红外视频动目标检测跟踪算法的研究及其DSP实现
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
红外视频动目标的检测跟踪是图像处理应用的一个重要领域。在计算机视觉、交通监控、可视预警、导航等诸多民用领域有着广泛的应用,同时在导弹电视制导等军用领域也发挥着重要作用。本论文所研究的感兴趣区域的动目标检测跟踪等技术,在第三代图像压缩技术中的视频信息检索等领域也都有着重要的应用。
    本论文首先根据系统的实时性特点,将系统分成四个基本的状态:搜索、捕获、跟踪和记忆。依照状态机制的激励和转换条件,实现四个状态的转换运行,从而达到动目标的实时检测与跟踪的目的。针对实际系统要求,进行了硬件和软件总体设计,并且讨论了系统硬件的各个功能模块和软件结构化分层设计。通过建立的动目标检测跟踪系统方程,探讨系统实现的关键技术,并且针对我们所要处理的红外视频图像序列的具体特点,我们提出基于背景运动估计和补偿的动目标检测跟踪关键技术。
    然后进一步研究了系统的四个基本状态的具体实现方案和技术路线,实现和分析四个状态模块的PC仿真系统和C6x系统,并给出了系统状态运行结果。
    论文最后研究了该系统的C6x技术实现及分析了实现的C6x系统数据流。本系统对相关研究所提供的多维图像数据进行交付性的试验,取得了良好的结果。系统实际的成功性实验也表明本论文提出的理论、技术及具体的实现设计是成功的。
Infrared Video Target Detection and Tracking System plays a great role in the field of digital image processing, and is widely used in civil and military affairs. The moving target detection techniques on the region of interesting (ROI) discussed in this dissertation can be also applied in image compression, information retrieval and interaction control by ROI.
    According to the real-time feature of this tracking system, it is divided into four states: search, capture, track and memory, and the states transition is fulfilled in the condition of the excitation of input and transition condition. According to system requirement, we design hardware and software system as a whole, and also discuss all modular functions of hardware system and software system architecture design at details. Through establishing the equation of moving target detection and tracking system, we find out the methods to solve the equation, and from the methods we come to discuss several crucial techniques of the implementation of system, and also according to the feature of the infrared video image sequence which we are processing and analyzing, we bring forward moving target detection and tracking technology based on background moving estimation and compensation.
    And then we go further into the implementation plan and technological path of the four fundamental states of this system, implement and analyze these four state model blocks on the PC simulation system and C6x system, also present the results.
    At last we discuss the DSP implementation of this system and analyze the data flow of DSP implementation. We have done the related experiments on the video sequences, which were provided by the relevant institute. The results of our experiments show that the system designed is correct, the key technologies have been implemented successfully in practical system and the design of system hardware and software is successful too.
引文
【1】 Shi Rong,Li Zaiming. "The Optical Flow Estimation with Displacement Compensation for High-speed Motion". IEEE communication, circuit and system International Conference,2002.6:595-599
    【2】 Qian Zhebin,Liu Xiaohong,Li Jie-gu J,Lingworth. "A Real-time System for Detecting Moving Point Target in Low SNR Image Sequences". IEEE International Conference on Intelligent Processing Systems,1997
    【3】 Ashraf Elnagar. "Prediction of Moving Objects in Dynamic Enviroments Using Kalman Filters". Proceedings of 2001 IEEE International Symposium on Computational Intelligence In Robotics and Automation,2001.7-9
    【4】 A.Del Bimbo. "Real-time Optical Flow Estimation". IEEE.
    【5】 Wouter Beck. "Image enhancement and moving target detection in IR image sequence". SPIE,Vol .2020,1993:187~195
    【6】 Demin Wang and Limin Wang. "Global Motion Parameters Estimation Using a Fast and Robust Algorithm". IEEE Trans on Circuits and Systems for Video Technology,1997,7(5):823~826
    【7】 K.R.Castleman. "Digital Video Processing" .Prentice Hall,1995
    【8】 郭旭平,李在铭. "跟踪系统中图像目标的快速提取技术研究" .电子科技大学学报,1998,27(1):17~21
    【9】 Vishal Markandey,Anthony Reid,Shenq Wang. "Motion Estimation for Moving Target Detection" .IEEE Transactons ON Aerospace and Electronic Systems,Vol.32,No.3:866~874
    【10】 Hyoungjun IM,Taejeong Kim."Optimization of Multiframe Target Dection Schemes".IEEE Transactons On Aerospace and Electronic Systems,Vol.35,No.1:176~187
    【11】 Yang Fan and Arjuna Balasuriya. "Optical Flow Based Speed Estimation in AUV Target Tracking" .MTS 0-933957-28-9:2377~2382
    张忠诚,孟庆华,沈振康. "红外目标特征分析" .激光与红外,1999.6,
    
    【12】 Vol.29,No.3
    【13】 魏波,李晓峰,李在铭. "任意指定图像目标的实时检测与跟踪".系统工程与电子技术,2000,22(5):61~64
    【14】 李金宗等. "图像序列的运动目标选择技术".电子学报,1998.9.26(9):34-39
    【15】 杨威,张田文,师海峰. "一种用于二值图像分割的快速聚类算法",计算机研究与发展,1998.8,Vol.35, No.8
    【16】 葛宝臻,秦玉文. "自适应运动目标跟踪系统及性能研究",天津大学学报, 1998.9,Vol.31, No.5
    【17】 赵红军. "背景运动估计与动目标检测" .西南民族学院学报自然科学版, 2001.2
    【18】 徐军,向健勇,林晓春,周翔. "背景预测方法在空中红外弱小目标检测、识别中的应用".西安电子科技大学学报,1998.8
    【19】 陈颖,李在铭. "运动背景中的动目标检测技术". 信号处理
    【20】 张建森,李宗杰,张桂林. "实时跟踪动目标的数学模型"
    【21】 李智勇,沈振康,杨卫平.《动态图像分析》.国防工业出版社,1998.8
    【22】 李在铭.《数字图像处理压缩与识别技术》.电子科技大学出版社,2000.11
    【23】 A.MURAT TEKALP.《数字视频处理》.电子工业出版社,1998.7
    【24】 K.R.Castlemen.《数字图像处理》.电子工业出版社,1998.8
    【25】 田捷,沙飞,张新生.《实用图像分析与处理技术》.电子工业出版社,北京,1994
    【26】 胡晓峰,李国辉.《多媒体系统》.人民邮电出版社,1997
    【27】 李在铭,张全芬,李晓峰. 《随机信号分析及工程应用》
    【28】 张雄伟.《DSP芯片的原理与开发应用》.电子工业出版社,1997
    【29】 任丽香,马淑芬,李方慧. 《TMS320C6000系列DSPs的原理与应用》.2000.7
    【30】 彭启琮,李玉柏.《DSP技术》.电子科技大学出版社,1997
    【31】 《TMS320C6000 Assembly Language Tools》.Texas Instruments.1999
    【32】 《TMS320C6202 CPU and Instruction Set》.Texas Instruments.1999
    【33】 《TMS320C6202 Programmer's Guide》.Texas Instruments.1999
    【34】 《TMS320C6202 Optimizing C Compiler》.Texas Instruments.1999

© 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