基于HSV分块颜色直方图的图像检索算法

针对基于颜色的图像检索问题，提出了一种新的基于HSV非均匀量化分块颜色直方图的图像检索方法。分块直方图算法把图像固定分成若干子块，分别计算子块的颜色特征。将含有重要空间信息的分块设置权值，并将不注重空间信息的分块权值设为零：此算法充分利用了HSV非均匀量化和图像分块的优点，改善了颜色特征缺乏空间信息的缺点。研究结果表明，基于HSV非均匀量化分块颜色直方图的算法相对于全局颜色直方图有较好的检索精度。

An image retrieval algorithm based on HSV color segment histograms

Aiming at the image retrieval based on color feature, a new image retrieval algorithm based on HSV non-uniform quantization color segment histogram was presented. The basic idea of segment histogram was to divide image space into fixed blocks according to a certain strategy, and then the color histogram of each block was calculated as the color feature of the corre- sponding block. Users assigned a weight to each block according to the importance of the space information that the block contains, and assigned zero to the unimportant of the block. This algorithm made use of HSV non-uniform quantization and the block of image. It improved the traditional histogram for its lack of space information. Compared with traditional histogram method, the results indicate that the proposed algorithm based on HSV non-uniform quantization color segment histogram has better retrieval precision.