Image analysis – Image compression or coding – Interframe coding
Reexamination Certificate
1998-12-10
2001-03-13
Couso, Jose L. (Department: 2724)
Image analysis
Image compression or coding
Interframe coding
C382S232000, C382S246000
Reexamination Certificate
active
06201896
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a picture encoding apparatus and more particularly, to a picture encoding apparatus for compressing video signals on a real-time basis.
2. Description of the Prior Art
When pictures/voices are transmitted/received by using a telephone line or a digital line as in the case of a videophone, a system in which data are compressed and transmitted in conformity with transmission performance of the line and the data thus transmitted are expanded at a reception side is generally used.
FIG. 9
is a block diagram showing a conventional picture compression apparatus.
In
FIG. 9
, a video signal of a subject which is picked up by camera
11
is supplied to picture input means
12
and then divided into plural blocks. A block having high correlation with reference picture data is searched every block by motion detection means
13
, and the differential between the block thus searched and the current block is calculated. The differential data are converted to frequency components in conversion means
14
, and then quantized in quantizer
15
.
The quantized data outputted from quantizer
15
are inversely quantized in inverse quantizer
18
, and then supplied to inverse conversion means
19
to be converted from the frequency components to pixel values. Thereafter, the differential data are restored to reference picture data in motion compensation means
20
, and then supplied to motion search means
13
. On the other hand, the quantized data are supplied to variable-length encoding means
16
to be subjected to variable-length encoding on the basis of run length or the like, and then transmitted to transmission path by transmission means
17
.
In order to enhance the picture quality when the above picture compression processing is carried out under such a condition that the transmission performance of the line is restricted, there is known a method of increasing the data amount per one frame. However, if a large data amount is allocated to one frame, it takes a long time to complete the transmission of the frame, and thus a time interval until the data of a next frame are compressed is increased. As the time interval is increased, intermittent moving pictures are obtained (i.e., moving pictures obtained are as if some frames thereof are intermittently missed (hereinafter referred to as “frame-missed moving pictures”).
On the other hand, if the data amount per one frame is reduced, the time interval until a next frame becomes shorter, and thus the motion becomes smoother. However, since the data amount per frame is small, block-shaped noises appear in pictures, resulting in degradation of picture quality.
Therefore, when the video data are transmitted through a line having low transmission performance, frame-missed moving pictures are transmitted or the picture quality is extremely degraded.
In order to avoid this phenomenon, it is required to improve the allocation of the data amount within one frame without increasing the data amount to be allocated to one frame and allocate a large data amount to portions to which viewers pay their attention.
Therefore, an apparatus which samples a picture of one frame and uses the correlation between adjacent pixels has been hitherto known as a picture encoding apparatus which enhances the picture compression efficiency as disclosed, for example, in JPA-8-242446. Further, there has been also known a picture encoding apparatus in which a still picture is sectioned into plural blocks and the fineness of gradation is adaptively varied every block to perform the optimum compression as disclosed, for example, in JPA-1-93278. Still further, there has been hitherto known a picture encoding apparatus in which the correlation between blocks is examined when a still picture is compressed and a high-correlation block are approximated by a preceding block as disclosed, for example, JPA-4-344771.
These conventional picture encoding apparatuses take no account of successive pictures or the enhancement in picture quality for the frame-missed moving pictures, and thus degradation of picture quality becomes remarkable when compressed pictures are intermittent. Further, in the conventional apparatus disclosed in JPA-8-242446, the data amount to be allocated to one frame is not taken into consideration, and a dedicated expanding apparatus must be provided to an expansion side.
In JPA-1-93278, there is no mention of the support of moving pictures, and also the data amount to be allocated to one frame is not taken into consideration. Further, in JPA-4-344771, a support of moving pictures is not taken into consideration, and also there is no mention of enhancement of picture quality.
SUMMARY OF THE INVENTION
In order to overcome the aforementioned disadvantages, the present invention has made and accordingly has an object to provide a picture encoding apparatus for performing data compression which reduces noises which are liable to attract viewer's attention when the number of the frames of compressed pictures to be transmitted per unit time is small due to the restriction of the transmission performance, whereby high-quality pictures can be obtained with no alteration of programs at an expansion side.
According to the present invention, there is provided a picture encoding apparatus for performing motion-compensated intra-frame picture coding on a block basis, which comprises: block data generating means for generating data of each block in a current frame from data in the current frame and data in a reference frame and a motion vector for each block by using motion-compensated prediction; a quantizer for quantizing data of each block in the current frame in accordance with a quantization scale factor; block type determining means for determining whether a mode of each block in the current frame is high-quality mode or normal mode on the basis of an interval between the current frame and the reference frame, a quantity of the motion vector of each block in the current frame, and an activity of each block in the current frame; data amount assignment means for determining the quantization scale factor for each block in the current frame on the basis of the mode of each block so that data amount of the output of the quantizer for each block is varied in accordance with the mode of each block; and variable-length encoding means for encoding the output of the quantizer into variable-length code; and reference frame data generating means for generating the data in the reference frame using the output of the quantizer.
REFERENCES:
patent: 6097737 (2000-08-01), Takizawa et al.
patent: 6108027 (2000-08-01), Andrews et al.
patent: 2 260 668 (1993-04-01), None
patent: 1-93278 (1989-04-01), None
patent: 4-344771 (1992-12-01), None
patent: 8-242446 (1996-09-01), None
Couso Jose L.
Do Anh Hong
Foley & Lardner
NEC Corporation
LandOfFree
Picture encoding apparatus 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 encoding apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Picture encoding apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2545449