一种综合颜色与边界方向直方图的图像检索方法的研究

Study on Image Retrieval Method Based on Color and Edge Direction Histogram

分析总结现有的基于内容的检索技术的基础上,通过比较及实验证明,提出了利用颜色直方图以及边界直方图进行图像检索的方法。为了颜色特征以及边界特征结合起来进行图像检索,提出对所提取的不同图像特征,定义不同的相似性度量方法来计算图像间的相似程度。本文中结合不同的子特征进行图像检索,两个图像之间的相似度定义为子特征间相似度的加权和。这样将两种特征有效的结合起来达到更好的检索效果。最后通过结合数据库以及MATLAB技术制作出一个系统模型,通过实验证明了结合颜色直方图和边界方向直方图进行检索的优越性和通用性。

Analyze and summarize the existing content-based retrieval technology based on mainly through comparison and experiments show that the proposed integrated use of color histogram and edge histogram image retrieval method. In order to color feature and edge fea- ture of image retrieval, is proposed for different image features extracted from the different definitions of the similarity measure to calculate the similarity between images. In order to retrieval images with different characteristics, the definition of similarity between two images as a weighted sum of sub feature similarity. The characteristics of two kinds of effective combination to achieve better retrieval effect. Combined with database and MATLAB technology to produce a system model, proved by experiments combined with the color histogram and edge di- rection histogram retrieval superiority and universality.