Matlab视频序列中的目标跟踪技术研究

摘要目标跟踪一直是计算机视觉研究领域的热点问题，其应用领域有视频监控、人机用户接口、虚拟现实等等。视觉跟踪要求在光照变化、遮挡等各种因素的干扰下，能准确有效地跟踪不同背景条件下的目标，如何对视频序列中的目标进行稳健、有效跟踪一直是目标跟踪的研究重点。
目前，目标跟踪的方法有很多，如帧间差分法、基于光流场的目标跟踪、粒子滤波跟踪算法、基于Snake模型的视频跟踪等，而基于目标颜色直方图来对目标进行跟踪的Mean Shift算法，因方法简洁、实时性好、能处理目标变形以及部分遮挡等困难情形，而得到了广泛的运用。传统的Mean Shift跟踪算法以颜色直方图为特征对目标进行跟踪，颜色直方图具有旋转不变性，缩放不变性等优点。本毕业论文对视频跟踪系统进行了总体设计、并利用Matlab进行目标跟踪的编程仿真，实现能在中等复杂环境下实现较为稳定的目标跟踪，每秒跟踪15-30帧，即算法满足实时性。
关键词目标跟踪；粒子滤波；颜色直方图；Mean Shift；实时性9383
毕业设计说明书（论文）外文摘要
Title    Study of Object Tracking in Video Sequences
Abstract
Object Tracking has always been a hot issue in Computer Version research, its application area include Video Surveillance, Human-Machine, Virtual Reality and so on. It requires accurate and effective track in different background, light changes, object block and so on. Track the target for stable is a hot research issue in the field of object tracking.
   There are a lot of object tracking methods now, such as Inter Frame difference method, object tracking based on optical flow field, particle filter tracking algorithm, Snake model-based video tracking and so on. The Mean Shift tracking algorithm based on the Mean Shift use the color histogram to track object, because of its simpleness, perfect real-time performance, the ability to process the target deformation and the block situation and other difficult situation, and also the high accuracy, so the Mean Shift video tracking algorithm has been widely used. The traditional Mean Shift tracking algorithm uses the color histogram to track object.The color-histogram has some excellences-the rotation inflexibility and scale inflexibility and so on. In this article I achieve the overall design of t e video tracking system and using the Matlab programming simulation for object tracking, achieving in a moderately complex environment to achieve a more stable object tracking, track 15-30 frames per second, that the algorithm for real-time.
Key words: Object tracking; Particle filter; Color histogram; Mean Shift;
目录
1 绪论   1
1.1 视频跟踪算法研究的背景与意义   1
1.2 目标跟踪方法综述   3
1.2.1 基于运动分析的方法   3
1.2.2 基于特征的方法   3
1.2.3 基于变形模板的方法   4
1.2.4 基于模型的方法   4
1.3 视频目标跟踪中的难点问题   5
1.4 本文的研究内容及主要工作   6
1.5 论文结构   7
2 经典视频跟踪算法   8
2.1 帧间差分法   8
2.2 基于光流场的目标跟踪   8
2.3 Condensation跟踪算法   10
2.4 基于Snake模型的视频跟踪   14
2.5 本章总结   15
3 均值移动算法理论   17
3.1 引言   17
3.2 参数密度估计和无参数密度估计   17
3.2.1 参数密度估计   17
3.2.2 无参密度估计   18 Matlab视频序列中的目标跟踪技术研究:http://www.751com.cn/jisuanji/lunwen_8096.html