摘要近年来，复杂背景下识别和跟踪运动目标，动态目标空间位置的测量和自动跟踪等研究引起国内外厂商和学者的重视并进行了一系列探索性的研究。本论文对Camshift的算法和TLD（tracking learning detection）跟踪算法进行了实现以及实验和研究。61150

Camshift算法是MeanShift算法的修改,是一种灵活的运用统计调查方法来寻找分布概率的模式。这是一个非常快捷的跟踪方法,因为Camshift算法跟踪物体的中心和大小的概率分布,前提是你提出的目标有良好的概率分布。这种算法充分的体现了目标跟踪及时性的特点，在动态图像中使用它来跟踪运动的物体方便而且快速。

TLD（tracking learning detection）算法其本质是一个框架，它将跟踪模块，机器学习模块和检测模块有机的结合在了一起，形成了一种高效的跟踪方法。其精度很高，可以实现实时跟踪，当目标离开摄像头范围再回来时又会被准确的检测到，继续跟踪。

毕业论文关键词   视频监控，Camshift，TLD，目标跟踪。 

毕业设计说明书（论文）外文摘要

Title   Design and implementation of target tracking algorithm in video surveillance                                             

Abstract In recent years, complex background to identify and track moving targets, dynamic target spatial position measurement and automatic tracking study attracted domestic and foreign manufacturers and attention of scholars and conducted a series of exploratory research. In this thesis, algorithm for TLD and Camshift tracking algorithm has been implemented and experimental and research.

Camshift algorithm is modification of MeanShift algorithm，is a flexible use of survey methods to find the probability distribution model. This is a very fast track approach because Camshift algorithm tracking the size of the object and the center of the probability distribution, provided that the goals you have a good probability distribution. This algorithm fully embodies the characteristics of timeliness target tracking in dynamic image using it to track moving objects easy and fast.

TLD algorithm by its very nature is a framework that will track modules, machine learning module and a detection module organically combine together to form an efficient tracking method. Its high accuracy, can achieve real-time tracking, leave the camera when the target range and then come back again be accurately detected, continue to follow.

Keywords    camshift,TLD,Target tracking

1 引言（或绪论） 1

1.1 应用领域 1

1.2工作重点 2

1.3本文的安排 2

2 图象处理 3

2.1 图像处理技术 3

2.1.1图像增强 3

2.1.2图像平滑 4

2.1.3图像数据的编码和传输 4

2.1.4边缘锐化 5

2.1.5图像分割 5

2.1.6图像识别 5

2.2机器视觉 6

2.2.1机器视觉的概念 6

2.2.2机器视觉的发展及应用 7

2.3 机器视觉系统 8

2.3.1视觉系统概念 8

2.3.2图像的获取