基于感兴趣区域的图像检索方法

摘要

A novel image retrieval method is presented based on the region of interest( ROI) of an image. The interesting corners are firstly detected by multi-scale curvature polynomial. Then the method for extracting the ROI is introduced by using salient points as clues. Finally, it is used the color and texture features of the ROI for image retrieval. Experimental results show that the method can effectively describe the ROI and achieve good performance.%提出了一种新的基于感兴趣区域的图像检索算法.首先基于多曲率多项式提取图像显著点,并依据显著点提取图像感兴趣区域,然后基于感兴趣区域的颜色和纹理特征进行图像检索.实验结果表明该方法可有效地提取图像感兴趣区域,并取得了较好的检索效果.