JPH0690442A - Leak forecast encoding system - Google Patents

Abstract

PURPOSE:To surely improve the precision of encoding while suppressing the extension of the circuit scale. CONSTITUTION:A difference means 11 which obtains the difference between each of transmission information given in series and the leak forecast value obtained based on preceding transmission information to obtain the forecast error and quantizes and sends this forecast error, a forecast means 13 which subjects the forecast error quantized by the difference means 11 to decoding processing based on the characteristic of transmission information and integral processing of the error included in the forecast error by quantization to obtain the forecast value of succeeding transmission information and multiplies this forecast value by a leak coefficient given as a positive value smaller than one, and a precision correction means 15 which matches the precision of the value obtained by multiplying the forecast value by the leak coefficient by the forecast means to the precision of transmission information to obtain the leak forecast value are provided to constitute this system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for predictively coding transmission information and transmitting the same by multiplying the predicted value of the transmission information obtained in the process of encoding by a positive coefficient of less than 1. The present invention relates to a leak predictive coding method for converging a bit error generated at a receiving end due to a bit error generated in 1 to a small value in a time series.

[0002]

2. Description of the Related Art In a system having a large amount of information such as an image transmission system, the transmission information is transmitted in order to improve the transmission efficiency by compressing the information amount and to effectively use the radio frequency and the transmission line. A predictive coding method is adopted in which subsequent transmission information is predicted based on the characteristic of 1 and the error component of the prediction is coded. However, in a system intended for real-time communication such as a videophone system, even if information is lost on the transmission path for some reason, it is not always possible to perform retransmission based on the communication procedure or the like. Therefore, in such a system, the leak predictive coding method is adopted in order to quickly converge a large bit error generated at the receiving end due to the above-mentioned loss of information to a small value.

FIG. 5 is a diagram showing a configuration example of a conventional leak predictive coding system. In the figure, in the encoding device, an integer value image (transmission) is input to the frame memory (FM) 51.
Given the information, its output is connected to one input of subtractor 52. The bus terminal of the frame memory 51 is connected to the bus terminal of the motion detector 53. The output of the subtractor 52 is connected to the transmission line and the input of the inverse quantizer 55 _S via the quantizer 54. The output of the inverse quantizer 55 _S is the adder 5
6 _S is connected to one input, and its output is connected to one input of a frame memory (FM) 57 _S. The output of the frame memory 57 _S is a cascaded multiplier (LF).
It is connected to the other input of the adder 56 _{S and} the other input of the subtractor 52 via 58 _S and the integer conversion unit 59 _S.
The output of the motion detector 53 is connected to the other terminal of the frame memory 57 _S and the transmission path.

In the decoding device, an inverse quantizer 55 _R is arranged to face the quantizer 54 of the encoder via a transmission line, and the output of the inverse quantizer 55 _R is one of the adders 56 _R. Connected to the input of. The output of adder 56 _R provides the decoded image information to one input of frame memory (FM) 57 _R. The output of the frame memory 57 _R is connected to the other input of the adder 56 _R through a cascade-connected multiplier (LF) 58 _R and an integer conversion unit 59 _R , and the frame memory 5 _R is connected.
The other input of 7 _R is connected to the output of the motion detection unit 53 of the encoding device via the above-mentioned transmission path.

In such a leak predictive coding system, image information to be coded is stored in the frame memory 51 for each frame given by the information. The motion detection unit 53 reads out the image information accumulated in this way and performs a calculation based on a predetermined procedure to obtain a motion vector indicating a relative motion of each frame given by the image information with respect to the preceding frame. It is detected and given to the frame memory 57 _S , and its motion vector is transmitted to the decoding device via the transmission path.

The subtractor 52 takes the difference between the image information for each pixel accumulated in the frame memory 51 and the predicted value of the image information given from the integerizing unit 59 _S based on the process described later, and the difference ( The prediction error) is quantized via the quantizer 54 and sent to the transmission line. The inverse quantization unit 55 _S converts the prediction error obtained by such quantization into a numerical value of the word length before the quantization, and the adder 56 _S calculates the converted prediction error and the above-described prediction value. By taking the sum, the error (quantization noise) caused by the above-mentioned quantization is integrated.
The frame memory 57 _S compensates for the motion of the image between the frames by accumulating the prediction error subjected to such integration processing for one frame and reading it out according to the motion vector described above. The multiplier 58 _S multiplies the prediction error given by such compensation by a predetermined leak coefficient LF (0 <LF <1) and outputs it. The integer conversion unit 59 _S rounds off the value of the first decimal place of the product thus obtained by the multiplier 58 _S or rounds down the value to the nearest decimal place to convert it into an integer, thereby converting the image information to be encoded. To obtain a predicted value with a matching precision.

Further, such a predicted value is obtained through a feedback circuit having the same configuration as that of the decoding device, as shown by a dotted frame in FIG. 5, and the feedback amount of the circuit is quantized as described above. It is obtained by integrating the quantization noise generated in the unit 54. Therefore, the codeword generated by smoothing the component of such quantization noise is transmitted to the transmission line.

In the decoding device, the inverse quantizer 55 _R inversely quantizes the prediction error quantized by the quantizer 54 of the encoder, and the adder 56 _{R and} the frame memory 57 _R , the multiplier 58 _R and The positive number conversion unit 59 _R includes a frame memory 57 _S on the side of the encoding device, a multiplier 58 _S, and a positive number conversion unit 59 _S.
Similar to _S , the predicted value of the subsequent image information is obtained. The adder 56 _R decodes the image information by adding the prediction value thus obtained and the above-described prediction error.

As described above, in the leak predictive coding method, the predictive value obtained according to the characteristic of the preceding transmission information is multiplied by the leak coefficient to be used for predictive coding,
Since the receiving end performs the decoding process corresponding to the encoding,
Even if a bit error occurs on the transmission line, the bit error to the decoded output converges to "0" within a fixed time during the decoding process. That is, in a system that adopts such an encoding method, the spread of bit errors to the decoded output due to the loss of bits or errors that occur in the transmission path is reduced, and the information obtained by the decoding performed at the receiving end is reduced. The reliability of is improved.

[0010]

By the way, in such a conventional leak predictive coding system, the accuracy of the numerical operation performed in the process of encoding (decoding) is generally determined by the hardware or software that performs the operation. Limited by the size of the wear, the required operation speed, and so on. Also, the integer conversion unit 59
_S (59 _R ) is effective in reducing the circuit scale by matching the accuracy of the image information to be transmitted and the encoded output (decoded output).

However, it is performed in the process of encoding (decoding).
Multiplier 58 out of the numerical operations_S(59_R) Other than multiplication
All numerical operations in are performed with integer precision. According to
For example, for simplicity, the image given by the transmission information is
Assuming that the image is a still image, the leak coefficient LF is 0.875 (= 1-
(1/2)³) And the adder 56 of the encoding device_SAhead
Data S that is output according to the image information that has been executed_iThe value of the
Is "2", but the addition of the decoding device corresponds to that value.
Vessel 56_RData output from R_iValue is on the transmission path
If it becomes "8" due to the bit error that occurred in
Subsequent image information X that takes a value of "2"_{i + 1}To the encoder
When given, the adder 56 corresponding to the image information_SOr
Data S output from_{i + 1}The value of is S_{i + 1}= X_{i + 1}-(LF x S_i) Integer part + (LF × S
_iThe integer part of) becomes "2" given by the equation, and similarly, the adder 56
_RData R output from_{i + 1}The value of R_{i + 1}= (LF x R_i) Integer part + X_{i + 1}-(LF x S
_iThe integer part of) becomes "8" given by the formula. That is, the above-mentioned
The error is an integer conversion unit 59 _S, 59_RWrong rounding
Because it does not converge to “0” due to the difference,
Bit error spreads.

An object of the present invention is to provide a leak predictive coding system which can surely improve the coding accuracy while suppressing an increase in circuit scale.

[0013]

FIG. 1 is a block diagram showing the principle of the invention described in claim 1. In FIG. According to the present invention, for each transmission information given in series, the difference means 11 which obtains a prediction error by taking the difference from the leak prediction value obtained based on the preceding transmission information, and quantizes and outputs the prediction error 11
Then, the prediction error quantized by the difference means 11 is subjected to a decoding process based on the characteristics of the transmission information, and an integration process of the error included in the prediction error accompanying the quantization is performed to obtain a prediction value of the subsequent transmission information. And the accuracy of the value obtained by multiplying the predicted value by the leak coefficient given by a positive number less than 1 and the accuracy obtained by the predicting means 13 by multiplying the predicted value by the leak coefficient to the accuracy of the transmission information. An accuracy correction means 15 for obtaining a leak prediction value is provided.

FIG. 2 is a block diagram showing the principle of the invention described in claim 2. According to the present invention, for each transmission information given in series, the difference between the leak prediction value obtained based on the preceding transmission information is obtained to obtain a prediction error, and the difference means 21 for quantizing the prediction error, and the difference The prediction error quantized by the means 21 is subjected to a decoding process based on the characteristics of the transmission information and an integration process of the error included in the leak prediction value due to the quantization to obtain the prediction value of the subsequent transmission information, A predicting unit 23 that multiplies the predicted value by a leak coefficient given as a positive number less than 1, and a predictive value of the value obtained by multiplying the predictive value by the leak coefficient by the predicting unit 23 matches the accuracy of the transmission information and the predicted leak value. In the leak predictive coding system including the accuracy correction means 25 for obtaining
It is characterized by including an error correction means 27 for detecting an error caused by the above and performing negative feedback to the prediction means 23.

[0015]

In the leak predictive coding system according to claim 1,
The prediction unit 13 performs a decoding process based on the characteristics of the transmission information with respect to the prediction error obtained by the difference unit 11 and quantized according to the preceding transmission information, and integrates the error included in the prediction error with the quantization. By performing processing, the predicted value of the subsequent transmission information is obtained, and the predicted value is multiplied by the leak coefficient. The accuracy correction unit 15 matches the accuracy of the value obtained by the multiplication with the leak coefficient with the accuracy of the transmission information, and thus the difference prediction unit 11 uses the leak prediction as a reference for obtaining the prediction error of the subsequent transmission information. Get the value.

That is, since no error occurs due to the accuracy adjustment performed by the accuracy correction means 15 in the process of calculating the above-described predicted value, the accuracy adjustment is performed in parallel in the above-described decoding process and integration process. The accuracy of leak predictive coding is improved as compared with the conventional example that has been performed.

In the leak predictive coding method according to the second aspect, the predicting means 23 performs a decoding process on the basis of the characteristics of the transmission information in the prediction error quantized by the difference means 21 in accordance with the preceding transmission information. And an integration process of an error included in the leak prediction value due to the quantization, a prediction value of subsequent transmission information is obtained, and the prediction value is multiplied by the leak coefficient. The accuracy correction unit 25 matches the accuracy of the value obtained by the multiplication of the predicted value and the leak coefficient with the accuracy of the transmission information, so that the difference unit 21 can be used as a reference for obtaining the prediction error of the subsequent transmission information. Get the leak prediction value to use.

Since the error correction means 27 detects an error caused by the adjustment of the accuracy performed by the accuracy correction means 25 in the process of calculating the above-mentioned leak prediction value and negatively feeds it back to the prediction means 23, The accuracy of leak predictive coding is improved as compared with the conventional example in which the accuracy adjustment is performed in parallel during the above-described decoding processing and integration processing.

[0019]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 3 is a diagram showing an embodiment corresponding to the invention described in claim 1.

In the figure, parts having the same functions and configurations as those shown in FIG. 5 are designated by the same reference numerals, and the description thereof will be omitted here. In the present embodiment, the configuration that characterizes the present invention is based on the coding prediction loop formed by the adder 56 _S , the frame memory 57 _S , the multiplier 58 _S, and the integerization unit 59 _S in the conventional coding device. The integer conversion unit 59 _S is removed, and an integer conversion unit 31 that replaces the integer conversion unit 59 _S is arranged between the new coding prediction loop formed in this way and the other input of the subtractor 52.
Furthermore, the decoding device is formed by removing the integerizing unit 59 _R from the conventional decoding device.

The correspondence relationship between this embodiment and the block diagram shown in FIG. 1 is as follows: frame memory 51, subtractor 5
2 and the quantizer 54 correspond to the difference means 11, and the inverse quantizer 55 _S , the adder 56 _S , the frame memory 57 _S (including the motion detector 53) and the multiplier 58 _S correspond to the predictor 13. The integer conversion unit 31 corresponds to the accuracy correction unit 15.

The operation of this embodiment will be described below. Coding
In the device, the integer conversion unit 59_SThe integer conversion unit 31 that changes to
Is placed outside the prediction loop and is an integer at the decoder
Conversion part 59_RBecause there is no
The problem on the transmission line under the same conditions as the “issue to be decided”
Error occurs and addition is caused by the bit error.
Vessel 56_SData S output from _iTakes a value of "2"
And adder 56_RData R output from_iIs "8"
In the state of taking a value, the subsequent image information X_{i + 1}Corresponds to (= 2)
Adder 56_SData S output from_{i + 1}The value of is S_{i + 1}= X_{i + 1}-(LF x S_i) Integer part + LF × S_i
 It becomes “2.75” given by the formula of ′ ′, and similarly, the adder 5
6_RData R output from_{i + 1}The value of R_{i + 1}= LF × R_i+ X_{i + 1}-(LF x S_i) Integer part
 It becomes "8" given by the formula of '. Thus data R
_{i + 1}Have the same value, but further subsequent image information X_{i + 2}of
If the value does not change, it is obtained according to the image information.
Data R_{i + 2}The value of is S_iIs preceded by
S obtained by_{i + 1}Substituting (= 2.75) into the above formula '
It becomes "7".

That is, in the process of the arithmetic operation performed in the coding prediction loop, since the integer conversion process by the positive number conversion unit 31 is not performed, the accuracy of the operation is higher than that of the conventional example, and the adder 56 is used. The value of the data R _i obtained as a decoded output from _R is the value of the newly given image data X _{i + 1 and} the value of the data S _i previously obtained based on the value based on the value. It is sequentially obtained by repeating the arithmetic operation shown in the above equation 'based on and.

That is, the conventional integer conversion unit 59_SSame structure as
Data R using the integer conversion unit 31 _iValue of image information X
_{i + 1}It is possible to reliably converge to the value of
Therefore, it occurred in the transmission line while suppressing the increase in circuit scale.
Definitely small error in decoded output due to bit error
Is limited to the value, and the spread of the error on the time axis is
Limited to a certain range.

FIG. 4 is a diagram showing an embodiment corresponding to the invention described in claim 2. In FIG. In the figure, parts having the same functions and configurations as those shown in FIG. 5 are designated by the same reference numerals, and the description thereof will be omitted here.

A feature of the present invention is that, in this embodiment, an adder 41 _S is arranged between the frame memory 57 _S and the multiplier 58 _S, and the output of the integer conversion unit 59 _S is changed to the input data of the integer conversion unit 59 _S. To the coding prediction loop via the subtracter 42 _S for taking the difference between the output data and the output data, the frame memory 43 _S operating in conjunction with the frame memory (FM) 57 _S , and the adder 41. is there.

The correspondence between this embodiment and the block diagram shown in FIG. 2 is as follows.
2 and the quantizer 54 correspond to the difference means 21, and include an inverse quantizer 55 _S , an adder 56 _S , a frame memory 57 _S (including the motion detector 53), an adder 41 _S and a multiplier 5.
8 _S corresponds to the prediction unit 23, the integer conversion unit 59 _S corresponds to the accuracy correction unit 25, and the subtractor 42 _S and the frame memory 43 _S (including the motion detection unit 53) include the error correction unit 2.
Corresponds to 7.

The operation of this embodiment will be described below. The integer conversion unit 59 _S is arranged in the coding prediction loop as in the conventional example, and the error component associated with the integer conversion performed by the integer conversion unit is subtracted by the subtracter 4.
Detected by 2 _S. The error component thus detected is given to the adder 41 _S via the frame memory 43 _S which operates in synchronization with the frame memory 57 _S, and the adder 41 _S encodes the error component thus given. Negative feedback is provided in the prediction loop.

That is, in this embodiment, similarly to the embodiment shown in FIG. 3, since the prediction value obtained by the arithmetic operation performed in the coding prediction loop does not include an error due to integerization, the transmission is performed. The error in the decoded output due to the bit error occurring on the path is reliably limited to a small value, and the spread of that error on the time axis is limited to a certain range.

Further, in the present embodiment, since the above-mentioned error loop is formed separately from the coding prediction loop via the adder 41 _S , the subtractor 42 _S and the frame memory 43 are formed.
By selecting whether or not _S is mounted, it is possible to flexibly adapt to the level of bit error occurring in the transmission path and the required transmission quality while standardizing the circuit configuration.

Regarding the configuration of the decoding device adapted to this embodiment, the same configuration as that of the decoding device shown in FIG.
As indicated by a dotted line frame, an error loop similar to the error loop in the encoder (subtractor 42 _R , frame memory 43 _R
And an adder 41 _R ).

In this embodiment, the error loop is an integer
Conversion part 59_SInput / output of the multiplier 58 _SPlaced in front of
Adder 41_SHowever, according to the present invention,
Not limited to error loops such as
If it is a method of negative feedback to some point, what kind of configuration
Very good.

Although each of the above-described embodiments is an image transmission system to which a compensation interframe coding system is applied, the present invention is not limited to such an image transmission system and a predictive coding system is applied. Any system can be applied regardless of the content of the transmission information or the transmission method of the transmission path.

Further, in each of the above-described embodiments, the integer conversion units 31 and 59 _S perform integer conversion processing in order to match the accuracy between the image information given as an integer value and the predicted value. However, the present invention is not limited to such processing,
Any processing may be applied as long as the accuracy of the transmission information and the accuracy of the predicted value corresponding to the information can be matched.

[0035]

As described above, according to the present invention, in the calculation process of calculating the leak prediction value which is the reference for obtaining the prediction error of the subsequent transmission information, the leak prediction value is collectively set in the final process of the calculation process. The accuracy of the leak prediction coding is improved by performing a process of matching the precision with the transmission information or by sequentially detecting and correcting an error that occurs when the process is performed in the calculation process.

That is, at the receiving end that receives the transmission information encoded in this way, the spread of bit errors to the decoded output due to the missing bits or errors that occurred on the transmission path is the value of the leak coefficient. Is surely suppressed according to.

Therefore, in the communication system to which the leak predictive coding method is applied, the transmission quality and reliability are improved.

[Brief description of drawings]

FIG. 1 is a principle block diagram of the invention according to claim 1.

FIG. 2 is a principle block diagram of the invention described in claim 2.

FIG. 3 is a diagram showing an embodiment corresponding to the invention described in claim 1.

FIG. 4 is a diagram showing an embodiment corresponding to the invention described in claim 2.

FIG. 5 is a diagram showing a configuration example of a conventional leak prediction coding system.

[Explanation of symbols]

 11, 21 Difference means 13, 23 Prediction means 15, 25 Accuracy correction means 27 Error correction means 31, 59 Integer conversion section 41, 56 Adder 42, 52 Subtractor 43, 51, 57 Frame memory (FM) 53 Motion detection section 54 Quantizer 55 Dequantizer 58 Multiplier (LF)

Claims (2)

[Claims] 1. For each transmission information provided in series,
Prediction error is obtained by taking the difference from the leak prediction value obtained based on the preceding transmission information, and the prediction error is quantized by the difference means (11) and the difference means (11). Decoding processing based on the characteristics of the transmission information to the prediction error, and integration processing of the error included in the prediction error to the prediction error to obtain the prediction value of the subsequent transmission information, the prediction value Prediction means (1 that multiplies the leak coefficient given by a positive number less than 1)
3) and accuracy correction means (15) for obtaining the leak prediction value by matching the accuracy of the value obtained by multiplying the prediction value by the leak coefficient by the prediction means (13) with the accuracy of the transmission information. A leak predictive coding method characterized by being provided. 2. For each transmission information provided in series,
Prediction error is obtained by taking the difference from the leak prediction value obtained based on the preceding transmission information, and the prediction error is quantized by the difference means (21) and the difference means (21). Decoding processing based on the characteristics of the transmission information to the prediction error, and integration processing of the error included in the leak prediction value accompanying the quantization are performed to obtain the prediction value of the subsequent transmission information, and the prediction value. And a prediction unit (23) for multiplying a leak coefficient given by a positive number less than 1, and an accuracy of a value obtained by the prediction unit (23) by multiplying the predicted value by the leak coefficient to the accuracy of the transmission information. In the leak predictive coding method, which comprises an accuracy correction unit (25) for obtaining the leak prediction value, an error generated in the accuracy correction unit (25) is detected in the leak prediction value to detect the prediction unit (23). Negative return to Leak predictive coding method characterized by comprising the error correction means (27) for.