Abstract: A technique is presented for compressing data which leverages the frequency of an escape symbol for better compression. The prefix of a data string is evaluated and the probability of all characters that might succeed it is predicted in tabular form. Symbols are designated “Hit” or “Miss” based upon whether they are in the table. A binary tree is generated by partitioning nodes into Zero and One groups based on a single bit value. A partition bit is chosen to maximize the difference of probability sums of Hit symbols in Zero and One groups, with exceptions for partitions having non Hit symbols in one of the groups. A probability value is assigned to each branch, based on the probabilities of Hit and Miss symbols. Encoding or decoding a symbol is facilitated by encoding or decoding the branch probabilities on the shortest path from the root to the leaf node containing the symbol using arithmetic encoding or decoding method.

Abstract: A method of compressing digital representations of images stores the images in multiple subsampling quality scales in a progressive manner such that a higher quality scale contains only data incremental to the data in an adjacent lower quality scale. The data in each quality scale is transformed, quantized, and entropy encoded. A discrete wavelet cosine transformation combining frequency transformation features of a discrete cosine transformation and spatial transformation features of a wavelet transformation is provided. Sequences of subsampling quality scales are provided for multi-scale representations of images. A novel context selection method is used which treats color components and coefficients of different positions differently. An image compressed in a given quality scale can be transmitted or decompressed progressively allowing progressive downloading or transmission over the Internet.

Abstract: A method of compressing digital representations of images stores the images in multiple subsampling quality scales in a progressive manner such that a higher quality scale contains only data incremental to the data in an adjacent lower quality scale. The data in each quality scale is transformed, quantized, and entropy encoded. A discrete wavelet cosine transformation combining frequency transformation features of a discrete cosine transformation and spatial transformation features of a wavelet transformation is provided. Sequences of subsampling quality scales are provided for multi-scale representations of images. A novel context selection method is used which treats color components and coefficients of different positions differently. An image compressed in a given quality scale can be transmitted or decompressed progressively allowing progressive downloading or transmission over the Internet.