Comparative Performance of SVD and Adaptive Cosine Transform in Coding Images
N. Garguir · IRE Transactions on Communications Systems · 1979
An adaptive coding for singular value decomposition (SVD) samples is described in order to reduce the amount of transmitted data for monochrome images. This method is compared to the discrete cosine transform in terms of mean square error. Low bit rates, as 0.8 bits/pixel, are attempted with a good visual quality and low RMS error (0.087%). The visual and mathematical results of the transmission errors are shown and given.