JPH02279080A - Picture coder and decoder - Google Patents

Abstract

PURPOSE:To prevent block distortion from being caused by applying area split to a picture in response to the property of each picture element of a digital picture signal and coding an outer periphery of each split picture area. CONSTITUTION:A picture area split means applying area split of a picture in response to the property of each picture element of a digital picture signal and a coding means coding the outer periphery of each split picture area by the picture area split means are provided. After the picture area of a digital picture signal is split in response to the property of each picture element, the coding is executed by coding the outer periphery of the area with the coding means. Thus, the quantization is not applied in the block unit as a conventional system, no block distortion is caused. Since the outer periphery of the area is coded, the compression efficiency is improved.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an encoding device suitable for encoding an image signal such as a color moving image including a still image, and an encoding method using the encoding device. The present invention relates to a decoding device that converts a code into an image deviation signal.

[Prior Art] Conventionally, this type of technology has been disclosed in the following documents.

``°゛Kosugi Matsuda, Sakai: A study of filters in motion compensated cosine exchange coding loops''.

IEICE technical research report IE86-100J Figures 2 (a) and 2 (b) are block diagrams showing the configurations of the image encoding device and decoding device, respectively, disclosed in the above-mentioned document. .

The encoding device will be explained based on FIG.

The original image is converted into an electrical signal representing the light and shade distribution by scanning the image pickup tube of the TV camera 201, and then sampled into pixels by an A/D converter (A/D>202) and converted into a digital image signal. be done.

Next, this digital image signal is further color separated by a color separation circuit (Y/C) 203 to obtain a luminance signal and a color difference signal.

The luminance signal/chrominance signal is detected by a motion vector detector (MV>
204, the contents of the frame memory 205 are compared with the contents of the frame memory 205, and the mXm pixel size processing block (e.g. 16x
16), a motion vector is calculated, motion compensation is performed using this vector, and a predicted value is obtained.

Next, the adder 206 subtracts the predicted value motion-compensated using the previously determined motion vector to obtain an inter-frame difference residual signal. At this time, the calculated motion vector is encoded and sent to the transmission path 209.

On the other hand, the inter-frame difference residual signal is processed by a discrete cosine transformer (DCT>207) into an nxn processing block (
For example, the signals are orthogonally transformed in 8×8) units and decorrelated. At this time, the transform coefficients in this orthogonal transform are quantized by a quantizer (Q>208, converted into a quantization level number, and sent to the transmission path 209).

The output of the quantizer 208 is processed by an inverse quantizer (Q'>210
The quantization level number is converted into an inverse quantum value by this inverse quantizer 210, and is further restored to an interframe difference residual signal by an inverse discrete cosine transformer (DCT'> 211). .

Then, an adder 212 adds the predicted value obtained by the adder 206 to the reconstructed interframe difference residual signal, and inputs the resultant signal to the frame memory 205 as a local decoder reproduction signal.

Next, the decoding device will be explained based on FIG. 2(b).

First, the quantization level number and motion vector of the residual signal are received from the transmission path 213.

The quantization level number is dequantized by an inverse quantizer 214, and this dequantized value is further transformed by an inverse discrete cosine transformer 215 to restore an interframe difference residual signal.

Next, the adder 21.6 adds a predicted value predicted by the motion vector received from the transmission path 213 to the reconstructed interframe difference residual signal, and local decoding on the encoder side It is converted into a reproduction signal that is exactly the same as the stopper reproduction signal.

This reproduced signal is input to frame memory 217 and color synthesis circuit 218. The color synthesis circuit (Y/C) 218
combines the luminance signal and color difference signal into a digital image signal. This combined digital image signal is converted into an analog image signal by a D/A converter 219 and reproduced by an image monitor 220.

[Problems to be Solved by the Invention] However, the above-mentioned conventional method has the following problems.

■ Since quantization and inverse quantization are performed on a block-by-block basis, if the characteristics of the quantizer differ for each block, the difference in quantization error will appear as block distortion in the image.

■ High-frequency components of orthogonal transform coefficients are lost due to quantization,
The outline of the image may become blurred.

The present invention was made against the above-mentioned background, and an object of the present invention is to provide an image encoding device and a decoding device that do not cause block distortion or blurred contours. It is.

[Means for Solving the Problems] The present invention solves the above problems by having the following configurations.

(1) An image encoding device that converts original image information into a digital image signal and encodes the same, comprising: an image area dividing unit that divides the image into areas according to the properties of each pixel of the digital image signal; and encoding means for encoding the outer periphery of each image area divided by the dividing means.

(2) In configuration 1, the image region dividing means has a function of dividing the image into regions by expressing the properties of each pixel as a vector, and classifying and expressing each pixel according to the value of this vector. A configuration characterized by being equipped with.

(3) In any of construction work or 2, the encoding means has a function of erasing pixels in the area where the encoding has been completed, and an area that has a cavity due to the erasure of the pixels. The configuration is also characterized in that it has a function of encoding only the outer periphery of the area.

(4) In any one of configurations 1 to 3, the encoding means tracks the outer periphery of the image area under a certain rule, and sets the tracking posture (direction) in one state during the tracking, A configuration characterized in that it has a function of a virtual tracking moving body (robot) that encodes changes in the tracking posture as state transitions by representing the state transitions with codes.

(5) In configuration 4, the encoding unit monitors the tracking start point of the tracking moving object, the current position during tracking, and the code generation situation, and detects the end of coding of each image area. A configuration characterized by having a detection function.

(6) In an image decoding device that inputs and decodes a code encoded by any one of the encoding devices described in component 5 to 5, the image decoding device inputs the encoded code and performs region segmentation by the encoding device. A configuration characterized by comprising: decoding means for decoding the outer periphery of each of the image areas, and image reproduction means for reproducing a digital image signal before being divided from the decoded image information.

(7) In configuration 6, the decoding means has a function of decoding only the outer periphery of the decoded area even if the decoded area has a cavity; The configuration is characterized in that it has a function of regenerating pixels having the same properties as the outer periphery of the finished area.

(8) In configuration 6, the image reproducing means has a function of reproducing digital image information according to the value of the vector when the property of each decoded pixel is expressed by a vector. A composition characterized by being something.

(9) In any one of configurations 6 to 8, the decoding means tracks the area specified by the input code, and sets the tracking attitude (direction) in one state during the tracking,
A configuration characterized in that it has a function of a virtual tracking mobile object (robot) that decodes changes in the tracking posture as state transitions by representing the state transitions in codes.

(10) In configuration 9, the decoding means detects the end of decoding of each image area by monitoring the tracking start point of the tracking moving object, the current position during tracking, and the code generation situation. A configuration characterized by having a detection function.

[Operation] According to the configuration 1, the image area dividing means:
The digital image signal is encoded by dividing the image into regions according to the properties of each pixel, and then encoding the outer periphery of these regions by the encoding means. That is, unlike the conventional method, quantization is not performed on a block-by-block basis, so block distortion does not occur. Moreover, since the outer periphery of the area is encoded, it is possible to improve compression efficiency.

According to configuration 2, the image region dividing means has a function of dividing the image into regions by expressing the properties of each pixel as a vector and classifying and expressing each pixel according to the value of this vector. Since it is equipped with this feature, area division can be done relatively easily.

According to configuration 3, the encoding means has a function of erasing pixels in the area where the encoding has been completed, and even if the area has a cavity due to the erasure of the pixels,
Since it has a function of encoding only the outer periphery of the area, efficient encoding is possible without erroneously encoding the inner periphery.

According to configuration 4, the encoding means tracks the outer periphery of the image area under a certain rule, sets the tracking attitude (direction) during the tracking to one state, and changes the tracking attitude. Since it has the function of a virtual tracking moving object (robot) that encodes state transitions by representing them as state transitions, efficient encoding using extremely simple code words is possible.

According to configuration 5, the encoding end detection unit detects the end of encoding of each image region by monitoring the tracking start point of the tracking moving object, the current position during tracking, and the code generation situation. Since this function is provided, it is possible to detect the end of encoding without requiring any special means for detecting the end of encoding.

Further, according to configurations 6 to 10, the configurations 1 to 5 described above
The code encoded by the described encoding device can be input and decoded to reproduce the original image.

[Embodiment] FIG. 1 is a block diagram showing the configuration of an image encoding device and a decoding device according to an embodiment of the present invention.
) shows an image encoding device, and FIG. 1(b) shows a decoding device. Note that an embodiment described in detail below shows an example in which the present invention is applied to a color moving image encoding device and a decoding device.

(Encoding Apparatus) First, an image encoding apparatus will be explained based on FIG. 1(a).

In the figure, numeral 101 is a TV camera (TV), numeral 1
02 is an A/D converter, and 103 is a color separation circuit, Y/C.
), code 104 is a vector quantizer (VQ), code 1
05 is an image memory (FM), 106 is a boundary tracking encoding circuit (C), and 107 is a transmission path.

The image signal input from the TV right camera 01 is converted into an original by the A/D converter 102 and becomes a digital image signal.

This digital image signal is separated into a luminance signal Y and two color difference signals CI by the color separation circuit 103.

It is separated into C2.

Next, these three types of signals are input to the vector quantizer 104.

The vector quantizer 104 receives the signal Y.

Each pixel value of C1 and C2 is regarded as an element of a three-dimensional vector X, and vector quantization is performed in the following procedure based on a codebook prepared in advance.

■ A three-dimensional vector number (Y, cl, C
2>.

■ The vector X and the representative vector V1 of the codebook
~■, Calculate the distance by brute force.

■ If the nearest representative vector is ■, that pixel I
Give vector number k to (x, y).

■ Perform steps from ■ to ■ for all pixels.

The vector number thus obtained is input to the image memory 105 and stored.

The vector number stored in the image memory 105 is read out and encoded by the boundary tracking encoding circuit 106 and sent to the transmission path 107.

The details of the contents of the image memory 105 and the operation of the boundary tracking encoding circuit 106 will be explained below.

FIG. 3 is an explanatory diagram of the storage state of vector numbers stored in the image memory. As shown in FIG. 3, the vector numbers are distributed so that each number constitutes a certain area, reflecting the properties of the input screen (brightness, color, etc. of the subject). The boundary tracing encoding circuit 106 performs encoding by tracing the outer periphery of this area.

The encoding process by the boundary tracking encoding circuit 106 is roughly divided into the following four processes.

1) Detection of m-th area AKm in vector number.

2) Tracking encoding of the outer periphery of area AKIIl.

3) Area AK1. Elimination of l.

4) Return to 1) with m=m+1. If there is no area in the vector number = +1. Set m=1, send a code indicating that encoding of all areas has been completed to the vector number, and return to 1).

FIG. 4(a) is a block diagram showing the internal configuration of the boundary tracking encoding circuit 106. In the figure, reference numeral 401
is a region detector (RD), 4°2 is a tracking simulator (Sim), and 403 is a counter. Below, the fourth
Of the above-mentioned processes 1) to 4), processes 1) to 3) will be described in detail with reference to FIG.

UΩZdanko 2 The area detection circuit 401 scans the contents of the image memory 1.5 and searches for the area AkIIlt corresponding to the vector number. Once discovered, a start signal is sent to the tracking simulator 402 and the process moves to process 2).

拉ΩZdanhuang 2 The tracking simulator 402 tracks the outer periphery of the area Akm on the image memory 105 by performing a tracking simulation using a virtual tracking robot as shown in FIG.

This virtual tracking robot tracks vector numbers based on the following rules.

*The robot's right hand must be on the area AklIl of the vector number.

*The robot's left hand must be outside the area Akm.

When a tracking robot is simulated using these two rules, the tracking robot maintains one of the four tracking postures (state S.-83) as shown in the state transition diagram of FIG. The vector numbers are tracked one after another. Then, the code “
°0'° to "2°" are output. The codes "0°" to "2" representing this tracking attitude are encoded in the area AkIIIt.

The operation of the tracking simulator 402 described above will be explained in more detail below.

(2) First, when a start signal is received from the area detector 401, the address of the pixel being scanned by the area detector 401 at that time is encoded as a start address A for area encoding, and sent to the transmission line 107.

■ As the initial state So, place the tracking robot's right hand on the start address A3.

■ Next, the tracking robot performs the following (a) to (C).
Execute one of the operations above and set the new address Aj where the right hand is placed.

(a) When pixel P1-, rotate 90” to the left around the left hand to get the sign “°1” and 5llod(.+3.a)
Transition to. Here, rmod(n+3.4)" is (
Represents the remainder when dividing n+3) by 4.

(b) When pixel p1≠k and pixel P2=, move forward, output code "0", and continue So.

(C) When pixel p1≠k and pixel p2≠, rotate 90° around the right hand to get the code “2”, 5I11°d(n
. 1, 4) Figures 7 (a) to (C) respectively show the above operation (a).
It is an explanatory diagram of operation corresponding to ~(C).

The codes "°0pa~"2' issued in the above operation.

is sent to the transmission line 107.

Further, at that time, when the code "2'°" is issued, the counter 403 is incremented by one.

Furthermore, when the code "0°" or "1" is issued, the counter 403 is reset to zero.

■ Next, the old address Aj-1 of the image memory 105
Since it has already been encoded, a pixel value of 0 is written to it.

■ The above ■ and ■ are repeated.

As a result of the above operations, when Aj = A3 (when the area has gone around once) or when the counter 403 has reached 4 (when the area consisting of one pixel has gone around once), the sign of the area AkIIl is Assuming that the process has been completed, the process moves on to the next process 3).

In the process 3), as shown in Figure 8, when one area Ak contains areas with several different vector numbers, the inner circumference of Ak is incorrectly tracked and encoded. So as not to,
AklIl! It is something to be erased. The details of the procedure will be explained below.

■ From the top left corner to the bottom right of the actual image data,
A pixel value of 0, which is a mark of encoding on the image memory 105,
Set the pixel value of the vector number adjacent to the pixel to 0.
shall be.

■ Perform the same action as ■ from the upper right corner to the lower left.

■ Do the same movement as ■ from the bottom right to the top left.

■ Perform the same action as ■ from the bottom left to the top right.

By performing the above operations ① to ②, the area AkIIl whose outer periphery was encoded in the process 2) is erased from the image memory 105. As a result, only the outer circumferences are encoded one after another without erroneously tracking and encoding the inner circumferences, so that waste can be avoided.

(Decoding device) Next, the image decoding device will be explained based on FIG. 1(b).

In the figure, 108 is a transmission path, 109 is a boundary tracking decoding circuit (D), and 110 is an image memory (FM).
, 111 is a vector inverse quantizer (Q-1), 112 is a color synthesis circuit (Y/C), 113 is a D/A converter (D/A>), and 114 is an image monitor.

The code from the transmission line 108 is sent to the boundary tracking decoding circuit 1
09, converted into a vector number, and stored in the image memory 110.

This stored vector number is stored in the inverse quantizer 111.
According to the codebook, (Y,
Then, these Y, C1°, C2 are color synthesized by the color synthesis circuit 112, and converted into an analog image signal by the D/A converter 113. and is reproduced on the image monitor 114.

The boundary tracking decoding circuit 109 will be described in detail below.

In short, this boundary tracing decoding circuit 109 reproduces the vector number area on the image memory 110 by following the reverse process of the boundary tracing encoding circuit 106 described above, and executes the following processes broadly divided. do.

1) Receive the start address A of the m-th area Akm as the vector number.

2) Track and decode the area Ak−k=.

3) Return to 1) above with m=m+1. If you receive a code indicating that the encoding of all areas in the vector number has been completed, set k = +1 and m = m + 1 (1) above.
Return to

4) Fill the inside of area AkITl with vector number k.

The above processes 1) to 4) are repeated to restore vector numbers one after another.

FIG. 4(b) is a block diagram showing the internal configuration of the boundary tracking decoding circuit 109. In the figure, reference numeral 404
is a tracking simulator (Sim), 405 is a controller (Cont), and 406 is a counter (Count).
). Hereinafter, each of the above-mentioned processes 1) to 4) will be explained in detail with reference to FIG. 4(b).

UΩ2danyo2 The controller 405 receives the start address information and sends a start signal to the tracking simulator 404 to start tracking decoding.

■Mu2danko2 The tracking simulator 404 performs the following operations to reproduce the outer periphery of the area corresponding to the vector number.

(2) First, a start signal is received from the controller 405, and a start address A for area encoding is received from the transmission line 108.

■ Initial state is S. Then, place the tracking robot's right hand on the start address A.

■ Next, the tracking robot performs the following (a) to (C).
Execute one of the operations above and set the new address Aj where the right hand is placed. Also, at that time, the code “2”
When the result appears, the counter 403 is incremented by one. Further, when the code 0° or 1 is output, the counter 403 is reset to 0.

(a) When the code “1pa” is received, consider the pixel p1”k,
Rotate 90° to the left using your left hand as an axis, and turn Smod (n+3°4
).

(b) When the code "0" is received, consider that pixel p1≠k and pixel ρ2"k, and move forward to continue Sn.

(C) When receiving the code "2pa, think that pixel P1.≠k and pixel p2≠, rotate 90 degrees around the right hand, and
Transition to IOd(..1,4).

FIGS. 7(a) to (C) respectively show the above operation (a).
It is an explanatory diagram of operation corresponding to ~(C).

■ Next, the old address Aj-1 of the image memory 105
, the pixel value k is written because it has been decoded.

■ The above ■ and ■ are repeated.

In process 3), Aj = A (when the area is cycled once), or the counter 406 becomes 4, indicating that the area AH-encoding has been completed by the controller 40.
5 is detected and the tracking simulator 404 and counter 406 are reset.

After that, return to the process 1) and create a new area Ak.
Perform trace decoding of m+1.

In addition, when a code indicating that the encoding of the area in the vector number is completed is received, the tracking simulator 404
The counter 406 is reset, and then the process 1) is returned to trace decoding of the new area Ak+11.

Further, when the completion of decoding for all vector numbers and all areas is detected, the process moves to the next process 4).

In UΩ1 Sanko Ward process 4), the area Ak surrounded by the vector number
The inside of lIl is filled with vector number k using the following procedure.

■ From the top left corner to the bottom right of the actual image data,
A pixel with a pixel value of 0 that is adjacent to a pixel with a pixel value that is a decoded mark on the image memory 110 is set as a new pixel value.

■ Perform the same action as ■ from the upper right corner to the lower left.

■ Do the same movement as ■ from the bottom right to the top left.

■ Perform the same action as ■ from the bottom left to the top right.

By performing the above operations ① to ②, the area AkIIl whose outer periphery was decoded in the process 2) is filled with vector numbers on the image memory 110.

In this way, the image transmitted from the transmission line 108 is reproduced on the image memory 110 using the vector number.

Thereafter, the vector inverse quantizer 111 converts the luminance signal Y and color difference signals C1 and C2, and sends them to the color synthesis circuit 112, D/A converter 113 and image monitor 114.
Through this, both images can be reproduced.

(Advantages of one embodiment) According to the embodiment described above, the following advantages can be obtained.

■ Think of the image as a collection of several areas, with similar colors,
Since the outer periphery of the bright area is encoded and transmitted, block-like distortion does not occur.

■ To encode a region, we consider a virtual tracking robot that tracks the outer periphery of the region, and encode changes in its tracking posture as state transitions.
“Tracking data can be composed of three code words of 2°. That is, it can be composed of short codes of 1 to 2 bits.

■ To encode a region, we consider a virtual tracking robot that tracks the outer periphery of the region, and encode changes in its tracking posture as state transitions, so the end of encoding for one region is the starting point. It can be automatically detected when the code "°2°°" that returns to the address appears four times in a row, and there is no need for a special code to indicate the end of encoding.

■ After encoding an area, the area is erased, so that the inner periphery of the area is not erroneously encoded, and encoding can be performed efficiently.

It should be noted that the present invention is not limited to the above-described embodiment, and can of course be applied to an encoding device and a decoding device for black and white moving images, or color or black and white still images.

[Effects of the Invention] As described in detail above, the present invention provides an image encoding device that converts original image information into a digital image signal and encodes it. A configuration comprising: an image region dividing means for performing division; and an encoding means for encoding the outer periphery of each image region divided by the image region dividing means; The image decoding device inputs and decodes the image information, the decoding means decoding the outer periphery of each image region divided into regions by the encoding device, and the digital image information before being divided from the decoded image information. An image encoding device and a decoding device that do not cause block distortion or blurred contours are obtained by having a configuration characterized by comprising an image reproducing means for reproducing an image signal. It is something that exists.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a conventional image encoding device and decoding device, and FIG. 3 is a vector number stored in the image memory. FIG. 4 is a block diagram showing the internal configuration of the boundary tracking encoding circuit and boundary tracking decoding circuit, FIG. 5 is an illustration of the operation of the virtual tracking robot, and FIG. 6 is a state transition diagram. , FIG. 7 is an explanatory diagram of a tracking operation by a virtual tracking robot, and FIG. 8 is an explanatory diagram of area encoding. 101...↑■Camera (TV), 102...A/D
Converter, 103... Color separation circuit (Y/C), 104.
... Vector quantizer (VQ), 105 ... Image memory (FM), 106 ... Boundary tracking encoding circuit (C),
107... Transmission line, 108... Transmission line, 109...
・Boundary O tracking decoding north circuit B (D), 110... Image memory (FM) 1... Vector inverse quantizer (VQ'') 112... Color synthesis circuit (Y/C) 113... D/A converter (D/A> 114... image monitor. Whistle diagram explaining the storage state of the vector number stored in the image memory Diagram explaining the operation of the virtual tracking robot Figure 5 (a) (b) Virtual tracking a*” 1F tracking operation explanatory diagram Figure 7 Figure 6 Figure 8 Procedure + F company official document (voluntary) Heisei 2.!, failure

Claims (10)

[Claims] (1) An image encoding device that converts original image information into a digital image signal and encodes the same, comprising: an image area dividing unit that divides the image into areas according to the properties of each pixel of the digital image signal; An image encoding device comprising: encoding means for encoding the outer periphery of each image region divided by the dividing means. (2) The image region dividing means has a function of dividing the image into regions by expressing the properties of each pixel as a vector and classifying and expressing each pixel according to the value of this vector. The image encoding device according to claim 1, characterized in that: (3) The encoding means has a function of erasing pixels in the area where the encoding has been completed, and even if the area has a cavity due to the erasure of pixels, only the outer periphery of the area is encoded. 3. The image encoding apparatus according to claim 1, wherein the image encoding apparatus has a function of encoding an image. (4) The encoding means tracks the outer periphery of the image area under a certain rule, sets a tracking attitude (direction) during the tracking as one state, and defines a change in the tracking attitude as a state transition. The image encoding device according to any one of claims 1 to 3, characterized in that the image encoding device has a function of a virtual tracking moving object (robot) that encodes the state transition by representing it with a code. . (5) The encoding means has an encoding end detection function that monitors the tracking start point, current position during tracking, and code generation situation of the tracking moving object to detect the end of encoding of each image area. The image encoding device according to claim 4, characterized in that: (6) An image decoding device that inputs and decodes a code encoded by the encoding device according to any one of claims 1 to 5, wherein the image decoding device inputs the encoded code and decodes the encoded code. The present invention is characterized by comprising: a decoding means for decoding the outer periphery of each image region divided into regions; and an image reproduction means for reproducing a digital image signal before division from the decoded image information. Image decoding device. (7) The decoding means has a function of decoding only the outer periphery of the decoded area even if the decoded area has a cavity, and the decoding means has a function of decoding only the outer periphery of the area where the decoding has been completed. 7. The image decoding apparatus according to claim 6, wherein the image decoding apparatus has a function of reproducing pixels having the same properties as those on the outer periphery of the area. (8) When the characteristics of each decoded pixel are expressed by a vector, the image reproducing means has a function of reproducing digital image information according to the value of this vector. The image decoding device according to claim 6, characterized in that: (9) The decoding means tracks the area specified by the input code, sets a tracking attitude (direction) during the tracking as one state, and uses a change in the tracking attitude as a state transition. 9. The image decoding device according to claim 6, wherein the image decoding device has a function of a virtual tracking moving object (robot) that decodes transitions by representing them with codes. (10) The decoding means includes a decoding completion detection function that monitors the tracking start point, current position during tracking, and code generation status of the tracking moving object to detect the completion of decoding of each of the image regions. The image decoding device according to claim 9, characterized in that: