Lossless video data compressor with very high data rate

Abstract

Lossless video data compression is performed in real time at the data rate of incoming real time video data in a process employing a minimum number of computational steps for each video pixel. Each pixel 8-bit byte is converted to a difference byte representing the difference between the pixel and its immediate predecessor in a serialized stream of the pixel bytes. This reduces the dynamic range of the data. Any carry bits generated in the subtraction process of two's complement arithmetic are discarded. This reduces the data by a factor of two. Finally, the 8-bit difference pixel bytes thus produced are subject to a maximum entropy encoding process. Such a maximum entropy encoding process may be referred to as a minimum length encoding process.

1 . A method using a computer of processing a video signal, said video signal including a stream of pixel values representing intensity of pixels of successive video frames, each of said video frames corresponding to a two-dimensional array of pixels, said method comprising:
 converting said stream of pixel values into a stream of difference values by performing a subtraction between each pixel value and a preceding pixel value in said stream;   performing a maximum entropy encoding process on the difference values on a block by block basis;   the maximum entropy process including replacing difference values with corresponding characters in accordance with a code table for each block, so as to produce a stream of characters comprising a losslessly compressed version of said video signal.   
     
     
         2 . The method of  claim 1  further comprising writing said stream of characters and the corresponding code table for each block. 
     
     
         3 . The method of  claim 2  further comprising retrieving said stream of characters and said code table for each block. 
     
     
         4 . The method of  claim 3  further comprising:
 converting each of said characters to a corresponding difference value in accordance with said code table for each block; 
 computing from each of said difference bytes a corresponding pixel value by adding a current one of the difference bytes and a pixel byte computed from a preceding one of the difference bytes. 
 
     
     
         5 . The method of  claim 1  wherein said code table corresponds to a Huffman code table. 
     
     
         6 . The method of  claim 1 , wherein the subtraction between each pixel value and a preceding pixel value are for pixels in different frames. 
     
     
         7 . The method of  claim 6 , wherein the pixels in different frames are in the same frame location. 
     
     
         8 . The method of  claim 6 , wherein the pixels in different frames are in different frame locations. 
     
     
         9 . The method of  claim 8 , wherein the different frame locations correspond to the same location in a moving image. 
     
     
         10 . A method using a computer of processing a video signal comprised of a stream of pixel intensity values, comprising:
 for each pixel intensity value, determining a difference between the pixel intensity value and another pixel intensity value;   performing a maximum entropy encoding process on the differences between pixel intensity values to obtain symbols of reduced length for the differences between pixel intensity values; and   storing in memory the symbols of reduced length for the differences between pixel intensity values and a table resulting from the maximum entropy encoding process.   
     
     
         11 . The method of  claim 10 , wherein the symbols are minimum length symbols. 
     
     
         12 . The method of  claim 10 , wherein determining the difference between the pixel intensity value and the other pixel intensity value comprises determining the difference between a pixel intensity value for a pixel in a first frame and a pixel intensity value for a pixel in a second frame, the first frame and the second frame being different frames. 
     
     
         13 . The method of  claim 12 , wherein a location of the pixel in the first frame and a location of the pixel in the second frame are different locations. 
     
     
         14 . The method of  claim 13 , wherein the different locations correspond to the same location in a memory image. 
     
     
         15 . The method of  claim 10 , wherein the symbols obtained by the maximum entropy encoding process for frequently occurring differences between pixel intensity values are smaller than symbols obtained by the maximum entropy encoding process for less frequently occurring differences between pixel intensity values. 
     
     
         16 . The method of  claim 10 , wherein the differences between pixel intensity values is determined by subtracting one pixel intensity value from another pixel intensity value, and discarding any carry bit. 
     
     
         17 . A method using a computer of processing a losslessly compressed video signal, comprising:
 receiving a plurality symbols corresponding to losslessly compressed video;   receiving a table including at least some differences between pixel intensity values for the losslessly compressed video;   using the table to correlate the symbols to differences between pixel intensity values for the compressed video;   computing pixel intensity values for a plurality of pixels by adding a difference in pixel intensity value to a previously computed pixel intensity value; and   providing the pixel intensity values to a memory.