Image Retrieval Based on HSV Feature and Edge Direction Feature
Yanxue Dong · 2015
With the development of multimedia and Internet technology, content-based image retrieval (CBIR) systems have been widely used.However, how to automatically detect and target images in an image database still remains challenging.Since the HSV color space is more benefiting with human visual perception mechanism, we convert the image from RGB color space to HSV color space firstly, obtain the color structure map and edge direction index map and then get the color feature and edge direction feature, finally accomplish the retrieval.Experimental results in Corel_10000 data set show that the proposed algorithm performs better both on precision rate and recall rate compared with two classic algorithms of image retrieval.