摘要基于计算机视觉的人机交互是自然人机交互技术领域的研究热点，它利用视觉信息从视频序列中捕获并理解人的肢体动作，使人与计算机之间的交互摆脱了鼠标及键盘的约束，人只需要通过肢体动作、手势、表情等自然信息就能与计算机进行交流，突破了从计算机诞生至今在人机交互过程中必须由人来适应机器的屏障，使人与计算机之间的交流更加自然、流畅。

本课题以手势识别为主要研究内容，在深入分析计算机视觉、人机交互、静态手势识别等方法基础上，研究手势动作与虚拟场景中三维物体的自然交互技术，并基于OpenCV库设计和实现了一个手势识别与虚拟三维空间人机交互系统。59673

毕业论文关键字  计算机视觉  人机交互  手势  识别

毕业设计说明书（论文）外文摘要

Title  Virtual Scene Interactive Technology Based on Gesture

Abstract

Human-computer interaction based on computer vision,which uses visual information captured from video sequences and understands the body movements,is a hotpot in natural Human-computer interaction technology area.It makes the interaction between human beings and the computer get rid of the constraints of mouse and keyboard.So people can communicate with the computer only by body movements, gestures, facial expressions and other natural information that breaks the barrier of people must adapt to the machine in the process of Human-computer interaction since the birth of the computer,which makes the communication between human beings and computers more natural and more smooth.

To take gesture recognition as the main contents,this task studied gestures and natural interaction techniques of three-dimensional objects in the virtual scenes on the basis of depth analysis of methods of computer vision,Human-computer interaction,recognition of the static- gestures,designed and implemented a gesture recognition and Human-computer interaction systems in virtual three-dimensional space based on the OpenCV library.

Keywords  Computer vision  Human-computer interaction  Gesture Recognition

1  引言（或绪论） 1

1．1  研究目的与意义 1

1．2  自然手势识别的研究现状 2

1．3  本论文的主要研究内容 3

2  自然手势识别技术的理论基础 4

2．1  自然手势识别技术综述 4

2．2  自然手势识别的处理流程 5

3  自然手势识别的算法实现 13

3．1  手势预处理的算法实现 13

3．2  手势分割的算法实现 14

3．3  模板的匹配的算法实现 16

4  基于手势的虚拟场景人机交互系统的设计与实现 18

4．1  系统框架 18

4．2  实验平台的搭建 20

4．3  系统的输入和输出 20

4．4  三维人机虚拟交互实例 22

5  总结与展望 24

5．1  全文总结 24

5．2  展望 24

结  论 26

致  谢