Coded data generation or conversion – Digital code to digital code converters – To or from bit count codes
Reexamination Certificate
1999-12-10
2001-01-09
JeanPierre, Peguy (Department: 2819)
Coded data generation or conversion
Digital code to digital code converters
To or from bit count codes
C341S050000
Reexamination Certificate
active
06172624
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATIONS
U.S. patent application Ser. No. 09/264,269 by Cooper, filed Mar. 8, 1999, entitled “Data Compression Method And Apparatus With Embedded Run-Length Encoding”.
U.S. patent application Ser. No. 09/300,810 by Cooper, filed Apr. 27, 1999, entitled “Data Compression Method And Apparatus With Embedded Run-Length Encoding Using Mathematical Run Processing”.
U.S. patent application Ser. No. 09/336,219 by Cooper, filed Jun. 19, 1999, entitled “LZW Data Compression/Decompression Apparatus And Method With Embedded Run-Length Encoding/Decoding”.
Said patent application Ser. Nos. 09/264,269; 09/300,810 and 09/336,219 are incorporated herein by reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to LZW data compression particularly with respect toe minimizing dictionary access when processing character runs.
2. Description of the Prior Art
Professors Abraham Lempel and Jacob Ziv provided the theoretical basis for LZ data compression and decompression systems that are in present day widespread usage. Two of their seminal papers appear in the IEEE Transactions on Information Theory, IT-23-3, May 1977, pp. 337-343 and in the IEEE Transactions on Information Theory, IT-24-5, September 1978, pp. 530-536. A ubiquitously used data compression and decompression system known as LZW, adopted as the standard for V.42 bis modem compression and decompression, is described in U.S. Pat. No. 4,558,302 by Welch, issued Dec. 10, 1985. LZW has been adopted as the compression and decompression standard used in the GIF and TIFF image communication protocols.
Further examples of LZ dictionary based compression and decompression systems are described in the following U.S. patents: U.S. Pat. Nos. 4,464,650 by Eastman et al., issued Aug. 7, 1984; 4,814,746 by Miller et al., issued Mar. 21, 1989; 4,876,541 by Storer, issued Oct. 24, 1989; 5,153,591 by Clark, issued Oct. 6, 1992; and 5,373,290 by Lempel et al., issued Dec. 13, 1994.
Another type of data compression and decompression, denoted as run-length encoding (RLE), compresses a repeating character run by providing a compressed code indicating the character and the length of the run. RLE is thus effective in encoding long strings of the same character. For example, RLE is effective in compressing a long sequence of blanks that may be included at the beginning of a data file. RLE is also effective in image compression where an image contains a long run of consecutive pixels having the same value, such as in the sky portion of a land-sky image.
When the above dictionary based LZ compression systems encounter a character run, numerous dictionary accesses are utilized to generate the compressed codes corresponding to the run. It is desirable in such systems to minimize the number of dictionary accesses so as to enhance system performance.
In the prior art, run-length encoding has been combined with LZ systems as exemplified in the following U.S. patents: U.S. Pat. Nos. 4,929,946 by O'Brien et al., issued May 29, 1990; 4,971,407 by Hoffman, issued Nov. 20, 1990; 4,988,998 by O'Brien, issued Jan. 29, 1991; 5,247,638 by O'Brien et al., issued Sep. 21, 1993; 5,389,922 by Seroussi et al., issued Feb. 14, 1995; and 5,861,827 by Welch et al., issued Jan. 19, 1999.
In some prior art systems, run-length encoding has been combined with an LZ system by applying the data to a run-length encoder and then applying the run-length encoded data to the LZ based system. In such an architecture, a run-length encoder is utilized at the front end of the compressor and a run-length decoder is utilized at the output end of the decompressor. Such a system suffers from the disadvantages of increased equipment, expense, control overhead and processing time. U.S. Pat. Nos. 4,971,407 and 4,988,998 exemplify such a system.
In the LZW based system of U.S. Pat. No. 5,389,922, certain output codes from the compressor are suppressed in the presence of a run of repeating input data characters but numerous dictionary accesses are nevertheless utilized. A special run enhancement engine is required at the input to the decompressor to regenerate the missing codes.
In the compressor of the system of U.S. Pat. No. 5,861,827, when a partial string W and a character C are found, a new string is stored with C as an extension character on the string PW where P was the string conveyed in the last transmitted output compressed code. With this compression algorithm, a run of characters is encoded in two compressed codes regardless of its length but, nevertheless, numerous dictionary accesses are utilized. The decompressor of this system uses a special unrecognized code process to maintain synchronism with the compressor.
In the system of U.S. Pat. No. 4,929,946, a run is indicated by transmitting a predetermined reserved reference value followed by a repeat count for the run. The requirement of the use of the reserved reference value in the compressed stream for every run that is detected tends to reduce the compression. U.S. Pat. No. 5,247,638 provides descriptions similar to those of U.S. Pat. No. 4,929,946.
Another data compression system involving the encoding of data character runs is disclosed in said patent application Ser. No. 09/264,269. In the compressor of this patent application, runs are processed by successively looking ahead into the input to determine if contiguous numerically increasing segments exist in the run.
Another data compression system involving the encoding of data character runs is disclosed in said patent application Ser. No. 09/300,810. In the compressor of this patent application, runs are processed by mathematically determining, from the length of the run, the respective output codes corresponding to the contiguous numerically increasing segments that exist in the run.
Another data compression and decompression system that involves the processing of data character runs is disclosed in said patent application Ser. No. 09/336,219. In the system of this patent application, run-length encoding/decoding is embedded in the LZW data compression/decompression system where the compressor and decompressor code counters are utilized in signalling and detecting that a character run has been encountered.
It is an object of the present invention to detect the presence of a character run in an LZW data compression system and to variously utilize run processing procedures described in said patent application Ser. Nos. 09/264,269 and 09/300,810 to process the run. Since these run processing procedures do not require numerous dictionary accesses, a performance improvement is effected.
SUMMARY OF THE INVENTION
The present invention enhances the well-known LZW data compression system by determining when a run of input data characters is about to occur. The run is processed by successively looking ahead into the input to determine if contiguous numerically increasing segments exist in the run and by generating a sequence of numerically increasing output codes corresponding to the numerically increasing contiguous run segments. Alternatively, the run is processed by determining the length of the run and mathematically determining, from the length of the run, the respective output codes corresponding to the contiguous numerically increasing segments that exist in the run. Specifically, an iterative mathematical algorithm or a quadratic equation algorithm are utilized to process the run.
REFERENCES:
patent: 4558302 (1985-12-01), Welch
patent: 4929946 (1990-05-01), O'Brien et al.
patent: 4971407 (1990-11-01), Hoffman
patent: 4988998 (1991-01-01), O'Brien
patent: 5247638 (1993-09-01), O'Brien et al.
patent: 5369605 (1994-11-01), Parks
patent: 5389922 (1995-02-01), Seroussi et al.
patent: 5448733 (1995-09-01), Satoh et al.
patent: 5463390 (1995-10-01), Whitting et al.
patent: 5525982 (1996-06-01), Cheng et al.
patent: 5610603 (1997-03-01), Plambeck
patent: 5764167 (1998-06-01), Adams et al.
patent: 5798718 (1998-08-01), Hadady
patent: 5861827 (1999-01-01), Welch et al.
Internet site: http://www.boutell
Adornato Rocco L.
Cooper Albert B.
Jean-Pierre Peguy
Starr Mark T.
Unisys Corporation
LandOfFree
LZW data-compression apparatus and method using look-ahead... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with LZW data-compression apparatus and method using look-ahead..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and LZW data-compression apparatus and method using look-ahead... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2451407