JP2862555B2 - Image coding control method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a system for encoding an image signal and performing digital transmission with a reduced bit rate, such as a video conference, a video phone, and digital television transmission. The present invention relates to a method of predictively encoding an image signal and encoding and transmitting a prediction error signal.

(Prior Art) There are many image communication services, such as a videophone and a videoconference, for encoding a moving image signal and digitally transmitting the image signal at a reduced bit rate. In such a moving image transmission method, in order to reduce the bit rate, when encoding the pixel or the pixel block, prediction is performed using pixels that have already been encoded and transmitted and decoded, and only the prediction error is calculated. Predictive coding for coded transmission is often used.

For predictive coding, depending on the position of the pixel used for prediction,
Broadly speaking, prediction is performed using decoded pixels that have already been encoded and transmitted in the same frame, and prediction is performed between rows that perform prediction using decoded pixels that have already been encoded and transmitted in the previous frame. It is known that the latter is more efficient for images with prediction and less motion. Also, as a technology to further reduce the bit rate,
Motion-compensated inter-frame prediction in which motion is detected from a moving image signal and pixels used for prediction are selected from the previous frame using the motion vector is also used.

(Problems to be solved by the invention) However, when the transmission bit rate is extremely low, all significant information cannot be transmitted, and the encoded image quality deteriorates. In this state, since the decoded pixels used for prediction contain large noise, the prediction accuracy itself deteriorates, the power of the prediction error signal increases, and the amount of information required to encode the signal increases. . In such a case, almost all of the data to be coded is noise. Therefore, despite the large amount of information required for coding, the coded image quality is not improved, and meaningless bits are consumed. There has been a problem that a vicious circle occurs in that an effective amount of information cannot be allocated to other parts to be encoded.

The present invention is directed to solving the above-described problem of the related art, and is an image coding control method that realizes coding control with little deterioration in coding image quality.

(Means for Solving the Problems) A first feature of the present invention is that a prediction is selected from a locally decoded image signal obtained from an encoded output signal and used as a predicted value of a pixel or a pixel block to be encoded. Calculating an error power value included in the prediction signal by comparing the signal with an original image signal corresponding to the pixel or the pixel block, and comparing the calculated error power value with at least one predetermined threshold value It is another object of the present invention to control the amount of encoded information of a prediction error in encoding according to the comparison result.

A second feature of the present invention is that, in a method of orthogonally transform-encoding a prediction error in pixel block units, a locally decoded pixel used for prediction of a corresponding pixel block is compared with an original signal, and the encoding error power is calculated. If this value is large, only the low-order transform coefficients among the transform coefficients of the prediction error of the corresponding pixel block are coded and transmitted, and if this value is small, up to the high-order transform coefficients are coded. The purpose is to control the coding so that it is transmitted.

(Principle of the Invention) The principle of the present invention will be described with reference to FIG. FIG. 1 is a block diagram of a predictive coding apparatus according to the present invention. Predicted signal VP selected by predictor 5 from local decoded signal VL
Is input to the encoder 2 as a prediction error VE by taking the difference from the input original signal VI by the subtractor 1 and the encoded output signal VC is encoded and transmitted. The difference from the signal VI 'is also calculated by the subtractor 7. The input source signal VI 'is an original image signal corresponding to the prediction signal VP, and the calculated difference value is a prediction signal error ep of the prediction signal VP. The prediction signal error ep is converted to error power by the power calculator 8, whereby the encoder 2 and the decoder 4 encode the corresponding pixel with a smaller amount of information when the error power is large, and when the value is small. Is controlled so that the corresponding pixel is encoded with a larger amount of information.

Further, the present invention is described as follows. It is assumed that a predicted signal selected from the input original image signal VI to be encoded and the local decoded image signal is VP, and an original image signal corresponding to a pixel used as the predicted signal is VI '. At this time, an encoding control signal Cn for controlling the information amount of encoding is defined by the following equation (1).

Cn = (VP−VI ′) ² (1) As means for controlling the amount of encoded information, a prediction error (VI−VP)
In many cases, when the present invention is applied to this quantizer control, q may be controlled by the following equation (2).

q = f (Cn) (2) where f is an increasing function, an example of which is shown in FIG.

Now, in the predictive coding, the orthogonal transform is performed by the orthogonal transform coder 10 for each prediction error signal m × n pixel block generated by the predictive coder 9 as shown in FIG. In recent years, a hybrid coding method is widely used.
In this, the invention is described as follows.

Now, each pixel value of the pixel block of the input image signal to be encoded is {VI}, each pixel value of the pixel block of the locally decoded image signal used for prediction is {VP}, and the corresponding pixel value of the original image signal is Let each pixel value of the pixel block be {VI ′}. At this time, the encoding control signal Cn is defined as in equation (1) and as shown in FIG. 4, the transform coefficient r _ij (1 ≦ i ≦
m, 1 ≦ j ≦ n), a range R of coefficients for transmitting 1 ≦ i ≦ k and 1 ≦ j ≦ l is defined by the following equation (3).

R = f ₁ (Cn) (3) Here, f ₁ is a function that determines the transmission range from the error power, and is defined such that as the coding control signal Cn increases, the range R decreases. That is, when the error power of the prediction signal increases, only lower-order transmission coefficients are transmitted, and higher-order coefficients are discarded without being transmitted (even if they have significant values). There are several ways to determine the transmission range of the coefficient,
FIG. 5 shows an example of f when an 8 × 8 pixel block is used. Here, R ₁ is a transmission coefficient range when Cn <Th ₁ , R ₂ is a transmission coefficient range when Cn <Th ₂ , R ₃ is a transmission coefficient range when Cn <Th ₃ , and R ₄ is Cn < transmission coefficient range for Th _4, R ₅ is Cn <transmission coefficient range for Th _5, the transmission factor range when R ₆ is a _{Cn <Th 6, Th i (} 1 ≦ i ≦ 6) is a threshold deli , Th ₁ <Th ₂ <Th ₃ <Th ₄ <T
h ₅ <a Th _6. As shown in the figure, in the present invention, the encoding control signal Cn is provided with a plurality of thresholds Th and controlled so as to change the range of the coefficient to be encoded according to the magnitude of the encoding control signal Cn.

As described above, the present invention eliminates useless bits that do not contribute to the encoded image quality by reducing the amount of information allocated to encoding when the error power of the prediction signal is large. By allocating to a portion that contributes to the improvement of the coded image quality, the entire coded image quality can be significantly improved.

(Embodiment) FIG. 6 shows an embodiment of the motion compensation hybrid coding according to the present invention, in which a portion surrounded by a chain line is a characteristic portion added by the present invention. The local decoded signal VL is written to the frame memory 24 for prediction. At the same time, the difference from the original image signal VI 'delayed by the frame memory 15 is obtained by the subtractor 23 and the squarer 22
After the conversion into power, the data is input to the accumulator 21 and added to the accumulator 21 for one pixel block. When the processing for one pixel block is completed, the contents of the accumulator 21 are stored in the error power memory 20. That is, the error power memory 20 has
The error power of the locally decoded image signal for one frame is stored for each pixel block. Now, when the motion vector MV is detected by the motion amount detector 26 and the prediction signal VP is selected thereby, the corresponding error power is read from the error power memory 20 at the same time, and the correction of the block boundary is performed by the corrector. 17
, And is input to the memory 18 in which the relationship between the error power and the transmission coefficient range is written. As a result, the transmission range pattern is output from the memory 18, the logical product of the transform coefficient and the transmission range pattern is calculated by the mask circuit 13, and only the transform coefficient within the transmission range is input to the quantizer 14 and the encoded output signal VC Is transmitted.

When the predicted portion Bp straddles the block boundary with the area ratio a: b: c: d as shown in FIG. 7, the error powers Ea, Eb, Ec, and Ed of each block are calculated as follows. The correction error power Ee is calculated by the equation (4).

Ee = (aEa + bEb + cEc + dEd) / (a + b + c + d)
(4) (Effects of the Invention) As described above, the present invention provides a prediction signal selected from a local decoded image signal obtained from an encoded output signal and used as a predicted value of a pixel or a pixel block to be encoded. And an original image signal corresponding to the pixel or the pixel block, thereby calculating an error power value included in the prediction signal, and comparing the calculated error power value with at least one predetermined threshold. The following effects can be obtained by controlling the amount of encoded information of the prediction error in encoding according to the comparison result. That is, the estimation is performed based on the S / N of a temporally past image, and the control of the image portion to be coded is not performed based on the estimation result. The error power value (prediction value) of the predicted image signal itself for the image portion to be coded now Since the coding control is performed by controlling the amount of coded information to be transmitted), the control becomes very accurate, and
Eliminating the consumption of meaningless bits that do not contribute to the encoded image quality, and allocating the portion to the portion that contributes to the improvement of the encoded image quality, can greatly improve the overall encoded image quality.

In addition, the present invention has a configuration as shown in FIG. 6, which has a circuit for calculating error power, a memory for storing error power, a ROM for storing a pattern of a coefficient transmission range, etc. Can be realized only by adding the storage means.

Accordingly, the present invention is to provide a video communication service that encodes a moving image signal and performs digital transmission with a reduced bit rate, such as a video conference, a video phone, and a digital television transmission. And the effect is great.

[Brief description of the drawings]

FIG. 1 is a block diagram of a predictive coding apparatus in which the present invention is applied to predictive coding. FIG. 2 is a diagram showing an example of a relationship between a quantization step width q in coding information amount control and a coding control signal Cn for controlling a coding information amount. FIG. 3 is a block diagram of hybrid coding. FIG. 4 is a diagram showing an example of a range of coefficients to be transmitted in hybrid coding. FIG. 5 is a diagram showing an encoding control signal Cn and a transmission range in hybrid encoding using a block of 8 × 8 pixels as an example. FIG. 6 is a device block diagram when the present invention is applied to motion compensation hybrid coding. FIG. 7 is a diagram showing an example relating to block boundary correction in the present invention in motion compensation hybrid coding. DESCRIPTION OF SYMBOLS 1 ... Subtractor, 2 ... Encoder, 3 ... Variable delay unit, 4 ... Decoder, 5 ... Predictor, 6 ... Adder, 7 ... Subtractor, 8 ... Power calculator, 9 ... Predictive encoder, 10 ... Orthogonal transform encoder, 11 subtractor, 12 orthogonal transformer, 13 mask circuit, 14 quantizer, 15 frame memory, 16 inverse quantizer, 17 corrector, 18 memory, 19 ... an inverse orthogonal transformer, 20 ... error power memory, 21 ... accumulator, 22 ... squarer, 23 ... adder, 24 ... frame memory, 25 ... subtractor, 26 ... motion amount detector.

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-219887 (JP, A) JP-A-2-198273 (JP, A) IEICE Technical Report, IE 87-83 (1987.10) .23) p. 53-60

Claims (3)

(57) [Claims] 1. An image coding control system for predictively coding a digital image signal and transmitting a coded output signal obtained by coding a prediction error obtained by the predictive coding, the method comprising the steps of: By comparing a predicted signal used as a predicted value of a pixel or a pixel block to be encoded with an original image signal corresponding to the pixel or the pixel block, which is selected from the locally decoded image signals thus obtained,
Calculating an error power value included in the prediction signal; and calculating the calculated error power value and at least one predetermined value.
An image coding control method, comprising: comparing two threshold values; and controlling the amount of coding information of the prediction error in the coding according to the comparison result. 2. The image encoding control method according to claim 1, wherein the prediction error is encoded with a smaller amount of encoded information as the error power value increases. 3. An image coding control system for predictively coding a digital image signal and transmitting a coded output signal obtained by orthogonally transforming a prediction error obtained by the prediction coding, wherein the error power value is large. Encodes only the lower-order transform coefficients among the transform coefficients of the prediction error, and encodes the higher-order transform coefficients among the transform coefficients of the prediction error when the error power value is small. The image encoding control method according to claim 1.