Efficient combination of color, texture and shape descriptor, using SLIC segmentation for image retrieval

Abstract

In this article we present a novel method of extraction and combination descriptor to represent image. First we extract a descriptor shape (HOG) from entire image, and in second we applied method of segmentation and then we extract the color and texture descriptor from each segment in order to have a local and global aspect for each image. These characteristics will be concatenate, stored and compared to those of the image query using the Euclidean distance. The performance of this system is evaluated with a precision factor. The results experimental show a good performance.