Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1997-10-09
2003-08-19
Lee, Young (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06608865
ABSTRACT:
FIELD OF THE INVENTION
This invention pertains to a processing technology of video signals. In particular, this invention pertains to a technology of compressed coding.
BACKGROUND OF THE INVENTION
The discrete cosine transform (DCT) is a means with high efficiency for:video signal compressed coding (N. Ahmed, et al.: [Discrete cosine transform] IEEE Trans. Comput., vol. C-23, pp. 90-93, 1974), and great efforts have been made to further increase the coding efficiency for the coding based on DCT. Consequently, it is necessary to perform adaptive processing for the transformation coefficient for the nonstatic property of the video signal.
In the adaptive DCT coding system, it is necessary to have the code adapted to the prescribed characteristics of the image block. The methods for realizing this include a variable block size method and adaptive quantization method. In the variable block size method, first of all, the image is decomposed to blocks having different sizes. The less fine region is represented by blocks with larger size. Then, the variable block size DCT is adapted. In this way, the bits of the less fine region can be reduced. Consequently, additional bits can be used for transmission of the highly fine region. Consequently, it is possible to improve the image quality by this method. On the other hand, the adaptive quantization method includes threshold coding and zonal coding. In the threshold coding, only the coefficient with its magnitude larger than a prescribed threshold is transmitted, and the transmitted coefficient's site and magnitude are coded. A typical method for coding such information is the run-level coding method adopted by JPEG and MPEG. On the other hand, in the zonal coding, only the coefficient present inside the prescribed region called the zone is coded. The shape of the zone has to be selected carefully as it significantly affects the coding efficiency. As a matter of fact, it is possible to code certain coefficients with a smaller magnitude, while to abandon the coefficients with a larger magnitude. This is because the shape of the zone is prescribed beforehand. This mechanism might be able to further improve the coding efficiency.
The aforementioned methods may be classified by means of the block classification method, which determines whether it is necessary to find out on the receiver side the class to which each block belongs, and whether it can be derived on the receiver side. When the size of the blocks is too small and the number of classes is too large, this overhead information can be transmitted directly without much penalty. However, in the other cases, other methods have to be used. For example, when 16 classes are used for the 8×8-pixel blocks, 4 bits are needed for each block. For SIF size image as one of the common formats of the image (352×240 pixel, 8 bits/pixel with respect to the brightness), suppose it has a chroma format of 4:2:0 and a compression ratio of 20:1, the total overhead information becomes 5280 bits, which is over 10% the compressed data. Sharing has to be adopted based on the coding gain for the overhead amount, that is, the information with the supporting role.
SUMMARY OF THE INVENTION
This invention provides an adaptive DCT coding based on the geometric edge display. In its algorithm, by means of the adaptive scanning, the zonal coding can be changed to a more compatible method, and it is possible to perform the variable block size method in a parallel manner. The geometric edge information including the site and direction of the edge can reduce the bits with respect to the information with the supporting role. On the other hand, it can replace the direct transfer of the block classification. This new algorithm can transfer the block classification information at a high efficiency, so that it might outdo the coding not of the adaptive type with respect to the performance.
In the coding method of this invention, there are the following steps: step in which the edge is extracted on one frame of the input image, step in which the aforementioned frame is classified into blocks where the aforementioned extraction edge is present and blocks where there is no said extraction edge, and a step in which for the aforementioned blocks with the aforementioned extraction edge present, based on the correlation between the edge direction and the distribution of the discrete cosine transform coefficient, the optimum scanning order is determined, and the scanning order is coded.
REFERENCES:
patent: 5410350 (1995-04-01), Kato et al.
patent: 5557330 (1996-09-01), Astle
patent: 5701368 (1997-12-01), Jung
patent: 5724097 (1998-03-01), Hibi et al.
patent: 5878161 (1999-03-01), Ishida et al.
patent: 6252993 (2001-06-01), Murayama
patent: 1 131 873 (1996-09-01), None
patent: 0 607 484 (1994-07-01), None
Kim J D: D Coding of edge blocks in digital video sequence coding Visual Communications and Image Processing '94, Chicago, IL. USA. Sep. 25-28, 1994, vol. 2308, pt. 3. pp. 1615-1626, XP002102940 ISSN 0277-786X, Proceedings of the SPIE—The International Society for Optical Engineering, 1994, USA * section 3 “ATBC for edge blocks in digital video coding” ** figures, 7,8.
Yo-sung Ho et al: “Classified Transform Coding of Images Using Vector Quantization” Multidimensional Signal Processing, Audio and Electroacoustics, Glasgow, May 23-26, 1989, vol. 3, No. Conf. 14. May 23, 1989, pp. 1890-1893, XP000089247 Institute of Electrical and Electronics Engineers * section 3 “Transform domain classification” ** figures 1,2 *.
Itoh Y: “An Edge-Oriented Progressive Image Coding” IEEE Transactions on Circuits and Systems for Video Technology, vol. 6, No. 2, Apr. 1, 1996, pp. 135-142, XP005383533 * section II “Hierarchical edge extraction” ** section III.D “Residual coding for smooth component” *.
Lee J H et al: “An Efficient Encoding of DCT Blocks with Block-Adaptive Scanning” IEICE Transactions on Communications, vol. 77-B. No. 12, Dec. 1, 1994, pp. 1489-1494, XP000498064 * p. 1489, right-hand column, line 27-line 42 ** section 2 “Block adaptive scanning” ** section 3 “Encoding algorithm” *.
Eisuke Nakasu et al: “Intra/Inter Mode Adaptive DCT Coding System of HDTV Signals” Signal Processing of HDTV, 3, Turin, Sep. 4-6, 1991, No. Workshop 4. Sep. 4, 1991. pp. 439-446. XP000379981 Yasuda H; Chiariglione L * sections 2.4 “Adaptive scanning and 2.5 Quantization”* * p. 443, line 14 line 15 * * figures 1-3.
Brady III W. James
Lee Young
Marshall, Jr. Robert D.
Telecky , Jr. Frederick J.
Texas Instruments Incorporated
LandOfFree
Coding method for video signal based on the correlation... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Coding method for video signal based on the correlation..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Coding method for video signal based on the correlation... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3074094