Image compression using conjugate symmetric sequency-ordered complex hadamard transform [CS-SCHT]

Abstract

In this report, image compression using a new transform called Sequency-ordered Complex Hadamard Transform (CS-SCHT) is presented with simulation results and computer based implementation demo. This compression method is specifically introduced for reducing computational complexity, archiving reasonably high-quality images of scanned documents with lowering cost for future real-time implementation. Transform matrix of the CS-SCHT possesses sequency ordering and its spectrum is a conjugated symmetric. Comparison of energy compaction capabilities of commonly used transforms such as discrete cosine transforms (DCT), sequence ordered Walsh transform (SOWT) and CS-SCHT are given. Computational load of the CS-SCHT is also compared against the above mentioned commonly used transforms. Derivation and proof of its properties are discussed and analyzed with applications in image compression. Mean-square-error (MSE) and Peak Signal to Noise Ratio (PSNR) performances are considered as standard measures of image quality. Simulation results from MATLAB and low level programming language C# proved that the CS-SCHT is promising to be developed in such application.