JPH0678294A - Coding method, coder and decoder - Google Patents

Abstract

PURPOSE:To improve the compression efficiency of picture data by dividing picture data into a basic picture element group and other picture element groups, calculating a prediction error of a predicted picture element group for the other picture element groups and coding the basic picture elements and the predicted picture elements. CONSTITUTION:A division circuit 1 decomposes picture data into a basic picture element group A and other picture element groups B-D, and prediction circuits 2b-2d generate respectively predicted picture element groups B'-D' predicting the other picture element groups B-D. Then computing elements 3b-3d calculate respectively predicted errors Be-De of the predicted picture element groups B'-D' with respect to the other picture element groups B-D respectively and coders 4a-4d encode respectively the basic picture element group A or the predicted errors Be-De.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coding method and code suitable for use in compressing and recording an image on a recording medium such as a disk or tape and expanding the image from the recording medium to reproduce it. The present invention relates to an encryption device and a decoding device.

[0002]

2. Description of the Related Art FIG. 14 is a block diagram showing a configuration of an example of a conventional encoder for encoding an image. First, for example, an image (image signal) composed of 720 pixels × 480 lines (720 pixels in the horizontal direction and 480 lines in the vertical direction) is divided into image data of blocks of 8 pixels × 8 lines, and the motion vector detection circuit 70 Is entered. The motion vector detection circuit 70 has a front original image portion 71.
a, an original image portion 71b, and a rear original image portion 71c, each of which has a frame memory 71 divided into three areas. Image data of a frame to be encoded (hereinafter, original image data) is stored in the frame memory 71. Original image section 7
1b, the image data of one frame before the original image data (hereinafter, forward original image data) is stored in the forward original image portion 71a, and the image data of, for example, one frame after the original image data (hereinafter , Rear original image data) is stored in the rear original image portion 71c.

Further, the motion vector detection circuit 70 includes the original image data stored in the original image portion 71b of the built-in frame memory 71 and the front original image data or the rear original image portion stored in the front original image portion 71a. The motion vector with respect to the backward original image data stored in 71c is detected, and the variable length coding circuit 75 and the motion compensation circuit 79 are detected.
And the sum of absolute values of the motion vectors are calculated and output to the prediction determination circuit 81.

Then, the original image data stored in the original image portion 71b of the frame memory 71 incorporated in the motion vector detecting circuit 70 is processed by the DCT circuit 73 through the arithmetic portion 72 including the arithmetic units 72a and 72b and the switch 72c. Is input to the digital camera, where it is subjected to DCT (discrete cosine transform) processing and converted into DCT coefficients. The DCT coefficient converted from the original image data by the DCT circuit 73 is stored in the quantization circuit 7
4 is quantized by a quantization step corresponding to the amount of data stored in the transmission buffer 76 (buffer storage amount), and then input to the variable length coding circuit 75.

Data quantized by the quantization circuit 74 (D
The CT coefficient) is also supplied to the dequantization 77, where it is dequantized according to the quantization step in the quantization circuit 74. The data inversely quantized by the inverse quantization circuit 77 is further input to the IDCT circuit 78, and the inverse DCT is performed.
It is processed. The data output from the IDCT circuit 78 is input to the motion compensation circuit 79 having the arithmetic unit 79a built therein.

When the image data input to the motion compensating circuit 79 is I picture (intra-coded image) data, the motion compensating circuit 79 uses the motion vector detecting circuit 70 for the input image data. Motion compensation corresponding to the output motion vector is performed, and is output to and stored in the frame memory 80 as a predicted image of the image input to the calculation unit 72.

When the image data input to the motion compensating circuit 79 is P picture (forward predictive coded image) data, the motion compensating circuit 79 first stores the data in the motion compensating circuit 79 in its built-in arithmetic unit 79a. The input image data is added to the image data of one frame before, which is the predicted image data of the image data input to the motion compensation circuit 79, which is stored in the forward prediction image unit 80a of the frame memory 80. Then, the added output is subjected to motion compensation corresponding to the motion vector output from the motion vector detection circuit 70, output to the frame memory 80 as a predicted image of the image input to the calculation unit 72, and stored therein. It

When the image data input to the motion compensating circuit 79 is B picture (bidirectional predictive coded image) data, the motion compensating circuit 79 first sends the motion compensating circuit 79 to the built-in arithmetic unit 79a. The input image data and the forward prediction image portion 80a of the frame memory 80
Image data one frame before stored as the predicted image data of the image data input to the motion compensation circuit 79, or the backward predicted image portion 80b of the frame memory 80.
The image data input to the motion compensation circuit 79 and the image data of one frame after as the predicted image data stored in are added. Then, the added output is subjected to motion compensation corresponding to the motion vector output from the motion vector detection circuit 70, output to the frame memory 80 as a predicted image of the image input to the calculation unit 72, and stored therein. It

As described above, the original image data obtained by decoding the data encoded by the arithmetic unit 72, the DCT circuit 73 and the quantizing circuit 74 is used as the predicted image data of the image data input to the arithmetic unit 72. It is stored in the frame memory 80.

On the other hand, the prediction determination circuit 81 is stored in the original image section 71b of the frame memory 71 incorporated in the motion vector detection circuit 70 based on the sum of the absolute values of the motion vectors output from the motion vector detection circuit 70. Determine a prediction mode for encoding the original image data. That is,
Which of the I, P, or B picture data the prediction determination circuit 81 encodes the original image data stored in the original image portion 71b of the frame memory 71 incorporated in the motion vector detection circuit 70 To decide. When the prediction determination circuit 81 determines to encode the original image data as I, P, or B picture data, the prediction determination circuit 81 sets the I, P, or B prediction mode to the switch 72c of the calculation unit 72, respectively. It outputs to the variable length coding circuit 75 and the motion compensation circuit 79.

In the motion compensation circuit 79, the prediction judgment circuit 81
According to the prediction mode output more, the prediction image data is generated from the input image data as described above.

When the prediction determination circuit 81 outputs the I prediction mode to the switch 72c of the arithmetic unit 72, the switch 72c selects the terminal a. As a result, the original image data itself stored in the original image portion 71b of the frame memory 71 incorporated in the motion vector detection circuit 70 becomes
By passing through the arithmetic unit 72 and being supplied to the DCT circuit 73, I picture data is generated.

When the P prediction mode is output from the prediction determination circuit 81 to the switch 72c of the arithmetic unit 72, the switch 72c selects the terminal b. As a result, in the arithmetic unit 72a, the original image data stored in the original image portion 71b of the frame memory 71 incorporated in the motion vector detection circuit 70 and the forward prediction image portion 8 of the frame memory 80.
0 of the already decoded original image data stored in 0a
The data obtained by subtracting the difference from the predicted image data before the frame is supplied to the DCT circuit 73, and the P picture data is generated.

When the B prediction mode is output from the prediction judgment circuit 81 to the switch 72c of the arithmetic unit 72, the terminal c is selected by the switch 72c. As a result, in the arithmetic unit 72b, the original image data stored in the original image portion 71b of the frame memory 71 incorporated in the motion vector detection circuit 70 and the forward prediction image portion 8 of the frame memory 80.
0 of the already decoded original image data stored in 0a
Predicted image data before frame or frame memory 8
The data which has been stored in the backward predicted image portion 80b of 0 and which has already been decoded and has one or both of the predicted image data of one frame after the original image data, or both, is stored in the DCT circuit 73. As a result, the B picture data is generated.

The variable length coding circuit 75 includes a quantization circuit 74.
Output data (data of I, P, or B picture) and quantization step, motion vector detection circuit 7
Motion vector output from 0 and the prediction determination circuit 8
The prediction mode output from 1 is converted into a variable length code such as Huffman code, and output to the transmission buffer 76. The data output from the variable equalization encoding circuit 75 is temporarily accumulated in the transmission buffer 76, and is recorded on a recording medium such as a disk (not shown) at a predetermined constant transmission rate.

Next, FIG. 15 is a block diagram showing the configuration of an example of a decoder for decoding the data encoded by the encoder of FIG. The encoded image data read from the disc is temporarily stored in the reception buffer 91 and then supplied to the variable length decoding circuit 92. The variable length decoding circuit 92 performs variable length decoding on the data supplied from the reception buffer 91, and outputs the motion vector and the prediction mode or the quantization step to the motion compensation circuit 95 or the inverse quantization circuit 93, respectively. , DCT processing and quantized image data are output to the inverse quantization circuit 93.

The inverse quantization circuit 93 is a variable length decoding circuit 9
The image data output from 2 is dequantized according to the quantization step also output from the variable length decoding circuit 92, and output to the IDCT circuit 94. Inverse quantization circuit 9
The data (DCT coefficient) output by 3 is IDCT
In the circuit 94, the inverse DCT processing is performed, and the result is supplied to the motion compensation circuit 95 including the arithmetic unit 95a.

In the motion compensation circuit 95, when the image data supplied from the IDCT circuit 94 is I picture data, that is, when the prediction mode supplied from the variable length decoding circuit 92 is the I prediction mode. For the data (I picture data), the variable length decoding circuit 9
Motion compensation corresponding to the motion vector output from 2 is performed. The motion-compensated data in the motion compensation circuit 95, that is, the decoded I-picture data is the predicted image of the image data (P- or B-picture data) input to the calculator 95a incorporated in the motion compensation circuit 95. The data is supplied to and stored in the frame memory 96.

When the image data output from the IDCT circuit 94 is P picture data in which the image data one frame before is the predicted image data, that is, the prediction mode output from the variable length decoding circuit 92 is P. In the prediction mode, in the motion compensation circuit 95, the image data supplied from the IDCT circuit 94 and the already decoded IDCT circuit stored in the forward prediction image section 96a of the frame memory 96 are first calculated in the arithmetic unit 95a. The image data supplied from 94 and the image data one frame before are added. Then, the added data is subjected to motion compensation corresponding to the motion vector output from the variable length decoding circuit 92. The data that has been subjected to motion compensation by the motion compensation circuit 95, that is, the decoded P picture data, is the predicted image of the image data (P or B picture data) that is input to the calculator 95a incorporated in the motion compensation circuit 95. The data is supplied to and stored in the frame memory 96.

The image data output from the IDCT circuit 94 is a B picture whose predicted image data is interpolated image data generated from the image data of one frame before, the image data of one frame after, or both. When the data is data, that is, when the prediction mode output from the variable length decoding circuit 92 is the B prediction mode, in the motion compensation circuit 95, first, in the built-in arithmetic unit 95a,
The image data input to the motion compensation circuit 95 and the image data one frame before stored as the predicted image data of the image data input to the motion compensation circuit 95, which is stored in the forward prediction image unit 96a of the frame memory 96, or The image data input to the motion compensation circuit 95, which is stored in the backward predicted image portion 96b of the frame memory 96, is added to the image data of one frame later as predicted image data. Then, the added output is subjected to motion compensation corresponding to the motion vector output from the variable length decoding circuit 92. In this way, the data that has been motion-compensated by the motion compensation circuit 95, that is, the decoded B picture data is also
It is supplied to and stored in the frame memory 96 as predicted image data of image data (P or B picture data) input to the arithmetic unit 95a incorporated in the motion compensation circuit 95.

The motion compensation circuit 95 outputs the
The decoded image data is D / A converted, for example.
Display after A / A conversion (neither shown)
Are supplied and displayed.

[0022]

By the way, when an attempt is made to encode a high-definition image such as a so-called high-definition image (hereinafter referred to as HD image) by the encoder shown in FIG. 14, the amount of data is enormous. In addition, the load on each block of the device becomes heavy, the compression efficiency decreases,
In addition, it may not be possible to encode.

Furthermore, the encoder shown in FIG.
Since there is no compatibility between a bitstream that encodes an image and a bitstream that encodes an image of a current television standard such as NTSC or SECAM (hereinafter referred to as CTV image), these are not compatible. For example, it was difficult to decode both with the decoder shown in FIG.

Further, the HD image recorded on the disc is
In the case of fast-forward playback (high-speed playback), the fast-forward playback (for example, FF / FR image) is generated by decoding the image every predetermined number of frames. By the way, since the fast-forward speed V is inversely proportional to the data amount α of the FF / FR image and is proportional to the data reading speed β from the disk (V = β / α), the data amount α of the FF / FR image is α.
The less the number, the smoother the fast forward reproduction can be performed.

However, the encoded HD image is
Even if the decoding is performed every predetermined number of frames, the data amount α of the FF / FR image is very large, and there is a problem that smooth fast-forward reproduction cannot be performed.

The present invention has been made in view of such a situation, and it is possible to improve the compression ratio of an image without degrading the image quality and to smoothly reproduce a highly compressed image. It is something to do.

[0027]

According to a first aspect of the present invention, there is provided an encoding method in which image data is divided into a basic pixel group and another pixel group, and another pixel group is predicted from the basic pixel group. Is generated, the prediction error of the prediction pixel group with respect to other pixel groups is calculated, and the basic pixel group and the prediction error are encoded.

The encoding method according to a second aspect of the present invention is characterized in that each pixel of the prediction pixel group is generated from horizontally adjacent pixels of the basic pixel group.

The encoding method according to the third aspect is characterized in that each pixel of the prediction pixel group is generated from vertically adjacent pixels of the basic pixel group.

The encoding method described in claim 4 is characterized in that each pixel of the prediction pixel group is generated from pixels adjacent to each other in the diagonal direction of the basic pixel group.

The encoding method according to the fifth aspect is characterized in that each pixel of the prediction pixel group is generated from a plurality of adjacent pixels in the basic pixel group.

According to a sixth aspect of the present invention, there is provided an encoding device comprising a basic pixel group, a division circuit 1 as a decomposition means for decomposing the image data into other pixel groups, and a basic pixel group output from the division circuit 1. Prediction circuits 2b to 2d as a generation unit that generates a prediction pixel group that predicts another pixel group, and a calculation unit that calculates a prediction error of the prediction pixel group generated by the prediction circuits 2b to 2d with respect to another pixel group. Computing unit 3b as
To 3d and the basic pixel group output from the division circuit 1,
It is characterized by comprising encoders 4a to 4d as encoding means for encoding the prediction error calculated by the arithmetic units 3b to 3d.

An encoding apparatus according to a seventh aspect is the encoder 4
Further, a decoder 11 as a decoding means for decoding the basic pixel group encoded by a is further provided, and the prediction circuits 2b to 2d are provided with the basic pixel group decoded by the decoder 11,
It is characterized in that a predicted pixel group that predicts another pixel group is generated.

According to the eighth aspect of the present invention, there is provided a coding device comprising: a basic pixel group, a dividing circuit 1 as a decomposing means for decomposing the image data into another pixel group, and a basic pixel group output from the dividing circuit 1. Prediction circuits 2b to 2d as a generation unit that generates a prediction pixel group that predicts another pixel group, and a prediction error that calculates a prediction error of the prediction pixel group generated by the prediction circuits 2b to 2d with respect to another pixel group. Arithmetic units 3b to 3d as calculation means, subsampler 41 as subsampling means for subsampling image data and outputting subsample data, subsample data output from subsampler 41, and output from dividing circuit 1 And a sub-sampler output from the sub-sampler. An encoder 42 as encoding means for encoding the pull data, the difference data calculated by the calculator 44, and the prediction error calculated by the calculators 3b to 3d, and the encoders 4a to 4d. And

An encoding apparatus according to a ninth aspect is the encoder 4
a, or difference data encoded by the encoder 42,
Alternatively, the decoder 51 or the decoder 43 as a decoding means for respectively decoding the sub-sample data, and the decoder 5
1 or a differential unit decoded by the decoder 43, and an arithmetic unit 52 as a reproducing unit for reproducing the basic pixel group from the sub-sample data, and the prediction circuit 2b
2d to 2d, a predictive pixel group in which another pixel group is predicted is generated from the basic pixel group reproduced by the calculator 52.

The encoding device according to the tenth aspect is characterized in that the image data is high definition wide television (so-called high-definition) image data.

The decoding device according to the eleventh aspect is the decoders 21a to 21d as decoding means for respectively decoding the transmission data into a basic pixel group or a prediction error, and the basic pixel output from the decoder 21a. Prediction circuits 22b to 22d as generation means for generating a prediction pixel group from the group
And the calculators 23b to 23d as calculating means for calculating the original pixel group from the prediction pixel group generated by the prediction circuits 22b to 22d and the prediction error output from the decoders 21b to 21d, and the decoder 21a. It is characterized in that the basic pixel group outputted by the above and the original pixel group calculated by the arithmetic units 23b to 23d are combined, and a combining circuit 24 as a combining means for generating image data is provided.

The decoding device according to the twelfth aspect further comprises a subsampler 25 as subsampling means for subsampling the basic pixel group output from the decoder 21a and outputting subsample data. .

A decoding device according to a thirteenth aspect is a decoder 6 as a decoding means for decoding transmission data into sub-sampled data, difference data or prediction error, respectively.
1 or the decoders 21a to 21d and the calculator 62 as the first calculating means for calculating the basic pixel group from the sub-sample data or the difference data output from the decoder 61 or 21a, respectively. Prediction circuits 22b to 22d as generation means for generating a prediction pixel group from the calculated basic pixel group, prediction pixel groups generated by the prediction circuits 22b to 22d, and prediction errors output from the decoders 21b to 21d. And the calculators 23b to 23d as second calculating means for calculating the original pixel group from
And the basic pixel group or the original pixel group calculated by the computing unit 62 or the computing units 23b to 23d, respectively, to synthesize the image data, and a synthesizing circuit 2 as a synthesizing unit.
And 4 are provided.

A decoding apparatus according to a fourteenth aspect is characterized by further comprising an upsampler 31 as upsampling means for upsampling a basic pixel group and outputting upsampled data.

[0041]

In the encoding method of the present invention, the image data is divided into a basic pixel group and another pixel group, and the prediction pixel group that predicts the other pixel group is a horizontal, vertical, or diagonal of the basic pixel group. It is generated from pixels that are adjacent in the direction. Then, the prediction error of the prediction pixel group with respect to the other pixel groups is calculated, and the basic pixel group and the prediction error are encoded. Therefore, the compression efficiency of image data can be improved.

In the encoding apparatus of the present invention, the image data is decomposed into the basic pixel group and the other pixel group, and the predicted pixel group which predicts the other pixel group is generated from the basic pixel group. Then, the prediction error of the prediction pixel group with respect to the other pixel groups is calculated, and the basic pixel group and the prediction error are encoded. Therefore, the compression efficiency of image data can be improved.

Further, in this encoding device, the basic pixel group encoded by the encoder 4a is decoded, and the predictor circuits 2b to 2d select another pixel group from the basic pixel group decoded by the decoder 11. A predicted pixel group that predicts is generated. Therefore, the prediction error of the prediction pixel group that is exactly the same as the prediction pixel group generated by the decoding device that decodes the encoded data with respect to the other pixel groups is calculated, and thus the deterioration of the image quality is prevented. can do.

Further, in the encoding apparatus of the present invention, the image data is decomposed into a basic pixel group and another pixel group, and a predicted pixel group which predicts another pixel group is generated from the basic pixel group, The prediction error of the prediction pixel group with respect to the other pixel groups is calculated, the image data is sub-sampled, the difference between the sub-sample data and the basic pixel group is calculated, and the difference data is calculated. Then, the subsample data, the difference data, and the prediction error are encoded. Therefore, it is possible to further improve the compression efficiency of image data, and it is possible to easily reproduce an image other than the standard of the original image from the sub-sample data on the decoding device side.

Further, in this encoding device, the difference data or sub-sample data encoded by the encoder 4a or the encoder 42 is decoded respectively, and the basic pixel group is reproduced from the difference data and the sub-sample data. To do. Then, the prediction circuits 2b to 2d are connected to the computing unit 5
From the basic pixel group reproduced in 2, a predicted pixel group in which another pixel group is predicted is generated. Therefore, the prediction error of the prediction pixel group that is exactly the same as the prediction pixel group generated by the decoding device that decodes the encoded data with respect to the other pixel groups is calculated, and thus the deterioration of the image quality is prevented. can do.

In the decoding device of the present invention, the transmission data is decoded into the basic pixel group or the prediction error,
A prediction pixel group is generated from the basic pixel group. Then, the original pixel group is calculated from the predicted pixel group and the prediction error, and the basic pixel group and the original pixel group are combined to generate image data.
Therefore, it is possible to easily decode high-definition images with a large amount of data, which have been post-arrival compression-coded, without degrading the image quality.

Further, in this decoding device, since the basic pixel group is sub-sampled and the sub-sample data is output, an image other than the standard of the original image can be easily reproduced from the sub-sample data. .

Further, in the decoding apparatus of the present invention, the transmission data is decoded into sub-sample data, difference data or prediction error, and a basic pixel group is calculated from the sub-sample data and the difference data, and the basic pixel thereof is calculated. A prediction pixel group is generated from the group. And a prediction pixel group,
The original pixel group is calculated from the prediction error, and the basic pixel group or the original pixel group is combined to generate image data. Therefore, it is possible to easily decode high-definition images with a large amount of data, which have been post-arrival compression-coded, without degrading the image quality.

Furthermore, in these decoding devices,
Since the basic pixel group is up-sampled and the up-sampled data is output, for example, when an image is fast-forwarded and played back, the amount of data of the image to be read is small, and smooth fast-forward playback can be performed.

[0050]

1 is a block diagram showing the configuration of an embodiment of an encoding apparatus according to the present invention. The division circuit 1 divides a pixel group forming one field or one frame HD image (for example, a high-definition image as a high-definition image) into four pixel groups A, B, C, or D, Four planes A, B, C, or D composed of each pixel group are generated. That is, the division circuit 1 is located at the upper left of a group of, for example, four adjacent pixels A, B, C, and D of an odd field or an even field in the HD image shown in FIG. 2A. Plane A consisting of A pixel groups that collect only pixels A, plane B consisting of B pixel groups that collect only pixel B located in the upper right of the group, and C pixel group that collects only pixels C located in the lower left of the group Plane C and a plane D composed of a D pixel group in which only the pixels D located at the lower right of the group are collected (FIG. 2B).

The predicting circuit 2b calculates the average value of, for example, the pixels A which are adjacent to each other in the horizontal direction of the A pixel group which constitutes the plane A output from the dividing circuit 1, and which is also calculated by the dividing circuit 1. It is supplied to the calculator 3b as a predicted value of a B pixel group forming the output plane B.

Therefore, in the prediction circuit 2b, as shown in FIG. 3 or 4, the prediction pixel B'as the prediction value of the pixel B is obtained.
Will be generated from the pixel A located on the left and right of the pixel B in the original HD image 1 field.

In the prediction circuit 2b, the original HD image 1
Instead of setting the average value of the pixels A located on the left and right of the pixel B in the field as the prediction pixel B ′, for example, the pixel A itself adjacent to the pixel B can be set as the prediction pixel B ′. .

Here, in FIG. 3 and FIG. 4, the shaded portions indicate pixels in the even field, for example, of the odd field or the even field in the HD image. Therefore, FIG. 3 shows how pixels B, C, or D in an odd field of an HD image are predicted from pixel A, and FIG. 4 shows pixels B, C, or D in an even field of an HD image. Shows the situation predicted from the pixel A.

The predicting circuit 2c calculates the average value of, for example, the pixels A which are adjacent to each other in the vertical direction in the A pixel group which constitutes the plane A output from the dividing circuit 1, and which is also calculated by the dividing circuit 1. It is supplied to the calculator 3c as a predicted value of a C pixel group forming the output plane C.

Therefore, in the prediction circuit 2c, as shown in FIG. 3 or 4, the prediction pixel C'as the prediction value of the pixel C is obtained.
Is generated from the pixel A located above and below the pixel C in the original HD image 1 field.

In the prediction circuit 2c, the original HD image 1
Instead of setting the average value of the pixels A located above and below the pixel C in the field as the prediction pixel C ′, for example, the pixel A itself adjacent to the pixel C can be set as the prediction pixel C ′. .

The predicting circuit 2d calculates the average value of two sets of pixels A which are adjacent to each other in the diagonal direction, for example, of the A pixel groups forming the plane A output from the dividing circuit 1,
This is also supplied to the calculator 3d as a predicted value of the D pixel group forming the plane D output from the division circuit 1.

Therefore, in the prediction circuit 2d, as shown in FIG. 3 or 4, the prediction pixel D'as the prediction value of the pixel D.
Will be generated from the pixels A located at the upper right, lower right, upper left, and lower left of the pixel D in the original HD image 1 field.

In the prediction circuit 2d, the original HD image 1
The average value of the pixels A located at the upper right, lower right, upper left, and lower left of the pixel D in the field is not set as the predicted pixel D ′, but the pixel A itself adjacent to the pixel D is predicted pixel D ′. Or the average value of the pixels A located at the upper right and lower left of the pixel D, or the average value of the pixels A located at the upper left and lower right of the pixel D, etc. You can

The computing unit 3b subtracts the prediction pixel group B'output from the prediction circuit 2b from the pixel group B output from the division circuit 1 to obtain the prediction error B of the prediction pixel group B'with respect to the pixel group B. Calculate _e . The calculator 3c subtracts the prediction pixel group C ′ output from the prediction circuit 2c from the pixel group C output from the division circuit 1 to calculate the prediction error C _e of the prediction pixel group C ′ with respect to the pixel group C. To do. Calculator 3d are calculated from pixel group D output from the dividing circuit 1, 'subtracted, with respect to the pixel group D, predicted pixel group D' predicted pixel group D output from the prediction circuit 2d prediction error D _e of To do.

The encoder 4a is, for example, the same as the encoder shown in FIG. 14 described above, and encodes the pixel group A output from the division circuit 1 and, for example, a recording medium such as a disk via a transmission line or the like. Supply to. The encoders 4b to 4d are
The same as the encoder 4a, calculator 3b to 3d prediction error B _e output from, C _e or D _e were each encoding, via a transmission path, for example, is supplied to a recording medium such as a disk It

In the encoders 4a to 4d, as shown in FIG. 5, first, pixel groups A1, B1 and C which are I-pictures.
1 or D1 is respectively encoded, after the data of I-picture I _A1, I _B1, I _C1 or I _D1, is output, the data of I-picture I _A1, I _B1, I _C1 or I _D1, Based on the pixel groups A2, B2, C2, which are P pictures,
Alternatively, D2 is encoded, and P picture data P _A2 , P _B2 , P _C2 , or P _D2 is output. further,
I-picture data I _A1 , I _B1 , I _C1 , or I _D1 and P-picture data P _A2 , P _B2 , P _C2 , or P _D2
Pixel groups A3, B3, C that become P pictures based on
3 or D3 is encoded respectively, and P picture data P _A3 , P _B3 , P _C3 , or P _D3 is output, and then P
Pixel groups A4, B4, C4, which are P pictures, based on the picture data P _A2 , P _B2 , P _C2 , or P _D2 and the P picture data P _A3 , P _B3 , P _C3 , or P _D3 .
Alternatively, D4 is encoded, and P picture data P _A4 , P _B4 , P _C4 , or P _D4 is output (FIG. 5).
(A)).

Then, the I picture data I _A1 , I _B1 ,
I _C1 , or I _D1 , P picture data P _A2 , P _B2 , P
_C2 or P _D2 , P picture data P _A3 , P _B3 ,
P _C3 or P _D3 , and P picture data P _A4 , P
_{Based on B4} , P _C4 , or P _D4 , the pixel group A5, B5, C5, or D5 to be a B picture is encoded, and the B picture data B _A5 , B _B5 , B _C5 , or B _D5 is encoded.
Is output, the data I picture data I _A1 and I picture I
_B1 , I _C1 , or I _D1 , P picture data P _A2 ,
P _B2 , P _C2 , or P _D2 , P picture data P _A3 , P
_B3 , P _C3 or P _D3 , and P picture data P _A4 ,
Based on P _B4 , P _C4 , or P _D4 , a pixel group A6, B6, C6, or D6 to be a B picture is encoded, and B picture data B _A6 , B _B6 , B _C6 , or B
_D6 is output (Fig. 5 (b)).

Therefore, from this encoding device, I _A1 , I
_B1 , I _C1 , I _D1 , P _A2 , P _B2 , P _C2 , P _D2 , P _A3 ,
P _B3 , P _C3 , P _D3 , P _A4 , P _B4 , P _C4 , P _D4 , B _A5 , B
_The encoded data is repeatedly output in the order of _B5 , B _C5 , B _D5 , B _A6 , B _B6 , B _C6 , and B _D6 .

In FIG. 5, for example, I _A1 , I _B1 ,
I _C1 and a dotted arrow in a square surrounding I _D1 indicate I picture data I _A1 in which the pixel group A1 is encoded.
Based on, (to be exact, the pixel group B1 to D1 each prediction error B _e to D _e) pixel group B1 to D1 data I _B1 of the I picture is encoded, I _C1 or I _D1, is generated Which indicates that.

Next, the operation will be described. First,
In the division circuit 1, for example, a pixel group forming a 1-field HD image (high-definition image as a high-definition image) is divided into four pixel groups A, B, C, or D, and the pixel group A is an encoder. 4a and prediction circuits 2b to 2
and the pixel groups B to D are supplied to the computing unit 3
b to 3d respectively.

In the prediction circuits 2b to 2d, from the A pixel group output from the division circuit 1, as described above,
Predicted values of the pixel groups B to D (predicted pixel groups B ′ to D ′) are calculated and supplied to the computing units 3b to 3d, respectively.

In the computing units 3b to 3d, the division circuit 1
The prediction pixel groups B ′ to D ′ output from the prediction circuits 2b to 2d are subtracted from the pixel groups B to D output by
For the pixel group B to D, the prediction error B _e to D _e of the prediction pixel group B 'or D' is calculated.

The prediction errors B _{e to} D _e are determined by the encoders 4b to 4
It is input to d, where it is encoded. At the same time, at the encoder 4a, output from the dividing circuit 1 pixel group A is encoded, along with the prediction error B _e to D _e encoded in the encoder 4b to 4d, the multiplexer circuit (not shown) The signals are multiplexed and supplied to a recording medium via a transmission line for recording.

By the way, in the pixel groups B to D encoded as described above, the pixel group A is decoded in the decoding device (for example, the decoding device shown in FIG. 7 described later), and this is decoded. Predicted pixel groups B ′ to D ′ are generated from the pixel group A as described above in the same manner as in the encoding device, and are decoded based on the generated predicted pixel groups. However, in this case, the pixel group A decoded in the decoding device is coded by the encoder 4a, and the encoder 4a converts the input data into the quantization circuit 74 (see FIG. Since it is quantized in 14), its output, that is, the encoded pixel group A, has a quantization error.

On the other hand, in the coding apparatus of FIG. 1, from the pixel group A before being coded by the encoder 4a to the predicted pixel group B '.
To D ′ are generated, the pixel group A used in the encoding device and the pixel group used in the decoding device in FIG. 1 when generating the prediction pixel groups B ′ to D ′. A
May not be the same, and the decoding device may not be able to correctly decode the image.

Therefore, the encoding device is constructed as shown in FIG. This encoding device has the same configuration as the encoding device of FIG. 1 except that the pixel group A encoded by the encoder 4a is input to the decoder 11. The decoder 11 decodes the pixel group A encoded by the encoder 4a and supplies the pixel group A to the prediction circuits 2b to 2d.

Therefore, in this coding apparatus, the decoder 11 outputs the same pixel group A as the pixel group A used in the decoding apparatus, and based on this, the prediction pixel groups B ′ to D ′ are output. Since they are generated in the prediction circuits 2b to 2d, the decoding device can accurately decode the image.

Next, FIG. 7 is a block diagram showing the configuration of an embodiment of a decoding device for decoding image (HD image) data encoded by the encoding device of FIG. 1 or 6.
The decoder 21a is, for example, the same as the decoder shown in FIG. 15 described above, and decodes the coded pixel group A supplied from the demultiplexing circuit (not shown) via the transmission line, It outputs to the prediction circuits 22b to 22d, the synthesis circuit 24, and the subsampler 25.

The decoders 21b to 21d are the decoders 21a.
Is the same as the above, and decodes the encoded prediction errors B ′ to D ′ supplied from the demultiplexing circuit via the transmission line,
It outputs to each of the computing units 23b to 23d.

The prediction circuits 22b to 22d are the same as the prediction circuits 2b to 2d shown in FIG.
From the pixel group A output from a, the predictive pixel groups B ′ to D ′ are calculated as described above. Computing unit 23b
To 23d are prediction pixel groups B ′ to D ′ calculated by the prediction circuits 22b to 22d, and decoders 21b to 21d.
The prediction errors B ′ to D ′ outputted by the above are added to generate original pixel groups (other pixel groups) B to D, respectively.

The synthesizing circuit 24 is composed of computing units 23b to 23d.
Pixel groups B to D generated by: and decoder 21a
The output pixel group A is combined to generate HD image data.

The sub-sampler 25 sub-samples (decimates) the pixel group A output from the decoder 21a, and C
TV image (for example, an image of the current television standard such as the NTSC system and SECAM system) is output. That is, the sub-sampler 25 is encoded by the encoding device in the order shown in FIG.
_A1 , I _B1 , I _C1 , I _D1 , P _A2 , P _B2 , P _C2 , P _D2 ,
P _A3 , P _B3 , P _C3 , P _D3 , P _A4 , P _B4 , P _C4 , P _D4 , B
_A5 , B _B5 , B _C5 , B _D5 , B _A6 , B _B6 , B _C6 , or B _D6
Of (FIG. 5), the data I _B1 , I _C1 , I _D1 , P _B2 ,
P _C2 , P _D2 , P _B3 , P _C3 , P _D3 , P _B4 , P _C4 , P _D4 , B
_B5 , B _C5 , B _D5 , B _B6 , B _C6 , and B _D6 (the shaded portions in FIG. 8) are ignored, so to speak, and data I _A1 , P _A2 , P _A3 , P _A4 , B _A5 and B _A6 to CTV
Generate an image.

Next, the operation will be described. In the decoder 21a, the coded pixel group A supplied from the demultiplexing circuit via the transmission path is decoded and output to the prediction circuits 22b to 22d, the synthesis circuit 24, and the subsampler 25. At the same time, in the decoders 21b to 21d, supplied from the demultiplexing circuit via the transmission line,
The encoded prediction errors B ′ to D ′ are decoded and output to the computing units 23b to 23d, respectively.

In the predicting circuits 22b to 22d, the predictive pixel groups B'to D'are calculated from the pixel group A output from the decoder 21a as described above, and are supplied to the computing units 23b to 23d. Computing units 23b to 23
In d, the prediction pixel groups B ′ to D ′ supplied from the prediction circuits 22b to 22d and the prediction errors B ′ to D ′ output from the decoders 21b to 21d are added, and the original pixel group (other Pixel groups B to D are generated respectively.

In the synthesizing circuit 24, the pixel groups B to D generated by the arithmetic units 23b to 23d and the pixel group A output from the decoder 21a are synthesized to generate HD image data.

On the other hand, the sub-sampler 25 sub-samples the pixel group A output from the decoder 21a and outputs CTV image data.

As described above, according to this decoding device,
Not only decoding encoded HD image data,
CTV image data can also be easily obtained.

Further, FIG. 9 is a block diagram showing the configuration of the second embodiment of the decoding apparatus of the present invention. This decoding device has the same configuration as the decoding device of FIG. 7 with an upsampler 31 added. Up sampler 31
Up-samples (interpolates) the pixel group A output from the decoder 21a.

Therefore, from the up sampler 31, HD
A simple image (FF / FR image) for high-speed reproduction, such as thinning out the image, is output.

Therefore, according to this decoding device, it is possible to reproduce the HD image smoothly and at high speed.

Next, FIG. 10 is a block diagram showing the configuration of the third embodiment of the coding apparatus of the present invention. In the figure, parts corresponding to those in FIG. 1 are designated by the same reference numerals. The sub-sampler 41 sub-samples HD image data, converts it into CTV image data, and encodes it with the encoder 4.
Output to 2. The encoder 42 is, for example, the same as the encoder of FIG. 14 described above, and encodes the CTV image data output from the subsampler 41. The decoder 43 is, for example, the same as the decoder shown in FIG.
The CTV image data encoded by is decoded and output to the computing unit 44. The calculator 44 subtracts the pixel group E forming the CTV image data output from the decoder 43 from the pixel group A output from the division circuit 1, and supplies the difference data to the encoder 4a.

In the thus constructed coding apparatus, the HD image data is subsampled by the subsampler 41, converted into CTV image data, and output to the encoder 42. The encoder 42 encodes the CTV image data output from the subsampler 41.

The CTV image data coded by the encoder 42 is input to the decoder 43, decoded there and supplied to the calculator 44. In the arithmetic unit 44, the pixel group E constituting the CTV image data output from the decoder 43 is subtracted from the pixel group A output from the division circuit 1, and the difference data is supplied to the encoder 4a.

As described above, in this encoding device, the CTV image data is first generated and the difference data between it and the pixel group A is calculated. Then, encoded with CTV image data and the difference data, together with the encoded prediction error B _e to D _e in the same manner as in FIG. 1, is recorded on the recording medium via a transmission path is multiplexed.

Therefore, the CTV image data is generated from the HD image data itself before being encoded, and the CTV image data is also encoded together with the HD image data. As compared with the case where the CTV image is generated only with the pixel group A as in the decoding device in FIG. 9, it is possible to obtain a CTV image with good image quality.

In the encoder 42 or the encoders 4a to 4d, first, as shown in FIG. 11, the pixel group E1 of the CTV image and the pixel groups A1 and B of the HD image, which are I pictures, are first displayed.
1, C1, or D1 is encoded respectively, and after the I picture data I _E1 , I _A1 , I _B1 , I _C1 or I _D1 is output, the I picture data I _E1 , I _A1 , I _B1 ，
CTV that becomes a P picture based on I _C1 or I _D1
Image pixel group E2, HD image pixel group A2, B2, C
2 or D2 is encoded, and P picture data P _E2 , P _A2 , P _B2 , P _C2 , or P _D2 is output. Further, the I picture data I _E1 , I _A1 , I _B1 , I
_C1 or I _D1 , and P picture data P _E2 ,
Based on P _A2 , P _B2 , P _C2 , or P _D2 , a pixel group E3 of a CTV image that becomes a P picture and a pixel group A of an HD image
3, B3, C3, or D3 are encoded respectively, and P
Picture data P _E3 , P _A3 , P _B3 , P _C3 , or P _D3
Is output, the P picture data P _E2 , P _A2 ,
P _B2 , P _C2 , or P _D2 , and P picture data P
_{Based on E3} , P _A3 , P _B3 , P _C3 , or P _D3 , the pixel group E4 of the CTV image that becomes the P picture, the pixel group A4, B4, C4, or D4 of the HD image is encoded,
P picture data P _E4 , P _A4 , P _B4 , P _C4 , or P
_D4 is output (FIG. 11 (a)).

Then, the I picture data I _E1 , I _A1 ,
I _B1 , I _C1 , or I _D1 , P picture data P _E2 , P
_A2 , P _B2 , P _C2 , or P _D2 , P picture data P _E3 , P _A3 , P _B3 , P _C3 , or P _D3 , and P picture data P _E4 , P _A4 , P _B4 , P _C4 , or Based on P _D4 , the pixel group E5, HD of the CTV image that becomes the B picture
The pixel group A5, B5, C5, or D5 of the image is encoded, and the data B _E5 , B _A5 , B _B5 , B of the B picture is encoded.
_{After C5} or B _D5 is output, similarly, I picture data I _E1 , I _A1 , I _B1 , I _C1 or I _D1 , P picture data P _E2 , P _A2 , P _B2 , P _C2 , Or P _D2 , P picture data P _E3 , P _A3 , P _B3 , P _C3 , or P _D3 ,
Also, based on the data P _E4 , P _A4 , P _B4 , P _C4 , or P _D4 of the P picture, the pixel group E6 of the CTV image, which is the B picture, the pixel group A6, B6, C6, or D6 of the HD image, respectively. The encoded B picture data B _E6 ,
B _A6 , B _B6 , B _C6 , or B _D6 is output (see FIG. 11).
(B)).

Therefore, from this encoding device, I _E1 , I
_A1 , I _B1 , I _C1 , I _D1 , P _E2 , P _A2 , P _B2 , P _C2 ,
P _D2 , P _E3 , P _A3 , P _B3 , P _C3 , P _D3 , P _E4 , P _A4 , P
_B4 , P _C4 , P _D4 , B _E5 , B _A5 , B _B5 , B _C5 , B _D5 ,
The encoded data is repeatedly output in the order of B _E6 , B _A6 , B _B6 , B _C6 , and B _D6 .

In FIG. 11, for example, I _E1 ,
The dotted arrows in the squares surrounding I _A1 , I _B1 , I _C1 , and I _D1 are based on the I picture data I _{E1 in} which the pixel group E1 is encoded. Pixel group A
1 and pixel group E1) is encoded, and I picture data I _A1 is generated. Based on this I picture data I _A1 , pixel groups B1 to D1 (more accurately, pixel group B
It indicates that 1 or D1 each of the prediction error B _e to D _e) is encoded I-picture data I _B1, I _C1 or I _D1, is generated.

By the way, in this coding apparatus, as in the case of the coding apparatus in FIG. 1, the pixel group A input to the prediction circuits 2b to 2d in the coding apparatus in FIG.
The pixel group A used in the decoding device for decoding the data encoded by this device may not be the same,
It may not be possible to accurately decode the HD image.

Therefore, the encoding device is constructed as shown in FIG. In this encoding device, the difference data encoded by the encoder 4a is input to the decoder 51, the outputs of the decoders 51 and 43 are added by the calculator 52, and the prediction circuit 2b is added.
2 to 2d, the configuration is the same as that of the encoding device in FIG.

The decoder 51 decodes the difference data encoded by the encoder 4a and outputs it to the calculator 52. The calculator 52 adds the decoded CTV image data output from the decoder 43 and the decoded difference data output from the decoder 51, and calculates the pixel group A as the prediction circuit 2
Output to b to 2d.

Therefore, in this encoding device, the same pixel group A as the pixel group A used in the decoding device is output from the calculator 52 to the prediction circuits 2b to 2d, and based on this, the prediction pixel is generated there. Since the groups B ′ to D ′ are generated, the decoding device that decodes this data can accurately decode the image.

Next, FIG. 13 is a block diagram showing the configuration of an embodiment of a decoding device for decoding image (HD image or CTV image) data encoded by the encoding device of FIG. 10 or 12. Is. In the figure, parts corresponding to those in FIG. 7 are designated by the same reference numerals. The decoder 61 is, for example, the same as the decoder shown in FIG. 15 described above, and decodes the encoded CTV image data supplied from the demultiplexing circuit via the transmission path. Calculator 62
Adds the CTV image data and the difference data decoded by the decoder 61 or 21a, respectively, and outputs the pixel group A
Is output to the prediction circuits 22b to 22d and the synthesis circuit 24.

In the thus constructed decoding device, the decoder 61 decodes the coded CTV image data supplied from the demultiplexing circuit via the transmission line, and the calculator 62 The CTV image data and the difference data decoded by the decoder 61 or 21a are added, and the pixel group A is output to the prediction circuits 22b to 22d and the synthesis circuit 24.

Then, in the same manner as in the decoding device of FIG. 7, the pixel groups B to D are output from the arithmetic units 23b to 23d to the synthesizing circuit 24, respectively, and are synthesized with the pixel group A to generate HD image data. Is output.

As described above, CTV image data with little deterioration in image quality can be obtained together with the HD image.

It should be noted that in the decoding apparatus of FIG.
Similar to the case of the decoding device of (1), the upsampler 31 can be provided at the subsequent stage of the arithmetic unit 62 to generate FF / FR image data.

In addition, the encoders 4a to 4 in the present embodiment.
d and the encoder 42, or the decoders 11, 21a to 2
1d, 43, 51, and 61 are shown in FIG.
It is not limited to the encoder or the decoder shown in FIG.

[0107]

As described above, according to the encoding method of the present invention, image data is divided into a basic pixel group and another pixel group,
A prediction pixel group that predicts another pixel group is generated from pixels that are adjacent to each other in the horizontal, vertical, or diagonal directions of the basic pixel group. Then, the prediction error of the prediction pixel group with respect to the other pixel groups is calculated, and the basic pixel group and the prediction error are encoded.
Therefore, the compression efficiency of image data can be improved.

According to the encoding apparatus of the present invention, image data is decomposed into a basic pixel group and another pixel group, and a predicted pixel group that predicts another pixel group is generated from the basic pixel group. Then, the prediction error of the prediction pixel group with respect to the other pixel groups is calculated, and the basic pixel group and the prediction error are encoded. Therefore, the compression efficiency of image data can be improved.

Further, according to this coding device, the basic pixel group coded by the coding means is decoded, and the generation means predicts another pixel group from the basic pixel group decoded by the decoding means. The predicted pixel group is generated. Therefore,
As a result, the prediction error of the same prediction pixel group as the prediction pixel group generated by the decoding device for decoding the encoded data with respect to the other pixel groups is calculated, so that the deterioration of the image quality is prevented. You can

Further, according to the encoding device of the present invention, the image data is decomposed into a basic pixel group and another pixel group, and a prediction pixel group for predicting another pixel group is generated from the basic pixel group. ,
The prediction error of the prediction pixel group with respect to the other pixel groups is calculated, the image data is sub-sampled, the difference between the sub-sample data and the basic pixel group is calculated, and the difference data is calculated. Then, the subsample data, the difference data, and the prediction error are encoded. Therefore, it is possible to further improve the compression efficiency of image data, and it is possible to easily reproduce an image other than the standard of the original image from the sub-sample data on the decoding device side.

Further, according to this encoding device, the difference data or the subsample data encoded by the encoding means are respectively decoded, and the basic pixel group is reproduced from the difference data and the subsample data. Then, in the generating means, from the basic pixel group reproduced by the reproducing means,
A prediction pixel group that predicts another pixel group is generated. Therefore, the prediction error of the prediction pixel group that is exactly the same as the prediction pixel group generated by the decoding device that decodes the encoded data is calculated with respect to the other pixel groups.
It is possible to prevent deterioration of image quality.

According to the decoding device of the present invention, the transmission data is decoded into the basic pixel group or the prediction error, and the predicted pixel group is generated from the basic pixel group. Then, the original pixel group is calculated from the predicted pixel group and the prediction error, and the basic pixel group and the original pixel group are combined to generate image data. Therefore, it is possible to easily decode high-definition images with a large amount of data, which have been post-arrival compression-coded, without degrading the image quality.

Further, according to this decoding device, since the basic pixel group is sub-sampled and the sub-sample data is output, it is possible to easily reproduce an image other than the standard of the original image from the sub-sample data. it can.

Further, according to the decoding apparatus of the present invention, the transmission data is decoded into the sub-sample data, the difference data or the prediction error, and the basic pixel group is calculated from the sub-sample data and the difference data. A prediction pixel group is generated from the pixel group. Then, the original pixel group is calculated from the predicted pixel group and the prediction error, and the basic pixel group or the original pixel group is combined to generate image data. Therefore,
It is possible to easily decode data in which a high-definition image having a large amount of data is arrear compression-coded without degrading the image quality.

Further, according to these decoding devices, since the basic pixel group is up-sampled and the up-sampled data is output, when the image is fast forward reproduced, the data amount of the image to be read out is small, You can perform smooth fast forward playback.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of an encoding device of the present invention.

FIG. 2 is a diagram for explaining the operation of the dividing circuit 1.

FIG. 3 Prediction circuits 2b to 2d or 22b to 22d
6 is a diagram for explaining the operation of FIG.

FIG. 4 Prediction circuits 2b to 2d or 22b to 22d
6 is a diagram for explaining the operation of FIG.

FIG. 5 is a diagram for explaining an order in which data is encoded in the embodiment of FIG.

FIG. 6 is a block diagram showing the configuration of a second embodiment of the encoding device of the present invention.

FIG. 7 is a block diagram showing a configuration of an embodiment of a decoding device of the present invention.

FIG. 8 is a diagram showing data used when CTV image data is decoded in the embodiment of FIG. 7.

FIG. 9 is a block diagram showing a configuration of a second embodiment of the decoding device of the present invention.

FIG. 10 is a block diagram showing the configuration of a third embodiment of the encoding device of the present invention.

11 is a diagram for explaining the order in which data is encoded in the embodiment of FIG.

FIG. 12 is a block diagram showing the configuration of a fourth embodiment of the encoding device of the present invention.

FIG. 13 is a block diagram showing the configuration of a third embodiment of the decoding device of the present invention.

FIG. 14 is a block diagram showing a configuration of an example of a conventional encoder.

FIG. 15 is a block diagram showing a configuration of an example of a conventional decoder.

[Explanation of symbols]

 1 division circuit 2b to 2d prediction circuit 3b to 3d arithmetic unit 4a to 4d encoder 11 decoder 21a to 21d decoder 22b to 22d prediction circuit 23b to 23d arithmetic unit 24 synthesis circuit 25 subsampler 31 upsampler 41 subsampler 42 encoder 43 Decoder 44 Operation unit 51 Decoder 52 Operation unit 61 Decoder 62 Operation unit 70 Motion vector detection circuit 71 Frame memory 71a Forward original image part 71b Original image part 71c Rearward original image part 72 Operation part 72a, 72b Operation device 72c Switch 73 DCT circuit 74 Quantization circuit 75 Variable length coding circuit 76 Transmission buffer 77 Inverse quantization circuit 78 IDCT circuit 79 Motion compensation circuit 79a Arithmetic unit 80 Frame memory 80a Forward prediction image section 80b Backward prediction image section 81 Prediction determination circuit 91 Reception buffer 92 Variable length decoding circuit 93 Inverse quantization circuit 94 IDCT circuit 95 Motion compensation circuit 95a Arithmetic unit 96 Frame memory 96a Forward prediction image part 96b Backward prediction image part

Claims (14)

[Claims] 1. The image data is divided into a basic pixel group and another pixel group, a predicted pixel group that predicts the other pixel group is generated from the basic pixel group, and the prediction is performed for the other pixel group. An encoding method characterized by calculating a prediction error of a pixel group and encoding the basic pixel group and the prediction error. 2. The encoding method according to claim 1, wherein each pixel of the prediction pixel group is generated from pixels adjacent to each other in the horizontal direction of the basic pixel group. 3. The encoding method according to claim 1, wherein each pixel of the prediction pixel group is generated from vertically adjacent pixels of the basic pixel group. 4. The encoding method according to claim 1, wherein each pixel of the prediction pixel group is generated from pixels adjacent to each other in the diagonal direction of the basic pixel group. 5. The encoding method according to claim 1, wherein each pixel of the prediction pixel group is generated from a plurality of pixels adjacent to each other in the basic pixel group. . 6. A decomposing unit that decomposes image data into a basic pixel group and another pixel group, and a predictive pixel group that predicts the other pixel group from the basic pixel group output from the decomposing unit. Generating means, calculating means for calculating the prediction error of the predictive pixel group generated by the generating means with respect to the other pixel group, basic pixel group output from the disassembling means, and calculated by the calculating means An encoding device, comprising: an encoding unit that encodes a prediction error. 7. The decoding means further comprises a decoding means for decoding the basic pixel group coded by the coding means, wherein the generating means selects the other pixel from the basic pixel group decoded by the decoding means. The encoding device according to claim 6, wherein a prediction pixel group that predicts a group is generated. 8. A decomposing unit that decomposes image data into a basic pixel group and another pixel group, and a predictive pixel group that predicts the other pixel group from the basic pixel group output from the decomposing unit. Generation means, prediction error calculation means for calculating a prediction error of the prediction pixel group generated by the generation means with respect to the other pixel group, and subsampling means for subsampling the image data and outputting subsample data A difference calculation means for calculating difference data by calculating a difference between the sub-sampled data output from the sub-sampling means and the basic pixel group output from the decomposition means, and a sub-output value output from the sub-sampling means The sample data, the difference data calculated by the difference calculation means, and the prediction error calculated by the prediction error calculation means are recorded. Encoding apparatus characterized by comprising an encoding means for reduction. 9. A decoding unit that respectively decodes the difference data or subsample data encoded by the encoding unit, the difference data decoded by the decoding unit, and the subsample data from the basic data. A reproducing unit for reproducing a pixel group is further provided, and the generating unit generates a predicted pixel group that predicts the other pixel group from the basic pixel group reproduced by the reproducing unit. 8. The encoding device according to item 8. 10. The encoding apparatus according to claim 6, wherein the image data is high definition wide television image data. 11. Decoding means for decoding the transmission data into a basic pixel group or a prediction error, respectively, and a generating means for generating a predictive pixel group from the basic pixel group output from the decoding means. Calculating means for calculating the original pixel group from the prediction pixel group generated by the means, and the prediction error output by the decoding means, the basic pixel group output by the decoding means, and the calculating means A decoding device, comprising: a synthesizing unit that synthesizes the calculated original pixel group to generate image data. 12. The decoding apparatus according to claim 11, further comprising subsampling means for subsampling the basic pixel group output from the decoding means and outputting subsample data. 13. Transmission data is sub-sampled data,
Decoding means for decoding the difference data or the prediction error, respectively, a first calculating means for calculating a basic pixel group from the subsample data and the difference data output from the decoding means, and the first calculating means. From the basic pixel group calculated by
Generating means for generating a predictive pixel group, second predicting means for calculating an original pixel group from the predictive pixel group generated by the generating means, and a predictive error output by the decoding means; A decoding device comprising: a synthesizing unit that synthesizes the basic pixel group or the original pixel group calculated by the first or second calculating unit to generate image data. 14. The decoding apparatus according to claim 11, further comprising an upsampling unit that upsamples the basic pixel group and outputs upsampled data.