JPH0810934B2 - Video coding device - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture coding device used in a videophone, a video conference system, or the like.

2. Description of the Related Art FIG. 8 shows a schematic configuration of a conventional moving picture coding device. In the conventional moving image coding apparatus, it is premised that a fixed speed line having a fixed transmission speed is used as a transmission medium.

In FIG. 8, a moving picture signal series 80 is a series conversion circuit.
After being input to 81 and undergoing sequence conversion, they are input to the quantization circuit 82. The quantization circuit 82 quantizes the sequence-converted coefficient with the quantization step width determined by the generated information amount control circuit 86. The entropy coding circuit 83 includes a quantization circuit 82.
The quantized representative value of is encoded by entropy, and the encoded data is output to the buffer 84. Buffer 84 is a fixed speed line
Entropy coding circuit while matching the transmission rate of 85
The data encoded in 83 is output to the fixed speed line 85.
The generated information amount control circuit 86, when the occupied amount of the coded data accumulated in the buffer 84 is large, the quantizing circuit 82
The amount of generated information is reduced by increasing the quantization step width of
The amount of generated information is increased by reducing the quantization step width of 82.

Therefore, the information generation rate by the encoding process can be controlled to be substantially equal to the transmission rate of the fixed rate line 85.

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention Generally, when the redundancy of a moving image signal is reduced by an encoding process, the information generation speed thereof largely fluctuates with time.

However, in the conventional moving picture coding apparatus, since the fixed speed line 85 having a fixed transmission speed is used as the transmission medium, the information generation speed by the coding process is set to the fixed speed line 85.
It is necessary to control the amount of generated information as described above so that the transmission rate becomes equal to the transmission rate of.

When such generated information amount control is performed, as is apparent from the velocity-distortion function in FIG. 7, the coding distortion (image quality) moves between A and B according to the activity of the moving image. However, a large image quality fluctuation occurs over time.

On the other hand, when a variable speed line is used as the transmission medium, it becomes possible to output information to the line at the information generation rate by the encoding process, and the speed-distortion function of FIG. Since the information generation speed moves between C and D, it is possible to keep the coding distortion (image quality) constant.

However, considering the efficient operation and billing of the communication network, it is desirable to average the information generation rates to some extent and to control the maximum information generation rate to some extent even when the variable speed circuit is used. Conceivable.

In view of such conventional problems, the present invention presupposes that a variable speed line is used as a transmission medium, and suppresses the fluctuation of the image quality by allowing the temporal fluctuation of the information generation speed, and at the same time, suppresses the information generation speed. A video that enables efficient operation of the communication network by controlling the average value of is close to the preset average information generation rate and by controlling the information generation rate so as not to exceed the preset maximum information generation rate. An object is to provide an image coding device.

Means for Solving the Problems In order to achieve the above object, the present invention uses a preset average generated information amount and a difference amount of information amount generated by an actual encoding process and a cumulative amount of the difference amount. Controlling the quantization step width of the quantization means, and adaptively controlling the characteristics of the quantization step width control means by using a preset maximum generated information amount and the information amount generated by the encoding process. It is the one.

Advantageous Effects of Invention According to the present invention, it is possible to suppress image quality fluctuations by a small amount, improve overall image quality, and make the average value of generated information amount close to a preset average generated information amount. Further, the generated information amount can be controlled so as not to exceed the preset maximum generated information amount, and the communication network can be efficiently operated.

Embodiments Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a moving picture coding apparatus according to the present invention.

In FIG. 1, 10 is a moving image signal series, and 11 is a circuit for converting the moving image signal series 10. 12 is a series conversion circuit
A quantizing circuit for quantizing the coefficient transformed by 11; 13
Is an entropy coding circuit for entropy coding the quantized representative value of the quantization circuit 12. Reference numeral 14 is a buffer for temporarily storing the encoded data from the entropy encoding circuit 13 and packetizing it to output it to the variable speed line 15.
16 is an information amount calculation circuit for calculating the difference amount between the preset average generated information amount and the information amount generated by the actual encoding process and its accumulated amount, and 17 is the difference amount calculated by the information amount calculation circuit 16. And a generated information amount control circuit for controlling the quantization step width of the quantization circuit 12 using the accumulated amount. 18 is a quantization step width control for controlling the information amount calculation circuit 16 and the generated information amount control circuit 17 so that the average value of the generated information amount by the encoding process is close to a preset average generated information amount. It is a means.

Numeral 19 is a quantization step width control means using the preset maximum generated information amount and the information amount generated by the encoding process.
It is a maximum generated information amount control circuit for controlling the generated information amount so as not to exceed a preset maximum generated information amount by adaptively controlling 18 characteristics.

Next, the operation of the above embodiment will be described with reference to FIGS. FIG. 2 is a graph showing the amount of information generated for each block that is a unit of quantization step width control, FIG. 3 is an explanatory diagram showing an example of frame division for quantization step width control, and FIGS. 4 and 5 are FIG. 6 is a graph showing the characteristics when determining the quantization step width, and FIG. 6 is a graph showing the characteristics of the quantization circuit of FIG.

First, FIG. 3 shows the case where one frame is divided into N blocks and the quantization step width is controlled for each block, 31 indicates the current frame to be encoded, and 32 Indicates the current block in the current frame 31 to be encoded. Further, 30 indicates a frame immediately before the current frame 31 to be encoded, and 33 indicates a block for one frame before the current block 32.

Here, FIG. 2 shows columnar graphs of the information amount R (k) generated for each block for all blocks from the start of communication to the last processed block n, and FIG. The total number of blocks is n, and the block to be processed next is the (n + 1) th block.

Now, calculate the average amount of generated information in preset frames.
Ra, and set the maximum amount of generated information in preset frame units
If Rp, the information amount calculation circuit 16 determines the difference amount α (n) between the information amount R (n) generated in the encoding process of the nth block and the preset average generated information amount Ra / N, and the difference. The cumulative amount C (n) from the start of communication of the amount is calculated by the following formula.

α (n) = R (n) -Ra / N = C (n) -C (n-1) (1) Next, the generated information amount control circuit 17 determines the quantization step width Δg (n) from the characteristics of FIG. 4 using α (n) and C (n) calculated by the information amount calculation circuit 16. , And from the quantization step width g (n) of the last encoded block n,
Next, the quantization step width g (n + 1) of the (n + 1) th block to be encoded is calculated by the following equation.

g (n + 1) = g (n) + Δg (n) (3) In the characteristic shown in FIG. 4, Th represents a threshold value for determining whether or not the quantization step width is updated. That is, when the cumulative absolute value of the difference amount between the preset average generated information amount and the actual generated information amount is Th or less, g (n + 1) = g (n) is set to prevent the image quality from changing. Further, by limiting the maximum value of the amount of change in the quantization step width between adjacent blocks to Δg _max , rapid image quality variation between adjacent blocks is avoided. In this way, the characteristics shown in FIG. 6 are obtained.

On the other hand, the maximum generated information amount control circuit 19 uses the information amount generated in the last one frame block before the current block 32 to be encoded, Is calculated, and θ (n + 1) is determined according to the characteristics shown in FIG. θ (n + 1) is Δg in the generated information amount control circuit 17
It is used as one parameter of the characteristic that determines (n).

As described above, according to the above-described embodiment, since it is premised that the variable speed line is used, it is possible to allow the time variation of the generated information amount, which is a feature of the moving image coding, and thus reduce the image quality variation. Can be suppressed.
Further, it is possible to efficiently operate the communication network for controlling the average generated information amount and the maximum generated information amount.

In the above embodiment, the case where one frame is divided into N as shown in FIG. 3 is shown as an example of division of the block which becomes the quantization step width control unit, but the present invention is limited to such division. In addition, the amount of information generated in the block for one frame before the current block 32 was used in the calculation of equation (4), but the present invention is not limited to using this amount of information. Absent.

EFFECTS OF THE INVENTION As described above, according to the present invention, by using a variable speed line as a transmission medium, it is possible to allow the time variation of the generated information amount in the moving image coding, and thus suppress the image quality variation of the coded image to be small. be able to. Also,
Since the average generated information amount control and the maximum generated information amount control are performed, efficient operation of the communication network can be achieved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a moving picture coding apparatus according to the present invention, FIG. 2 is a graph showing generated information amount for each block which is a unit of quantization step width control, and FIG. Explanatory diagram showing an example of frame division for step width control, FIGS. 4 and 5 are graphs showing characteristics when determining a quantization step width, and FIG. 6 shows characteristics of the quantizing circuit of FIG. FIG. 7 is a graph showing a velocity-distortion function in moving image activity, and FIG. 8 is a block diagram showing a schematic configuration of a conventional moving image coding apparatus. 10 ... Moving image signal sequence, 11 ... Sequence conversion circuit, 12 ... Quantization circuit, 13 ... Entropy coding circuit, 14 ... Buffer, 15 ... Variable speed line, 16 ... Information amount calculation circuit, 17
...... Generated information amount control circuit, 18 ・ ・ ・ Quantization step width control means, 19 ・ ・ ・ Maximum generated information amount control circuit.

Claims (1)

[Claims] 1. A quantizer for quantizing a coefficient obtained by systematically converting a moving image signal, a difference amount between a preset average generated information amount and an information amount generated by an actual encoding process, and an accumulated amount of the difference amount. Using the quantization step width control means for controlling the quantization step width of the quantization means, and the quantization step width control using the preset maximum generated information amount and the information amount generated by the actual encoding process. And a maximum generated information amount control means for controlling the characteristic of the means so that the amount of information generated by the encoding process does not exceed the maximum generated information amount.