HVS-weighted progressive image transmission using the lapped orthogonal transform

Progressive transmission of images based on lapped orthogonal transform (LOT), adaptive classification, and human visual sensitivity (HVS) weighting is proposed. HVS weighting for LOT is developed using a general technique that can be applied to any orthogonal transform. The order in which the transform coefficients are transmitted is based on the estimated variances of these coefficients weighted by the human visual system sensitivity, measured in the 2D-LOT domain. The method is compared with discrete cosine transform (DCT)-based progressive image transmission (PIT). It is shown that LOT-based PIT yields subjectively improved images compared to those based on DCT.<>