METHOD FOR COMPRESSING IMAGE SIGNALS

Original image signal components representing an image are subjected to component number decreasing processing wherein, along each of block lines which are set in parallel on an image, blocks each comprising a single picture element in the image are set at predetermined intervals, or blocks each comprising several picture elements are set continuously or at predetermined intervals, such that the phases of the blocks located along a block line are shifted from the phases of the blocks located along a neighboring block line. Only the respresentative image signal components determined from the image signal components representing the picture elements in the respective blocks are sampled as new image signal components. The new image signal components are classified into main components, which have been sampled at appropriate sampling intervals, and interpolated components other than the main components. The interpolated components are subjected to interpolation prediction encoding processing based on the main components.