Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1996-12-24
2001-07-03
Kelley, Chris (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06256349
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a picture signal encoding method and apparatus, a picture signal transmitting method, a picture signal decoding method and apparatus and a recording medium having thereon recorded signal decodable by the picture decoding apparatus. More particularly, it relates to a picture signal encoding method and apparatus, a picture signal transmitting method, a picture signal decoding method and apparatus and a recording medium, that may be employed with advantage for recording moving picture signals on a recording medium, such as a magneto-optical disc or a magnetic tape, transmitting the picture signals over a transmission channel, or reproducing the moving picture signals.
2. Description of the Related Art
In a system for transmitting moving picture signals to a remote place, such as a video conferencing system or a video telephone system, picture signals are encoded for data compression by utilizing line-to-line correlation or frame-to-frame correlation of picture signals for efficient utilization of the transmission channel.
If line-to-line correlation (two-dimensional correlation) is utilized, data compression can be achieved by processing picture signals by, for example, digital cosine transform (DCT) for concentrating the signal power in specified frequency components.
If frame-to-frame (or field-to-field) correlation is utilized, the picture signals can be further compressed for encoding. That is, since pictures of temporally neighboring frames are not vitally changed from each other, difference signals obtained on computing the difference between these frames are of smaller values. Therefore, the amount of the codes can be compressed by encoding these difference signals.
However, if only the difference signals are transmitted, the original picture cannot be restored. Thus, the frame pictures are classed into three different sorts, that is an I-picture, a p-picture and a B-picture, and the picture signals are encoded for compression. This compression technique is used in the encoding standards by the Moving Picture Image Experts Group (MPEG). This MPEG, discussed in ISO-IEC/JTC1/SC2/WG11 and proposed as a draft standard, is of a hybrid system consisting in the combination of the motion-compensated predictive coding and the discrete cosine transform (DCT), as proposed in, for example, the U.S. Pat. No. 5,155,593, issued on Oct. 13, 1992.
For example, picture signals of 17 frames of from frame F
1
to frame F
17
are grouped together as a group-of-picture (GOP) asa processing unit. The picture signals of the leading frame F
1
, second frame and the third frame are encoded as I-, B- and P-pictures, respectively. The fourth to seventeenth frames F
4
to F
17
are processed as B- and P-pictures alternately.
The picture signal, encoded as an I-picture, are directly transmitted as picture signals of a full frame or field. As for the picture signals, encoded as a P-picture, basically the difference from the temporally previous I- or P-picture is transmitted. As for the picture signals, encoded as a B-picture, basically the difference from an average value of a temporally previous frame F
1
and a temporally posterior frame F
3
is computed and encoded.
Thus, in encoding moving picture signals, the first frame F
1
is processed as an I-picture, and hence is directly transmitted as transmission data on the transmission channel (intra-frame coding). On the other hand, since the second frame F
2
is processed as a B-picture, basically the difference from an average value between the temporally previous frame F
1
and the temporally posterior frame F
3
is computed and the difference is transmitted.
More specifically, the encoding as the B-picture is subdivided into four sorts of encoding. The first processing is transmitting data of the original frame F
2
directly as transmission data (intra-coding or intra-picture predictive coding). This processing is similar to that for an I-picture. The second processing is computing the difference from the temporally posterior frame F
3
and transmitting the difference (backward predictive coding). The third processing is transmitting the difference from the temporally previous frame F
1
(forward predictive coding). The fourth processing is generating the difference from an average value between the temporally previous frame F
1
and the temporally posterior frame F
3
and transmitting the difference as transmission data (bidirectional predictive coding).
Of these four methods, such a method as will minimize the volume of the transmission data is used.
Meanwhile, when transmitting the difference data, a motion vector x
1
between the picture of an original frame and the picture of a frame the difference of the original frame from which is computed (prediction picture) which, in the case of forward prediction, is a motion vector between the frames F
1
and F
2
, a motion vector x
2
which, in the case of backward prediction, is a motion vector between the frames F
3
and F
2
, or both the vectors x
1
and x
2
, are transmitted along with the difference data.
As for the frame F
3
of the P-picture, basically the temporally previous frame F
1
is used as a prediction picture, difference data between the frame F
3
and the temporally previous frame F
1
and the motion vector x
3
are computed, and the difference data and the motion vector x
3
are transmitted as transmission data (forward predictive coding).
The processing as the P-picture is subdivided into two processing sorts. The first processing is transmitting the difference of the frame F
3
from the temporally previous frame F
1
(forward predictive coding). The second processing is directly transmitting data of the original frame F
3
(intra-coding or intra-picture predictive coding). Which of these processing methods is used is determined depending which method gives a smaller amount of transmission data, as in the case of the B-picture.
Meanwhile, with the conventional picture signal encoding and decoding method or apparatus, transform coding, such as discrete cosine transform (DCT) is used for encoding. In executing the DCT, which is the real-number precessing operation, the processing is terminated with a finite word length if an actual DCT circuit is used. Thus, if an IDCT processing is done on an IDCT circuit after DCT processing on a DCT circuit, the information of the original signal IS partially lost.
Therefore, if the real-number processing, such as DCT, is used on an encoding/decoding apparatus, reversible coding, which is a technique required for precision picture signals, cannot be actually used.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to enable reversible coding even in picture signal encoding/decoding employing real-number processing, such as DCT.
In one aspect, the present invention provides a picture signal encoding method for outputting a signal obtained on processing a picture signal with encoding accompanying real-number calculations. The encoding method includes a step of processing the input picture signal with encoding accompanying real-number calculations for generating encoded data, a step of locally decoding the encoded data for generating locally decoded data, and a step of comparing the locally decoded data to picture data prevailing prior to performing the real-number calculations for calculating a difference.
In another aspect, the present invention provides a picture signal encoding apparatus for outputting a signal obtained on processing a picture signal with encoding accompanying real-number calculations. The encoding apparatus includes means for processing the input picture signal with encoding accompanying real-number calculations for generating encoded data, means for locally decoding the encoded data for generating locally decoded data, and means for comparing the locally decoded data to picture data prevailing prior to performing the real-number calculations for calculating a difference.
In another aspect, the present invention provides a pict
Suzuki Teruhiko
Yagasaki Yoichi
Frommer William S.
Frommer Lawrence & Haug LLP.
Kelley Chris
Philippe Gims
Smid Dennis M.
LandOfFree
Picture signal encoding method and apparatus, picture signal... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Picture signal encoding method and apparatus, picture signal..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Picture signal encoding method and apparatus, picture signal... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2561240