A New Coding System for Monochromatic Images Based on Wavelet Transform and Singular Value Decomposition (HDWTSVD)

In this paper the HDWTSVD algorithm to encode monochromatic images is proposed. The algorithm combines DWT and SVD techniques. The input image is divided into tiles of 64x64 pixels. A criterion based on the average standard deviation of 8x8 subblocks is used to choose DWT or SVD. If the tile exhibits a high average standard deviation, it is compressed by using SVD otherwise by DWT. Eigenvalues and eigenvectors are scalar and vector quantized respectively. If the MSE of the highest three eigenvectors, with respect to the vectors in the codebooks, is greater than a threshold then the eigenvectors are scalar quantized and sent, otherwise the indices of de codevectors are sent. The remaining eigenvectors are quantized using small codebooks. The maximum number of levels of subband decomposition for each tile is 3. Each subband is coded using the HC-RIOT. Each subband is coded using HC-RIOT. The proposed algorithm is applied to image coding and its performance is discussed.