Abstract: The paper presents a low bit-rate video compression algorithm for digital monochrome image sequence transmission. Presented method is based on wavelet transform and an original image segmentation algorithm for intra- and inter-frames. The proposed algorithm divides an image into rectangular blocks, which may overlap. The width and height of these blocks are set independently and can have optical values from a present range. Blocks are filled with a mean value of pixels form original image and their seizes are increased until the mean square error value for the block is smaller than the preset value. The algorithm is strongly parameterized, so it can be used for compressing the subimages created by the discrete wavelet transform. In 2D wavelet transform, an image is interpreted as a sum of details that appear at different resolution. As the result each subimage produced by WT can be compressed and transmitted separately to decrease the sensitivity to errors that may occur in the channel. The paper also presented the results obtained during off-line image compression. These results show better quality of restored images in compare to H.261 and H.263 algorithms. The presented algorithm achieves good restored image quality with high compression ratio and ability of simple hardware implementation. !15
展开▼
