JPH0251917A - Inter-frame predictive coding transmission method - Google Patents

Abstract

PURPOSE:To eliminate the effect of an error caused on a transmission line and to improve the picture quality by multiplying a value of the unity or below with an inter-frame prediction signal of an inter-frame coding and decoding device is response to the presence of the occurrence of a transmission line error. CONSTITUTION:Inter-frame predictive coding and decoding devices A, B are opposed to each other with lines L1, L2 sending a compressed picture data and lines L3, L4 sending error occurrence information on the transmission line inbetween. The control gives multiplication of the unity as a prediction coefficient to its inter-frame prediction signal. If any error takes place on the transmission line, the error is detected and transmission line error detection information is sent to the opposite side and the control multiplying the value below the unity is applied to an output signal of a frame memory of the opposite and its own inter-frame prediction coding and decoding devices thereby applying the leakage integration type inter-frame predictive coding. Thus, the effect of an error occurred on the transmission line is eliminated without almost deteriorating the transmission efficiency and the picture quality is improved.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a coding transmission method for highly efficiently coding and transmitting a television signal, and particularly for frames in which the influence of errors on the transmission path propagates to the next frame. The present invention relates to an inter-predictive coding transmission method.

[Conventional technology]

Conventionally, in this type of interframe predictive coding transmission method, the basic configuration of the interframe predictive coding circuit and interframe predictive decoding circuit is that the coefficient to be multiplied by 1 is used in the prediction coefficient circuit 22 or 54 shown in FIG. If a transmission path error occurs, the effect will propagate to the next frame and will not be resolved forever.
Intraframe predictive coding has been used to deal with such transmission path errors.

(Problems to be Solved by the Invention) In the conventional interframe predictive coding transmission method described above, as a measure against errors occurring on the transmission path, intraframe predictive coding is sometimes used for transmission. The disadvantage was that the transmission efficiency deteriorated.

[Means to solve the problem]

In order to eliminate such drawbacks, the interframe predictive coding transmission method according to the present invention detects transmission path errors using a parity signal added to the image signal, and performs an error function depending on whether or not a transmission path error occurs. Therefore, the interframe prediction signal of the interframe prediction encoding/decoding device connected to the transmission path where the error occurred is multiplied by 1 or a value less than 1.

[Effect]

In the interframe predictive coding transmission method according to the present invention, the influence of errors occurring on the transmission path can be eliminated without substantially reducing transmission efficiency, and as a result, image quality can be improved.

〔Example〕

FIG. 1 is a diagram illustrating the principle of the interframe predictive coding transmission method according to the present invention. In the figure, two interframe predictive coding/decoding devices A and B use lines L1 and L2 for transmitting compressed image data, and lines L3 and L4 for transmitting error occurrence information on the transmission path. They are facing each other through the

In FIG. 1, consider the case where an error occurs on the transmission path. In the interframe predictive coding/decoding devices A and B, the interframe predictive signal is normally multiplied by "1" as a prediction coefficient, and the frame memory output signal simply becomes the interframe predictive signal as it is. Interframe predictive coding is performed. The prediction coefficient is a coefficient by which the frame memory output signal is multiplied.
The frame memory output signal multiplied by this becomes the interframe prediction signal.

If an error occurs on the transmission path, the error is detected, transmission path error detection information is transmitted to the other party,
Leak integration type interframe prediction is performed by controlling the output signals of the frame memory of the interframe predictive coding circuit on the other side and the interframe predictive decoding circuit on the own side to be multiplied by a value less than "1" as a prediction coefficient. Perform encoding. In this way, by performing leakage integration type interframe predictive coding for errors on the transmission path, the effects of transmission path errors can be efficiently suppressed without substantially reducing transmission efficiency.

As an embodiment of the interframe predictive coding transmission method according to the present invention, two examples will be shown in which two devices A and B are facing each other. FIG. 2 is a block diagram for explaining the first embodiment of the interframe predictive coding transmission method according to the present invention, and FIG. 3 is a block diagram for explaining the second embodiment.

Figure 2 shows two devices A and B facing each other via a transmission line 15. In the figure, 1 and 11 are input terminals into which image signals are input, 2 and 12 are interframe predictive coding circuits, 3 and 14 are multiplexing circuits, 4.8 is a separation circuit, and 5 and 9 are interframe prediction A decoding circuit, 6.13 is a prediction coefficient control circuit, 7.10 is an output terminal to which an image signal is output, 15 is a transmission path, and 16.17 is a delay circuit.

The configurations of the interframe predictive coding circuits 2 and 12 and the interframe predictive decoding circuit 5.9 are shown in FIG.
The configuration of the prediction coefficient control circuit 6.13 is shown in FIG. As shown in FIG. 4, the interframe predictive coding circuit 2.12 and the interframe predictive decoding circuit 5.9 have a general interframe coding/decoding loop, as well as a predictive coefficient circuit 22.54 and a parity calculation circuit. It has a circuit 25.55.

The prediction coefficient circuit 22 is controlled by a prediction coefficient control signal a, and outputs a prediction signal C by multiplying the output signal of the frame memory 23, which is an input signal, by a prediction coefficient.

FIG. 6 shows, as an example of the prediction coefficient circuit 22, the values of the prediction coefficient α for the values al to a4 of the prediction coefficient control signal a. In the example shown in the figure, the prediction coefficient control signal a
If the value of is 21 or less, the prediction coefficient is set to 1'', the output signal of the frame memory 23 is used as the prediction signal C,
When the value of the prediction coefficient control signal a exceeds al, as the value increases, the prediction coefficient is set to a small value less than "1" and leakage is applied more strongly. The prediction coefficient circuit 54 is also a similar circuit.

The parity calculation circuit 25 calculates parity for each bit brain in one frame period for the local decoded signal d and outputs a parity signal e, and the parity calculation circuit 55 also does the same.

The delay circuit 16 shown in FIG. 2 is a circuit for delaying the prediction coefficient control signal a to obtain timing with the difference image signal f, and the same applies to the delay circuit 17.

The prediction coefficient control circuit 6 receives two parity signals g and h.
The error bit detection circuit 62 and the weighting circuit 61 detect transmission path errors and output a prediction coefficient control signal according to the weight of the error, as shown in FIG. configured. This error bit detection circuit 62 compares the parity signal g and the parity signal ri, taking into account delay, and detects the presence or absence of an error for each bit brain. Furthermore, a circuit 61 performs different weighting depending on the presence or absence of an error for each bit brain, and outputs a prediction coefficient control signal i as a result.

For example, as a simple example, in the case of an N-bit image signal,
The weighting for the parity error of the 0th bit (1≦n≦N) is w(nl (however, w (1)≦W(2)≦・
...5w (N-1) 5w (N) ), and the 0th
The presence or absence of an error in the digit bit (1≦n≦N) is determined by e(n) (however, when e (nl = 1, there is an error, e (n)
= 0, there is no error), then the prediction coefficient control signal a, which is the output thereof, can be defined as follows.

a=Σe　(nl　w　(nl The prediction coefficient control circuit 13 also operates in a similar manner.

Below, the operation when an error occurs on the transmission path Lb from the device B side to the device A side will be described.

In FIG. 2, data j sent from the opposite station (device B) is input to a separation circuit 4, and prediction coefficient control signals a,
The compressed image signal is separated into a parity signal g and a prediction coefficient control signal l.

In FIG. 2, consider the case where an error occurs on the transmission path Lb. In the error bit detection circuit 62 (see FIG. 5) of the prediction coefficient control circuit 6, the parity signal g separated by the separation circuit 4 and the parity signal ri calculated by the parity calculation circuit 55 of the interframe predictive decoding circuit 5 are processed. are compared in consideration of delay, errors are detected for each bit brain, and a circuit 61 applies different weights to the results for each bit, and outputs the result as a prediction coefficient control signal i. This prediction coefficient control signal i is transmitted through the transmission path La
It is transmitted to the device B side through.

The prediction coefficient control signal i transmitted to the device B side is sent to the separation circuit 8
is separated from the transmission data as a prediction coefficient control signal m, which is manually inputted to the prediction coefficient circuit 22 of the interframe predictive encoder 12 to set the prediction coefficient to "1" or "1".
For example, in the above example, if a relatively high-order bit error occurs, the prediction coefficient is controlled to a value less than “
Set it to a small value less than "1" and apply strong leakage, and set the prediction coefficient to "1" for lower bit errors and ignore the bit error, or set it to a value less than "1" and close to "1" and leak it. Control is performed to apply weakly.

On the other hand, this prediction coefficient control signal m is timed with the compressed image data n by a delay circuit 17 and is sent to the device A side through the transmission line Lb as a prediction coefficient control signal p. This prediction coefficient control signal p is separated as a prediction coefficient control signal 1 by the separation circuit 4 in the device A, and is input to the prediction coefficient circuit 54 of the interframe predictive decoding circuit 5 to control the prediction coefficient.

In this way, if no transmission path error occurs, the prediction coefficient is set to "1" and normal interframe coding is performed, and if a transmission path error occurs, both the transmitting side and the receiving side leak and transmit. The influence of road error propagation can be suppressed.

The above description also applies to the case where an error occurs on the transmission path La.

Compared to the first embodiment, the second embodiment described below transmits the prediction coefficient control signal to the interframe predictive coding circuit of the opposite station, and also transmits the prediction coefficient control signal to the interframe predictive coding circuit of the own station. The difference is that the prediction coefficients of the predictive decoding circuit are controlled.

FIG. 3 is a block system diagram of the second embodiment. In the figure, except for delay circuits 18 and 19, each block operates in the same manner as in the first embodiment.

In device A, the prediction coefficient control signal 11 is transmitted to the encoding circuit 12 of the opposite station to control the prediction coefficient, and
In order to control the prediction coefficients of the decoding circuit 5 of its own station, the delay circuit 18 delays the prediction coefficient control signal 11 by a predetermined amount.

Let us consider a case where an error occurs on the transmission line Lb as in the first embodiment. This error is detected by the prediction coefficient control circuit 6 by comparing the two parity signals g and parity signal 1. Further, the prediction coefficient control signal 11, which is the output of the prediction coefficient control circuit 6, is transmitted to the transmission line L.
a to the device B side, and controls the prediction coefficients of the interframe predictive coding circuit 12 as a prediction coefficient control signal m. On the other hand, the prediction coefficient control signal i1 is delayed by a predetermined time by the delay circuit 18 in order to match the timing when the prediction coefficients of the interframe predictive coding circuit 12 are controlled, and the prediction coefficient control signal q The prediction coefficients of the interframe predictive decoding circuit 5 of the local station are controlled as follows.

In this way, similarly to the first embodiment, the prediction coefficients on the transmitting side and the receiving side can be controlled depending on the presence or absence of a transmission path error.

〔Effect of the invention〕

As explained above, according to the present invention, transmission path errors are detected by multiplying the interframe prediction signal of the interframe encoding/decoding device by 1 or a value less than 1, depending on whether or not a transmission path error has occurred. If this does not occur, set the prediction coefficient to 1.
If a transmission path error occurs, detect the error bit by bit and apply a strong leak for errors in the upper bits, and apply a weak leak for errors in the lower bits. As a result, the effects of errors occurring on the transmission path can be eliminated with almost no reduction in transmission efficiency, and as a result, image quality can be improved.

[Brief explanation of the drawing]

FIG. 1 is a principle diagram illustrating an interframe predictive coding/decoding device to which the present invention is applied, and FIGS. FIG. 4 is a block system diagram showing an interframe predictive coding/decoding device for explaining the second embodiment, FIG. 4 is a block system diagram showing an interframe predictive coding circuit and an interframe predictive decoding circuit, and FIG. FIG. 6 is a block diagram showing the prediction coefficient control circuit, and a characteristic diagram showing the characteristics of prediction coefficients versus prediction coefficient control signals. 1.11... Input terminal, 2.12... Inter-frame predictive coding circuit, 3.14... Multiplexing circuit, 4.8...
- Separation circuit, 5, 9... interframe predictive decoding circuit,
6.13... Prediction coefficient control circuit, 7.10... Output terminal, 15... Transmission line, 16.17... Delay circuit. Patent applicant: NEC Corporation

Claims (1)

[Claims] An interframe predictive coding/decoding device that detects a transmission path error using a parity signal added to an image signal, and is connected to the transmission path where the error occurs, depending on whether or not the transmission path error occurs. 1 for the interframe prediction signal of
Or, an interframe predictive coding transmission method characterized by multiplying by a value less than 1.