基于分块主色调和纹理特征的彩色图像检索

Image Retrieval based on Main Color and Texture Feature of Each Block

提出一种基于分块主色调和纹理特征的彩色图像检索方法。该方法首先提取占彩色图像比重较大的前几种主色调,根据主色调将图像划分为不同分块,并以主色调作为各分块图像的颜色特征。然后,采用灰度共生矩阵算法,提取各分块的纹理特征。综合考虑各分块图像的颜色特征和纹理特征,从图像数据库中检索相似图像。实验结果表明,该方法具有较高的检索效率。

A new image retrieval algorithm based on the main color and texture of image blocks was presented. Firstly, the main kinds of color of each image were extracted as the image＇ s color feature. Then images were partitioned into sub _ images. Secondly, the texture of each block was extracted by gray symbiotic matrix algorithm as the image＇ s texture feature. Color feature and texture feature were considered to retrieve similar images. Experiment results show that the new algorithm is more effective in image retrieval.