MODIFIED HUFFMAN ENCODE/DECODE SYSTEM WITH SIMPLIFIED DECODING FOR IMAGING SYSTEMS

The K most frequently occurring symbols in an image represented by an alphabet of N symbols are mapped to a set of K codewords. The length of each codeword is an inverse function of the frequency of occurrence of the corresponding symbol in the image and the longest codeword is of length d bits. The remaining N-K symbols are mapped in order of their magnitude to a set of supplementary codewords of a uniform maximum length D bits. The d most significant bits of each supplementary codeword form a prefix which is uniform within each set and specifies whether the corresponding symbol is positive or negative. The remaining or least significant D-d bits of each supplementary codeword comprise a suffix and are mapped in order of bit position to progressively more narrow ranges of symbol values of the remaining N-K symbols.