基于分块颜色直方图的快速图像检索  

作　　者：陈迪凯 陈鹏[1] 

机构地区：[1]三峡大学计算机与信息学院,湖北宜昌443002

出　　处：《信息通信》2017年第12期85-87,共3页Information & Communications

基　　金：国家重点研发计划项目(编号:2016YFC0802503)资助

摘　　要：针对传统颜色直方图不能反映图像空间颜色位置信息的问题,提出一种基于分块颜色直方图的检索方法,该方法先对一幅图像进行合理的分块,然后提取包含了图像位置信息的分块颜色特征,对提取的颜色特征采用相似度计算的方法,计算出两幅图像之间的相似度。实验结果表明,该方法不仅提高了图像检索的速度,同时也提高了检索的准确度。The traditional color histogram canJt show theimage position and color information. To solve this problem, my dissertation come up with the searching method based on blocking color histogram. The dissertation presents blocking an image in a proper way and extracting the color characteristics of every block, including the location information of the image. As for the color characteristics, similarity calculation is used to calculate the similarities between two images. The experiment shows the result that the method presented in this dissertation has increased the speed and accuracy of the image retrieval.

关 键 词：图像检索 颜色直方图 相似度计算 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]