西南交通大学学报 2012, 47(3) 439-445 DOI:   10.3969/j.issn.0258-2724.2012.03.014  ISSN: 0258-2724 CN: 51-1277/U

本期目录 | 下期目录 | 过刊浏览 | 高级检索                                                            [打印本页]   [关闭]
论文
扩展功能
本文信息
Supporting info
PDF(0KB)
[HTML全文]
参考文献
服务与反馈
把本文推荐给朋友
加入我的书架
加入引用管理器
引用本文
Email Alert
文章反馈
浏览反馈信息
本文关键词相关文章
车辆运动检测
背景提取
小波分解
模历史图像
本文作者相关文章
屈桢深
于萌萌
姜永林
闻帆
王常虹
PubMed
Article by Que,Z.S
Article by Yu,M.M
Article by Jiang,Y.L
Article by Wen,f
Article by Yu,C.H

应用小波模历史图像的运动车辆视频检测

屈桢深, 于萌萌, 姜永林, 闻帆, 王常虹

哈尔滨工业大学空间控制与惯性技术研究中心, 黑龙江 哈尔滨 150080

摘要

为提高车辆目标检测的稳定性和准确性,提出了基于背景减除和小波分解模历史图像的运动车辆检测算法.首先对原始图像进行小波分解,对低频分量用混合高斯模型和纹理特征相结合的方法,自适应更新背景并标记运动目标初始区域;然后,基于高频分量计算模值,并通过逐帧历史累积得到模历史图像;最后,利用车辆目标与阴影相比富含边缘细节的特点,对目标进行倾斜校正后,将目标边缘分别沿图像xy方向投影,利用投影曲线将边缘信息与目标初始区域信息迭代融合,得到最终检测结果.实验结果表明,用本文方法检测车辆的捕获率达到99.0%,有效率为92.5%;与使用单一自适应背景提取方法相比,在实际交通场景中可有效处理阴影导致的多目标粘连问题,检测结果更准确.

关键词 车辆运动检测   背景提取   小波分解   模历史图像  

Vision-Based Detection of Moving Vehicles Using Wavelet Modulus History Images

QU Zhenshen, YU Mengmeng, JIANG Yonglin, WEN Fan, WANG Changhong

Space Control and Inertial Technology Research Center, Harbin Institute of Technology, Harbin 150080, China

Abstract:

To improve the robustness and accuracy of vehicle detection, a new detection method of moving vehicles was put forward based on background subtraction and wavelet decomposition modulus history images. First, wavelet decomposition was performed on the original image. In low frequency component, the Gaussian mixture model was used in conjunction with textural features to adaptively update the background image and label initial regions of moving objects. The high frequency component was used to calculate modulus value and obtain modulus history image through history frame accumulation. In view of the fact that vehicle objects have richer edge details than shadow regions, the edges were projected to x and y axes after object slant correction. Using the projection curves, the edge information was iteratively integrated with initial object regions to get final detection results. Experiment results show that compared with commonly used adaptive background extraction methods, the proposed method could detect vehicle objects accurately in practical traffic applications with a capture rate of 99.0% and an effective rate of 92.5%, and could effectively process the object conglutination caused by shadow with higher accuracy.

Keywords: vehicle motion detection   background extraction   wavelet decomposition   modulus history image  
收稿日期 2010-07-14 修回日期  网络版发布日期 2012-05-29 
DOI: 10.3969/j.issn.0258-2724.2012.03.014
基金项目:

国家自然科学基金资助项目(70971030)

通讯作者:
作者简介: 屈桢深(1973-),男,副教授,博士,研究方向为视觉信息处理及空间控制,电话:13704803558,E-mail:miraland@hit.edu.cn

参考文献:

[1] STAUFFER C, GRIMSON W L. Learning patterns of activity using real-time tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8): 747-757.
[2] 朱碧婷,郑世宝. 基于高斯混合模型的空间域背景分离法及阴影消除法[J]. 中国图象图形学报,2008,13(10): 1906-1909. ZHU Biting, ZHENG Shibao. Space-domain background subtraction and shadow elimination based on Gaussian mixture model[J]. Journal of Image and Graphics, 2008, 13(10): 1906-1909.
[3] 金立生,贾敏,孙玉芹,等. 日间高速公路侧后方车辆识别方法[J]. 西南交通大学学报,2010,45(2): 231-237. JIN Lisheng, JIA Min, SUN Yuqin, et al. Detection of backside vehicle on freeway in daytime[J]. Journal of Southwest Jiaotong University, 2010, 45(2): 231-237.
[4] HAN B, COMANICIU D, ZHU Y, et al. Sequential kernel density approximation and its application to real-time visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(7): 1186-1197.
[5] TAI J C, TSENG S T, LIN C P, et al. Real-time image tracking for automatic traffic monitoring and enforcement applications[J]. Image and Vision Computing, 2004, 22(6): 485-501.
[6] GIL S, MILANESE R, PUN T. Comparing features for target tracking in traffic scenes[J]. Pattern Recognition, 1995, 29(8): 1285-1296.
[7] BRADSKI G, DAVIS J. Motion segmentation and pose recognition with motion history gradients[J]. Machine Vision and Applications, 2002, 13(3): 174-184.
[8] DAVIS J, BOBICK A. The representation and recognition of human movement using temporal templates //Proceedings of Computer Vision and Pattern Recognition. San Juan: IEEE Press, 1997: 928-934.
[9] 刘光耀. 基于模型的交通流信息检测系统研究. 杭州:浙江大学,2005.
[10] MARTEL-BRISSON N, ZACCARIN A. Learning and removing cast shadows through a multi-distribution approach[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(7): 1133-1146.
[11] KAEWTRAKULPONG P, BOWDEN R. A real time adaptive visual surveillance system for tracking low-resolution colour targets in dynamically changing scenes[J]. Image and Vision Computing, 2003, 21(10): 913-929.
[12] ZHOU D, ZHANG H, RAY N. Texture based background subtraction//International Conference on Information and Automation. Changsha: IEEE, 2008: 601-605.
[13] 姜永林,屈桢深,王常虹. 基于纹理及统计特征的视频背景提取研究[J]. 光学精密工程,2008,16(1): 172-177. JIANG Yonglin, QU Zhenshen, WANG Changhong. Video background extraction based on textural and statistical features[J]. Opt. Precision Eng., 2008, 16(1): 172-177.
[14] MALLAT S. A wavelet tour of signal processing: the sparse way[M]. 3rd ed. New York: Academic Press, 2009: 170-181.
[15] 戴小文,蔡志平,钟桂英. 基于小波变换及熵的视频镜头分割检测方法[J]. 西南交通大学学报,2008, 43(3): 314-318. DAI Xiaowen, CAI Zhiping, ZHONG Guiying. Video shot detection method based on wavelet transform and wavelet entropy[J]. Journal of Southwest Jiaotong University, 2008, 43(3): 314-318.
[16] 张桂林,陈益新,曹伟垣,等. 基于跑长码的连通区域标记算法[J]. 华中理工大学学报,1994,22(5): 11-14. ZHANG Guilin, CHEN Yixin, CAO Weihuan, et al. A connected component labeling algorithm using the run-length code[J]. Journal of Huazhong University of Science and Technology, 1994, 22(5): 11-14.
[17] CANNY J F. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 11(6): 679-698.
[18] BRADSKI G, KAEBLER A. Learning OpenCV[M]. Sebastopol: O'Reilly Media Inc., 2008: 266-293.

本刊中的类似文章
1.彭 强;李 华 .基于块直方图分析的视频背景提取方法 [J]. 西南交通大学学报, 2006,41(1): 48-53

文章评论 (请注意:本站实行文责自负, 请不要发表与学术无关的内容!评论内容不代表本站观点.)

Copyright 2008 by 西南交通大学学报