［J］. Neurocomputing, 2015, 149(PB): 473-482.

[1]	张智,郑锦. 结合帧间目标回归网络的无人机视频车辆检测[J]. 西安电子科技大学学报, 2021, 48(4): 151-158.
[2]	王艳芬,朱绪冉,云霄,孙彦景,石韫开,王赛楠. 面向公共安全监控的多摄像机车辆重识别[J]. 西安电子科技大学学报, 2019, 46(4): 190-196.

构建多部件关系概率模型的车辆检测方法

Vehicle detection using the location relationship model between multi-components

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

Metrics

本文评价

推荐阅读 0