Quadtree-based reversible integer time-domain lapped transform for progressive lossy-to-lossless images compression

Abstract. Reversible integer time-domain lapped transform (RTDLT) has been proven to be efficient in progressive lossy-to-lossless image compression. We introduced the quadtree-based RTDLT (QRTDLT) to further improve the performance of block-transform methods.Traditional block transforms are usually conducted with a fixed size of 8×8 or 16×16. This is an oversimplified and hardware-friendly implementation, but it is not optimal for different areas of various images. In the proposed QRTDLT method, we first partitioned the image into macroblocks. Then, the macroblocks are iteratively processed using quadtree splitting while abiding by special regulation. These blocks are transformed by RTDLTs with different-length basis functions. With limited complexity increase, the QRTDLT is still suitable for online applications of image compression. Simulation results demonstrated that the proposed method yields more than 0.5-dB performance improvement in lossy compression. At the same time, the QRTDLT also performs better than the RTDLT in lossless compression.

[1]  Trac D. Tran,et al.  Context-based entropy coding of block transform coefficients for image compression , 2001, Optics + Photonics.

[2]  Pengwei Hao,et al.  Matrix factorizations for reversible integer mapping , 2001, IEEE Trans. Signal Process..

[3]  T. Ebrahimi,et al.  The Jpeg Xr Image Coding Standard , .

[4]  Gary J. Sullivan,et al.  Overview of the High Efficiency Video Coding (HEVC) Standard , 2012, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  Ricardo L. de Queiroz,et al.  Extended lapped transform in image coding , 1995, IEEE Trans. Image Process..

[6]  Philippe Michel Cassereau.  A new class of optimal unitary transforms for image processing , 1985 .

[7]  Henrique S. Malvar.  Biorthogonal and nonuniform lapped transforms for transform coding with reduced blocking and ringing artifacts , 1998, IEEE Trans. Signal Process..

[8]  Truong Q. Nguyen,et al.  Multiplierless approximation of transforms with adder constraint , 2002, IEEE Signal Processing Letters.

[9]  Saverio Salzo,et al.  Lossless hyperspectral compression using KLT , 2004, IGARSS 2004. 2004 IEEE International Geoscience and Remote Sensing Symposium.

[10]  A NikaraJari,et al.  Discrete cosine and sine transforms , 2006 .

[11]  Iso-Iec Jtc Sc Wg,et al.  FCD14495, lossless and near-lossless coding of continuous tone still images ({JPEG-LS}) , 1997 .

[12]  Truong Q. Nguyen,et al.  The GenLOT: generalized linear-phase lapped orthogonal transform , 1996, IEEE Trans. Signal Process..

[13]  Faouzi Kossentini,et al.  Reversible integer-to-integer wavelet transforms for image compression: performance evaluation and analysis , 2000, IEEE Trans. Image Process..

[14]  Guangming Shi,et al.  Lossy-to-lossless image compression based on multiplier-less reversible integer time domain lapped transform , 2010, Signal Process. Image Commun..

[15]  Michael T. Orchard,et al.  A comparative study of DCT- and wavelet-based image coding , 1999, IEEE Trans. Circuits Syst. Video Technol..

[16]  Feng Wu,et al.  Lifting-Based Directional DCT-Like Transform for Image Coding , 2007, IEEE Trans. Circuits Syst. Video Technol..

[17]  Feng Wu,et al.  Adaptive Directional Lifting-Based Wavelet Transform for Image Coding , 2007, IEEE Transactions on Image Processing.

[18]  William A. Pearlman,et al.  Efficient, low-complexity image coding with a set-partitioning embedded block coder , 2004, IEEE Transactions on Circuits and Systems for Video Technology.

[19]  Feng Wu,et al.  Directional Lapped Transforms for Image Coding , 2008, DCC.

[20]  Masaaki Ikehara,et al.  Integer fast lapped transforms based on direct-lifting of DCTs for lossy-to-lossless image coding , 2013, EURASIP J. Image Video Process..

[21]  Henrique S. Malvar.  Lapped transforms for efficient transform/subband coding , 1990, IEEE Trans. Acoust. Speech Signal Process..