图像技术的人类识别.docVIP

PAGE 27 图像技术的人类识别 摘 要 当今社会是一个科技技术和经济飞速发展的社会，图像处理技术水平也在砥砺前行。随着科技进步和生活现代化程度的提高，图像技术的应用也变的更加成熟。当遇到危险状况或自然灾害时，利用图像处理技术提高图像检测的精确度，快速有效的识别出人类，将是图像检测技术未来的发展方向。图像技术检测人体和轮廓面临着许多的困难 ，包括人型态的多样性，穿衣服的不同，拍摄的天气因素，拍摄的角度变化以及所处的背景不同等多个方面。文本通过分析图像分割和目标轮廓提取的一些基本概念和算法，然后详细的阐述了视频监控中检测识别域内的人类方法，经过单高斯背景建模、目标检测这样的两个步骤进行，最后通过MATLAB对系统设计进行了分析仿真实验。 关键词：背景差分；运动目标检测；图像分割；目标提取；MATLAB Image technology for human recognition Abstract Todays society is a society with rapid development of science and technology and economy, and the level of image processing technology is also forging ahead. With the progress of science and technology and the improvement of life modernization, the application of image technology has become more mature. When encountering dangerous situations or natural disasters, it will be the future development direction of image detection technology to use image processing technology to improve the accuracy of image detection and quickly and effectively identify human beings. There are many difficulties in detecting the human body and silhouette with image technology, including the diversity of human types, the difference in clothing, the weather factors, the Angle changes and the different backgrounds. The text analyzes some basic concepts and algorithms of image segmentation and target contour extraction, and then elaborates the human method in the detection and recognition domain in video surveillance in detail. After the two steps of single gaussian background modeling and target detection, the system design is analyzed and simulated by MATLAB. Keywords: 背景差分；运动目标检测；图像分割；目标提取；MATLAB 目 录 TOC \o 1-3 \h \u 8028 1前言 1 27083 1.1本设计的简介和意义 1 6471 1.2本设计的发展概况及存在的问题 2 2983 2图像预处理技术 4 27263 2.1图像技术的过程 4 17093 2.2 图像灰度化 4 22392 2.3图像几何变换 6 3327 2.4图像增强 7 27896 3图像分割技术 9 30991 3.1图像分割技术定义 9 1975 3.2基于阈值的分割方法 10 24103 4图像轮廓提取技术 12 16909 4.1Roberts算法的轮廓提取方法 12 21865 4.2Canny算法的轮廓提取算法 15 16407 5图像识别技术 17 11260 5.1目标识别 17 22004 6实验设计 22 31430