基于信息熵和分块颜色矩的图像检索

Image retrieval based on information entropy and block color moment

提出一种基于信息熵和分块颜色矩的图像检索方法,采用HSV(hue,saturation,value)颜色空间进行非均匀模糊量化,改进的信息熵描述图像的全局颜色特征.通过对图像的重叠分块、计算各子块的颜色矩来反映图像的局部颜色特征,利用循环队列的数据结构使得固定分块具有旋转性,并采用加权欧氏距离计算图像相似度.实验结果表明,该方法具有较好的检索性能.

This paper proposes an image retrieval method based on information entropy and segment color moment. It selects the non-equal fuzzy quantification to quantize HSV color space. And the improved information entropy is used to describe the global color feature of an image. It adopts the thought of overlapping blocks, calculates color moment of each sub-block to reflect the local color feature of an image,and uses the data structure technique of circular queue to make the fixed block rotation-invariant. Euclidean distance with weights is used to calculate the similarity of images. The experimental results show that this method has a high retrieval performance.