JPH07107479A - Picture encoding device - Google Patents

Abstract

PURPOSE:To provide a good picture quality by using a rate buffer with a relatively small capacity with the use of the non-linear function with low or high change rate of encoding amount control parameter for the storage amount in the area where the code storage amount of the rate buffer is low or high. CONSTITUTION:A video signal inputted from an input terminal 1 is encoded by an encoder 2 and the outputted code is outputted from an output terminal 4 by the constant transmission rate through a rate buffer 3. The storage amount of the buffer 3 is inputted to an arithmetic unit 5 and it calculates the quantization step width from the storage amount of the buffer 3 based on the input/ output characteristic decided in advance, controlling the quantization step width to be used in the encoder 2. The I/O characteristic of the arithmetic unit 5 becomes a nonlinear function so that the quantization step for the code storage amount is represented by the function with the small inclination of the width in the area with the small code storage amount of the buffer 3 and it is represented by the function with the large inclination of the quantization step width in the area with the large storage amount of the buffer 3.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding apparatus used for digital video information communication and broadcasting, recording digital video information and the like.

[0002]

2. Description of the Related Art As a conventional image coding apparatus, IEEE TRA
NSACTIONS ON COMMUNICATIONS, Vol.COM-32, NO.3, MAR
The paper "Scene Adap" described in P 225-P232 of CH 1984
1 is shown in FIG. 1. In this image encoding device, the code amount control is executed by controlling the normalization factor according to the code storage amount of the rate buffer. ing.

As a conventional image coding apparatus, there is also known one which executes code amount control by controlling the quantization step width according to the code storage amount of the rate buffer. In the conventional code amount control, as shown in FIG. 3, the relationship between the quantization step width with respect to the code accumulation amount of the rate buffer is expressed by a linear function, and the quantization step from the code accumulation amount of the rate buffer is used. Width is required (MPEG (Motion Picture Image Coding Experts
Group) SM (see MPEG 90/041 document).

[0004]

However, in the method of obtaining the quantization step width from the code storage amount of the rate buffer by using the relation that the quantization step width with respect to the code storage amount of the rate buffer is expressed by a linear function. , There are the following problems.

That is, it is necessary to increase the capacity of the rate buffer in order to widely support an image with a small code amount to an image with a large code amount and maintain a preferable image quality.
Increasing the capacity of the rate buffer in this way increases the delay time.

On the other hand, in order to reduce the capacity of the rate buffer in order to shorten the delay time, it is necessary to increase the control sensitivity of the quantization step width with respect to the rate buffer storage amount. If the control sensitivity of the quantization step width to the rate buffer storage amount is increased, the fluctuation of the quantization step width becomes large even in an image with a small code amount, and the image quality deteriorates.

It is an object of the present invention to provide an image coding apparatus capable of shortening the delay time due to the rate buffer and maintaining the high image quality in the part which has a great influence on the image quality.

[0008]

An image coding apparatus according to the present invention is based on an encoder, a late buffer in which a code output from the encoder is stored, and an encoding based on a code storage amount of the late buffer. In an image coding apparatus provided with an arithmetic unit for determining a code amount control parameter of a rate buffer, the arithmetic unit has a rate of change of the code amount control parameter with respect to the code accumulation amount of the late buffer in an area where the code accumulation amount of the late buffer is small. In the region where the code accumulation amount of the late buffer is large and the code accumulation amount of the late buffer is large, a nonlinear function that increases the rate of change of the code amount control parameter with respect to the code accumulation amount of the late buffer is used to calculate the code amount control parameter from the code accumulation amount of the late buffer. It is characterized by seeking.

The code amount control parameter includes a quantization step width, a normalization coefficient and the like.

[0010]

In the area where the code storage amount of the late buffer is small, the rate of change of the code amount control parameter with respect to the code storage amount of the late buffer is small, and in the region where the code storage amount of the late buffer is large, the code storage amount of the rate buffer is small. The code amount control parameter is obtained from the code storage amount of the late buffer using a non-linear function that increases the rate of change of the amount control parameter. Therefore, good image quality can be obtained even if a late buffer having a relatively small capacity is used.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the structure of an image coding apparatus.

The video signal supplied to the input terminal 1 is encoded by the encoder 2. The code output from the encoder 2 is guided to the rate buffer 3 and output from the output terminal 4 at a constant transmission rate.

The amount stored in the rate buffer 3 is sent to the calculator 5. The arithmetic unit 5 calculates the quantization step width from the storage amount of the rate buffer 3 based on a predetermined input / output characteristic, and controls the quantization step width used in the encoder 2.

FIG. 2 shows the input / output characteristics of the arithmetic unit 5.

The input / output characteristic of the computing unit 5 is represented by a function (straight line a) in which the slope of the quantization step width with respect to the code storage amount is small in the region where the code storage amount of the rate buffer 3 is small. In the area where the storage amount is large, the function is a non-linear function as indicated by the function (straight line b) having a large slope of the quantization step width with respect to the storage amount of code.

Therefore, since a rate buffer having a relatively small capacity can be used as compared with the rate buffer for the conventional arithmetic unit 5 having the input / output characteristic of FIG. 3, the delay time can be shortened. In addition, since the parameter can be gently changed in a portion having a small step size that affects the image quality, a good image quality can be obtained.

In the above embodiment, the arithmetic unit 5 controls the quantization step width based on the code storage amount of the rate buffer 3, but controls the normalization coefficient based on the code storage amount of the rate buffer 3. An arithmetic unit may be used. In such an arithmetic unit, its input / output characteristic is represented by a function having a small inclination of the normalization coefficient with respect to the code accumulation amount in the area where the code accumulation amount of the rate buffer 3 is small, and In a large area, a non-linear function is used in which the slope of the normalization coefficient with respect to the code storage amount is represented by a large function.

[0018]

According to the present invention, since a late buffer having a relatively small capacity can be used, the delay time can be shortened. In addition, high image quality can be maintained in a portion that greatly affects the image quality.

[Brief description of drawings]

FIG. 1 is an electrical block diagram showing a configuration of an image encoding device.

FIG. 2 is a graph showing input / output characteristics of a computing unit.

FIG. 3 is a graph showing input / output characteristics of a conventional arithmetic unit.

[Explanation of symbols]

 2 encoder 3 late buffer 5 arithmetic unit

Claims (3)

[Claims] 1. An encoder, a rate buffer in which a code output from the encoder is accumulated, and an arithmetic unit which determines a code amount control parameter of the encoder based on the code accumulation amount of the rate buffer. In the provided image coding apparatus, the arithmetic unit has a small rate of change of the code amount control parameter with respect to the code accumulation amount of the late buffer in a region where the code accumulation amount of the late buffer is small, and an area where the code accumulation amount of the late buffer is large. An image coding apparatus, wherein the code amount control parameter is obtained from the code storage amount of the late buffer by using a non-linear function such that the rate of change of the code amount control parameter with respect to the code storage amount of the rate buffer increases. 2. The image coding apparatus according to claim 1, wherein the code amount control parameter is a quantization step width. 3. The image coding apparatus according to claim 1, wherein the code amount control parameter is a normalization coefficient.