JPH06132823A - Encoding system - Google Patents

Abstract

PURPOSE:To simplify processing required for encoding by efficiently compressing the signal of a transmitting object, to considerably shorten processing time for considerably reducing transmission cost, to miniaturize an encoder and to reduce the cost. CONSTITUTION:Information concerning feature points such as the maximal point, minimal point and inflecting point of a signal to be transmitted inputted to a signal input terminal 4 is extracted by an encoder 1, this information is sent onto a transmission line 2 and sent to a decoder 3, this signal to be transmitted is restored by this decoder 3 while using a three-dimensional spline function based on the information sent from the encoder 1, and this signal is outputted from a signal output terminal 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding system for encoding (transmitting (or recording)) a signal and then decoding the signal. Particularly, on the encoding side, characteristic points of a waveform (for example, a maximum point, Minimum points, inflection points, etc.) are extracted, encoded and transmitted, and on the decoding side, the signals between the transmission data are interpolated by using a spline function to be restored, and this should be transmitted. The present invention relates to an encoding method for compressing information.

SUMMARY OF THE INVENTION The present invention relates to a coding system for coding a signal for transmission or recording, and then decoding the signal. Particularly, on the coding side, characteristics of a signal to be transmitted are described. A point is selected, its position and signal value are encoded and transmitted, and the decoding side uses the position and signal value for the transmitted characteristic point to interpolate the spline function and transmit the transmitted signal in the transmitted signal. The part which did not exist is restored and the amount of information to be transmitted is reduced by this.

[0003]

2. Description of the Related Art Conventionally, as a coding apparatus for audio signals and image signals, transform coding such as a DPCM method for coding the difference between the signals by utilizing the correlation between the signals and a DCT method focusing on the energy concentration of the signal. A coding method such as a method, a vector quantization method for patterning and transmitting a signal is known.

Further, when the object to be encoded is limited to the image signal, there is also known a motion compensation technique for increasing the degree of correlation by predicting a moving part, and these techniques are selected according to the characteristics of the signal to be transmitted. In many cases, each coding method and motion compensation technology are used in combination.

As a data interpolation technique, it is known that by interpolating the measured data using a spline function, good characteristics can be obtained as compared with ordinary polynomial approximation, and CG (computer In the field of graphics), it is often used to express surface shapes.

[0006]

However, the above-mentioned conventional coding system has the following problems.

That is, most of the conventional coding systems are
Due to the nature of the signal, such as the degree of energy concentration, and the visual characteristics of human beings, only the information that can be judged to be a redundant part that needs to be transmitted is appropriately separated from the data to be transmitted. Because it is a technology for
It requires a lot of complicated processing and judgment.

Since many of these complicated processes and judgments fall into the category of "waveform coding" for coding the signal waveform itself, when a general hardware method or software method is used, There is a problem that not only the scale tends to be large, but also it takes a long time to process and it is difficult to increase the encoding speed.

In view of the above circumstances, the present invention can efficiently compress the signal to be transmitted to simplify the processing required for encoding, and can significantly reduce the processing time. It is an object of the present invention to provide a coding method that can significantly reduce the transmission cost and can achieve the downsizing and cost reduction of the coding apparatus.

[0010]

In order to achieve the above object, the coding method according to the present invention is such that, on the coding side, the positions of feature points selected from the transmitted signal and the signal values thereof are coded and transmitted. On the decoding side, the characteristic of reconstructing the transmitted signal by interpolating the signal values other than the characteristic points by the spline function using the information of the transmitted characteristic points and the signal values thereof is characterized. There is.

[0011]

In the above structure, the encoding side encodes and transmits the position of the feature point selected from the transmitted signal and its signal value, and the decoding side transmits the position of the transmitted feature point and its signal. By using the value information to reconstruct the transmitted signal by interpolating the signal values other than the characteristic points with the spline function, the signal to be transmitted is efficiently compressed and the processing required for encoding is simplified. At the same time, the processing time is significantly shortened, the transmission cost is significantly reduced, and the size and cost of the encoding device are reduced.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The basic principle of the coding system according to the present invention will be described prior to a detailed description of the coding system according to the present invention.

First, in the encoding method according to the present invention, rather than transmitting the signal waveform, only the particularly important one is extracted from the signal waveform and transmitted, and the receiving side "similarly" to this. A basic principle is to use a technique of synthesizing waveforms, which is so-called “waveform approximation coding”, to facilitate the deviceization of the encoding unit, the decoding unit, and the like.

In this case, specifically, as a characteristic point of the waveform, at least one of the (m-1) th order differential coefficients of the transmitted signal is inverted, for example, the maximum point of the transmitted signal, The minimum point (the point where the primary differential coefficient is "0" and the sign is inverted before and after it) and the inflection point (the point where the secondary differential coefficient is "0" and the code is inverted before and after it) are extracted and characterized. As points, their positions and signal values are encoded and transmitted.

Then, on the receiving side, the transmitted information is interpolated by using an m-th order spline function to restore the transmitted signal.

This is equivalent to automatically performing the variable density sampling, and the deterioration of the high frequency band can be reduced as compared with the usual interpolation technique.

An encoding method according to the present invention using the above-mentioned basic principle will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing an example of a signal transmission system to which an embodiment of an encoding system according to the present invention is applied.

The signal transmission system shown in this figure comprises an encoding device 1, a transmission line 2 and a decoding device 3, and features of the transmitted signal input to the signal input terminal 4 by the encoding device 1. The points are extracted and transmitted to the transmission line 2 to be transmitted to the decoding device 3. The decoding device 3 decodes the information transmitted from the encoding device 1 to restore the transmitted signal. , Which is output from the signal output terminal 5.

In this case, the signal input terminal 4 receives as the transmitted signal, for example, an audio signal, an image signal, a television signal, another one-dimensional signal, a two-dimensional image signal, a three-dimensional signal, and the like. Suitable pre-processing for DC (DC with motion compensation)
A signal that has been subjected to T, zigzag scan conversion, DPCM, etc.) is input.

The encoding device 1 differentiates a transmitted signal input through the signal input terminal 4 to calculate a first-order differential coefficient value, and a first differential operation circuit. The primary differential coefficient value output from 6 is differentiated again to 2
A second differential calculation circuit 7 for calculating a secondary differential coefficient value, and the transmitted data based on the primary differential coefficient value and the secondary differential coefficient value output from the first and second differential calculation circuits 6 and 7. From the control signal generation circuit 8, a control signal generation circuit 8 that generates a control signal necessary for extracting characteristic point information of the signal, and the received signal that is supplied via the signal input terminal 4 are fetched. And a transmission signal forming circuit 9 for extracting characteristic point information of the transmitted signal based on the output control signal, encoding the extracted characteristic point information, and transmitting the encoded characteristic point information to the transmission path 2.

Then, the primary differential coefficient value and the secondary differential coefficient value of the transmitted signal input to the signal input terminal 4 are detected, the characteristic point information of the transmitted signal is extracted, and this is extracted on the transmission line 2. Send to.

The transmission line 2 takes in the encoded feature point information output from the encoding device 1 and supplies it to the decoding device 3.

The decoding device 3 uses a cubic spline function using the information from the encoding device 1 supplied via the transmission line 2 to obtain a maximum point of the transmitted signal input to the signal input terminal 4 and A cubic spline interpolation filter circuit 10 for restoring the transmitted signal by generating a C ² -class function passing through a minimum point, an inflection point, or an additional transmission point is provided, and the restored transmitted signal is transmitted to the signal. Output from the output terminal 5.

Next, the encoding operation and the decoding operation of this embodiment will be sequentially described with reference to the block diagram shown in FIG.

First, when the signal to be transmitted is input to the signal input terminal 4, the signal to be transmitted is differentiated by the first differential operation circuit 6 provided in the encoding device 1 to obtain the first derivative (1 A second derivative value is calculated and the second derivative value obtained by the first derivative operation circuit 6 is further differentiated by the second derivative operation circuit 7 to obtain a second derivative (second derivative value). ) Is calculated and these primary differential coefficient values and secondary differential coefficient values are supplied to the control signal generating circuit 8.

Then, the control signal generating circuit 8 outputs the primary differential coefficient value output from the first differential operation circuit 6 and the secondary differential coefficient value output from the second differential operation circuit 7. Based on the maximum and minimum points of the transmitted signal, inflection points are detected based on these maximum and minimum points,
A control signal necessary for extracting the characteristic point information of the transmitted signal is created based on the inflection point and is supplied to the transmission signal forming circuit 9.

In parallel with the above-described operation, the transmission signal forming circuit 9 takes in the transmitted signal input to the signal input terminal 4.

Then, based on the control signal output from the control signal generating circuit 8 by the transmission signal forming circuit 9, the positions of the maximum point, the minimum point and the inflection point of the transmitted signal and their signal values are obtained. After being extracted, these pieces of information are encoded, sent out on the transmission line 2, and transmitted to the decoding device 3.

Then, the signal passes through the maximum point, the minimum point, the inflection point or the additional transmission point of the transmitted signal by a cubic spline function using the information supplied through the transmission line 2 by the decoding device 3. A class C ² function is generated to restore the transmitted signal, which is output from the signal output terminal 5.

As described above, in this embodiment, the characteristic points of the transmitted signal input to the signal input terminal 4 by the encoding apparatus 1 are extracted, and the characteristic points are transmitted to the transmission line 2 to be transmitted to the decoding apparatus 3. The decoding device 3 decodes the information sent from the coding device 1 to restore the transmitted signal and outputs it from the signal output terminal 5. Therefore, the coding device 1 With this, most of the information carried by the transmitted signal can be reduced and transmitted to the decoding device 3, whereby the signal to be transmitted is efficiently compressed and the processing required for encoding is simplified. In addition, the processing time can be shortened significantly.

As a result, the transmission cost can be significantly reduced, and the encoding device 1 can be downsized and reduced in cost.

Further, in this embodiment, since the technique of waveform interpolation from the thinned-out data is basically used, the present method has a larger amount of information than the conventional interpolation by the linear filter. In the region where the signal is required, for example, the region including the high-frequency component or the edge portion of the image, the thinning interval can be automatically narrowed, whereby a large amount of information can be transmitted and a better interpolation signal can be obtained. it can.

FIG. 2 is a diagram showing an example of encoding a transmitted signal according to this embodiment. As you can see from this figure,
As the transmitted signal, a chirp signal whose frequency increases with position (Y = 1-cos {2π (x /
200) ² }. Here, x is a coordinate on the horizontal axis. ), The original data 64
For 0, the maximum and minimum points indicated by black circles in the figure,
Since there are 38 inflection points (data to be transmitted),
If each of these data is represented by 8 bits, the amount of information of "0.95" bits is sufficient for each data, and therefore the data amount can be compressed to about 1/8 in the example of FIG. it can.

At this time, as can be seen from the solid line near the horizontal axis, when the S / N of the transmitted signal is about "48" dB, the S / N of the transmitted signal restored by the decoding device 3 is obtained. N
Became "43.3" dB.

FIG. 3 is a block diagram showing an example of a signal transmission system to which another embodiment of the coding system according to the present invention is applied. In this figure, parts corresponding to those in FIG. 1 are designated by the same reference numerals.

The signal transmission system shown in this figure differs from the system shown in FIG. 1 in that an error coding section 11 is added to the coding apparatus 1 side and a decoding error canceling section 12 is added to the decoding apparatus 3 side. Then, the decoding error when the cubic spline function is used on the encoding device 1 side is detected, and this is encoded and sent out on the transmission line 2 to be transmitted to the decoding device 3 side, and this decoding is performed. This is because the device 3 restores the decoding error and cancels the decoding error of the transmitted signal restored by the decoding process.

The error coding unit 11 uses the information output from the transmission signal forming circuit 9 to form a local spline function using a cubic spline function to input a maximum point, a minimum point, and an inflection point of the transmitted signal input to the signal input terminal 4. Or C ² passing through the additional transmission point
A cubic spline interpolation filter circuit 13 for generating a class function to restore the transmitted signal, a transmitted signal including a restoration error output from the cubic spline interpolation filter circuit 13, and the signal input terminal 4 And a subtraction circuit 14 for calculating a difference with a transmitted signal input via the subtraction circuit 14 to generate a restoration error signal, and the restoration error signal output from the subtraction circuit 14 is encoded and sent to the transmission line 2. , And an error signal encoding circuit 15 for transmitting to the decoding device 3.

When the transmitted signal input to the signal input terminal 4 is encoded by the encoder 1 and transmitted to the decoder 3, the error encoder provided on the encoder 1 side. The information output from the transmission signal forming circuit 9 of the encoding device 1 by 11 and the signal input terminal 4
On the basis of the transmitted signal input to the decoder 3, the decoding device 3 detects the restoration error generated when the transmitted signal is restored based on the information output from the transmission signal forming circuit 9, It is digitized, sent out on the transmission line 2, and transmitted to the decoding device 3.

The decoding error canceling section 12 also includes an error signal decoding circuit 16 for decoding a restoration error based on the error information supplied from the encoding device 1 via the transmission line 2, and the error signal decoding circuit 16. An addition circuit 1 for adding the restoration error output from the digitization circuit 16 and the restored transmission signal output from the third-order spline interpolation filter circuit 10 to cancel the restoration error included in the transmission signal.
7 is provided, and when the transmitted signal is restored based on the characteristic information of the transmitted signal output from the encoding device 1 supplied via the transmission line 2, the transmission line 2
The restoration error is decoded on the basis of the error information from the encoding device 1 supplied via the, and the restored transmission signal output from the cubic spline interpolation filter circuit 10 based on the restoration error. Then, the restoration error contained in is canceled and output from the signal output terminal 5.

As described above, in this embodiment, the characteristic points of the transmitted signal input to the signal input terminal 4 by the encoding device 1 are extracted in the same manner as in the above-described embodiment, and are extracted on the transmission line 2. The information is transmitted and transmitted to the decoding device 3, and the decoding device 3 decodes the information transmitted from the encoding device 1 to restore the transmitted signal and output it from the signal output terminal 5. Therefore, most of the information that the transmitted signal has can be reduced by the encoding device 1 and transmitted to the decoding device 3, whereby the signal to be transmitted can be efficiently compressed and encoded. It is possible to simplify the processing required for the conversion, and it is possible to significantly reduce the processing time.

Furthermore, in this embodiment, the error coding unit 11 is added to the coding device 1 side, and the decoding error canceling unit 12 is added to the decoding device 3 side, so that the coding device 1 side performs 3 coding. The decoding error when the next spline function is used is detected, and this is encoded and sent out on the transmission line 2 to be transmitted to the decoding device 3 side, and this decoding device 3 restores the decoding error. Since the decoding error of the transmitted signal restored by the decoding process on the decoding device 3 side is canceled, the restoration error of the transmitted signal restored on the decoding device 3 side can be significantly reduced. it can.

Further, in this embodiment, the information output from the transmission signal forming circuit 9 of the encoding device 1 and the error information output from the error encoding unit 11 are separately transmitted to the transmission line 2. However, these may be collectively encoded using an appropriate format and transmitted on the transmission line 2.

In each of the above-described embodiments, the amount of signal sent out on the transmission line 2 is, as described above, at least the positions of the poles and the inflection points and the sizes of these poles and the inflection points. However, if there is a margin in the capacity of the transmission line 2, additional information, for example, information about a pole, a point near the inflection point, an end point, or the like may be added.

When sending out to the transmission line 2,
Various generalized coding methods such as vector quantization and DPCM may be used together.

[0046]

As described above, according to the present invention, it is possible to efficiently compress a signal to be transmitted and simplify the processing required for encoding, and it is possible to greatly reduce the processing time. As a result, the transmission cost can be significantly reduced, and the size and cost of the encoding device can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a signal transmission system to which an embodiment of an encoding method according to the present invention is applied.

FIG. 2 is a table diagram showing an example of a decoded waveform of the signal transmission system shown in FIG. 1 and its error waveform.

FIG. 3 is a block diagram showing an example of a signal transmission system to which another embodiment of the encoding system according to the present invention is applied.

[Explanation of symbols]

 1 Encoding Device 2 Transmission Line 3 Decoding Device 4 Signal Input Terminal 5 Signal Output Terminal 6 First Differentiation Arithmetic Circuit 7 Second Differentiation Arithmetic Circuit 8 Control Signal Generating Circuit 9 Transmission Signal Forming Circuit 10 Third-Order Spline Interpolation Filter Circuit 11 Error Encoding Unit 12 Decoding Error Canceling Unit 13 Third-Order Spline Interpolation Filter Circuit 14 Subtraction Circuit 15 Error Signal Encoding Circuit 16 Error Signal Decoding Circuit 17 Addition Circuit

Claims (5)

[Claims] 1. The encoding side encodes and transmits the position of a feature point selected from the transmitted signal and its signal value, and the decoding side transmits information on the position of the transmitted feature point and its signal value. An encoding method characterized by reconstructing a transmitted signal by interpolating signal values other than the characteristic points by using a spline function using. 2. A characteristic point on the encoding side is that the sign of at least one differential coefficient among the m−1 differential coefficients of the first to (m−1) th order of the transmitted signal is inverted. ,
The encoding system according to claim 1, wherein the decoding side performs interpolation using an m-th order spline function. 3. The encoding side is provided with an m-th order spline interpolation filter circuit similar to that used on the decoding side, and an error signal between the transmitted signal and the output of the m-th order spline interpolation filter circuit is provided. 3. The encoding method according to claim 1, wherein the encoded signal is transmitted, and the decoding side reconstructs the transmitted signal from the output of the m-th order spline interpolation filter circuit and the error signal. 4. The encoding side includes two differential operation circuits for sequentially differentiating the transmitted signal, and a control signal for generating a control signal indicating the position of the feature point based on the output of each of the differential operation circuits. It includes a generating circuit and a transmission signal forming circuit that generates characteristic information of the transmitted signal based on a control signal output from the control signal generating circuit, and the decoding side is output from the encoding side. The encoding system according to claim 1 or 2, further comprising a cubic spline interpolation filter circuit that performs interpolation using a spline function that uses feature information. 5. The cubic side spline interpolation filter circuit, which performs interpolation using a spline function using the characteristic information output from the transmission signal forming circuit, on the encoding side,
A subtraction circuit for generating an error signal between the reconstructed transmitted signal output from the cubic spline interpolation filter circuit and the transmitted signal to be transmitted, and an error signal output from the subtraction circuit to be encoded. An error signal encoding circuit for transmitting, wherein the decoding side outputs an error signal decoding circuit for decoding error information output from the error signal encoding circuit, and an output from the error signal decoding circuit. Error signal and decoding side 3
3. An addition circuit that adds the reconstructed transmitted signal output from the next spline interpolation file circuit and cancels an error component included in the reconstructed transmitted signal. 3. The encoding method according to any one of 3 above.