Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1996-09-19
2002-05-21
Rao, Andy (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240030
Reexamination Certificate
active
06393058
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a moving picture coding system wherein a quantizer is controlled by the amount of generated bits.
2. Description of the Related Art
In a conventional moving picture coding system, a quantization step size is controlled in accordance with the amount of generated bits as disclosed, for example, in “A Study on Control Method for Low Bit Rate Video Coding”, Collection of Drafts D-311 for the Spring National Meeting of the Electronic Information Communications Society in 1990.
The conventional moving picture coding system disclosed is shown in block diagram in FIG.
4
. Referring to
FIG. 4
, the moving picture coding system shown includes a generated bit amount calculation section
14
which calculates and supplies a generated bit amount I to a quantization variation value calculation section
15
by way of a signal line
20
. The quantization variation value calculation section
15
first multiplies the generated bit amount I received by way of the signal line
20
and a first differential value (differential value of a generated bit amount between different frames) I′ of the generated bit amount I by a sensitivity coefficient to calculate a variation value &Dgr;Q of a quantization step size and sends the thus calculated variation value &Dgr;Q to a quantization determination section
16
by way of a signal line
21
. The quantization determination section
16
adds the quantization variation value received by way of the signal line
21
to a quantization step size of the last frame to determine a new quantization step size for a current frame and sends the thus determined quantization step size to a coding section
17
by way of a signal line
22
. The coding section
17
performs coding using the quantization step size thus received by way of the signal line
22
.
In the conventional moving picture coding system described above, the quantization step size is determined only based on a quantitative value of a generated bit amount with equal sensitivity for any scene, without considering the scene contents whether the active images with a great amount of motion accordingly attach greater importance to the motion smoothness than to the image clarity or the still images with a small amount of motion accordingly attach greater importance to the image clarity than to the motion smoothness. Therefore, where the sensitivity of the quantization step size to the variation of the bit amount is low, in active frames with a great amount of motion, the variation of the quantizer is so low that dropped or missing frames are increased, resulting in loss of motion smoothness, but in still frames with small motion, much time is required until the quantization step size is reduced low and consequently the image clarity does not rise rapidly.
On the contrary where the sensitivity of the quantization step size to the variation of the bit amount is high, the quantizer varies in response to a small amount of variation of the bit amount, and consequently, the quantizer step size is unstable. In this manner, where the quantization step size is determined only based on a bit amount, it is difficult to determine a sensitivity coefficient adaptively for any scene.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a moving picture coding system which allows moving pictures to be observed subjectively easily whether frames of the moving pictures attach great importance to the motion smoothness or the image clarity.
In order to attain the object described above, according to the present invention, there is provided a moving picture coding system which includes a quantizer for quantization processing, which comprises generated bit amount calculation means for calculating a generated bit amount for each coded frame, quantization variation value calculation means for calculating a quantization variation value which is a variation of a quantization step size from the generated bit amount calculated by the generated bit amount calculation means, variation value correction frame determination means for determining a frame as a frame for which correction of the quantization variation value should be performed when the frame involves a large amount of motion and the quantization variation value is positive in sign or when the frame involves a small amount of motion and the quantization variation value is negative in sign, variation value correction means for correcting, for a frame which is determined as a frame for which correction should be performed by the variation value correction frame determination means, the quantization variation value calculated by the quantization variation value calculation means so as to increase the variation of the quantizer, and quantization determination means for adding the quantization variation value corrected by the variation value correction means to a quantization step size of the last frame to calculate a new quantization step size for a current frame.
The variation value correction frame determination means may include moving block determination means for calculating the sum of the absolute values of differential signals between the last frame and the current frame for each block to determine moving blocks, moving block number calculation means for calculating the number of moving blocks based on the results of determination of the moving block determination section for each block, and correction frame determination means for determining a frame as a frame for which correction of the quantization variation value should be performed when the number of moving blocks of the frame calculated by the moving block number calculation section is higher than a predetermined threshold value for a moving frame and the quantization variation value calculated by the quantizer variation value calculation section is positive in sign or when the number of moving blocks is lower than a predetermined threshold value for a stationary frame and the quantization variation value is negative in sign.
Where the moving picture coding system further comprises a motion compensation interframe prediction section, the variation value correction frame determination section may include moving block number calculation means for counting the number of blocks whose motion vector is determined to be a non-zero vector by the motion compensation interframe prediction section, and correction frame determination means for determining a frame as a frame for which correction of the variation value should be performed when the number of moving blocks calculated by the moving block number calculation section is higher than a predetermined threshold value for a moving frame and the quantization variation value calculated by the quantization variation value calculation section is positive in sign or when the number of moving blocks is lower than a predetermined threshold value for a stationary frame and the quantization variation value is negative in sign.
In the moving picture coding system, where the quantization variation value calculated by the quantization variation value calculation means is represented by &Dgr;Q, the variation value correction means corrects the quantization variation value &Dgr;Q so that it may be increased in absolute value. The variation value &Dgr;Q is corrected when the frame is a moving frame and &Dgr;Q>0 or when the frame is a stationary frame and &Dgr;Q<0. Consequently, in active frames with a large amount of motion, a coarser quantization step size is selected to achieve smooth motion. But in the still frames with a small amount of motion, a finer quantization step size is selected to get image clarity.
Since the variation value of the quantization step size is corrected in accordance with scene contents and a quantizer is controlled adaptively depending upon whether the scene attaches great importance to the motion smoothness or to the image clarity and subjective image quality is improved.
The above and other objects, features and advantages of the present invention wi
LandOfFree
Moving picture coding system having a quantizer controlled... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Moving picture coding system having a quantizer controlled..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Moving picture coding system having a quantizer controlled... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2879995