图像检索系统设计与研究
海 南 大 学
毕 业 论 文(设计)
题 目:学 号:姓 名:年 级:学 院:系 别:专 业:指导教师:完成日期:
图像检索系统的设计与研究 vv
www 2010级 vv vv vv
年 月 日
图像检索系统设计与研究
摘 要
本次研究主要是基于颜色和形状两种特征的图像检索系统的设计,其包括特征提取和相似度计算两个主要的技术,当颜色、形状等图像特征被提取后,形成了特征向量,图像检索系统的关键在于计算出数据库图像特征与图像检索的距离,当下使用较为广泛的相似度提取技术是基于欧氏距离函数来测度的。 图像检索系统采用visual C++ 6.0运行环境下开发实现,设置几种检索途径,颜色、形状以及综合两种物理特征的三种方法,输入待检索图像利用上述特征从图像库里面预先存在的多种图像检索出与之相似的图像,并按一定顺序排列出一组结果,将结果显示出来。
关键字:颜色空间转换;颜色特征;形状特征;不变矩;欧氏距离;OTSU法
图像检索系统设计与研究
Abstract
This research is mainly about color and shape of image retrieval system ,which is designed based on two kinds of features, including feature extraction and similarity calculation of two main technology, when the color, shape and other image features are extracted to form the feature vector, image retrieval system, the key is to calculate the database image features and image retrieval distance at present, using similarity extraction techniques are widely used to measure based on Euclidean distance function. Image retrieval system using visual C++ 6 runtime environment development environment, set several retrieval approach, three kinds of method of color, shape and two physical characteristics, input to the variety of image retrieval using the features from the image repository of pre-existing images are similar, and are arranged according to a certain order of a group the results, the results are displayed.
Key words:Color space conversion;color feature;shape feature;invariant moment ;Euclidean distance;OTSU method
图像检索系统的设计与研究毕业论文(设计)
![](/skin/haowen/images/icon_star.png)
![](/skin/haowen/images/icon_star.png)
![](/skin/haowen/images/icon_star.png)
![](/skin/haowen/images/icon_star.png)