Abstract: A method for compressing digital data from an original bit stream into a coded bit stream useful for facsimile data transmission and digitized document image storage. The method comprises the steps of dividing the original bit stream into a plurality of data blocks and then comparing each of the data blocks with a reference block having a predetermined bit pattern. A coded bit stream is formed that comprises data blocks, compressed blocks and index blocks according to the following procedure. When a data block in the original bit stream is a mismatch with the reference block, the data block is included as is, without modification, in the coded bit stream along with an index bit indicating a mismatch in the index block. When a data block in the original bit stream is matched to the reference block, then compressed block is included in the coded bit stream along with an index bit in the index block indicating that the data block is matched.