An Efficient Sparse 8×8 Orthogonal Transform Matrix for Color Image Compression

Abstract

This paper presents an efficient orthogonal sparse 8×8 transform matrix for color image compression particularly at lower bit rate applications. The transform matrix is made sufficiently sparse by appropriately inserting additional zeros into the matrix proposed by Bouguezel. The algorithm for fast computation is also developed. It is shown that the proposed transform matrix provides a 7% reduction in computation over the matrix by Bouguezel, and 45% over signed discrete cosine transform (SDCT). By using various natural test images, it is shown that the rate-distortion performance is comparable with the matrix proposed by Bouguezel and approximated DCT at low bit rates. Further, it outperforms SDCT by a large margin almost at all bit rates.