A study on using the homogeneous texture descriptor in compressed domain

Abstract

A technique is proposed for generating the MPEG-7 descriptor in compressed image/video data. Image processing in transform domain is a much interesting area recently because compressed image and video data are becoming widely available with the data format like MPEG or JPEG. In general, processing in transform domain requires smaller data quantities, and lower computation complexity than that in spatial domain. In this paper, I propose a generation algorithm of the MPEG-7 metadata in compressed domain. I have developed an algorithm to get the homogeneous texture descriptor in compressed domain. The original homogeneous texture descriptor is generated in frequency domain. In this paper, proposed algorithms have extracted texture descriptor directly from a compressed domain, i.e., Block DCT domain. Experiments were performed with the MPEG-7 texture data set. First, all images in the data set are changed into the JPEG format. Then, I computed the MPEG-7 texture descriptors with a compressed image using the proposed method. For comparison, we also computed the descriptors in conventional frequency domain. Experimental results show that the computation time of the MPEG-7 descriptor using the proposed method is faster and more efficient than that of the conventional extraction method with compressed data. The proposed method can provide the efficient MPEG-7 metadata generation for compressed image/video in the contents based multimedia indexing/retrieval system.