Abstract: A method for encoding a block of video signal which includes a plurality of frames, each of the frames is divided into a multiplicity of nonoverlapping blocks and each of the blocks contains K.times.K pixels, K being a positive integer. The method determines two reconstruction values and a bit plane for the block, scans K.times.K binary pixels included in the bit plane in a zigzag order for providing run-length and level values, encodes the run-length and the level values by using a predetermined variable length coding table, thereby providing a coded bit plane, determines a selection signal denoting either the bit plane or the coded bit plane which is represented with a fewer number of bits; choosing either the bit plane or the coded bit plane in response to the selection signal, and combines the two reconstruction values, and either of the bit plane and the coded bit plane which was selected.