基于模式识别与ST-MRF相结合的车辆检测方法

周君; 包旭; 高焱; 李耘; 姜晴

doi:10.3963/j.jssn.1674-4861.2021.02.012

基于模式识别与ST-MRF相结合的车辆检测方法

doi: 10.3963/j.jssn.1674-4861.2021.02.012

淮阴工学院交通工程学院江苏淮安 223003

基金项目:

国家自然科学基金项目 51808248

江苏省高校自然科学重大项目 17KJA580001

详细信息

通讯作者:
周君(1980—)，博士，副教授.研究方向：智能交通. E-mail: joujou1980@163.com

中图分类号: U491.1+16
计量
- 文章访问数: 395
- HTML全文浏览量: 263
- PDF下载量: 13
- 被引次数: 0
出版历程
- 收稿日期: 2020-07-07

A Vehicle Detecting Method Based on Pattern Recognition Combined with ST-MRF

Faculty of Transportation Engineering, Huaiyin Institute of Technology, Huai'an 223003, Jiangsu, China

摘要

摘要: 车辆检测技术的主要难点是在于解决车辆之间的遮挡，以及由于光照变化引起的车辆与其阴影之间的遮挡问题，这些问题将直接影响检测的精度。针对这个问题，在原ST-MRF方法上研究了基于模式识别与ST-MRF相结合的车辆检测方法。模式识别技术分割相互遮挡的2辆车之间的边界，并识别相互遮挡车辆的边缘间隙以及边界信息，模式识别结果反馈给ST-MRF算法，算法对相互遮挡车辆重新分配标号，优化处理并融合不完整的分割部分，确定单个车辆信息。路段车辆检测实验结果表明，在检测区域行驶的325辆车，用原始ST-MRF算法跟踪统计到的车辆数为258辆，成功率为79%，采用模式识别技术与ST-MRF相结合算法统计到车辆315辆，成功率为97%；交叉口车辆检测实验结果表明，该方法在机动车与非机动车混行，公交车与小汽车相互遮挡的交叉口场景下，能较准确地得到车辆检测结果。
- 智能交通 /
- 车辆检测 /
- 模式识别 /
- 时空马尔可夫随机场 /
- 遮挡 /
- 能量函数
Abstract: The main difficulty of the vehicle detecting technology lies in overcoming the inaccuracy of detecting, which is directly affected by video shaking, mutual occlusion between vehicles, occlusion between the vehicles, and its shadow due to illumination changes. Based on the traditional spatial-temporal Markov random field (ST-MRF), the vehicle detecting method combined with pattern recognition and ST-MRF is proposed to address the problem. The boundary between two vehicles occluding each other is segmented by the pattern recognition technology, with the edge clearances and boundary information of occluded vehicles identified. Then, the results of pattern recognition are fed back to ST-MRF algorithm, which reassigns labels to occluded vehicles, integrates incomplete segmentation, and determines individual vehicle information. The results in the road section show 325 vehicles driving in the test area, vehicles tracked by the original ST-MRF algorithm, and the success rate is 79%. When the pattern recognition technology combined with ST-MRF algorithm to calculate 315 vehicles, the success rate is 97%. The results at the intersection show that the method can obtain more accurate results of vehicle detection at the intersection where motor vehicles and non-motor vehicles are mixed, and buses and cars are obstructed by each other.
- intelligent transportation /
- vehicle detecting /
- pattern recognition /
- spatial-temporal Markov random field /
- occlusion /
- energy function

HTML全文

图 1 研究路线图

Figure 1. Flow for the study

下载: 全尺寸图片幻灯片

图 2 2车相互遮挡示意图

Figure 2. Mutual occlusion between vehicles

下载: 全尺寸图片幻灯片

图 3 间隙类型

Figure 3. Gap type

下载: 全尺寸图片幻灯片

图 4 图像扫描示意图

Figure 4. Image scanning

下载: 全尺寸图片幻灯片

图 5 边界数

Figure 5. Number of borders

下载: 全尺寸图片幻灯片

图 6 边缘间隔评估值

Figure 6. Border interval evaluation

下载: 全尺寸图片幻灯片

图 7 边缘

Figure 7. Border

下载: 全尺寸图片幻灯片

图 8 车辆检测结果

Figure 8. Vehicle detection results

下载: 全尺寸图片幻灯片

图 9 路段处2种算法车辆跟踪结果比较

Figure 9. Vehicle tracking results of two algorithms in the road section

下载: 全尺寸图片幻灯片

图 10 交叉口处2种算法车辆跟踪结果比较

Figure 10. Vehicle tracking results of two algorithms at the intersection

下载: 全尺寸图片幻灯片

参考文献(21)

[1]	GAO Fei, GE Yisu, LU Shufang, et al. Online vehicle detection at nighttime-based tail-light pairing with saliency detection in the multi-lane intersection[J]. IET Intelligent Transport Systems, 2019, 13(3): 515-522. doi: 10.1049/iet-its.2018.5197
[2]	ZHAO Huijing, WANG Chao, LIN Yuping, et al. On-road vehicle trajectory collection and scene-based lane change analysis: part I[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 18(1): 1-14. http://ieeexplore.ieee.org/document/7515360
[3]	杨敏, 裴明涛, 王永杰, 等. 一种基于运动目标检测的视觉车辆跟踪方法[J]. 北京理工大学学报, 2014, 34(4): 370-375. https://www.cnki.com.cn/Article/CJFDTOTAL-BJLG201404009.htm YANG Min, PEI Mingtao, WANG Yongjie, et al. Video-based vehicle tracking based on moving object detection[J]. Transactions of Beijing Institute of Technology, 2014, 34(4): 370-375. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BJLG201404009.htm
[4]	袁俊. 基于角点SIFT特征匹配的车辆跟踪方法研究[D]. 杭州: 浙江理工大学, 2013. YUAN Jun. Study on a new vehicle tracking method based on SIFT features matchment of corners[D]. Hangzhou: Zhejiang Sci-Tech University, 2013. (in Chinese)
[5]	刘阳, 王忠立, 蔡伯根, 等. 复杂环境基于多信息融合的车辆跟踪方法[J]. 交通运输系统工程与信息, 2015, 15(6): 74-81. doi: 10.3969/j.issn.1009-6744.2015.06.012 LIU Yang, WANG Zhongli, CAI Baigen, et al. Multiple information fusion-based vehicle tracking in complex environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2015, 15(6): 74-81. (in Chinese) doi: 10.3969/j.issn.1009-6744.2015.06.012
[6]	王文龙, 李清泉. 基于蒙特卡罗算法的车辆跟踪方法[J]. 测绘学报, 2011, 40(2): 200-203. https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201102013.htm WANG Wenlong, LI Qingquan. A vehicle tracking algorithm with Monte-Carlo method[J]. Acta Geodaeteca et Cartographica Sinica, 2011, 40(2): 200-203. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201102013.htm
[7]	WEN Longyin, CAI Zhaowei, LEI Zhen, et al. Robust online learned spatio-temporal context model for visual tracking[J]. IEEE Transactions on Image Processing, 2014, 23 (2): 785-796. doi: 10.1109/TIP.2013.2293430
[8]	周君, 程琳. 基于ST-MRF的自适应车辆跟踪算法研究[J]. 交通运输系统工程与信息, 2013, 13(3): 65-70. doi: 10.3969/j.issn.1009-6744.2013.03.010 ZHOU Jun, CHENG Lin. Adaptive vehicle tracking algorithm based on ST-MRF model[J]. Journal of Transportation Systems Engineering and Information Technology, 2013, 13(3): 65-70. (in Chinese) doi: 10.3969/j.issn.1009-6744.2013.03.010
[9]	周君, 程琳. 基于反向ST-MRF模型的车辆遮挡分割算法[J]. 公路交通科技, 2013, 30(4): 107-111. doi: 10.3969/j.issn.1002-0268.2013.04.019 ZHOU Jun, CHENG Lin. A segmentation algorithm of vehicle occlusion based on reversed ST-MRF model[J]. Journal of Highway and Transportation Research and Development. 2013, 30(4): 107-111. (in Chinese) doi: 10.3969/j.issn.1002-0268.2013.04.019
[10]	周君, 高尚兵, 包旭, 等. 基于ST-MRF模型的电动自行车与汽车交通冲突检测方法: 201611030277.4[P]. 2019-01-22. ZHOU Jun, GAO Shangbing, BAO Xu, et al. Traffic conflict detection based on vehicle trajectories of electric bicycles and cars: 201611030277.4[P]. 2019-01-22. (in Chinese)
[11]	OLIVERA R, OLIVERA R, VITE O, et al. Application of the three state Kalman filtering for moving vehicle tracking[J]. IEEE Latin America Transactions, 2016, 14 (5) : 2072-2076. doi: 10.1109/TLA.2016.7530397
[12]	AHMADI P, TABANDEH M, GHOLAMPOUR I. Abnormal event detection and localisation in traffic videos based on group sparse topical coding[J]. IET Image Processing, 2016, 10(3): 235-246. doi: 10.1049/iet-ipr.2015.0399
[13]	侯进辉, 曾焕强, 蔡磊, 等. 基于随机遮挡辅助深度表征学习的车辆再辨识[J]. 控制理论与应用, 2018, 35(12): 1725-1730. https://www.cnki.com.cn/Article/CJFDTOTAL-KZLY201812004.htm HOU Jinhui, ZENG Huanqiang, CAI Lei, et al. Random occlusion assisted deep representation learning for vehicle re-identification[J]. Control Theory & Applications. 2018, 35(12): 1725-1730. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KZLY201812004.htm
[14]	TALMALE S, UNNIKRISHNAN S, LANDE B K. Distance increasing mapping for variable distance block code[J]. IET Communications, 2020, 14(9): 1495-1501. doi: 10.1049/iet-com.2019.0875
[15]	ZAKARIA N J, SHAPIAI M I, FAUZI H, et al. Gradient-based edge effects on lane marking detection using a deep learning-based approach[J]. Arabian Journal for Science and Engineering, 2020(45): 10989-11006. doi: 10.1007/s13369-020-04918-4
[16]	张大奇, 曲仕茹, 石爽. 基于序列图像运动分割的车辆边界轮廓提取算法[J]. 交通运输工程学报, 2009, 9(3): 1671-1637. https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC200903025.htm ZHANG Daqi, QU Shiru, SHI Shuang. Edge detection algorithm of moving vehicle base on sequential image motion segmentation[J]. Journal of Traffic and Transportation Engineering, 2009, 9(3): 1671-1637. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC200903025.htm
[17]	CAO Xianbin, LIN Renjun, YAN Pingkun. Visual attention accelerated vehicle detection in low-attitude airborne video of urban environment[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2012, 22(3): 366-378. doi: 10.1109/TCSVT.2011.2163443
[18]	ELLENRIEDER K D. Dynamic surface control of trajectory tracking marine vehicles with actuator magnitude and rate limits[J]. Automatica, 2019(105): 433-442. http://www.sciencedirect.com/science/article/pii/S0005109819301785
[19]	王宇宁, 庞智恒, 袁德明. 基于YOLO算法的车辆实时检测[J]. 武汉理工大学学报, 2016, 38(10): 41-46. doi: 10.3963/j.issn.1671-4431.2016.10.008 WANG Yuning, PANG Zhiheng, YUAN Deming. Vehicle detection based on YOLO in real time[J]. Journal of Wuhan University of Technology, 2016, 38(10): 41-46. (in Chinese) doi: 10.3963/j.issn.1671-4431.2016.10.008
[20]	赖见辉, 王扬, 罗甜甜, 等. 基于YOLO_V3的侧视视频交通流量统计方法与验证[J]. 公路交通科技, 2021, 38(1): 135-142. doi: 10.3969/j.issn.1002-0268.2021.01.017 LAI Jianhui, WANG Yang, LUO Tiantian, et al. A YOLO_V3 based road-side video traffic volume counting method and verification[J]. Journal of Highway and Transportation Research and Development. 2021, 38(1): 135-142. (in Chinese) doi: 10.3969/j.issn.1002-0268.2021.01.017
[21]	徐子睿, 刘猛, 谈雅婷. 基于YOLOv4的车辆检测与流量统计研究[J]. 现代信息科技, 2020(15): 98-100+103. https://www.cnki.com.cn/Article/CJFDTOTAL-XDXK202015031.htm XU Zirui, LIU Meng, TAN Yating. Research on vehicle detection and traffic statistics based on YOLOv4[J]. Modern Information Technology, 2020(15): 98-100+103. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XDXK202015031.htm