运动目标图像的获取、传输与跟踪
|
摘要
运动目标检测与跟踪涉及到计算机图像处理、视频图像处理、模式识别、以及人工智能等诸多领域,广泛地应用于军事、工业、生活等各个方面。对此方向开展研究,具有实际的应用价值。
本文研究了一定空域中缓慢降落目标的检测与跟踪。通过视频的采集与压缩,将观测目标的视频数据由无线传输链路发送到后端控制中心,在后端的图像处理计算机上实现对观测目标的检测和跟踪处理,同时实现对前端图像采集系统的控制。
首先,设计了图像采集与压缩的硬件平台。分析了系统的实际需要,确定了系统所需的镜头和摄像机的类型和参数。分析了原始数字图像的特点和传输系统的带宽限制,在保证一定图像质量的条件下,确定了图像的压缩格式为MPEG-4,并阐述了MPEG-4图像压缩格式的特点。
其次,建立了无线传输链路。根据前端图像的分辨率和帧率以及图像的压缩比,确定了图像传输的带宽;根据实际需要,设计了基于ISM波段的无线收发模块ADF7025的无线收发系统。围绕ADF7025划分了系统的各部分功能,给出了部分电路的工作原理图。
最后,实现图像的检测与跟踪。本文主要研究在简单背景下运动目标信息的获取和跟踪,介绍了图像序列中运动目标的检测方法,详细研究了目前运动目标检测的一些常用方法,包括相邻帧差法、背景差法、光流场法,分析比较了这些常用算法的优缺点。考虑到算法的稳定性和对干扰的抑制性,本文利用光流法来实现了对运动目标的检测和跟踪。
Moving target detection and tracking are related to computer image processing, video and image processing, pattern recognition and artificial intelligence, and widely used in military, industrial, and other aspects of life. This research on the field is of the actual value.
In this paper, the research is emphasized on a slow dropping target detection and tracking in certain airspace. The target video is acquired and compressed with a vidicon and a video-capture card DVP1412, and would be sent to a back-end control center with a wireless transmission link. And then video data are processed to detect and track the target. At the same time, the direction of the vidicon is modulated with a servo system.
First, the hardware platform is designed of image acquisition and compression. According to the actual need, the lens and parameters of a vidicon are analyzed and designed, and the bandwidth limitation of an image transmission system is discussed. Under image quality guaranteed, MPEG-4 is adopted for the image compression format, and the characteristic of MPEG-4 image compression format is described.
Second, a wireless link is designed for image data transmission. According to the front-end image resolution, frame rate and image compression ratio, the transmission bandwidth is determined. And according to actual needs, a wireless transceiver system is designed on the ISM-band wireless transceiver module ADF7025. The circuit schematic is given in detail.
Finally, the image detection and tracking are realized. The study is focused on acquisition and track of a moving target on a simple background and detection methods are introduced in this paper. Some common methods, such as the Inter-frame difference method, background subtraction method and optical flow method, are analyzed and compared their advantages and disadvantages. Taken into account the stability and the interference suppression of an algorithm, optical flow is adopted to achieve the detecting and tracking of a moving target in the paper.
本文研究了一定空域中缓慢降落目标的检测与跟踪。通过视频的采集与压缩,将观测目标的视频数据由无线传输链路发送到后端控制中心,在后端的图像处理计算机上实现对观测目标的检测和跟踪处理,同时实现对前端图像采集系统的控制。
首先,设计了图像采集与压缩的硬件平台。分析了系统的实际需要,确定了系统所需的镜头和摄像机的类型和参数。分析了原始数字图像的特点和传输系统的带宽限制,在保证一定图像质量的条件下,确定了图像的压缩格式为MPEG-4,并阐述了MPEG-4图像压缩格式的特点。
其次,建立了无线传输链路。根据前端图像的分辨率和帧率以及图像的压缩比,确定了图像传输的带宽;根据实际需要,设计了基于ISM波段的无线收发模块ADF7025的无线收发系统。围绕ADF7025划分了系统的各部分功能,给出了部分电路的工作原理图。
最后,实现图像的检测与跟踪。本文主要研究在简单背景下运动目标信息的获取和跟踪,介绍了图像序列中运动目标的检测方法,详细研究了目前运动目标检测的一些常用方法,包括相邻帧差法、背景差法、光流场法,分析比较了这些常用算法的优缺点。考虑到算法的稳定性和对干扰的抑制性,本文利用光流法来实现了对运动目标的检测和跟踪。
Moving target detection and tracking are related to computer image processing, video and image processing, pattern recognition and artificial intelligence, and widely used in military, industrial, and other aspects of life. This research on the field is of the actual value.
In this paper, the research is emphasized on a slow dropping target detection and tracking in certain airspace. The target video is acquired and compressed with a vidicon and a video-capture card DVP1412, and would be sent to a back-end control center with a wireless transmission link. And then video data are processed to detect and track the target. At the same time, the direction of the vidicon is modulated with a servo system.
First, the hardware platform is designed of image acquisition and compression. According to the actual need, the lens and parameters of a vidicon are analyzed and designed, and the bandwidth limitation of an image transmission system is discussed. Under image quality guaranteed, MPEG-4 is adopted for the image compression format, and the characteristic of MPEG-4 image compression format is described.
Second, a wireless link is designed for image data transmission. According to the front-end image resolution, frame rate and image compression ratio, the transmission bandwidth is determined. And according to actual needs, a wireless transceiver system is designed on the ISM-band wireless transceiver module ADF7025. The circuit schematic is given in detail.
Finally, the image detection and tracking are realized. The study is focused on acquisition and track of a moving target on a simple background and detection methods are introduced in this paper. Some common methods, such as the Inter-frame difference method, background subtraction method and optical flow method, are analyzed and compared their advantages and disadvantages. Taken into account the stability and the interference suppression of an algorithm, optical flow is adopted to achieve the detecting and tracking of a moving target in the paper.
引文
[1]章毓晋.图像处理与分析.北京:清华大学出版社.1993年3月.
[2]黄贤武.数字图像处理与压缩解码技术.成都:电子科技大学出版社,2000年.
[3]冈萨雷斯.数字图像处理.第二版.北京:电子工业出版社,2004年6月.
[4]朱志刚等译著.数字图象处理.北京:电子工业出版社,1998.
[5]McKenna S,Jabri S,Dude Z,Rosenfeld A,Wechsler H.Tracking Groups of People.Computer Vision and Image Understanding,2000,80(1):42-56.
[6]Arseneau.S,Cooperstock J.Real-Time Image Segmentation for Action Recognition.Proc.IEEE Pacific Rim Conference on Communications,Computers and Signal Processing,Victoria,Canada,1999,86-89.
[7]Sun H,Feng T,Tan T.Robust extraction of moving objects from image sequences.In:Proc the Fourth Asian Conference on Computer Vision,Taiwan,2000,961-964.
[8]张江山,朱光喜.一种基于Kalman滤波的视频对象跟踪方法.中国图像图形学报,Vol.7(A),NO.6.
[9]万文静.基于光流的图像跟踪方法研究西北工业大学硕士论文2006.
[10]熊国清,于起峰.用于实时跟踪的快速匹配算法.计算机辅助设计与图形学报,2002,14(1):41-43.
[11]刘铁军,肖颖杰等.一种改进的区域相关跟踪方法.山东科技大学学报(自然科学版),2001,20(3):44-48.
[12]杨响,张桂林.一种新的相关跟踪算法的设计与实现.红外与毫米波学报,2000,19(5):377-380.
[13]张岩,崔智杜等.图像序列中机动目标的形心跟踪.航空学报,2001,2(4):312-316.
[14]张嫣,姚耀文等.运动图像序列中多目标跟踪的研究与实现.计算机应用研究,2002,(1):74-76.
[15]张坤华,王敬儒等.基于扩展目标的不变矩跟踪算法.强激光与粒子束,2002,14(1):6-10.
[16]McKerma.S.J Raja.Y.Tracking color objects using adaptive mixture models.Image andVision Computing,1999(17):225-231.
[17]Mazor.E,Averbuch.A.Interacting multiple model methods in target tracking:a survey.IEEE Trans.On Aerospace and Electronic Systems,1998,34(1):103-123.
[18]Toyama.K,Blake.A.Probabilistic Tracking with Exemplars in a Metric Space International Journal of Computer Vision,2002,48(1):9-19.
[19]谭仕勇,戴琼海,陈慧蓉MPEG-4视频的实时传输 计算机工程,2001年第27卷,第11期,141-142页.
[20]钟玉琢,王琪,贺玉文.基于对象的多媒体数据压缩编码国际标准—MPEG-4及其校验模型.2000.北京:科学出版社.
[21]Image Communication Tutorial Issue on MPEG-4,December,1999.
[22]MPEG Video Group.The MpEG-4 video standard verification models IEEE Trans.Circuits and systems for video technology,vol.7,no.l,February1997.
[23]ISO/IEC 14496-2:2000 MPEG-4 Visual[s].
[24]张宗念,余英林,马义德.MPEG-4的视频“框架”及其“等级”技术.电视技术.第209期.1999年11月.
[25]www.advantech.com.
[26]刘志东.3G视频编解码器软件设计和优化实现 浙江大学硕士论文.
[27]黄智伟.无线通信集成电路北京:北京航空航天大学出版社,2005.
[28]www.analog.com.
[29]高文,陈熙霖.计算机视觉.北京:清华大学出版社.1999.
[30]王全.基于差分图像的多运动目标的检测与跟踪.中国图像图形学报,Vol.4(A),1999.
[31]汪亚明,楼正国等.一种非刚体运动图像序列的特征点对用方法.中国图像图形学报.Vol.5(A),No.3 March.2000.
[32]任金昌等.一种基于自适应阈值的复杂背景下自动目标跟踪方法.计算机应用研究,第四期.2002.
[33]周西汉,刘勃,周荷琴.一种基于对称差分和背景消减的运动检测方法.计算机仿真,2005,22(4):117-119.
[34]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,中国图像图形学报,第4卷(A版)第6期,1999,6,470-475.
[35]R.Cutler and L.Davis,View-based detection and analysis of Periodic Motion,International Conference Pattern Recognition,Brisbane,Australia,1998,490-500.
[36]G.P.Stein,Tracking from multiple view Points:Self-calibration of spaceman Time tracking,Computer Vision and Pattern Recognition,Fort Collins,Colorado,1999,521-527.
[37]Stauff.C,Grimson.WEL Adaptive Background Mixture Models for Real-Time Tracking IEEE Computer Society Conference on Computer Vision and Patten Recognition,June 1999 vol.2:23-25.
[38]马颂德,张正友.计算机视觉计算理论与算法基础.北京:科学出版社,1998.235-239.
[39]B.K.Hom,B.G.Schunck,Determining Optical Flow,A117(1951)185-203.G.
[40]D.Terzopoulos,Regularization of Inverse Visual problems Involving Discontinuities,IEEE Trans.PAMI,8(1986),413-424.
[41]H.H.Nagel,On the Estimation of Optical Flow:Relation Between Different Approaches and Some New Results,Al,33(1987),299-324.
[42]Verri A,Poggio T.Against quantitative optical flow.Proc,of First Conference Computer Vision[C],London,England,IEEE,1987.171-180.
[43]许长桥,黄明和.DirectShow中构建过滤器图表技术。计算机应用研究,2002,12(2):156-157.
[2]黄贤武.数字图像处理与压缩解码技术.成都:电子科技大学出版社,2000年.
[3]冈萨雷斯.数字图像处理.第二版.北京:电子工业出版社,2004年6月.
[4]朱志刚等译著.数字图象处理.北京:电子工业出版社,1998.
[5]McKenna S,Jabri S,Dude Z,Rosenfeld A,Wechsler H.Tracking Groups of People.Computer Vision and Image Understanding,2000,80(1):42-56.
[6]Arseneau.S,Cooperstock J.Real-Time Image Segmentation for Action Recognition.Proc.IEEE Pacific Rim Conference on Communications,Computers and Signal Processing,Victoria,Canada,1999,86-89.
[7]Sun H,Feng T,Tan T.Robust extraction of moving objects from image sequences.In:Proc the Fourth Asian Conference on Computer Vision,Taiwan,2000,961-964.
[8]张江山,朱光喜.一种基于Kalman滤波的视频对象跟踪方法.中国图像图形学报,Vol.7(A),NO.6.
[9]万文静.基于光流的图像跟踪方法研究西北工业大学硕士论文2006.
[10]熊国清,于起峰.用于实时跟踪的快速匹配算法.计算机辅助设计与图形学报,2002,14(1):41-43.
[11]刘铁军,肖颖杰等.一种改进的区域相关跟踪方法.山东科技大学学报(自然科学版),2001,20(3):44-48.
[12]杨响,张桂林.一种新的相关跟踪算法的设计与实现.红外与毫米波学报,2000,19(5):377-380.
[13]张岩,崔智杜等.图像序列中机动目标的形心跟踪.航空学报,2001,2(4):312-316.
[14]张嫣,姚耀文等.运动图像序列中多目标跟踪的研究与实现.计算机应用研究,2002,(1):74-76.
[15]张坤华,王敬儒等.基于扩展目标的不变矩跟踪算法.强激光与粒子束,2002,14(1):6-10.
[16]McKerma.S.J Raja.Y.Tracking color objects using adaptive mixture models.Image andVision Computing,1999(17):225-231.
[17]Mazor.E,Averbuch.A.Interacting multiple model methods in target tracking:a survey.IEEE Trans.On Aerospace and Electronic Systems,1998,34(1):103-123.
[18]Toyama.K,Blake.A.Probabilistic Tracking with Exemplars in a Metric Space International Journal of Computer Vision,2002,48(1):9-19.
[19]谭仕勇,戴琼海,陈慧蓉MPEG-4视频的实时传输 计算机工程,2001年第27卷,第11期,141-142页.
[20]钟玉琢,王琪,贺玉文.基于对象的多媒体数据压缩编码国际标准—MPEG-4及其校验模型.2000.北京:科学出版社.
[21]Image Communication Tutorial Issue on MPEG-4,December,1999.
[22]MPEG Video Group.The MpEG-4 video standard verification models IEEE Trans.Circuits and systems for video technology,vol.7,no.l,February1997.
[23]ISO/IEC 14496-2:2000 MPEG-4 Visual[s].
[24]张宗念,余英林,马义德.MPEG-4的视频“框架”及其“等级”技术.电视技术.第209期.1999年11月.
[25]www.advantech.com.
[26]刘志东.3G视频编解码器软件设计和优化实现 浙江大学硕士论文.
[27]黄智伟.无线通信集成电路北京:北京航空航天大学出版社,2005.
[28]www.analog.com.
[29]高文,陈熙霖.计算机视觉.北京:清华大学出版社.1999.
[30]王全.基于差分图像的多运动目标的检测与跟踪.中国图像图形学报,Vol.4(A),1999.
[31]汪亚明,楼正国等.一种非刚体运动图像序列的特征点对用方法.中国图像图形学报.Vol.5(A),No.3 March.2000.
[32]任金昌等.一种基于自适应阈值的复杂背景下自动目标跟踪方法.计算机应用研究,第四期.2002.
[33]周西汉,刘勃,周荷琴.一种基于对称差分和背景消减的运动检测方法.计算机仿真,2005,22(4):117-119.
[34]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,中国图像图形学报,第4卷(A版)第6期,1999,6,470-475.
[35]R.Cutler and L.Davis,View-based detection and analysis of Periodic Motion,International Conference Pattern Recognition,Brisbane,Australia,1998,490-500.
[36]G.P.Stein,Tracking from multiple view Points:Self-calibration of spaceman Time tracking,Computer Vision and Pattern Recognition,Fort Collins,Colorado,1999,521-527.
[37]Stauff.C,Grimson.WEL Adaptive Background Mixture Models for Real-Time Tracking IEEE Computer Society Conference on Computer Vision and Patten Recognition,June 1999 vol.2:23-25.
[38]马颂德,张正友.计算机视觉计算理论与算法基础.北京:科学出版社,1998.235-239.
[39]B.K.Hom,B.G.Schunck,Determining Optical Flow,A117(1951)185-203.G.
[40]D.Terzopoulos,Regularization of Inverse Visual problems Involving Discontinuities,IEEE Trans.PAMI,8(1986),413-424.
[41]H.H.Nagel,On the Estimation of Optical Flow:Relation Between Different Approaches and Some New Results,Al,33(1987),299-324.
[42]Verri A,Poggio T.Against quantitative optical flow.Proc,of First Conference Computer Vision[C],London,England,IEEE,1987.171-180.
[43]许长桥,黄明和.DirectShow中构建过滤器图表技术。计算机应用研究,2002,12(2):156-157.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |