文档介绍：毕业设计(论文)
题目:基于中介真值程度度量的图像匹配方法研究
专业:
学生姓名:
班级学号:
指导教师:
指导单位:
日期:2008年 3 月 17 日至 2008年 6 月 18 日
摘要
图像匹配是指通过一定的匹配算法在两幅或多幅图像之间识别同名点,最终达到图像识别的目的。灰度图像匹配是基于像素的。本文首先对经典的灰度匹配算法进行了分析和实现,然后以中介真值程度的数值化度量为基础,通过引入距离比率函数来衡量图像像素点间的相似程度,建立了基准图像与输入图像某子区域的中介相似性测度,设计了基于中介真值程度度量的新的灰度图像匹配算法,并将其应用到动态图像序列识别和指纹图像识别中。经仿真实验证明,与经典的图像匹配算法结果比较,新的图像匹配算法能较好的进行模板匹配,并具有良好的抗噪性、抗失真性和较高的匹配精度和匹配速度。
关键词:图像匹配;灰度图像;中介真值程度度量;相似性测度
ABSTRACT
Image matching is to identify points of the same name in between two or more images through a matching algorithm for image recognition. Gray image matching algorithm is based on the pixels. In this paper, the classic gray matching algorithms are analyzed and plished firstly, next, based on the intermediary true extent of the value measure, the distance ratio function of image pixels is introduced to measure the degree of similarity between points, and the intermediation similar measure between the base image and one of the sub regions of input image is established, finally, the new gray image matching algorithm based on the true extent of intermediary measure is given, which is applied to dynamic image sequence and fingerprint image recognition. The simulation experiments show pared with the classic image matching algorithm results, the new image matching algorithm can match a template better, and has a good anti-noise, anti-distortion and high matching precision and speed.
Key words: image matching; gray image; intermediary true extent metric; similar measure
目录
第一章绪论 1
1
3
3
第二章图像匹配技术简介 5
5
6
6
7
7
10
10
11
12
14
14
14
15
第三章中介数学系统简介 16
16
16
18
19
第四章基于中介真值程度度量的图像匹配理论研究 20
20
21
22
第五章基于中介真值程度度量的图像匹配方法 23