The application of images and video has increased significantly in recent years. It is crucial to develop indexing techniques for searching images and video based on their content. Due to the lower computational complexity, compressed domain indexing techniques are becoming popular. Among the compression techniques, discrete-wavelet-transform based techniques have become popular because of its excellent energy compaction and multi-resolution capability, and has been adopted in the JPEG2000 image compression standard. In this paper, an efficient indexing scheme in the JPEG2000 framework is proposed. Here, the information on the numbers of bit-planes in code-blocks will be directly extracted from the packet header of a JPEG2000 compressed image without decompressing the bitstream. A vector comprising variances and means of a few bit-planes of the code-blocks is used as the image index. Image retrieval is performed by matching the index of the query and candidate images from the database. Experimental results show that the proposed scheme provides a good indexing performance.
