基于OpenCV的运动目标检测与跟踪
|
摘要
计算机视觉和数字图像处理技术可以广泛地应用于工业、医疗保健、航空航天、军事等各个领域,其中针对视频连续图像中运动物体的分析是其中应用前景最为广泛的一个方向,在机器人导航、智能视觉监控系统、医学图像分析、工业检测、视频图像分析以及军事雷达视频信号的处理上都有占有重要地位。
其中基于视频图像的运动目标分析也是计算机视觉和数字图像处理技术最为复杂的一个方向。它是一个庞大的工程,既要对视频信号进行采集,又要对采集的图像进行处理,还要针对具体的应用编写程序进行开发。对于要完成某项应用的工程人员来说,如果所有底层的算法都要自己编码实现,所有的图像处理函数都要从头编写,既造成时间和精力上的浪费,又难以保证稳定性、实用性和通用性。
OpcnCV(Open Source Computer Vision Library)是一种用于数字图像处理和计算机视觉的函数库,由Intel微处理器研究实验室(Intel's MicroprocessorResearch Lab)的视觉交互组(The Visual Interactivity Group)开发。采用的开发语言是C++,可以在Windows系统及Linux系统下使用,该函数库是开放源代码的,能够从Intel公司的网站免费下载得到。OpenCV提供了针对各种形式的图像和视频源文件(如:bitmap图像,video文件和实时摄像机)的帧提取函数和很多标准的图像处理算法,这些函数都可以直接用在具体的视频程序开发项目中。
针对在复杂背景中检测出多批特定运动目标并实施分配批号实行标记跟踪,本文利用OpenCV的运动物体跟踪的数据结构、函数以及基本框架,建立了一个由人机交互界面模块;运动物体的前景检测模块;运动物体的团块特征检测模块;运动物体的团块跟踪模块轨迹生成模块;轨迹后处理模块组成的视频图像运动目标分析系统。
在本文的最后阶段,使用此视频图像运动目标分析系统进行了大量实验,并全面分析了实验现象和数据。通过这些现象和数据可以得出结论:本文基于OpenCV设计的视频图像运动目标分析系统具有良好的实时性,能够正确的进行运动物体的实时检测和跟踪,并具有良好的鲁棒性。由于该系统在Windows下开发,如何将该系统移植到其他的系统或者是嵌入式平台并进一步提高系统的通用性和鲁棒性就成了今后研究工作的重点。
Detection and tracking of moving object in image sequence is one important area in the domain of digital image processing and computer vision. It's applications include fields like robot navigation, vision-based supervision, ecurity surveillance, medical image analysis, industrial automation, video processing, etc...
Detection and tracking of moving object in image sequence is also the most complicated direction of computer vision and digital image processing techniques. It is a ample project that including video collection, digital image processing and application development.It's too complex for a engineer to accomplish all these works, at the same time the stability, practicability and general availability is difficult to guaranteed.
OpenCV(Open Source Computer Vision Library) is a library for digital image processing and computer vision. It is developed by the visual interactivity group in Intel's microprocessor research Lab with C++ language. It can be used in window system and Linux system. The library is open source and can be free download from Intel's Web site. OpenCV provides various functions to extract frame from image sequence and video source ( such as: bitmap images, video files and real-time video camera ) and many of the standard image processing algorithms, these functions can be directly used in the video program for specific development projects.
Aim at detecting, tracking and marking multipule specific targets in complex background. We use the basic framework of object tracking of OpenCV to establish a multi-modules system. The system is composed of Human-computer interaction interface module, foreground of moving objects detection module, blocks feature of moving objects detection module, blocks tracking module and track analysis module.
We use this system doing a lot of experiment under various contexts, and tht experiments results are well wxamined. The results demonstrate that the system can do automatic motion detection and tracking with high real-time performance and moderate robustness. Since we develop this system under windows platform, how to migrate to other systems or embedded platform and further improve the versatility and robustness has become the focus of future research work.
其中基于视频图像的运动目标分析也是计算机视觉和数字图像处理技术最为复杂的一个方向。它是一个庞大的工程,既要对视频信号进行采集,又要对采集的图像进行处理,还要针对具体的应用编写程序进行开发。对于要完成某项应用的工程人员来说,如果所有底层的算法都要自己编码实现,所有的图像处理函数都要从头编写,既造成时间和精力上的浪费,又难以保证稳定性、实用性和通用性。
OpcnCV(Open Source Computer Vision Library)是一种用于数字图像处理和计算机视觉的函数库,由Intel微处理器研究实验室(Intel's MicroprocessorResearch Lab)的视觉交互组(The Visual Interactivity Group)开发。采用的开发语言是C++,可以在Windows系统及Linux系统下使用,该函数库是开放源代码的,能够从Intel公司的网站免费下载得到。OpenCV提供了针对各种形式的图像和视频源文件(如:bitmap图像,video文件和实时摄像机)的帧提取函数和很多标准的图像处理算法,这些函数都可以直接用在具体的视频程序开发项目中。
针对在复杂背景中检测出多批特定运动目标并实施分配批号实行标记跟踪,本文利用OpenCV的运动物体跟踪的数据结构、函数以及基本框架,建立了一个由人机交互界面模块;运动物体的前景检测模块;运动物体的团块特征检测模块;运动物体的团块跟踪模块轨迹生成模块;轨迹后处理模块组成的视频图像运动目标分析系统。
在本文的最后阶段,使用此视频图像运动目标分析系统进行了大量实验,并全面分析了实验现象和数据。通过这些现象和数据可以得出结论:本文基于OpenCV设计的视频图像运动目标分析系统具有良好的实时性,能够正确的进行运动物体的实时检测和跟踪,并具有良好的鲁棒性。由于该系统在Windows下开发,如何将该系统移植到其他的系统或者是嵌入式平台并进一步提高系统的通用性和鲁棒性就成了今后研究工作的重点。
Detection and tracking of moving object in image sequence is one important area in the domain of digital image processing and computer vision. It's applications include fields like robot navigation, vision-based supervision, ecurity surveillance, medical image analysis, industrial automation, video processing, etc...
Detection and tracking of moving object in image sequence is also the most complicated direction of computer vision and digital image processing techniques. It is a ample project that including video collection, digital image processing and application development.It's too complex for a engineer to accomplish all these works, at the same time the stability, practicability and general availability is difficult to guaranteed.
OpenCV(Open Source Computer Vision Library) is a library for digital image processing and computer vision. It is developed by the visual interactivity group in Intel's microprocessor research Lab with C++ language. It can be used in window system and Linux system. The library is open source and can be free download from Intel's Web site. OpenCV provides various functions to extract frame from image sequence and video source ( such as: bitmap images, video files and real-time video camera ) and many of the standard image processing algorithms, these functions can be directly used in the video program for specific development projects.
Aim at detecting, tracking and marking multipule specific targets in complex background. We use the basic framework of object tracking of OpenCV to establish a multi-modules system. The system is composed of Human-computer interaction interface module, foreground of moving objects detection module, blocks feature of moving objects detection module, blocks tracking module and track analysis module.
We use this system doing a lot of experiment under various contexts, and tht experiments results are well wxamined. The results demonstrate that the system can do automatic motion detection and tracking with high real-time performance and moderate robustness. Since we develop this system under windows platform, how to migrate to other systems or embedded platform and further improve the versatility and robustness has become the focus of future research work.
引文
[1]高守传,姚领田等.Visual C++数字图像处理与工程应用篇[M].北京:中国铁道出版社.2006,1
[2]陈兵旗,孙明.使用数字图像处理与分析[M].中国农业大学出版社,清华大学出版社.2008,2
[3]王亮,胡卫明,谭铁牛.人运动的视觉分析综述[J].计算机学报.2002(3):1-16
[4]朱仲杰 蒋刚毅.目标基视频编码中的运动目标提取与跟踪新算法[J].电子学报.2003,9(9):1408-1426.
[5]张益贞,刘滔.Visual C++实现MPEG/JPEG编解码技术[M].北京:人民邮电出版社.2002
[6]Maggioni C and Kammerer B." Gesture Computer:history,design,and applications Computer Vision of Human-Machine Interaction[J]".Cambridge Univ.Press,1998.
[7]Freeman W and Weissman C."Television control by hand gesture.In:Proc International Conference on Automatic Face and Gesture Recognition[J]".Zurich,Switzerland,1993:179-183
[8]C.Anderson,PeterBurt,and G.van der Wal."Change detection and tracking using Pyramid transformation techniques[J]".InProceedings of SPIE-Intelligent Robots and Computer Vision,1985,Vol.579:72-78.
[9]Haritaoglu,Larry S.Davis,and D.Harwood."w4:who?when?where?what?A real time system for detecting and tracking people[C]".Japan:International conferenee on face and gesture recognition,1998.
[10]C.Wren,A.Azarbayejani,T.Darrell et al."Pfinder:Real-time tracking of the human body[C]".IEEE Transactions on Pattern Analysis and Machinelntelligence,1997,19(7):780-785.
[11]J.Barron,D.Fleet,5.Beauchemin." Performance of opticalflow techniques[C]".International Journal of Computer Vision,1994,12(1):42-77
[12]周兵,李波,怀进鹏.运动对象检测及其在视频监控中的应用[J].计算机应 用技术.2003,2
[13]吕国亮,赵曙光,赵俊.基于三帧差分和连通性检验的图像运动目标检测新方法[J].液晶与显示,2007.2.Vol.22:87-92
[14]Surendra Gupte,Osama Masoud,Robert F.K.Martin,and Nikolaos P.Papanikolopoulos."Detection and Classification of Vehicles[J]".IEEE Transactions on intelligent transportation system[J].2002,3(1):37-47
[15]邱德润,朱明早,伍宗富.一种基于帧间差分与背景差分的运动目标检测新方法[J].湖南文理学院学报(自然科学版).2007,3.Vol.19 No.1:81-86
[16]J.B.Kim,H.J.Kim."Efficient region-based motion segmentation for video monitoring system[J]".PatternRecognition Letters.2003,(24):113-128
[17]陈伟杰,王睿.复杂场景下灰度图像的运动目标跟踪[J].计算机与现代化.2007,10:110-113
[18]Dorin Comaniciu,Visvanathan Ramesh,PeterMeer."Kernel-based object tracking[J]" IEEE Trans on Pattern Analysis and Machine Intelligence,2003,25(5):564-577.
[19]M.Kass,A.Wiktin and D.Terzopoulos."Snakes:Active contour models.Int.Conf.Computer Vision[J]".1987,vol.53:211-218
[20]OpenCV.http://www.intel.com/technology/computing/opencv/index.htm
[21]VXL.http://vxl.sourceforge.net/
[22]CH OpenCV.http://www.softintegration.com/products/thirdparty/opencv/
[23]刘瑞祯,于仕琪.OpenCV教程-基础篇[M].北京航空航天大学出版社,2007
[24]Chiao-Fe Shu,Brown L,Hampapur A,Connell J et al.IBM smart sureillance system(S3):an open and extensible architecture for smart video surveillance.
[25]The ObjectVideo Company.http://www.objectvideo.com
[26]Haritaoglu I,Harwood D,Davis L."W4:real-time surveillance of people and their activities[J]".IEEE Trans Pattern Analysis and Machine Intlligence,2000,22(8):809-830
[27]R.C.Gonzalez,R.E.woods.数字图像处理(第二版)[M].阮秋琦,阮宇智等译.北京:电子工业出版社.2003.
[28]Jeong B,Lee Y H."Design of weighted order statistic filter using the perception algorithm[J]".IEEE Transactions on Signal Processing.1994.
[29]秦鹏,丁润涛.一种基于排序阈值的开关中值滤波方法[J].中国图象图形学报.2004.4:412-416
[30]祈大勇,韩月秋.一种快速图像中值滤波算法[J].北京理工大学学报.1996.8.Vol.16:454-456
[31]郑岩,战晓苏.数据结构与算法C++版(第3版)[M]清华大学出版社.2006.1
[32]张丽,陈志强,高文焕等.均值加速的快速中值滤波算法[J].清华大学学报(自然科学版),2004(9):1572-1159
[33]朱虹.数字图像处理基础[M].北京:科学出版社,2005.
[34]董付国,王平勤.分治法在中值滤波快速算法中的应用研究.电脑开发与应用[J]2007.6.Vol.20:42-43
[35]周奇,张永光,徐健健.基于直方图特性的图像去噪方法[J]电子测量技术第30卷第1期2007,1:40-42
[36]刘西成,冯燕.一种基于脉冲噪声检测的图像去噪方法[J].计算机仿真.2007年02月第24卷第02期:187-190
[37]A.J.Bardos."SELECTIVE VECTOR MEDIAN FILTERING OF COLOUR IMAGES.Conference Publication No.443[C]".IEEE,July 1997:708-711.
[38]周燕,曹文.面向彩色手术显微图像的去噪算法改进[J].微计算机信息.2007年第23卷第4-3期:287,288,294
[39]王建卫.彩色图像的中值滤波算法的改进与应用[J].哈尔滨商业大学学报(自然科学版)2006年8月第22卷第4期:67-69
[40]吴燕,张晨燕.基于目标特征的灰度直方图重分布图象增强算法[J]计算机与信息技术.2007.6:14-15,19
[41]张懿,刘旭,李海峰.自适应图像直方图均衡算法[J].浙江大学学报(工学版)Vol.41.No.4.Apr.2007:630-633
[42]李弼程,彭天强,彭波.智能图像处理技术[M].北京:电子工业出版社.2004.
[43]杜振华,张艳宁,郑江滨,须明.基于视频增强的昏暗背景下目标检测方法[J].微电子学与计算机.2007年第24卷第7期:16-19
[44]Otsu N."A threshold selection method from gray21evel histograms[J]".IEEE Transactions on System Man and Cybernetic,1979,9(1):62-66.
[45]刘健庄,粟文青.灰度图像的二维Otsu自动阈值分割法[J].自动化学报,1993,19(1):101-105.
[46]Gong Jian,Li Li yuan,Chen Wei nan."Fast recursive algorithm for two 2dimensional thresholding[J]".Pattern Recognition,1998,31(3):295-300.
[47]吕国亮,赵曙光,赵俊.基于三帧差分和连通性检验的图像运动目标检测新方法[J]液晶与显示.Vol.22,Nol,Feb.2007:87-93
[48]Surendra Gupte,Osama Masoud,Robert F.K.Martin,and Nikolaos P.Papanikolopoulos," Detection and Classification of Vehicles[J],IEEE Transactions on intelligent transportation system[J]".2002,3(1):37-47
[49]郭海峰,张亮,粟毅.视频图像序列中的运动目标实时检测[J].科技信息(学术研究).2008.1:1-3.
[50]William K.Pratt.数字图像处理[M].北京:机械工业出版社,2005.[51]Margrit Betke,Esin Haritaoglu,Larry S.Davis."Real-time Multiple Vehiele Detection And Tracking From A Moving Vehiele[J]".Machine Vision and Applications(2000)12:69-83.
[52]何斌,马天宇等.ViusalC++数字图像处理(第二版)[M].北京:人民邮电出版社.2002.
[53]伏思华,张小虎.基于序列图像的运动目标实时检测方法[J].光学技术,2004,30(2):215-217.
[54]周志宇,汪亚明,黄文清.基于动态图像序列的运动目标跟踪[J].浙江工程学院学报,2002,19(3):165-170
[55]王栓,艾海舟,何克忠.基于差分图像的多运动目标的检测与跟踪[J].中国图象图形学报.1999,4(A6):470-475
[56]马颂德,张正友.计算机视觉.计算理论与算法基础[M].北京:科学出版社.2003.
[57]关海英,阮秋琦.序列图像中运动目标的检测[J].铁道学报.1996,18(6):33-38
[58]杨威,张田文.复杂景物环境下运动目标检测的新方法[J].计算机研究与发展.1998,8(35).
[59]付永会,张风超,张宪民.一种改进的基于颜色直方图的实时目标跟踪算法[J]数据采集与处理,2001年9月第16卷第3期:309-314
[60]陆宗骐,金登男.Visual C++.Net图像处理编程[M],清华大学出版社.2006
[61]章毓晋.图像处理和分析基础[M].北京:高等教育出版社,2002.
[62]朱虹等.数字图像处理基础[M].北京:科学出版社,2005.
[63]M.Pctrou,P.Bosdogianni.数字图像处理疑难解析[M],赖剑煌,冯国灿等译.北京:机械工业出版社,2005.
[64]喻擎苍,翁秀娟,赵匀H.H.Cheng交互式开放结构计算机视觉平台 Ch OpenCV计算机工程与应用 2006.23
[65]Simon Haykin.自适应滤波器原理[M]郑宝玉译.电子工业出版社.2003.7
[66]董士崇,王天珍,许刚 视频图像中的运动检测[J].武汉理工大学学报.信息与管理工程版 2004Aug.Vol.26.No.4
[2]陈兵旗,孙明.使用数字图像处理与分析[M].中国农业大学出版社,清华大学出版社.2008,2
[3]王亮,胡卫明,谭铁牛.人运动的视觉分析综述[J].计算机学报.2002(3):1-16
[4]朱仲杰 蒋刚毅.目标基视频编码中的运动目标提取与跟踪新算法[J].电子学报.2003,9(9):1408-1426.
[5]张益贞,刘滔.Visual C++实现MPEG/JPEG编解码技术[M].北京:人民邮电出版社.2002
[6]Maggioni C and Kammerer B." Gesture Computer:history,design,and applications Computer Vision of Human-Machine Interaction[J]".Cambridge Univ.Press,1998.
[7]Freeman W and Weissman C."Television control by hand gesture.In:Proc International Conference on Automatic Face and Gesture Recognition[J]".Zurich,Switzerland,1993:179-183
[8]C.Anderson,PeterBurt,and G.van der Wal."Change detection and tracking using Pyramid transformation techniques[J]".InProceedings of SPIE-Intelligent Robots and Computer Vision,1985,Vol.579:72-78.
[9]Haritaoglu,Larry S.Davis,and D.Harwood."w4:who?when?where?what?A real time system for detecting and tracking people[C]".Japan:International conferenee on face and gesture recognition,1998.
[10]C.Wren,A.Azarbayejani,T.Darrell et al."Pfinder:Real-time tracking of the human body[C]".IEEE Transactions on Pattern Analysis and Machinelntelligence,1997,19(7):780-785.
[11]J.Barron,D.Fleet,5.Beauchemin." Performance of opticalflow techniques[C]".International Journal of Computer Vision,1994,12(1):42-77
[12]周兵,李波,怀进鹏.运动对象检测及其在视频监控中的应用[J].计算机应 用技术.2003,2
[13]吕国亮,赵曙光,赵俊.基于三帧差分和连通性检验的图像运动目标检测新方法[J].液晶与显示,2007.2.Vol.22:87-92
[14]Surendra Gupte,Osama Masoud,Robert F.K.Martin,and Nikolaos P.Papanikolopoulos."Detection and Classification of Vehicles[J]".IEEE Transactions on intelligent transportation system[J].2002,3(1):37-47
[15]邱德润,朱明早,伍宗富.一种基于帧间差分与背景差分的运动目标检测新方法[J].湖南文理学院学报(自然科学版).2007,3.Vol.19 No.1:81-86
[16]J.B.Kim,H.J.Kim."Efficient region-based motion segmentation for video monitoring system[J]".PatternRecognition Letters.2003,(24):113-128
[17]陈伟杰,王睿.复杂场景下灰度图像的运动目标跟踪[J].计算机与现代化.2007,10:110-113
[18]Dorin Comaniciu,Visvanathan Ramesh,PeterMeer."Kernel-based object tracking[J]" IEEE Trans on Pattern Analysis and Machine Intelligence,2003,25(5):564-577.
[19]M.Kass,A.Wiktin and D.Terzopoulos."Snakes:Active contour models.Int.Conf.Computer Vision[J]".1987,vol.53:211-218
[20]OpenCV.http://www.intel.com/technology/computing/opencv/index.htm
[21]VXL.http://vxl.sourceforge.net/
[22]CH OpenCV.http://www.softintegration.com/products/thirdparty/opencv/
[23]刘瑞祯,于仕琪.OpenCV教程-基础篇[M].北京航空航天大学出版社,2007
[24]Chiao-Fe Shu,Brown L,Hampapur A,Connell J et al.IBM smart sureillance system(S3):an open and extensible architecture for smart video surveillance.
[25]The ObjectVideo Company.http://www.objectvideo.com
[26]Haritaoglu I,Harwood D,Davis L."W4:real-time surveillance of people and their activities[J]".IEEE Trans Pattern Analysis and Machine Intlligence,2000,22(8):809-830
[27]R.C.Gonzalez,R.E.woods.数字图像处理(第二版)[M].阮秋琦,阮宇智等译.北京:电子工业出版社.2003.
[28]Jeong B,Lee Y H."Design of weighted order statistic filter using the perception algorithm[J]".IEEE Transactions on Signal Processing.1994.
[29]秦鹏,丁润涛.一种基于排序阈值的开关中值滤波方法[J].中国图象图形学报.2004.4:412-416
[30]祈大勇,韩月秋.一种快速图像中值滤波算法[J].北京理工大学学报.1996.8.Vol.16:454-456
[31]郑岩,战晓苏.数据结构与算法C++版(第3版)[M]清华大学出版社.2006.1
[32]张丽,陈志强,高文焕等.均值加速的快速中值滤波算法[J].清华大学学报(自然科学版),2004(9):1572-1159
[33]朱虹.数字图像处理基础[M].北京:科学出版社,2005.
[34]董付国,王平勤.分治法在中值滤波快速算法中的应用研究.电脑开发与应用[J]2007.6.Vol.20:42-43
[35]周奇,张永光,徐健健.基于直方图特性的图像去噪方法[J]电子测量技术第30卷第1期2007,1:40-42
[36]刘西成,冯燕.一种基于脉冲噪声检测的图像去噪方法[J].计算机仿真.2007年02月第24卷第02期:187-190
[37]A.J.Bardos."SELECTIVE VECTOR MEDIAN FILTERING OF COLOUR IMAGES.Conference Publication No.443[C]".IEEE,July 1997:708-711.
[38]周燕,曹文.面向彩色手术显微图像的去噪算法改进[J].微计算机信息.2007年第23卷第4-3期:287,288,294
[39]王建卫.彩色图像的中值滤波算法的改进与应用[J].哈尔滨商业大学学报(自然科学版)2006年8月第22卷第4期:67-69
[40]吴燕,张晨燕.基于目标特征的灰度直方图重分布图象增强算法[J]计算机与信息技术.2007.6:14-15,19
[41]张懿,刘旭,李海峰.自适应图像直方图均衡算法[J].浙江大学学报(工学版)Vol.41.No.4.Apr.2007:630-633
[42]李弼程,彭天强,彭波.智能图像处理技术[M].北京:电子工业出版社.2004.
[43]杜振华,张艳宁,郑江滨,须明.基于视频增强的昏暗背景下目标检测方法[J].微电子学与计算机.2007年第24卷第7期:16-19
[44]Otsu N."A threshold selection method from gray21evel histograms[J]".IEEE Transactions on System Man and Cybernetic,1979,9(1):62-66.
[45]刘健庄,粟文青.灰度图像的二维Otsu自动阈值分割法[J].自动化学报,1993,19(1):101-105.
[46]Gong Jian,Li Li yuan,Chen Wei nan."Fast recursive algorithm for two 2dimensional thresholding[J]".Pattern Recognition,1998,31(3):295-300.
[47]吕国亮,赵曙光,赵俊.基于三帧差分和连通性检验的图像运动目标检测新方法[J]液晶与显示.Vol.22,Nol,Feb.2007:87-93
[48]Surendra Gupte,Osama Masoud,Robert F.K.Martin,and Nikolaos P.Papanikolopoulos," Detection and Classification of Vehicles[J],IEEE Transactions on intelligent transportation system[J]".2002,3(1):37-47
[49]郭海峰,张亮,粟毅.视频图像序列中的运动目标实时检测[J].科技信息(学术研究).2008.1:1-3.
[50]William K.Pratt.数字图像处理[M].北京:机械工业出版社,2005.[51]Margrit Betke,Esin Haritaoglu,Larry S.Davis."Real-time Multiple Vehiele Detection And Tracking From A Moving Vehiele[J]".Machine Vision and Applications(2000)12:69-83.
[52]何斌,马天宇等.ViusalC++数字图像处理(第二版)[M].北京:人民邮电出版社.2002.
[53]伏思华,张小虎.基于序列图像的运动目标实时检测方法[J].光学技术,2004,30(2):215-217.
[54]周志宇,汪亚明,黄文清.基于动态图像序列的运动目标跟踪[J].浙江工程学院学报,2002,19(3):165-170
[55]王栓,艾海舟,何克忠.基于差分图像的多运动目标的检测与跟踪[J].中国图象图形学报.1999,4(A6):470-475
[56]马颂德,张正友.计算机视觉.计算理论与算法基础[M].北京:科学出版社.2003.
[57]关海英,阮秋琦.序列图像中运动目标的检测[J].铁道学报.1996,18(6):33-38
[58]杨威,张田文.复杂景物环境下运动目标检测的新方法[J].计算机研究与发展.1998,8(35).
[59]付永会,张风超,张宪民.一种改进的基于颜色直方图的实时目标跟踪算法[J]数据采集与处理,2001年9月第16卷第3期:309-314
[60]陆宗骐,金登男.Visual C++.Net图像处理编程[M],清华大学出版社.2006
[61]章毓晋.图像处理和分析基础[M].北京:高等教育出版社,2002.
[62]朱虹等.数字图像处理基础[M].北京:科学出版社,2005.
[63]M.Pctrou,P.Bosdogianni.数字图像处理疑难解析[M],赖剑煌,冯国灿等译.北京:机械工业出版社,2005.
[64]喻擎苍,翁秀娟,赵匀H.H.Cheng交互式开放结构计算机视觉平台 Ch OpenCV计算机工程与应用 2006.23
[65]Simon Haykin.自适应滤波器原理[M]郑宝玉译.电子工业出版社.2003.7
[66]董士崇,王天珍,许刚 视频图像中的运动检测[J].武汉理工大学学报.信息与管理工程版 2004Aug.Vol.26.No.4