Image analysis – Image compression or coding – Lossless compression
Reexamination Certificate
1999-09-09
2004-06-01
Johns, Andrew W. (Department: 2621)
Image analysis
Image compression or coding
Lossless compression
C382S250000, C382S253000
Reexamination Certificate
active
06744928
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to according method, a coding apparatus and a recording medium for coding input image data, and to a decoding method, a decoding apparatus and a recording medium for decoding coded image data.
RELATED ART
In accordance with the improvement in digital signal processing technology, apparatuses capable of compressing, coding and recording digital image signals and capable of decoding, decompressing and reproducing digital image signals have been accomplished; and DVC (Digital Video Cassette) can be taken as an example. The format of the DVC is described in the Specifications of “Consumer-Use Digital VCRs using 6.3 mm magnetic tape” compiled by the HD Digital VCR Association.
In digital image apparatuses including DVC, since the amount of input image data is large, the amount of data is decreased by compression and coded, and the coded image data is decoded so as to be reconstructed to original images by decoding, as a general practice.
FIG. 25
is a block diagram illustrating the configuration of a coding portion in a conventional digital image apparatus. In
FIG. 25
, the numeral
2501
represents a first input terminal, the numeral
2502
represents a first signal format conversion portion, the numeral
2503
represents a switch, the numeral
2504
represents a second input terminal, the numeral
2505
represents a second signal format conversion portion, the numeral
2506
represents a shuffling portion, the numeral
2507
represents an orthogonal transform portion, the numeral
2508
represents a variable-length coding portion, and the numeral
2509
represents a coded image output terminal.
An example of an image signal to be coded for recording is a YUV
422
component signal comprising a luminance (Y) signal, a first color difference (U) signal and a second color difference (V) signal in a ratio of 4:2:2. When this YUV
422
component signal is input from the first input terminal
2501
, it is converted into a YUV
411
component signal (hereinafter referred to as a YUV format signal) comprising four Y signals, one U signal and one V signal as the number of pixels per frame by the first signal format conversion portion
2502
as shown in FIG.
26
.
On the other hand, an image signal other a YUV format signal, for example, a digital RGB component signal (hereinafter referred to as an RGB format signal) comprising a red (R) signal, a green (G) signal and a blue (G) signal may sometimes become an input signal. In this case, the RGB format signal input from the second input terminal
2504
is required to be converted into a YUV format signal by the second signal conversion portion
2505
. The RGB format signal has horizontal pixels in the ratio of 4:4:4 per frame as shown in FIG.
27
. In the second signal format conversion portion
2505
, by using the RGB pixel values on the respective coordinates,
Y=
0.30
R+
0.59
G+
0.11
B
U=
0.70
R−
0.59
G−
0.11
B
V=
−0.30
R−
0.59
G+
0.89
B
are obtained; furthermore, the numbers of the horizontal pixels of the U and V signals were thinned out to ¼ to obtain a YUV format signal.
The YUV format signal obtained by the second signal format conversion portion
2505
is supplied to the shuffling portion
2506
via the switch
2503
, and processed hereinafter in the same way as described above.
The Y signal, U signal and V signal of the YUV signal sent tothe shuffling portion
2506
are each divided into a block comprising M horizontal pixels and N vertical pixels (usually, M=N=8). Four blocks of Y signal, one block of U signal and one block of V signal located in the same region of a display screen are defined as a macro block. By using this macro block as a unit, a sync block used as a coding unit is formed of five macro blocks located at separate positions in a frame as shown in FIG.
28
.
The shuffled image signal is sent to the orthogonal transform portion
2507
, and subjected to orthogonal transform (usually, discrete cosine transform) in block units. The image signal subjected to orthogonal transform is sent to the variable-length coding portion
2508
, and coded so that the amount of codes in the above-mentioned sync block is not more than a specific value. By carrying out the above-mentioned shuffling, the amount of codes required for each sync block is averaged for the entire frame, and coding can be carried out efficiently; furthermore, even if errors remain during reproduction, they disperse on the whole display screen, whereby the errors become less conspicuous. The coded image signal is output from the coded image output terminal
2509
.
The variable-length coding portion carries out variable-length coding for a set of zero run, i.e., the number of continuous 0s, and value, i.e., the value of non-zero coefficient following the zero run, for the coefficient string subjected to orthogonal transform.
FIG. 29
is a variable-lengthcoding table for DVC.
In the variable-length code for DVC, the code length is 3 bits or more and 16 bits or less, and the code length is determined uniquely by its high-order 8 bits. Furthermore, it is characterized that a code word having a shortercode length is assigned as the probability of occurrence is higher. The end of a series of code words is referred to as EOB.
Variable-length coding operation will be described below by using
FIGS. 29 and 30
.
It is assumed that coding has been completed up to coefficient A (coefficient value 9) in FIG.
30
. The portions to be coded next are 3 continuous 0s and non-zero coefficient 2 (coefficient group B) following thereto. At this time, the zero run is 3, and the value is 2. According to FIG
29
, the coefficient group B is coded as “111001000.”
Following the coefficient group B, coefficient C located immediately after the coefficient group B is coded. Since “0” is not present between the coefficient group B and the coefficient C, the zero run is 0 at this time. Since the value is −6, the coefficient C is coded to “101111” according to FIG.
29
.
FIG. 31
is a block diagram illustrating decoding for decoding variable-length coded image signals to obtain an ordinary image signal. In
FIG. 31
, the numeral
3101
represents a coded image input terminal, the numeral
3102
represents a variable-length decoding portion, the numeral
3103
represents an inverse orthogonal transform portion, the numeral
3104
represents a deshuffling portion, the numeral
3105
represents a first signal format conversion portion, the numeral
3106
represents a first signal output terminal, the numeral
3107
represents a second signal format conversion portion, and the numeral
3108
represents a second signal output terminal.
A coded image signal (a code word string) having been input to the coded image input terminal
3101
is decoded by the variable-length decoding portion
3102
. In the case when a code word string is decoded according to the variable-length coding table of
FIG. 29
, the following methods can be considered.
A first method is a method wherein a code word string is scanned bit by bit until its code length (code word) is determined, and the zero run and value of the determined code word are output referring to the table. This method will be described by using FIG.
32
.
In
FIG. 32
, 3-bit data from the head of a code word is taken as a candidate, and a judgment as to whether the code length is determined is made (corresponding to the description indicated by code a in FIG.
32
). If the code length is not determined, a code word additionally having the next 1-bit data is taken as a candidate, and a judgment as to whether the code length is determined is made (corresponding to the description indicated by code b in FIG.
32
). This operation is repeated until the code length is determined, whereby the code word is determined (corresponding to the description indicated by code c in FIG.
32
).
Next, the table is made reference to with respect to the determined code word, and its zero run and value are obtained (corresponding to
Juri Tatsuro
Ono Tadashi
Alavi Amir
Johns Andrew W.
Matsushita Electric - Industrial Co., Ltd.
Smith , Gambrell & Russell, LLP
LandOfFree
Variable-length decoding method and apparatus employing code... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Variable-length decoding method and apparatus employing code..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Variable-length decoding method and apparatus employing code... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3317057