Local extrema co-occurrence pattern for color and texture image retrieval

Abstract

A real world problem of image retrieval and searching is considered in this paper. In modern generation, managing images from a large storage medium is not a straightforward job. Many researchers have worked on texture features, and produced diverse feature descriptors based on uniform, rotation invariant, edges and directional properties. However, most of them convert the relationship of the center pixel and the boundary pixel into a local pattern, and use histogram to represent the local pattern as a feature vector. In this work, we propose a new image retrieval technique; local extrema co-occurrence patterns (LECoP) using the HSV color space. HSV color space is used in this method to utilize the color, intensity and brightness of images. Local extrema patterns are applied to define the local information of image, and gray level co-occurrence matrix is used to obtain the co-occurrence of LEP map pixels. The local extrema co-occurrence pattern extracts the local directional information from local extrema pattern, and convert it into a well-mannered feature vector with use of gray level co-occurrence matrix. The presented method is tested on five standard databases called Corel, MIT VisTex and STex, in which Corel database includes Corel-1k, Corer-5k and Corel-10k databases. Also, this algorithm is compared with previous proposed methods, and results in terms of precision and recall are shown in this work.