Pulse or digital communications – Bandwidth reduction or expansion
Reexamination Certificate
2000-02-04
2001-06-05
Rao, Andy (Department: 2713)
Pulse or digital communications
Bandwidth reduction or expansion
C382S172000, C382S181000, C382S190000
Reexamination Certificate
active
06243416
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to methods for coding and decoding an image by dividing the image into blocks, and to an image coder and image decoder employing these methods.
Data compression is a field of intense interest, because it is an enabling technology for multimedia. Image data compression methods have been standardized by the Telecommunication Standardization Sector of the International Telecommunication Union (ITU-T), and by the International Organization for Standardization and the International Electrotechnical Commission (ISO/IEC). For still images, the method advocated by the Joint Photographic Experts Group (JPEG) has been adopted in ITU-T recommendations T.81 and T.83 and ISO/IEC standard 10918. For moving images, methods advocated by the Moving Picture Experts Group (MPEG) have been adopted in ISO/IEC standard 11172 (MPEG1) and ISO/IEC standard 13818 (MPEG2).
The JPEG and MPEG methods divide an image into blocks, and compress each block separately by a combination of the discrete cosine transform, quantization, and entropy coding. These techniques require complex calculations and much memory access, however, so their speed is limited, and an image coding and decoding apparatus (codec) employing the JPEG and MPEG methods is comparatively large and expensive.
In addition, the JPEG and MPEG methods are not directly applicable when the decoded image is reduced or enlarged by changing the image resolution; that is, by changing the number of picture elements (pixels) in the image. Resolution conversion has to be performed as a separate operation, following image decoding.
SUMMARY OF THE INVENTION
It is accordingly an object of the present invention to provide high-speed methods of coding and decoding an image by dividing the image into blocks.
Another object of the invention is to provide a small and inexpensive image coder.
Another object is to provide a small an inexpensive image decoder.
Yet another object is to provide an image decoder that can change the image resolution while decoding the image.
The invented image coding method employs a table of predetermined patterns identical in size and shape to the blocks into which the image divided. One pattern is made up of a single region; the other patterns are made up of two or more regions, defined by internal edges. The patterns are identified by pattern identifiers.
Each block in the image is matched with one of the patterns in the table by detecting internal edges in the block. The block is thereby partitioned into regions matching the regions in the pattern. One representative pixel value, e.g. an average value, is calculated for each region in the block. The block is coded as the pattern identifier of the matching pattern, and the representative pixel value of each region.
The invented image coder comprises a coding parameter memory having the above table, an edge detector, a partitioner for partitioning a block into regions as in the matching pattern, a pixel value calculator for calculating the representative pixel values, and a code assembler for assembling the coded image information.
The invented image decoding method employs at least one table of predetermined patterns, similar to the table used in the invented coding method. Each block is divided into regions according to the pattern identifier given in the coded image information, and each region is filled with the corresponding representative pixel value.
The image decoding method may employ a plurality of tables, each storing patterns of a different size. For example, one table may store the same patterns as used in coding the image, another table may store smaller patterns, and another table may store larger patterns. One of the tables is selected in response to mode information.
The invented image decoder comprises a decoding parameter memory having the above table or tables, and a fill processor for filling regions with representative pixel values.
The invented image coding and decoding methods are fast because they require comparatively little memory access, and involve no complex mathematical operations. For the same reasons, the invented image coder and decoder are small and inexpensive.
If the decoder has a plurality of tables, the image can be reduced or enlarged, as selected by the mode information, with no extra operations.
REFERENCES:
patent: 5396584 (1995-03-01), Lee et al.
patent: 5453787 (1995-09-01), Hancock et al.
patent: 5465118 (1995-11-01), Hancock et al.
patent: 5604499 (1997-02-01), Miyagoshi et al.
patent: 5613015 (1997-03-01), Suzuki et al.
patent: 5617333 (1997-04-01), Oyamada et al.
patent: 5668646 (1997-09-01), Katayama et al.
patent: 5724097 (1998-03-01), Hibi et al.
patent: 5768438 (1998-06-01), Etoh
patent: 5798719 (1998-08-01), Wise et al.
patent: 5815287 (1998-09-01), Yamada
patent: 5854853 (1998-12-01), Wang
Matsushiro Nobuhito
Yoshida Yoshihiko
Akin Gump Strauss Hauer & Feld L.L.P.
An Shawn
Oki Data Corporation
Rao Andy
LandOfFree
Image coding and decoding methods, image coder, and image... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image coding and decoding methods, image coder, and image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image coding and decoding methods, image coder, and image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2452273