JP3400428B2 - Image transmission method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] The present invention relates to an image transmission method.
And the encoding process is particularly efficient.
The information to properly decode the decoded image data
It is about the method. [0002] 2. Description of the Related Art Image coding has a long history, and ITU-T
Excellent standardization proposals such as H.261, ITU-T H.263, ISO MPEG1 / 2
Has been established. The image encoding method can be roughly classified into orthogonal
Encoding method using transform and prediction using prediction function
There is a predictive coding method for coding an error with a value. Orthogonal
Although the encoding method by transform is complicated in calculation,
When obtaining a coded signal with the number of bits,
High quality coding can be performed. JPEG, MPEG, etc.
DCT (discrete code)
Sign conversion). Less with DCT
It is known that it can be encoded by the number of bits.
High precision multiplication is required, and the calculation becomes complicated.
The problem is that reversible encoding is not possible. Therefore,
DCT operation cannot be used in fields where inversion is required.
It is supposed to be. On the other hand, predictive coding is easy to calculate.
And that reversible encoding is also possible.
You. Facsimile is a reversible image coding method.
MMR (Modified Modified Read) used in millimeters
Is famous. This is CCITTRec.T6 "Facsimilie Codi
ng Schemes and Coding Control Functions for Group
4 Facsimile Apparatus "
Change point of the pixel value of the immediately preceding scan line
The difference value in the horizontal direction from the pixel change point of the
This is a method of performing variable length coding. In addition, this MMR is further improved
MMMR (Modified MMR) is used as an evaluation model for MPEG4.
(ISO / IEC JTC / SC29 / WG11N1277, July
 1996). On the other hand, an image signal is separated for each object and separated.
If the image signal is handled with the shape of each object as an arbitrary shape,
Images can be manipulated and composited on an object-by-object basis, improving efficiency
Signal transmission. Also, the number of bits
For applications that are restricted, using this information
Transmission and recording of important objects selectively
Is also possible. However, in the prior art, any shape
Coding of objects with shape is not considered
Was. An image signal having an arbitrary shape in ISO MPEG4
Standardization of signal coding is in progress. MPEG4 now VM
3.0 (described in ISO / IEC JTC1 / SC29 / WG11 N1277)
Evaluation model is created, which has an arbitrary shape
Known to be able to encode image signals
This is the only image coding method. [0005] An image signal of an arbitrary shape generally has the shape of an object.
Shape information indicating the shape and the pixel value that is each pixel value inside the object
Information (color information). And shape information
As a report, each pixel value is significant (inside the shape)
Binary shape information indicating whether it is insignificant (outside the shape)
Or the ratio of each pixel when combining with other images (object
Using the transparency information, which is the ratio of hiding the background)
it can. Also, there are only two types of transmittance, 0% and 100%.
The shape information and the transparency information match, so
Represents an image signal of an arbitrary shape with two pieces of shape information and pixel value information
it can. FIG. 53 is a diagram for explaining such information.
It is. The image of the fish shown in FIG. 53 (a) is combined with another image.
When using each pixel of this image as what ratio
Transparency information indicates whether or not the images are combined. FIG. 53 (a)
The values of the transmittance information in the horizontal scanning line direction indicated by dotted lines in FIG.
This is shown in FIG. The outside of the fish is completely transparent. here
Defines a transparency of 0 as completely transparent for convenience.
Therefore, outside the fish, the value of the transparency information is 0,
Then, the inside of the fish becomes a non-zero value. Therefore, two kinds of transmittance, 0 and non-zero, are used.
FIG. 53 (c) shows the transparency information binarized.
In FIG. 53 (c), pixels having a non-zero transmittance are symbols of pixel value information.
However, the pixel value information of a pixel having a transparency of 0 is not
It is important to encode the pixel value information
The transmitted transmittance information is very important. On the other hand, FIG.
The components of transparency information that cannot be represented by binary information as shown in
Is multi-valued information called gray scale.
Shape information expressed as multi-valued information is pixel value information
Waveform encoding similar to the above. In image coding, spatial correlation is used.
Based on intra-frame (frame) coding and temporal correlation
Use and use inter-frame (frame) coding
Or to be done. Of these, inter-screen coding is performed.
In this case, the motion on the adjacent screen is detected and
Motion compensation is performed for the current moment. One for motion compensation
Generally, a motion vector is used. In VM3.0 mentioned above,
Intra- and inter-screen coding adaptively in block units
Switching, performs motion compensation similar to that of MPEG1 / 2, and
The rate is being improved. As described above, the shape information and the pixel value information
Used to encode images when encoding
For shape information, use motion vector of pixel value information
And then perform motion compensation coding of the shape information,
ISO encoding is more efficient than tangent encoding
/ IEC JTC1 / SC29 / WG11 N1260 Reported in March 1996
You. Also, when performing motion detection and motion compensation, the shape information
Is separated into a binary shape information component and a multi-valued information component.
The value information component performs the same waveform encoding together with the pixel value information.
Is considered to be efficient and treated as such
Was. [0010] SUMMARY OF THE INVENTION As described above,
Technology-based image coding and associated image decoding
Has the following problems. As mentioned above
In addition, MMR coding is a typical example of lossless coding.
However, because it is reversible, it is not visually important.
Greatly improves the compression ratio by allowing for poor image quality
It is impossible to do. In addition, MMR encodes in-screen
This is a method of improving the compression ratio using correlation between screens.
Not considered. And MMR and its improvement method
In an MMMR, the change point of the current scan line
Only the difference from the change point is used, and the vertical straight line
Redundancy of connection (correlation) has not been sufficiently removed
No. Therefore, if the pixel value changes along the scan line,
Encode if coding efficiency is good but not along scanline
Efficiency gets worse. Also, MMR or MMMR is the change of the immediately preceding scan line.
Encode a pixel that cannot be encoded as the difference value
Horizontal coding mode that does not use vertical correlation at all.
Have a code. This horizontal encoding mode is also
There is room for further efficiency improvement by using correlation.
You. Further, in the conventional MMR and MMMR, some
Hierarchically by decoding the stream
Cannot be played back, and hierarchical image playback is possible
In other methods, the coding efficiency is not good,
Has the disadvantage of increasing the number of images, enabling effective hierarchical image playback.
Does not exist. Also, the shape information and
When encoding an image consisting of image information by motion compensation
Conventionally, the same motion vector as image information is used for shape information.
Motion compensation using torque, but for example, the sphere rotates
Shape, the shape does not change, but the shape drawn on the sphere is
In general, motion vectors and shapes of image information
It does not match the motion vector of the state information. Therefore such a place
That good coding cannot be performed
Was a problem. Also, as described above, in VM3.0,
Adaptive switching between coding and inter-screen coding in block units
To improve coding efficiency, but within the screen / screen
Judgment for inter-coding is adaptive switching in MPEG1 / 2
As in the case of, it is determined based on the pixel value information,
Appropriately and appropriately form information whose properties differ greatly from pixel value information
It has been difficult to code efficiently. The present invention
This has been done in view of the
Image that enables appropriate decoding of the encoded signal
It is an object to provide an image transmission method. [0013] Means for Solving the Problems The present inventionToSuch image transmission method
The law isIndicates the shape of the objectShape information and, Each object
Pixel color informationImage signal including pixel value information
Is an encoded signal ofImage transmission for transmitting image encoded signals
Sending method,Supports blocks consisting of a predetermined number of pixels
To doBlocked shape information and pixel value information
Encoded identification information that identifies the encoding mode for
Is included in the image coded signal and transmitted.
Each of the above shape information and pixel value information
Encoding mode corresponding toToRespectively,In-screen encoding mode
Mode or inter-screen coding mode
Is characterized byThings. [0014] [0015] BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described.
Will be explained. Embodiment 1 FIG. Image coding according to Embodiment 1 of the present invention
The device performs a predetermined range in performing the predictive coding.
By selecting a difference value with a short coding length within the box,
It performs rate coding. FIG. 1 shows an embodiment of the present invention.
FIG. 3 is a block diagram showing a configuration of an image encoding device according to Embodiment 1.
is there. In the figure, reference numeral 1 denotes an input signal, which is a binary image
The signal is input to the image encoding device as a signal. 2 is a change pixel
Detector whose pixel value changes with respect to input signal 1
Is detected and output as a detected change pixel. 3 is
Encoding and decoding that is a memory and used as a reference image
The stored image signal is temporarily stored. 4 is change pixel prediction
The pixel value of the reference image changes based on
To predict the changed pixel output from the changed pixel detector 2,
Output the measured change pixel. The prediction that the change pixel predictor 4 can use
As a measurement method, for example, the vertical direction of a two-dimensional image signal is
Based on strong correlation, same as the change pixel of the upper scan line
Typical prediction method for predicting that there is a changing pixel at the horizontal position
Etc. can be used. Numeral 5 is a difference value calculator.
Value of the changed pixel detected by the transformed pixel detector 2 and the difference value of the predictor 4
Calculate D. 6 is preset as a rounding error tolerance.
Is the value e given to the difference value rounder. 7 is the difference value
It is a rounder, within the range defined by the tolerance e.
Then, the difference value D is corrected, and a corrected difference value D ′ is output.
Reference numeral 8 denotes an encoder, which encodes a difference value. 9 is encoding
This is an encoded signal output from the device 8. 11 is the difference value addition
A modified difference value D ′ output from the difference rounder 7,
Add the predicted changed pixel output from the changed pixel predictor 4
You. Reference numeral 10 denotes a changing pixel decoder, and a difference value adder 11
Decodes the binary pixel value using the addition result output from
You. According to the first embodiment configured as described above.
The operation of the image encoding apparatus will be described. Binary
When an input signal 1, which is an image signal, is input to the device, a change occurs.
The pixel detector 2 receives an input signal 1 as an input and outputs a binary pixel value.
Are detected. On the other hand, the change pixel predictor 4
The reference image stored in the memory 3 is read out, and the input
Predict the changing pixels in the signal. The changed pixel detector 2 detects
The output result is output to the difference value calculator 5 as a detected change pixel.
Then, the changed pixel predictor 4 regards the result of the prediction as a predicted changed pixel.
And outputs it to the difference value calculator 5. And the difference value calculator
5 subtracts the predicted change pixel from the detected change pixel, and
The difference value D corresponding to the raw prediction error is obtained. Difference value meter
The arithmetic unit 5 outputs the difference value D to the difference value rounding unit 7. The difference value rounder 7 is provided with a preset value.
The allowable value e and the prediction error output from the difference value calculator 5
The difference value D corresponding to the difference is compared with the difference value D, and the difference value D becomes the allowable value e.
If it does not exceed the value x, the value x satisfies D-e ≦ x ≦ D + e.
Is the smallest number of bits when the value x is encoded
And outputs it as a modified difference value D '
You. On the other hand, the difference value D corresponding to the prediction error is
If the difference exceeds the value e, the corrected difference value is calculated based on the allowable value e.
D ′ is obtained and output to the encoder 8. And fix
The difference value D ′ is encoded by the encoder 8 and
Become. The corrected difference output from the difference value rounder 7
The value D ′ is also output to the difference value adder 11. Addition of difference value
The modified difference value D ′ of the change pixel predictor 4
The pixel is added by adding the predicted change pixel to be output.
The changing pixel of the value is calculated and the result is the changing pixel decoder 1
Output to 0. The changing pixel decoder 10 performs the changing pixel prediction.
From the decoded pixel indicated by the output of the measuring instrument 4, a difference value adder
11 to the input change pixel.
Then, the pixel value is decoded and stored in the memory 3. This
As a result, the content stored in the memory 3 is used as a reference image.
You. The above operation will be specifically described with reference to FIG.
I do. FIG. 2 shows each pixel value as a model of a binary image signal.
It is represented as a rectangle of white and black (fine diagonal lines)
Here, for simplicity of explanation, one pixel at a time is sequentially processed.
The processing procedure will be described. Figure 2 (a) shows the input signal
Scans from the upper left to the right and moves to the lower right
Processing shall proceed. On one line (scanning line)
The pixel whose pixel value changes (white → black or black → white)
It is a changed pixel. Pc in FIG. 2B is the last pixel that has been encoded,
u is the change pixel in the upper scan line, and
Shaded area indicates pixels that have not been encoded yet
And The changing pixel detector 2 receives the input signal shown in FIG.
Of the unencoded part shown in FIG. 2 (b)
To determine the changing pixel whose pixel value changes,
P1 is detected, and this is used as a detected change pixel.
Output to On the other hand, the change pixel predictor 4 uses the above-described method.
Therefore, the change pixel is predicted, and the change pixel Pu of the scanning line at the upper position is predicted.
Predict pixel P0 as if it were at the same horizontal position as
This is output to the difference value calculator 5 as a predicted change pixel. difference
The value calculator 5 calculates the detected change pixel P1 and the predicted change pixel P0
D = 1 is output to the difference rounder 7 as the difference value of. here
Then, the settings of the image encoding device according to the first embodiment include:
A code with a shorter code length is assigned as the difference value from P0 is smaller.
Encoding. And the rounding error tolerance e
It is assumed that 1 has been given. To the difference value calculator
Since the difference between P1 and P0 is less than or equal to e,
The difference value rounder 7 outputs
D "= 0. As a result, the changed pixel is rounded,
Because the pixel is subjected to the decoding process, the encoded and decoded pixel values
Is as shown in FIG. On the other hand, the input signal is shown in FIG.
If it is, the prediction change is performed as shown in FIG.
The difference value D indicated by the difference between the conversion pixel P0 and the detection change pixel P1 is 2
In this case, the difference value D exceeds the allowable value e.
Become. Therefore, the difference value rounder 7 calculates the prediction error (difference value)
To avoid exceeding the allowable range.
And the difference value -1 corresponding to the changed pixel P2 is output.
Power. As a result, the encoded and decoded pixel values are
It looks like 2 (f). Thus, the image of the first embodiment is
The image coding apparatus includes a difference value rounder 7 and detects
The difference value between the changed pixel and the predicted changed pixel is
Using the tolerance 6 obtained, the prediction error less than the tolerance is used.
And the corrected difference value that minimizes the code length of the error (difference value)
Is selected, and this is output.
In some cases, greatly reduce the number of bits required for encoding
be able to. Further, the image encoding apparatus according to the first embodiment
The encoded signal 9 obtained in
The decoding process is possible in the device. Embodiment 2 FIG. According to Embodiment 2 of the present invention
Image encoding device based on the prediction by the frame
Prediction with motion compensation by coding and reference frame
That performs processing by adaptively switching between encoding based on
It is. FIG. 3 is a diagram showing image coding according to the second embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of the device. In the figure, 2
0 is a motion compensator, which is a coded and decoded reference frame.
Motion compensation for the image signal of
To achieve. Reference numeral 21 denotes a mode selector, which is an image of the frame.
The difference value when the prediction based on the image signal is performed and the reference frame
And the difference value when prediction is performed based on the
The number of bits required for encoding is small.
Select the one without the encoding mode. 22 is switching
To the encoding mode selected by the mode selector 21.
Select and output the corresponding difference value. About symbols 1-9
1 and the description is the same as that of the first embodiment.
Since there is, it is omitted here. The image of the second embodiment thus configured is
The operation of the image encoding device will be described. Binary image
When an input signal 1, which is a signal, is input to the apparatus, the input signal
1 is input to the change pixel detector 2 and the memory 3
Is also input to the memory 3 and stored in the memory 3.
Coded and decoded reference in the frame
Used as an image. The change pixel detector 2 receives the input signal 1
Is used as an input to detect a pixel whose binary pixel value changes
You. The changed pixel detector 2 uses the detection result as a detected changed pixel.
And outputs it to the difference value calculators 5a and 5b. On the other hand,
The coded pixel predictor 4a calculates the corresponding frame stored in the memory 3.
Read the encoded and decoded reference image of
The change pixel is predicted based on the force signal, and the result is predicted and changed.
It is output to the difference value calculator 5a as a converted pixel. And the difference
The value calculator 5a subtracts the predicted change pixel from the detected change pixel.
To obtain the difference value D. Output D of difference value calculator 5a
Is based on the encoded and decoded pixels in the frame.
Is equivalent to the prediction error of the changed pixel
The difference value calculator 5a uses the difference value D as a mode selector
21 and a switch 22. The motion compensator 20 is stored in the memory 3
Motion compensation is performed on the encoded and decoded images of the reference frame.
The change pixel estimator 4b compensates for this motion-compensated image.
The change pixel of the input signal is predicted based on the
The result is output to the difference value calculator 5b as a predicted change pixel.
The difference value calculator 5b calculates a predicted change pixel from the detected change pixel.
Subtraction is performed to obtain a difference value D ". Output of the difference value calculator 5b
D "is based on the encoded and decoded pixels in the reference frame.
To the prediction error of the changed pixel predicted with motion compensation.
The difference value calculator 5b calculates the difference value
D "is output to the mode selector 21 and the switch 22. The mode selector 21 includes a difference value calculator 5a,
Difference value D and difference value D "input from
The code length when encoding each (bits required for encoding)
Number of bits), and can be encoded with fewer bits.
Measurement method and use that identification signal as the encoding mode
Output. When the mode selector 21 encodes the difference value D
If it is determined that the code length of the
The code length when encoding the frame "" and the difference value D "is short
If it is determined that the encoding mode "reference frame",
It outputs to the switch 22 and the encoder 8a. The switch 22 is the output of the mode selector 21
Corresponding to the encoding mode "the frame",
The difference value D output from the value calculator 5a is converted into an encoding mode
If it is a "reference frame", the difference value calculator 5b outputs
The difference value D "is output to the encoder 8b. The encoder 8a
The encoding mode selected by the mode selector 21 is encoded,
The encoder 8b encodes the output difference values, and
And output as encoded signals 9a and 9b. Of this implementation
Encoding by the image encoding apparatus according to mode 2 involves rounding errors.
Lossless coding, and the input image
The signal 1 is a pixel value which is coded and decoded up to the change pixel.
And stored in the memory 3. The above operation will be specifically described with reference to FIG.
I do. FIG. 4 is used for explanation in the first embodiment.
Each pixel value is represented as a model of a binary image signal as in FIG.
It is represented by white and black rectangles, and here also the embodiment
Similarly to 1, for simplicity of description, one pixel at a time
The processing procedure will be described as processing. Smell
4 (a) is an input signal, and FIG. 4 (b) is an image of a reference frame.
Signal. FIG. 4C illustrates prediction based on the frame.
It is a figure for clarification. P1 is the same as in the first embodiment.
It is a detection change pixel detected by the conversion pixel detector 2. Pc
Decoded final pixel position, Pu is the pixel value of the upper scan line
And the coarse shaded area is still encoded.
It is assumed that it represents a pixel that has not been processed. Change pixel predictor 4
a is a method similar to the method of predicting a changed pixel in the first embodiment.
, Based on the change pixel Pu of the scanning line at the upper position,
Predicts using the relationship and is at the same horizontal position as Pu
Let P0 be a predicted change pixel based on the frame. The image signal of the reference frame shown in FIG.
After the motion compensation by the compensator 20
Then, the changed pixel predictor 4b acquires the predicted changed pixel Pr.
You. Therefore, the difference value D by the difference value calculator 5a is P1 and P0
And the difference value D "by the difference value calculator 5b.
Is 0, which is the difference between P1 and Pr. Image code according to the second embodiment
As for the setting of the encoding device, P0 and
Code that assigns a code with a shorter code length as the difference value of
Encoding, the difference between P1 and P0 is encoded.
Also, encoding the difference between Pr and P1 reduces the code length.
Become. Therefore, the selection of the mode selector 21 changes the difference value D "
It becomes the "reference frame" to be output, and the encoding mode "reference
The frame "and the difference value D" are encoded, and
It becomes a coded signal output from the image coding device of state 2. Figure
4 (e) is a decoding obtained by decoding this encoded signal.
The results are shown. As described above, the image coding of the second embodiment
In the device, the memory 3 and the change pixel predictors 4a and 4b
And the difference value calculators 5a and 5b and the motion compensator 20
Prediction based on the frame
And prediction with motion compensation based on the reference frame.
The difference between each predicted value and the detection result,
Mode selector 21, switcher 22, encoder 8
a and 8b, the frame
Prediction based on the motion-compensated reference frame
By comparing the difference value with the prediction, the one with the minimum code length
You can select and encode,
By using elementary correlation, the bits required for encoding
The number can be greatly reduced. The image coding apparatus according to the second embodiment uses
Indicates that the input signal 1 is input in block units,
Set the coding mode to be selected in block units.
That is, based on prediction by the frame in units of blocks
Prediction with motion compensation by coding and reference frame
Adaptively switch between encoding based on
And the above effects can be obtained. Further, in the image coding apparatus according to the second embodiment,
Are changed pixel detector 2, changed pixel predictors 4a and 4b
Now, let's output the distance (number of pixels) to the changed pixel.
However, "the next pixel is a change pixel" and "the next pixel
Is not a change pixel. "
For example, “0” and “1” are output, and the difference value
The arithmetic units 5a and 5b calculate the difference value between the binary signals.
It can also be. However, in this case,
Instead of encoding the distance as in
The output of the difference value calculators 5a and 5b is
Will be encoded. With such a setting,
Change pixel detector 2, change pixel predictors 4a and 4b
Output is binary, so that the encoding process can be simplified.
The effect is obtained. Embodiment 3 FIG. According to the third embodiment of the present invention.
The image decoding device according to the second embodiment
For an encoded signal efficiently encoded by
It performs appropriate decryption. FIG. 5 shows an embodiment of the present invention.
FIG. 24 is a block diagram illustrating a configuration of an image decoding device according to a third embodiment.
is there. In the figure, reference numerals 30a and 30b denote the symbols in FIG.
Encoded signals corresponding to the encoded signals 9a and 9b,
A signal obtained by encoding the encoding mode and a difference value are encoded.
Signal. 31a and 31b are respectively
Encoded signal of encoding mode and difference value encoded
And decoding the prediction signal and the decoded difference value.
The decoder to get. 32 is taken by the decoder 31a.
According to the obtained prediction mode signal, the predicted value of the changed pixel is
It is a switching device for switching. 34 is the decoded image
Signal. The memory 3, the changing pixel decoder 10, and the difference
1 is the same as that of FIG.
The device 20 is the same as that of FIG.
Since it is similar to the first and second embodiments, the description is omitted here.
I do. The present embodiment shown in FIG.
The operation of the image decoding apparatus according to Embodiment 3 will be described.
I do. In the image coding apparatus according to Embodiment 2,
The signal 9a obtained by encoding the selected encoding mode is the input signal 3
0a is input to the image decoding apparatus of the third embodiment.
And decoded by the decoder 31a.
To indicate “the frame” or “reference frame”.
A measurement mode signal is obtained. The decoder 31a operates in the prediction mode.
Output to the switch 32. Also, the embodiment
2, the selected difference value is encoded.
In this embodiment, the decoded signal 9b is used as the input signal 30b.
3 and is inputted to the decoder 31b.
Decrypted to obtain a decryption difference value.
You. The decoder 31b adds the decoded difference value to the difference value adding means 11
Output to On the other hand, the change pixel predictor 4a stores
Read the decoded reference image of the frame
And predicting a changed pixel based on the image signal.
As a predicted change pixel based on the frame
32. In addition, the motion compensator 20 stores
Motion compensation is performed on the decoded image of the stored reference frame.
The change pixel estimator 4b compensates for this motion-compensated image.
The change pixel of the input signal is predicted based on the
The result as a predicted change pixel based on the reference frame
Output to the device 32. From the change pixel predictors 4a and 4b, respectively.
The switch 22 that outputs the predicted change pixel is input to the switch 22.
Switching is performed according to the predicted mode signal thus set. Follow
Then, the switch 22 determines that the input prediction mode signal is
If it indicates “the frame”, the change pixel predictor 4
predicted change pixel based on the frame output from a
Indicates that the input prediction mode signal indicates a `` reference frame ''.
If it is, the reference output from the change pixel predictor 4b
Select the predicted change pixel based on the frame and add the difference value
Output to the means 11. The difference value adding means 11 is provided with a switch 22
The predicted change pixel obtained from the
By adding the obtained decoding difference values, the changed pixel
And outputs the result to the changing pixel decoder 10.
You. The changing pixel decoder 10 outputs the changing pixel predicting means 4a.
Predicted change pixel and change obtained from difference value adding means 11
Based on the pixels, the pixel values between them are decoded. this
The result of decoding is input to the memory 3 and stored.
In the third embodiment, as the decoded image signal 34,
Is output from the image decoding device. For example, in the embodiment
In FIG. 2, the coded signal described with reference to FIG.
In the case of a force signal, the decoding result as shown in FIG.
Fruit is obtained. FIG. 6 shows an image according to an application example of the third embodiment.
It is a block diagram which shows the structure of an image decoding apparatus. As shown in FIG.
The difference from the image decoding device that has been
11a and 11b, and the switch 33 is
Instead of the outputs of the change pixel prediction means 4a and 4b,
A configuration for switching the outputs of the adding means 11a and 11b;
It is a point that has become. Even in such a configuration,
Signal output from the image coding apparatus according to the second embodiment.
Signal in accordance with the encoding mode at the time of encoding.
Can be decoded. Further, changing pixel decoding
And a switching device.
As a position to receive the outputs of the plurality of changing pixel decoders 10
Has the same effect. As described above, the image restoration according to the third embodiment is performed.
In the encoding device, the encoded signal in the encoding mode is decoded.
A decoder 31a for decoding a coded signal of the difference value;
Predicts a changed pixel based on the encoder 31b and the frame
Pixel prediction means 4a,
Change pixel prediction means for predicting change pixels with motion compensation
4b and a difference value for performing a decoding process based on the predicted change pixel
An adder 11 and a changing pixel decoder 10 are provided.
Switching according to the prediction mode acquired by the encoder 31a
The encoder switches, so that the encoding mode at the time of encoding is changed.
In the prediction mode corresponding to the
Decoding using a predicted value based on
The decoding using the measured values is adaptively switched and performed.
Appropriate coded signal efficiently coded in form 2
Can be decoded. In the second and third embodiments,
Prepare a plurality of frames as reference frames.
The above prediction mode may be used. Ma
Also, in the third embodiment, the coding mode is
To process an encoded signal that has been encoded by selecting
Input a signal in block units, and
Acquire measurement mode and perform processing corresponding to encoding mode
Thus, it is possible to perform decoding appropriately. Embodiment 4 FIG. According to Embodiment 4 of the present invention.
Image encoding apparatus, the code based on the prediction by horizontal scanning
And encoding based on prediction by vertical scanning
The processing is performed by switching. FIG. 7 shows an embodiment of the present invention.
Block diagram showing the configuration of an image encoding device according to Embodiment 4.
It is. In the same figure, 40a and 40b are horizontal scanning
And 41a and 41b are vertical scanners. other
Are the same as those in FIG.
Since it is the same as the state 2, it is omitted here. The image of the fourth embodiment thus configured is
The operation of the image encoding device will be described. Binary image
When an input signal 1, which is a signal, is input to the apparatus, the input signal
1 is changed by being horizontally scanned by the horizontal scanner 40a.
Is input to the converted pixel detector 2a and the vertical scanner 41a
Scans in the vertical direction and enters the changing pixel detector 2b.
Is forced. Further, the input signal 1 is also input to the memory 3.
And stored in the memory 3, the frame
As encoded and decoded reference images in the
Can be. The changing pixel detector 2a is scanned in the horizontal direction
Pixel whose binary pixel value changes with input signal 1 as input
Is detected. The change pixel detector 2b is scanned in the vertical direction.
An image in which a binary pixel value changes with the input signal 1
Detect element. The change pixel detectors 2a and 2b detect
The difference value calculators 5a, 5
b. On the other hand, the horizontal scanner 40b stores the data in the memory 3.
The encoded and decoded reference image of the frame
The image is read out, scanned in the horizontal direction, and changed pixel predictor 4a
To enter. The change pixel predictor 4a predicts a change pixel,
The result is output to the difference value calculator 5a as a predicted change pixel.
I do. Then, the difference value calculator 5a calculates the difference
Subtract the predicted change pixel and calculate the difference value Dh
get. The output Dh of the difference value calculator 5a is horizontal scanning.
Is equivalent to the prediction error of the changed pixel predicted by
Therefore, the difference value calculator 5a selects the difference value Dh in the mode.
To the switch 21 and the switch 22. On the other hand, the vertical scanner 41b stores the data in the memory 3.
The encoded and decoded reference image of the frame
The image is read and scanned in the vertical direction to change the pixel predictor 4b.
To enter. The change pixel predictor 4b predicts a change pixel,
The result is output to the difference value calculator 5b as a predicted change pixel.
I do. Then, the difference value calculator 5b calculates the difference
Subtract the predicted change pixel and calculate the difference value Dv by the vertical scanning.
get. The output Dv of the difference value calculator 5b is vertical scanning
Is equivalent to the prediction error of the changed pixel predicted by
Accordingly, the difference value calculator 5b selects the mode of the difference value Dv.
To the switch 21 and the switch 22. The mode selector 21 includes a difference value calculator 5a,
And the difference value Dv and the difference value Dv input from
The code length when encoding each (bits required for encoding)
Number of bits), and can be encoded with fewer bits.
Measurement method and use that identification signal as the encoding mode
Output. When the mode selector 21 encodes the difference value Dh
If it is determined that the code length of the
Direction is determined to be short when the difference value Dv is encoded.
Switch the encoding mode "vertical direction"
To the encoder 22 and the encoder 8a. The switch 22 is an output of the mode selector 21
Corresponding to the encoding mode "horizontal direction", the difference value
The difference value Dh output from the calculator 5a is converted into the encoding mode “vertical
Direction ”, the difference value Dv output from the difference value calculator 5b is
Output to the encoder 8b. The encoder 8a selects the mode
The encoding mode selected by the encoder 21 is encoded, and the
b encodes the output difference values, and encodes each encoded signal.
Output as Nos. 9a and 9b. Image of Embodiment 4
The encoding by the image encoding device is performed without loss
Input image signal 1 as described above.
As the pixel value of the encoded pixel up to the pixel of the
Stored in the memory 3. FIG. 8 shows an image coding apparatus according to the fourth embodiment.
FIG. 6 is a diagram for explaining switching of the scanning direction according to FIG.
The image signal is correlated in the horizontal and vertical directions.
This correlation can also be used in image coding by
Compression had been used. And conventional
Regarding the use of correlation by the technology of
Horizontal or vertical, as is sometimes the case
Encoding based solely on the correlation in one of the directions
Was to do. However, if you look at the image partially,
Or one of the vertical correlations may be stronger than the other.
You. For example, as shown in FIG. 8, the correlation in the horizontal direction is the correlation in the vertical direction.
Stronger than horizontal based on vertical prediction
Based on the prediction of the direction, the prediction of the change pixel of the pixel position
The error is smaller, and the encoding efficiency can be further improved.
It becomes possible. Therefore, depending on the nature of the image,
Switch between vertical and horizontal predictions.
Thus, the coding efficiency can be greatly improved. As described above, the image coding of the fourth embodiment
In the apparatus, horizontal scanners 40a and 40b and vertical scanning
Detectors 41a and 41b and change pixel detectors 2a and 2
b, the memory 3, the change pixel predictors 4a and 4b,
Having difference value calculators 5a and 5b
The prediction by horizontal scanning and the prediction by vertical scanning
And calculate the difference between the predicted value and the detection result.
And a mode selector 21, a switch 22,
By having the containers 8a and 8b,
Difference between prediction by vertical scanning and prediction by vertical scanning
And select the code with the smallest code length to encode
The horizontal and vertical correlation of the image
Encoding by taking advantage of local changes in
, The number of bits required for the operation can be greatly reduced. The image coding apparatus according to the fourth embodiment has
Also, input signal 1 is input in block units
And set the encoding mode to be selected in block units.
In other words, block-by-block prediction based on horizontal scanning
And coding based on prediction by vertical scanning.
The above effects can be adaptively switched
Is obtained. In addition, the image coding apparatus of the fourth embodiment
Also, as in the second embodiment, the changed pixel detector 2
The change pixel predictors 4a and 4b calculate the distance to the change pixel.
(Pixel number) is output instead of pixel
It is also possible to set to output a binary signal.
It is possible to reduce the processing load. Embodiment 5 According to Embodiment 5 of the present invention
The image decoding apparatus according to the fourth embodiment
For an encoded signal efficiently encoded by
It performs appropriate decryption. FIG. 9 shows an embodiment of the present invention.
FIG. 24 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 5.
is there. In the same figure, 40b and 41b are the same as in FIG.
, And other symbols are the same as those in FIG.
Since this is the same as Embodiments 4 and 3,
It is omitted here. In the fifth embodiment configured as described above,
The operation of the image decoding apparatus according to the first embodiment will be described. Implementation
In the image coding apparatus according to aspect 4, the selected code
A signal 9a obtained by encoding the encoding mode is used as an input signal 30a.
Input to the image decoding apparatus according to the fifth embodiment,
31a, the "horizontal direction"
A prediction mode signal indicating “direction” or “vertical” is acquired.
You. The decoder 31a switches the prediction mode signal to the switch 32
Output to Further, the image coding apparatus according to Embodiment 4
, The signal 9b obtained by encoding the selected difference value is input
A signal 30b is input to the image decoding apparatus according to the fifth embodiment.
And decoded by the decoder 31b.
Thus, a decoding difference value is obtained. The decoder 31b decodes
The difference value is output to the difference value adding means 11. On the other hand, the horizontal scanner 40b stores the data in the memory 3.
The encoded and decoded reference image of the frame
The image is read out, scanned in the horizontal direction, and changed pixel predictor 4a
To enter. The change pixel predictor 4a predicts a change pixel,
The result is output to the switch 22 as a predicted change pixel.
I do. On the other hand, the vertical scanner 41 b is stored in the memory 3.
Read the encoded and decoded reference image of the frame
Start, scan in the vertical direction and input to the change pixel predictor 4b
I do. The change pixel predictor 4b predicts a change pixel, and
The result is output to the switch 22 as a predicted change pixel. Each of the change pixel predictors 4a and 4b
The switch 22 that outputs the predicted change pixel is input to the switch 22.
Switching is performed according to the predicted mode signal thus set. Follow
Then, the switch 22 determines that the input prediction mode signal is
If it indicates "horizontal direction", the change pixel predictor 4a
The predicted change pixel based on the horizontal scan output from
The input prediction mode signal indicates "vertical direction".
To the vertical scanning output from the change pixel predictor 4b.
And selects the predicted change pixel based on the
Output. The difference value adding means 11 is provided with a switch 22
The predicted change pixel obtained from the
By adding the obtained decoding difference values, the changed pixel
And outputs the result to the changing pixel decoder 10.
You. The changing pixel decoder 10 outputs the changing pixel predicting means 4a.
Predicted change pixel and change obtained from difference value adding means 11
Based on the pixels, the pixel values between them are decoded. this
The result of decoding is input to the memory 3 and stored.
In the fifth embodiment, the decoded image signal 34 is
Is output from the image decoding device. As described above, the image restoration according to the fifth embodiment is performed.
In the encoding device, the encoded signal in the encoding mode is decoded.
A decoder 31a for decoding a coded signal of the difference value;
Predicts a change pixel based on the horizontal encoder and the encoder 31b
Pixel predicting means 4a that changes based on vertical scanning.
Pixel predicting means 4b for predicting a transformed pixel,
Difference value adding means 11 for performing a decoding process based on a prime,
With the changing pixel decoder 10 and obtained by the decoder 31a.
Switcher switches according to the prediction mode to be changed
This allows prediction corresponding to the encoding mode at the time of encoding.
Mode uses predicted values based on horizontal scanning
Decoding and decoding using predicted values based on vertical scanning
Is adaptively switched, and the efficiency is improved in the fourth embodiment.
To properly decode the encoded signal
It becomes possible. In the fifth embodiment, the implementation
Embodiment 3 has a configuration similar to the configuration shown in FIG.
Although the image decoding apparatus has been described as
Therefore, the configuration shown in FIG.
Further, as described in the embodiment, the switching device
Receive the output of the changing pixel decoder.
It is also possible to perform appropriate decoding as well.
it can. Further, in the fifth embodiment, in units of blocks,
Select an encoding mode and process the encoded signal
When inputting a signal in block units,
For each, the prediction mode was acquired and the encoding mode was supported.
By performing the processing, it is possible to decrypt properly
is there. Embodiment 6 FIG. According to the sixth embodiment of the present invention.
Image encoding apparatus efficiently encodes a multi-level image signal.
Things. FIG. 10 shows an image according to the sixth embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an image encoding device. In the figure
In this embodiment, the input signal 1a is a multi-valued image signal according to the present embodiment.
6 image encoding apparatus. 8a and 8b are symbols
It is an encoder. In this way, multi-level signals are input and processed.
And the configuration with two encoders.
Unlike Embodiment 1, the rest is the same as FIG.
Since it is the same as the first embodiment, the description is omitted here. In the sixth embodiment configured as described above,
The operation of the image coding apparatus according to the first embodiment will be described. input
When the signal 1a is input, the changing pixel detector 2 detects the multi-value
For the input signal, the pixel value of the final encoding / decoding position,
Compare the pixel value of the position following the position with the pixel value, and
At this time, it is determined that there is "change" or "no change". Soshi
The number of pixels that are determined to be "changed"
Is calculated, and the number of changed pixels is compared with a predetermined value. here
Is a predetermined value of 60. Judge as "changed"
The pixel whose number of changed pixels is 60 or more is referred to as a changed pixel.
Judge and detect the pixel value and position of the changed pixel
The difference value calculator 5, the changing pixel decoder 10, and the code
Output to the converter 8a. The change pixel predictor 4 stores the data in the memory 3.
Read the encoded and decoded reference image of the frame
However, based on this, the change pixel is predicted, and this prediction is
Difference value calculator 5 and difference value adder 1 as predicted change pixels
1 and output to the changing pixel decoding means. Difference value calculator
5 is obtained by subtracting the predicted change pixel from the detected change pixel.
The difference value is output to the encoder 8b and the difference value adder 11.
You. The difference value adder 11 calculates a difference between the input predicted change pixel and
The added value is output to the changed pixel decoder 10 to obtain the changed image.
The elementary decoder 10 calculates the pixels up to the change pixel based on the input.
The value and the pixel value of the changed pixel are decoded and stored in the memory 3.
I do. The encoders 8a and 8b respectively transmit the input transforms.
The pixel value of the coded pixel and the difference value are coded and the coded signal 9
a and 9b are output. As described above, the image code according to the sixth embodiment
In the decoding device, in the same configuration as in the first embodiment,
It is judged as "changed" by checking the presence or absence of change for each element
Calculate the number of pixels, and if the number of pixels above the threshold is
Pixels that are determined to be "Yes" are determined to be changed pixels
By doing so, not only binary images but also multi-valued images
Can also enable similar encoding. Embodiment 7 FIG. According to the seventh embodiment of the present invention.
The image decoding apparatus according to the sixth embodiment
Decoding of the encoded signal encoded by
To obtain a multi-valued image signal. Fig. 11
Showing the configuration of an image decoding apparatus according to Embodiment 7 of the present invention.
It is a lock figure. In the figure, a decoder 31a has a
Decoding of the coded signal obtained by coding the pixel value of the coded pixel
And the decoder 31b outputs a code obtained by encoding the prediction difference value.
Decode the encoded signal. Others are the same as FIG.
Are the same as in the third embodiment, and a description thereof will not be repeated. According to the seventh embodiment configured as described above,
The operation of the image decoding apparatus according to the first embodiment will be described. Implementation
In the image coding apparatus according to the sixth aspect, the image of the changed pixel is
The signal 9a obtained by encoding the prime value is used as the input signal 30a.
Input to the image decoding apparatus according to the seventh embodiment,
a, a decoded pixel value is obtained.
The decoded pixel value is input to the changing pixel decoder 10.
It is. Also, in the image encoding device according to the sixth embodiment,
The signal 9b obtained by encoding the prediction difference value is the input signal 30b
Is input to the image decoding apparatus according to the seventh embodiment.
By being decoded in the encoder 31b,
A difference value is obtained, and the decoded difference value is sent to the difference value adder 11.
Is entered. On the other hand, the change pixel predictor 4 accumulates in the memory 3
Read out the decoded reference image, and
And predicts the changed pixel based on the result.
Output to the change pixel decoder 10 and the difference value adder 11
Power. The difference value adder 11 receives the predicted change pixel
And the difference value, and outputs the result to the changing pixel decoder 10.
The changing pixel decoder 10 performs processing up to the changing pixel based on the input.
Of the pixel value of the pixel and the pixel value of the changed pixel
It is output as a signal 34 and stored in the memory 3.
You. As described above, the image decoding apparatus according to the seventh embodiment
Now, for the coded signal obtained by coding the pixel value of the changed pixel,
Decoder 31a that performs decoding by decoding, and encodes the prediction difference value
And a decoder 31b for decoding the encoded signal thus obtained.
, The image encoding device according to Embodiment 6
Multi-valued by properly decoding the encoded signal
Can be obtained. Embodiment 8 FIG. According to Embodiment 8 of the present invention.
The image coding apparatus which transmits the image indicates a ratio at which the image is synthesized.
An image signal consisting of an excessive signal and a pixel value signal is input to an input signal.
And encode this input signal with reference to the reference image
Things. FIG. 12 shows an image according to the eighth embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an image encoding device. In the figure
60a is a pixel value signal, and 60b is a transmittance signal.
And constitute the image signal.
Input as an input signal to the image encoding apparatus according to aspect 8.
You. Reference numeral 61 denotes a memory, which is an encoding used as a reference image.
And temporarily store decoded data such as image signals.
You. Reference numerals 62a and 62b denote motion detectors.
Perform motion detection on the image and output a motion vector.
63a and 63b are motion compensators, which perform coding and decoding.
Motion compensation for the image signal of the
To generate a reference pixel value. 64a and 64b are differences
A value calculator that calculates the input signal and the signal with motion compensation
Is calculated and a difference value is output. 65a and 67b
An encoder that encodes a motion vector. 67a
And 65b are encoders for encoding the difference values. 6
6a and 68b are coded signals for coding motion vectors.
It is a thing. 67a and 65b are encoded signals.
That is, the difference value is encoded. In the eighth embodiment configured as described above,
The operation of the image coding apparatus according to the first embodiment will be described. image
The signals are a pixel value signal 60a and a transmittance signal 60b,
This is input to the image encoding device according to the eighth embodiment. here
The transmittance signal is a diagram used in the description of the related art.
53 (b), when combining with another image
Represents the ratio of each pixel to be synthesized.
is there. The pixel value signal 60a is stored in the memory 61, the motion detector 6
2a and the difference value calculation means 64a.
The signal 60b includes a memory 61, a motion detector 62b, and a difference
It is input to the value calculation means 64b. The motion detectors 62a and 62b are respectively
And the input image and the reference image read from the memory 61.
Comparison with the coded pixel values of the image.
And motion detection, and calculate the motion of each signal.
Get a vector. Image acquired by the motion detector 62a
The motion vector of the raw signal is calculated by the encoder 65a
Is output to the device 63 a and the memory 61. Motion compensator 6
3a represents the pixel value indicated by the motion vector of the pixel value signal.
Read from the memory 61 and calculate the motion compensation value of the pixel value signal.
The value is output to the value calculator 64a. The difference value calculator 64a is
The difference between the input pixel value signal and the motion compensation value is obtained by calculation.
And outputs it to the encoder 67a. Pixel value signal
The motion vector is encoded by the encoder 65a, and
The difference value is encoded by the encoder 67a.
To be a coded signal 68a. Similarly, the transparency obtained by the motion detector 62b is obtained.
The motion vector of the excessive signal is calculated by the encoder 67b,
Output to the device 63b and the memory 61. Motion compensator 6
3b performs motion compensation of the transparency signal and obtains the obtained motion compensation.
The compensation value is output to the difference value calculator 64a. And the difference value
The calculator 64b encodes the acquired difference value in the same manner as 64a.
Output to the container 67a. Like the pixel value signal, the transmittance signal
Is encoded by the encoder 67b.
The difference value is encoded by the encoder 65b.
To become a coded signal 66b. In the eighth embodiment, a reversible code
This is an example of encoding, and the encoded input signal is stored in the memory 61.
Remembered and used for encoding for subsequent image signals
(Not shown). As described above, the image code according to the eighth embodiment
In the encoding device, motion detection is performed to process the pixel value signal 60a.
Unit 62a, motion compensator 63a, difference value calculator 64a, code
The encoder 65a, the encoder 67a, and the transmission signal 60
b, a motion detector 62b, a motion compensator 63b,
Difference value calculator 64b, encoder 65b, and encoder 6
7b, the pixel value signal 60a and the transmittance signal
60b, and perform motion detection separately
It is possible to acquire motion vectors and perform motion compensation.
You. As described in the description of the prior art,
In image coding according to the conventional technology, shape information and pixel value information are used.
When encoding an image consisting of
Image information to improve coding efficiency.
Motion compensation of shape information using motion vector of prime information
Encoding. Therefore, in the eighth embodiment,
When encoding a signal such as an input image signal, the pixel value signal
Motion compensation code of the transparency signal using the motion vector
Will be performed. However, the transmission signal is
, But the motion vector is not necessarily the pixel
It does not match the motion vector of the value signal. Rotate
The shape of the disk is unchanged, but the pattern drawn on the disk is
Is an example. Therefore, in such a case,
The motion vector of the raw signal and the motion vector of the transparency signal
Is large, the motion vector of the pixel value signal is
The motion error when performing motion compensation of the transparency signal using
And the coding length of the difference value becomes longer, resulting in higher coding efficiency.
Will decrease. On the other hand, the image code of the eighth embodiment
The coding device calculates the motion value of the pixel value signal for the transmittance signal.
Motion compensation using motion vectors detected separately from the vector
By doing so, the motion compensation signal
Can be approximated with higher accuracy, and the motion compensation error is small.
As a result, the coding efficiency is improved. In addition, the real
Also in the image coding apparatus according to Embodiment 8, the input signal is
It is input in lock units and moves in block units.
Compensation and coding can be set, and the above effects
Is obtained. Embodiment 9 According to Embodiment 9 of the present invention.
The image coding apparatus according to the third embodiment has the
And an image signal consisting of a pixel value signal as an input signal,
This input signal is encoded with reference to a reference image.
is there. FIG. 13 shows an image code according to the eighth embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of a conversion apparatus. In the figure,
The reference numerals are the same as those in FIG.
The same is true. In the image coding apparatus according to the ninth embodiment,
Is obtained by the motion detector 62a for the pixel value signal 60a.
The motion vector of the obtained pixel value signal is converted into a transmittance signal 60b.
Output to the motion detector 62b
Is near the motion vector of the input pixel value signal.
According to the eighth embodiment, the motion detection of the transmission signal is performed.
This is a difference from the configuration of the image encoding apparatus. In addition, this implementation
As for the operation of the image coding apparatus according to the ninth embodiment, the motion detection
The output unit 62a performs the output described above, and the motion detector 62b
Except for detecting the predicate, the same as that of the eighth embodiment is used.
Operation. As described above, the image code according to the ninth embodiment
In the encoding device, based on the configuration of the eighth embodiment, the pixel value signal
The pixel value obtained by the motion detector 62a for the
The motion vector of the signal is used for motion detection with respect to the transmittance signal 60b.
Output to the output unit 62b, and the motion detector 62b outputs
In the vicinity of the motion vector of the pixel value signal, the transmittance signal
The motion of the transmission signal
For the detection, use the result of motion detection in the pixel value signal.
Is what it is. Motion vector between pixel value signal and transmittance signal
Are significantly different from those in the example shown in Embodiment 8.
, But almost match in many images. So
Here, when detecting the motion vector of the transmittance signal,
Motion vector of the transparency signal only in the vicinity of the motion vector of the value signal.
Is detected, it can be performed completely independently of the pixel value signal.
The number of calculations required for motion detection can be reduced
You. Compared to the case of detecting motion independently of the pixel value signal
Then, the number of selectable motion vectors is limited
The motion compensation error of the transmission signal increases slightly.
But the proportion is small. Therefore, this implementation
In the image coding apparatus according to the ninth embodiment, the same as in the eighth embodiment
In addition, by performing appropriate motion compensation for each signal,
In addition to improving coding efficiency, the number of calculations for motion detection
Can be reduced. Note that the image coding apparatus of the ninth embodiment uses
Is the motion of the pixel value signal when detecting the motion of the transmittance signal.
Although the vector is used, in the embodiment shown in FIG.
Based on the configuration of the image encoding apparatus according to mode 8, the transmission signal
The motion detector 62b for the signal 60b
The motion vector of the signal is used to detect the motion of the pixel value signal 60a.
Output to the output unit 62a, and the motion detector 62a outputs
In the vicinity of the motion vector of the transparency signal, the pixel value signal
Motion detection of the pixel value signal
The result of motion detection in the transparency signal is used
It is also possible to adopt a configuration with
Can be reduced. Also, depending on the settings
The point that coding in block units is possible is
Same as state 8. Embodiment 10 FIG. Embodiment 10 of the present invention
Is the same as in Embodiments 8 and 9.
Input an image signal consisting of a transmittance signal and a pixel value signal.
This input signal is signed with reference to the reference image as a force signal.
It becomes something. FIG. 13 shows a tenth embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an image encoding device. In the figure
70 is a motion vector difference calculator.
Motion vector of pixel value signal obtained from detector 62a
And the motion vector of the transmittance signal acquired from the motion detector 62b.
Get the difference vector with the vector. The encoder 67b
In the eighth embodiment, the motion vector of the transparency signal is encoded.
However, in the tenth embodiment, the difference value calculator 70
The difference vector of the motion vector to be encoded is encoded. Other marks
The symbols are the same as in FIG. 12, and the description is the same as in Embodiment 8.
It is like. The image encoding apparatus according to the tenth embodiment
Regarding the operation, the motion detectors 62a and 62b
The vector is output to the difference value calculator 70, and the difference value calculator 7
0 obtains the above difference vector and outputs it to the encoder 67b.
And the encoder 67b calculates the difference vector of the motion vector
Except for encoding, the operation is the same as that of the eighth embodiment.
Becomes As described above, the image code of the tenth embodiment
The encoding device is based on the configuration of the image encoding device of the eighth embodiment.
And a configuration in which a motion vector difference value calculator 70 is added
To encode the motion vector of the transparency signal
Instead, the motion vector of the pixel value signal and the
The difference vector from the current vector is encoded. Embodiment
As described in Section 9, the motion vectors of both signals are correlated.
In many cases, there is a difference vector between the motion vectors of both signals.
If the vector is encoded, the frequency of occurrence of the difference vector is zero.
Concentrate near the kutor. As a result, the difference around the 0 vector
Variable length code that assigns short code length code to minute vector
Encoding improves coding efficiency and reduces
It becomes possible to encode by the number of bits. The image encoding apparatus according to the tenth embodiment
Now, instead of encoding the motion vector of the transparency signal
And encode the difference vector between the motion vectors of both signals
But the difference vector obtained by the difference value calculator 70
Output to the encoder 65b instead of the encoder 67b
Instead of encoding the motion vector of the pixel value signal
And encode the difference vector between the motion vectors of both signals
Thereby, the same effect can be obtained. In addition,
The fact that coding in block units is possible by
This is the same as Embodiment 8. Embodiment 11 FIG. Embodiment 11 of the present invention
Is an image decoding apparatus according to the eighth embodiment.
Coded signal efficiently coded by the coding device
Thus, appropriate decoding is performed. FIG.
A block diagram showing a configuration of the image decoding apparatus according to the eleventh embodiment.
FIG. In the figure, 82a and 82b are:
Codes corresponding to the encoded signals 68a and 66b in FIG.
Signal of the pixel value signal and the transmittance signal.
This is a signal obtained by encoding the difference value. 80a and 80b
Corresponds to the encoded signals 66a and 68b in FIG.
These are coded signals, each of which is a pixel value signal and a transmittance signal.
This is a signal obtained by encoding these motion vectors. 83a, and
And 83b are decoders, each of which has a pixel value signal and a transmittance.
Decode the coded signal of the difference value of each signal, and
A value signal and a decoded difference value of the transmittance signal are output. 81
a and 81b are decoders, which are pixel value signals, and
Decode the motion signal coded signal of each transparency signal
Into the pixel value signal and the decoding motion vector of the transparency signal.
Output file. 61 is a memory used as a reference image
Data such as encoded and decoded image signals
To remember. 63a and 63b are motion compensators.
Then, motion compensation is performed using the decoded motion vector. 84
Reference numerals a and 84b denote difference value adders, which use decoded difference values.
And performs an addition process. 85a and 85b are decrypted
Image signal. The eleventh embodiment configured as described above
The operation of the image decoding apparatus according to the above will be described. Real
In the image coding apparatus according to Embodiment 8, the pixel value signal
Signal that encodes the difference value of the signal
68a and 66b are input signals 82a and 82b and
And input to the image decoding apparatus according to the eleventh embodiment.
And decoded by decoders 83a and 83b,
The difference between the pixel value signal and the decoded difference value of the transmittance signal is
It is output to value adders 84a and 84b. Also implemented
In the image coding apparatus according to aspect 8, the pixel value signal;
Signal that encodes the motion vector of each
Nos. 66a and 68b are input signals 80a and 80b
As input to the image decoding apparatus according to the eleventh embodiment.
And decoded by decoders 81a and 81b,
Pixel value signal, and the decoded motion vector of the transparency signal,
It is output to the motion compensators 63a and 63b. The motion compensators 63a and 63b are respectively
The pixel value indicated by the inputted motion vector is stored in the memory 6.
1 to perform motion compensation and calculate the motion compensation value as a difference.
Output to the value adders 84a and 84b. Difference value adder
84a and 84b are input decoding difference values, respectively.
And the motion compensation value are added, and the decoded image signal 8
5a and 85b, and the memory 61
Will be stored. As described above, the image according to the eleventh embodiment is
The decoding device processes the encoded signal of the pixel value signal.
Decoder 81a, decoder 83a, motion compensator 63
a, the difference value calculator 84a, and the
Decoders 81b and 83b that perform processing, motion
With the provision of the compensator 63b and the difference value calculator 84b,
The coded signals 80a and 82a of the prime value signal and the transparency signal
Signal coded signals 80b and 82b, respectively.
It is possible to perform the decryption process separately,
It is possible to obtain an image signal by decoding. The image decoding according to the eleventh embodiment
In the image encoding apparatus according to Embodiment 8,
Although the obtained coded signal is to be decoded,
Encoded signal obtained in image encoding apparatus according to mode 9
Can be properly decoded in the same way.
is there. Also, in Embodiment 8 or 9,
For a coded signal input and coded in units,
By inputting and decoding each block,
Thus, the decoding can be appropriately performed. Embodiment 12 FIG. Embodiment 12 of the present invention
The image decoding apparatus according to the tenth embodiment
Coded signal efficiently coded by the coding device
Then, appropriate decoding is performed. FIG.
15 shows a configuration of an image decoding apparatus according to Embodiment 12 of the present invention.
It is a block diagram. In the figure, reference numeral 86 denotes a motion vector
Is a difference value adder, and a decoded motion vector and a decoded difference
Perform motion vector addition processing. Other symbols are the same as in FIG.
Since the description is the same as that of the eleventh embodiment,
It is omitted here. The twelfth embodiment configured as described above
The operation of the image decoding apparatus according to the above will be described. Return
The encoder 81a obtains by decoding the input signal 80a.
The decoded motion vector of the pixel value signal
And the motion vector difference value adder 86
Also output to The decoder 80b includes the
Coded signal of the motion vector of the transparency signal as if
Is not input, but the difference motion vector of the tenth embodiment is used.
The encoded signal 68b (FIG. 14) of the vector is input and decoded.
The decoder 80b performs decoding as in the eleventh embodiment.
Instead of obtaining the motion vector of the
Get the minute vector and move the decoded differential motion vector
It outputs to the vector difference value adder 86. Output
The signal difference motion vector is the same as the motion vector of the pixel value signal.
Since it is a difference vector from the motion vector of the excessive signal,
This difference vector is used as the pixel value in the difference value adder 86.
Addition to the decoded motion vector of the signal results in transmission
A motion vector of the degree signal is obtained. Decryption
The motion vector of the transmitted transmittance signal is calculated by the motion compensator 63b.
Is output to Other operations are described in the eleventh embodiment.
It is equivalent to the processing in the image decoding device, and the pixel value signal
And a decoded signal 85b of the transparency signal.
Output from the device. As described above, the image decoding of the twelfth embodiment
In the decoding device, the configuration of the image decoding device of the eleventh embodiment
Based on this configuration, a motion vector difference value adder 86 is added.
The decoded motion vector and the decoded difference vector
And the difference vector as an encoded signal.
Embodiment 10 Output Code of Embodiment 10 for Outputting an Encoded Signal of Torr
The decoded signal can be appropriately decoded. The actual
In Embodiment 10, coding is performed in block units.
That it is possible to deal with
This is the same as Embodiment 11. Embodiment 13 FIG. Embodiment 13 of the present invention
The image coding device based on
Shape signal indicating whether the prime value is significant, and pixel value
Signal into a block-shaped image signal
Use this as an input signal and refer to the reference image to sign this input signal.
It becomes something. FIG. 17 shows Embodiment 13 of the present invention.
1 is a block diagram illustrating a configuration of an image encoding device according to the present invention.
In the figure, 60a is a pixel value signal, 60b is a shape signal
And constitute the image signal.
Input as an input signal to the image coding apparatus of the eighth embodiment.
It is. 69a and 69b are decoders, and
Encoded signals of difference values output from 67a and 65b
Is decrypted. 75a and 75b are difference value adders.
Add the decoded difference value and the motion compensation value.
And store it in the memory 61. Figure for other symbols
The description is the same as that of the eighth embodiment.
Here, the description is omitted. The thirteenth embodiment configured as described above
The operation of the image coding apparatus according to the first embodiment will be described. Entering
The block-shaped image signal, which is a force signal, is
In the present embodiment, the prime value signal 60a and the shape signal 60b
13 image encoding devices. Where the shape signal
Is the one shown in FIG. 53 used in the description of the prior art.
And the binary information shown in FIG. 53 (c) or the binary information shown in FIG.
It becomes the multi-value information shown. If it is multi-valued information, the form
It is the same as the transmittance signal in state 8. Of this implementation
In the image coding apparatus according to the thirteenth embodiment,
By the same processing, the pixel value signal and the shape signal
Encoded signal of motion vector of pixel value signal
66a, encoded signal 68a of difference value of pixel value signal, shape
Encoded signal 66b of signal motion vector and shape signal
Is obtained. In the apparatus according to the eighth embodiment, the encoded
Signal is input to the memory 61.
In 13, the encoded difference values are supplied to the decoders 69, respectively.
a, and 69b, and the difference value adder 75a,
And 75b, and the difference value adders 75a and 75b
b, output from the motion compensators 63a and 63b.
The added motion compensation value is input to the memory 61 after being added.
You. Therefore, the reference image used for encoding is
Is decoded and the motion compensation value is added.
This is different from the eighth embodiment. As described above, the image according to the thirteenth embodiment is
In the encoding device, the image encoding device according to Embodiment 8
Based on the configuration, the decoders 69a and 69b and the difference value
A configuration in which adders 75a and 75b are added
Thus, as in the eighth embodiment, the motion compensation error is reduced.
In addition to improving coding efficiency,
Therefore, the processing load will be slightly increased.
The image is encoded and decoded, and the motion compensation value is added.
The use of a more appropriate signal
The compensation error can be reduced. Note that this implementation
Encoded signal output by image encoding apparatus according to aspect 13
Is similar to that of the eighth embodiment.
Image decoding device, it is possible to perform appropriate decoding.
Wear. Embodiment 14 FIG. Embodiment 14 of the present invention
Is an image encoding device similar to that of the thirteenth embodiment.
An image signal composed of a shape signal and a pixel value signal is defined as an input signal.
Then, the input signal is encoded with reference to a reference image.
It is. FIG. 18 is a view according to Embodiment 14 of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an image encoding device. In the figure
And the reference numerals are the same as those in FIG.
The same as in embodiment 13. Image code according to Embodiment 14
In the image forming apparatus, the pixel value signal 60a
Is detected by the motion detector 62a.
Motion detector 62 for the transmission signal 60b.
b, and the motion detector 62b outputs the input pixel value signal.
In the vicinity of the motion vector of the signal,
The point that the output is performed is different from that of the image coding apparatus according to the thirteenth embodiment.
This is a difference in configuration. In addition, the image according to the fourteenth embodiment
As for the operation of the encoding device, the motion detector 62a
And the motion detector 62b performs the above detection.
Otherwise, the operation is the same as that of the thirteenth embodiment. As described above, the image according to the fourteenth embodiment is
In the encoding device, based on the configuration of the thirteenth embodiment,
The motion detector 62a for the value signal 60a
Move the motion vector of the raw signal to the shape signal 60b
The motion detector 62b outputs to the detector 62b.
In the vicinity of the motion vector of the pixel value signal
The same as the ninth embodiment.
In the motion detection of the shape signal,
Using the results of motion detection,
The number can be reduced. The motion vector of the shape signal
Detects motion vector of pixel value signal near tor
The point which can be configured to be
And a code obtained by the image coding apparatus of the fourteenth embodiment.
The decoded signal is decoded by the image decoding apparatus according to the eleventh embodiment.
It can be encoded in the same manner as in the thirteenth embodiment. Embodiment 15 FIG. Embodiment 15 of the present invention
Image encoding apparatus according to Embodiments 13 and 14,
Similarly, an image signal composed of a shape signal and a pixel value signal is
Use this as the input signal, refer to the reference image,
Is to be encoded. FIG. 19 shows a structure according to the fifteenth embodiment.
1 is a block diagram illustrating a configuration of an image encoding device. Same figure
In FIG. 13, the motion vector difference calculator 70 is shown in FIG.
This is the same as Embodiment 10. Also, the encoder 67b
As in the tenth embodiment, the difference value calculator 70 acquires
Encode the difference vector of the motion vector. Other signs are
This is the same as FIG. 17, and the description is the same as that of the thirteenth embodiment.
It is. The image coding apparatus according to the fifteenth embodiment
Regarding the operation, the motion detectors 62a and 62b
The vector is output to the difference value calculator 70, and the difference value calculator 7
0 obtains the above difference vector and outputs it to the encoder 67b.
And the encoder 67b calculates the difference vector of the motion vector
Encoding is the same as in the tenth embodiment.
Is the same as that of the thirteenth embodiment. As described above, the image code of the fifteenth embodiment
In the encoding device, the configuration of the image encoding device of the thirteenth embodiment is
And a motion vector difference value calculator 70 is added.
To encode the motion vector of the shape signal
Instead, the motion vector of the pixel value signal and the motion of the shape signal
Encode the difference vector with the vector. Therefore, the implementation
By performing variable-length coding in the same manner as in Embodiment 10,
It is possible to further improve the coding efficiency. The motion vector of the pixel value signal is encoded.
Instead of encoding the difference vector.
This is the same as in the tenth embodiment. Ma
The image output by the image coding apparatus according to the fifteenth embodiment.
The coded signal is the same as that in the tenth embodiment.
In the image decoding apparatus according to the twelfth aspect, decoding is performed appropriately.
Can be Embodiment 16 FIG. Embodiment 16 of the present invention
Are the same as those in Embodiments 13 to 15.
Input an image signal consisting of a shape signal and a pixel value signal
This input signal is encoded by referring to the reference image as a signal.
Is what you do. FIG. 20 shows Embodiment 16 of the present invention.
FIG. 2 is a block diagram showing a configuration of an image encoding device according to the first embodiment. same
In the figure, reference numeral 90 denotes a motion detection determiner,
No. 60b and the motion of the pixel value signal output from the motion detector 62a.
The motion vector of the pixel value signal.
The motion compensation of the shape signal by the
Whether or not to perform motion detection on the shape signal motion detector 62b
Output instructions. The sixteenth embodiment configured as described above
The operation of the image coding apparatus according to the first embodiment will be described. Book
Pixel input to image encoding apparatus according to Embodiment 16
The processing of the value signal 60a is performed by the motion detector 62a.
The motion vector of the obtained pixel value signal is used as a motion detection
The processing is the same as in the thirteenth embodiment, except that the
And the motion vector coded signal 66a of the pixel value signal
And an encoded signal 68a of a difference value of the pixel value signal are obtained.
You. On the other hand, regarding the input shape signal 60b,
Input to the motion detection determiner 90. Motion detection judgment unit 9
0 is input using the motion vector of the input pixel value signal.
The motion of the force shape signal 60b is compensated, and the motion compensated
By comparing the shape signal with the input shape signal 60b,
Find out if you are. And if they match
The motion vector of the pixel value signal is supplied to the motion detector 61b.
Is output, and the motion detector 61b detects the motion of the shape signal.
Output without executing the motion vector of the input pixel value signal.
Thus, the motion vector of the shape signal is used. Conversely, motion detection
If the result of the comparison by the outgoing judgment unit 90 does not match,
The detection determiner 90 performs a motion detection on the motion detector 61b.
And outputs a motion vector by the motion detector 61b.
The calculation of torr is performed. For shape signals,
Is the same as that of the thirteenth embodiment,
b and 68b are obtained. As described above, the image according to the sixteenth embodiment is
In the encoding device, the image encoding device according to Embodiment 13
In addition, by adopting a configuration in which a motion detection determiner 90 is added,
Using the motion vector of the input pixel value signal,
When it is determined whether or not motion compensation can be performed,
In this case, do not perform motion detection for the input shape signal.
By reducing calculation, the processing load can be reduced
Can be achieved. If it is determined that it is impossible,
Similar to the thirteenth embodiment, the motion detection for the shape signal is performed.
Effect on the coding efficiency and the image quality of the coded signal.
I can't. In Embodiment 16, motion compensation is performed.
When the shape signal matches the input shape signal
Motion detection for the shape signal was not performed.
A slight decrease in coding efficiency due to an increase in motion compensation error is allowed.
If possible, the error due to motion compensation in the decision
Even if is less than a predetermined value, motion detection is not performed.
Can be set, further reducing the processing load
Becomes possible. Embodiment 17 FIG. Embodiment 17 of the present invention
Are the same as those in the thirteenth to sixteenth embodiments.
Input an image signal consisting of a shape signal and a pixel value signal
This input signal is encoded by referring to the reference image as a signal.
Is what you do. FIG. 21 shows Embodiment 17 of the present invention.
FIG. 2 is a block diagram showing a configuration of an image encoding device according to the first embodiment. same
In the figure, reference numeral 93 denotes a switching decision unit, which is a pixel value signal.
Input the motion vector of the signal and the shape signal, and
Similarly to the state 16, the motion vector of the pixel value signal is used.
Judge whether motion compensation of shape signal is possible and respond to the judgment.
Then, an instruction to the switch 94 is issued. Switch
94 is a difference value corresponding to the instruction of the switching judgment unit 93.
The output to the calculator 70 is the motion vector of the pixel value signal.
Or motion vector of shape signal
You. The motion vector memory 95 is output from the switch 94.
The input motion vector is delayed and input to the difference detector 70.
A temporary memory is used to save power. The other symbols are those in FIG.
The description is the same as that of the fifteenth embodiment. Operation of the device in the seventeenth embodiment
Also, the difference value calculator 7
Except that the difference motion vector obtained by 0 is different
Is the same as in the fifteenth embodiment.
Only the operation will be described. The switch judging device 93 outputs
Of the pixel value signal of the decoded input signal
Input coded signal of motion and motion vector of shape signal
To compare the input signal
The motion vector of the shape signal is coded.
Find out if it is. That is, the signal encoded immediately before is processed.
By processing, the differential motion obtained from the encoder 67b is obtained.
The coded signal of the motion vector is the motion vector of the pixel value signal.
And the motion vector of the shape signal
Or the difference vector between the motion vectors of the shape signals.
Find out what is. Then, the motion vectors of the shape signal
If the difference vector has been encoded, the switch
By sending an instruction to the memory 94, the
The motion vector detected from the state signal is input. Therefore
In this case, the difference value calculator 70 reads from the delay memory 95.
The motion vector of the shape signal coded immediately before
And the motion vector detected from the input shape signal.
And the encoder 67b calculates the difference vector
Is encoded. On the other hand, in the immediately preceding input signal, the pixel value
Difference between signal motion vector and shape signal motion vector
If the vector has been encoded, switch 94
By transmitting the instruction, as in the fifteenth embodiment,
The difference vector of the motion vector of the signal is encoded
Become. As described above, the image code of the seventeenth embodiment
In the encoding apparatus, the configuration of the image encoding apparatus according to Embodiment 15 is described.
Based on the result, a switching decision unit 93 and a switching unit 94
And a delay memory 95 are added.
Immediately before the motion vector of the shape signal is coded.
In that case, the motion vector and the detected motion vector
The difference vector of
By using the difference between motion vectors of different shape signals,
It is possible to improve the coding efficiency. Note that this implementation
Regarding form 17, regarding the motion vector of the shape signal
Such judgment and encoding of the difference vector are performed.
However, such a determination is made for the motion vector of the pixel value signal.
It is also possible to encode the difference vector.
Thus, the coding efficiency can be similarly improved. Embodiment 18 FIG. Embodiment 18 of the present invention
Image encoding device, the shape information and the transparency information
An image signal composed of at least one of the
Signal as an input image signal, in a mode suitable for each signal.
Is encoded. FIG. 22 shows an embodiment of the present invention.
FIG. 24 is a block diagram illustrating a configuration of an image encoding device according to aspect 18.
is there. In FIG. 1, reference numeral 101 denotes an input image signal.
Yes, at least one of shape information and transparency information,
And pixel value information. 102 is a blocker
Yes, block the input image signal 101 and block
Shape signal 103, blocked transparency signal 1
05, and a pixel value signal 107 that is blocked
You. 110 is a shape encoding mode determiner, 114 is transmittance
An encoding mode determiner 116 determines a pixel value encoding mode.
And a shape signal 103 and a transmittance signal 10 respectively.
5, and an appropriate encoding mode for the pixel value signal 107.
Is determined, the shape encoding mode 111, the transparency encoding
The mode 115 and the pixel value encoding mode 119 are output.
You. 112 is a shape encoder, 116 is a transparency encoder,
Reference numeral 120 denotes a pixel value encoder, each of which is a mode determination unit.
According to the judgment of the above, each signal is encoded and shape encoded
Signal 113, transparency encoded signal 117, pixel value encoded signal
No. 121 is output. 122 is a mode encoder,
The encoding modes 111, 115, and 119 are collectively
And outputs a mode coded signal 123. The eighteenth embodiment configured as described above
The operation of the image coding apparatus according to the first embodiment will be described. Ma
, And consists of shape information, transparency information, and pixel value information.
The input image signal 101 is an image code according to the eighteenth embodiment.
Input to the decoding device. Here, the transparency information and the shape information
Is explained with reference to FIG.
I will tell. Transparency information refers to the image shown in FIG.
When combining with the image, what ratio is used to combine each pixel
This is basically shown in FIG. 53 (d).
Is multi-valued information. The shape information is shown in FIG.
The binary information shown, where the transparency information is 0 or non-zero.
The object is "present / not present"
Information. Note that if the transparency information is completely transparent and completely
If there are only two types of transmission,
It can be expressed only by state information, and does not require transparency information. Follow
In this case, only the shape information and pixel value information are encoded.
Alternatively, decoding may be performed. The block generator 102 receives the input image signal 10
1, the image of shape information / transparency information and pixel value information
Multiple pixels are integrated and blocked based on the correspondence
And block signal 103, block
Transparency signal 105 and blocked pixels
The value signal 107 is output. Shape signal 103 is shape coded
The mode determiner 110 and shape encoder 112
The signal 105 is transmitted by the transparency coding mode determiner 114 and the transparency.
To the encoder 116 and the pixel value signal 107
Prime value encoding mode determiner 118 and pixel value encoder 120
Is output to Shape coding mode determiner 110, transparency code
Coding mode determiner 114 and pixel value coding mode determination
The detector 118 receives the input shape signal 103,
For the degree signal 105 and the pixel value signal 107,
The encoding mode is determined, and the shape encoding mode 111
Excessive encoding mode 115 and pixel value encoding mode 11
9 is output. Each encoding mode is output to each encoder.
At the same time, it is also output to the mode encoder 122.
Shape encoder 112, transparency encoder 116, and pixel
The value encoder 120 outputs the input encoding mode.
Outputs the input signals corresponding to the
Encoding signal 113, transparency encoding signal 117, and pixel value
An encoded signal 121 is output. On the other hand, mode encoder 1
22: Encodes each input encoding mode collectively
Then, the mode encoding signal 123 is output. Shape coded signal
No. 113, the transmittance coded signal 117, the pixel value coded signal
121 and the mode coded signal 123
18 is the encoded output of the image encoding device. As described above, the image according to the eighteenth embodiment is
In the encoding device, the input image signal is divided into blocks and
Signal, transmittance signal, and pixel value signal.
And a coding mode suitable for each signal
Encoding mode determiners 110, 114, and 1
18 and a code for coding each signal according to the coding mode.
Encoders 112, 116, and 120 and the encoding mode
And a mode encoder 122 for collectively encoding.
Performs encoding in a mode suitable for each separated signal.
Information about the selected mode is collectively
Can be realized. Shape information, transparency information, and
Pixel value information is often correlated with each other,
Therefore, the same encoding mode is easily selected. There
In the variable mode, the code in the same mode has a shorter code length
By performing coding, the number of bits of the mode-coded signal can be reduced.
The effect of being able to reduce is obtained. Embodiment 19 FIG. Embodiment 19 of the Invention
The image coding apparatus according to
State information and / or transparency information, and pixel value information.
The image signal composed of the
The encoding is performed in a mode suitable for the signal. Figure
23 is an image coding apparatus according to Embodiment 19 of the present invention.
FIG. 3 is a block diagram showing the configuration of FIG. In FIG.
The encoding mode determiner 110 is adapted to the shape signal 103
The encoding mode is determined, and the determination result is referred to as the encoding mode.
To the shape encoder 112 and the mode encoder 122
Output, and the transparency encoding mode determiner 130,
And the pixel value encoding mode determiner 132. So
Then, the transparency encoding mode determination unit 130
A determination is made with reference to the shape encoding mode 111, and the determination is made.
The transmission result is used as the encoding mode with the transparency encoder 116,
Output to the mode encoder 122 and the pixel value code.
It is also output to the encryption mode determiner 132. Pixel value encoding mode
The mode determiner 132 receives the input shape encoding mode 11
1 and the transparency encoding mode 115 to make a determination.
U. Operation of image encoding apparatus according to Embodiment 19
Are the same as those of the first embodiment except for the judgments of the respective mode judgment units.
8, and the shape encoded signal 113, the transmission
Degree encoded signal 117, pixel value encoded signal 121, and
The code encoded signal 123 is output. As described above, the image according to the nineteenth embodiment is obtained.
In the encoding device, referring to the shape encoding mode,
Transparency coding mode determination to determine signal coding mode
130, a shape encoding mode, and a transparency encoding mode.
The encoding mode of the pixel value signal with reference to both
With the provision of the pixel value encoding mode determiner 132
The selected mode is likely to be the same. Follow
To assign a shorter code if the modes match.
In code encoder 122, it is more possible than in Embodiment 18.
The efficiency of variable-length coding has been
The effect that the number can be reduced is obtained. Embodiment 20 FIG. Embodiment 20 of the present invention
The image coding apparatus according to
It is intended to improve the coding efficiency of a code-coded signal. Figure
24 is an image coding apparatus according to Embodiment 20 of the present invention.
FIG. 3 is a block diagram showing the configuration of FIG. In FIG.
The encoding mode determiner 118 adapts to the pixel value signal 107.
Of the selected encoding mode, and the result of the
Pixel value encoder 120 and mode encoder 1
22 and a transparency encoding mode decision unit 1
36, and also output to the shape encoding mode decision unit 138.
You. Then, the transparency encoding mode determiner 136 receives the input
A determination is made with reference to the pixel value encoding mode 119 that has been set.
The result of the decision is used as the encoding mode as a transparency encoder.
116 and the mode encoder 122
, And also output to the shape encoding mode determiner 138. shape
The encoding mode determiner 138 determines whether or not the input pixel value has been encoded.
Mode 119 and the transparency encoding mode 115.
To make a judgment. Operation of the image coding apparatus according to the twentieth embodiment
The actual operation is the same except for the judgment of each mode judgment unit.
This is the same as the eighteenth embodiment.
13, the transmittance coded signal 117, the pixel value coded signal 12
1 and the mode coded signal 123 are output. Thus, the image according to the twentieth embodiment is
In the encoding device, referring to the pixel value encoding mode,
Transparency encoding mode judgment to determine the encoding mode of the degree signal
136, a pixel encoding mode, and a transparency encoding mode.
The encoding mode of the shape signal with reference to both
With the shape encoding mode determiner 138,
The selected modes are likely to be the same. Therefore,
Mode code that assigns a short code if the modes match
In the encoder 122, the variable length is further
The coding efficiency is improved and the number of bits of the mode
The effect that reduction can be obtained is obtained. Embodiment 21 FIG. Embodiment 21 of the present invention
The image decoding apparatus according to the
Coded signal efficiently coded by the coding device
Then, appropriate decoding is performed. FIG. 25 shows the present invention.
Showing the configuration of an image decoding apparatus according to Embodiment 21 of the present invention.
It is a lock figure. In the figure, input signals 113, 11
7, 119 and 123 are images according to the eighteenth embodiment.
Shape encoded signal 113 output from the encoding device,
The transparency coded signal 117, the pixel value coded signal 119, and
And the mode encoded signal 123. 150 is mode decoding
And decodes the mode coded signal 123 to form
Coding mode 151, transparency coding mode 153, and
And the pixel value encoding mode 155 is output. 156 is the shape
A decoder 158 is a transparency decoder, and 160 is a pixel value decoder.
And input from the mode decoder 150, respectively.
Depending on the encoding mode applied, the shape encoded signal 11
3, the transparency encoded signal 117, and the pixel value encoded signal 1
19, shape-decoded signal 157, transparency decoding
The signal 159 and the decoded pixel value signal 161 are output.
Numeral 162 denotes an inverse blocker which outputs the shape decoded signal 15
7, the decoded transparency signal 159, and the decoded pixel value signal 1
61, these are integrated, and the decoded image signal 163 is
Output. The twenty-first embodiment configured as described above
The operation of the image decoding apparatus according to the above will be described. Book
The image decoding apparatus according to the twenty-first embodiment has
No. 113, the transmittance coded signal 117, the pixel value coded signal
119 and the mode coded signal 123 are input,
The shape decoder 156, the transparency decoder 158,
Input to prime value decoder 160 and mode decoder 150
Is done. The mode decoder 150 outputs the mode encoded signal 1
23, the shape encoding mode 151, the transparency code
Encoding mode 153 and pixel value encoding mode 155,
Each of the shape decoder 156, the transparency decoder 158,
And to the pixel value decoder 160. Shape decoder 1
56, transparency decoder 158 and pixel value decoder 16
0 corresponds to the input coding mode.
The input coded signal is decoded, and the shape decoded signal 15 is decoded.
7, the decoded transparency signal 159, and the decoded pixel value signal 1
61 is output to the deblocker 162. Reverse blocking
The unit 162 integrates the input decoded signal and generates a decoded image.
The signal 163 is output. As described above, the image according to the twenty-first embodiment is
In the decoding device, a mode decoder 150 and shape decoding
156, transparency decoder 158, pixel value decoder 16
0 and the provision of the deblocker 162,
To the encoded signal obtained by the image encoding apparatus according to aspect 18.
On the other hand, the decoded image signal is appropriately decoded and integrated.
163 can be obtained. In addition, Embodiment 2
In the first image decoding apparatus, the image code according to the eighteenth embodiment is used.
The encoded signal obtained by the encoding device shall be decoded.
Is the image encoding apparatus according to Embodiments 19 and 20,
Similarly, appropriate decoding is performed on the obtained encoded signal.
It is possible to do. Embodiment 22 FIG. Twenty-second embodiment of the present invention
Image encoding device inputs intra / inter-frame encoding
The coding is performed by switching according to the signal. Figure
26 is an image coding apparatus according to Embodiment 22 of the present invention.
FIG. 3 is a block diagram showing the configuration of FIG. In the figure, 178
Is an intra-screen / inter-screen encoding determiner for pixel value encoding,
Regarding the encoding mode of the pixel value signal,
The determination between the surfaces is performed, and the encoding mode 119 of the pixel value signal is set.
Output. Reference numeral 138 denotes a shape encoding mode determiner.
24, shape encoding in embodiment 20 shown in 138
It corresponds to a mode determiner. 170 and 176 are switches.
Switch according to the output of the decision unit 178.
Thus, the coding mode of the shape signal is determined. 172 is the shape
It is an in-screen / inter-screen coding determiner for coding, and a shape signal
The intra-screen or inter-screen for the encoding mode of
Is performed, and the coding mode 17 of the shape signal is determined based on the result of the determination.
3 is output. Other symbols are the same as in FIG.
The description is the same as that of the eighteenth embodiment. The twenty-second embodiment constructed as described above
The operation of the image coding apparatus according to the first embodiment will be described. Ma
First, the input image signal 101 corresponds to the image code of the twenty-second embodiment.
When input to the decoding device, the blocker 102
Blocking and signal separation are performed as in Embodiment 18.
A pixel value signal and a shape signal are output. Separated pixel value signal
No. 107 is an intra-screen / inter-screen coding determiner for pixel value coding
When input to the pixel value signal 178, the decision unit 178
7 for either intra-screen coding or inter-screen coding
It is determined whether encoding should be performed based on the
Pixel value code indicating either “in-plane” or “between screens”
As the encoding mode 119, the pixel value encoder 120 and the mode
Encoder 122 and shape encoding mode determiner 138
Output to The shape encoding mode determiner 138
Switches 170 and 176 are in the pixel value encoding mode 119.
Can be switched accordingly. If the pixel value encoding mode is "Screen
If "inside" is indicated, it should not be input to the decision unit 178.
In the case where “between screens” is indicated, the
Can be switched. Therefore, the pixel value encoding mode
When 119 indicates intra-frame encoding, the decision unit 138 outputs
A shape determination mode 111 indicating intra-screen encoding is output
It will be. On the other hand, the pixel value encoding mode 119 is
In the case of indicating the encoding, the corresponding
Whether the shape signal is coded within the screen or between screens
Intra / inter-screen encoding determiner 172 as to whether encoding should be performed by
Is determined, and the determination result is set as the shape encoding mode 111.
Output. In any case, shape encoding mode 1
11 is a shape encoder 112 and a mode encoder 122
Is output to Then, the pixel value encoder 120, the shape
The operations of the encoder 112 and the mode encoder 122 are actual.
As in the eighteenth embodiment, each encoded signal is output
You. Since the above operation is performed, the present embodiment
22, the pixel value signal is converted to an intra-screen code.
When encoded, the shape signal is always encoded in the screen.
The Rukoto. Generally, when pixel values do not match,
Since the states do not match, the pixel value signal is encoded in the screen
Power, that is, when the correlation is small in time,
Is the encoding of the shape signal encoding as in the twenty-second embodiment.
Even if the number of modes is limited, the coding efficiency of shape signal coding
Hardly deteriorates. Also, as in the case of an inset image (composite image),
Pixel value to be synthesized even if shape signal during synthesis is constant
May change. In such a case, the pixel value signal
Therefore, even if inter-picture coding is selected, the shape signal
Inter-picture coding is not always appropriate. This embodiment
22, the inter-picture coding is selected for the pixel value signal.
If selected, the shape signal is within / between screens
The shape signal
Encoding can also select the encoding mode of intra-screen encoding.
Yes, inappropriate inter-screen coding in shape signal coding
Will greatly reduce the coding efficiency.
Can be prevented. Further, when the shape signal or the pixel value signal is small,
If at least one is inter-coded,
Many necessary for motion compensation etc. performed in encoding
Additional information is required. Image according to the twenty-second embodiment
In the encoding device, only the shape signal is inter-coded.
In-screen encoding is selected for pixel value signal
In this case, the number of bits can be saved. General
For the additional information, the pixel value signal is
Less than the number of bits when
Cannot be ignored compared to the number of bits required for intra-plane coding
Since the number of bits is of the order, the effect is also great. This
As described above, according to the image coding apparatus of the twenty-second embodiment, the
An intra-screen / inter-screen encoding determiner 178 for prime value encoding;
Includes in-screen / inter-screen coding determiner 172 for shape coding
By providing the shape encoding mode determiner 138
Shape when the encoding mode of the raw signal is intra-screen encoding
The encoding mode of the signal is set to intra-screen encoding, and the pixel value signal
When the encoding mode is inter-screen encoding, the shape signal encoding
Since the decision is made by making a decision about the
Value signal encoding mode 119 and shape signal encoding mode
111 to increase the number of bits of the mode-coded signal.
Reduce the selection and suppress the selection to perform inter-screen coding
The number of bits of additional information for motion compensation
Can also be suppressed. Embodiment 23 FIG. Embodiment 23 of the present invention
Is a motion vector in encoding.
That performs encoding by switching the number of
It is. FIG. 27 shows an image according to Embodiment 23 of the present invention.
FIG. 2 is a block diagram illustrating a configuration of an encoding device. Smell
188 is a motion vector number determiner for pixel value encoding.
Motion vector in the pixel value signal encoding mode.
The number of pixels is determined, and the sign of the pixel value signal is determined.
The conversion mode 119 is output. 138 is a shape encoding mode
24 is a determinator, which is different from the twentieth embodiment shown at 138 in FIG.
It corresponds to a shape encoding mode determiner in the present embodiment. 180, and
And 186 are switches which respond to the output of the decision unit 188.
To determine the coding mode of the shape signal.
You. Reference numeral 182 denotes a motion vector number determiner for shape encoding.
The motion vector for the shape signal encoding mode
Judge how many numbers to make
The signal encoding mode 183 is output. About other signs
The description is the same as that of FIG.
It is like. The twenty-third embodiment constructed as described above
The operation of the image coding apparatus according to the first embodiment will be described. Ma
First, the input image signal 101 corresponds to the image code of the twenty-third embodiment.
When input to the decoding device, the blocker 102
Blocking and signal separation are performed as in Embodiment 18.
A pixel value signal and a shape signal are output. Separated pixel value signal
Reference numeral 107 denotes a motion vector number determiner 188 for pixel value encoding.
Are input to the pixel value signal 107,
To determine how many motion vectors to encode.
And the result of the determination is referred to as a pixel value encoding mode 119.
Then, the pixel value encoder 120 and the mode encoder 12
2 and output to the shape encoding mode determiner 138. FIG. 28 illustrates the number of motion vectors.
FIG. The motion is complicated near the contour of the object.
Using one motion vector (MV) per block
Then, it is difficult to make the motion compensation error sufficiently small.
In such a case, the block is divided and
It is desirable to assign a motion vector to
The number of motion vectors adaptively according to the nature of the image
It is known that the coding efficiency can be improved by changing
You. Therefore, in the image coding apparatus according to Embodiment 23,
As shown, one motion vector (MV
1) Use motion compensation or divide the block into four
One for each divided block, for a total of four motion vectors
(MV1, MV2, MV3, MV4)
Shall be switched accordingly. Therefore, in the decision unit 188
Determines whether the number of motion vectors is one or four.
And the encoding mode of the pixel value signal is “1” or
"4" is output as above. In the shape encoding mode determination unit 138, the switch
Switches 180 and 186 switch to the pixel value encoding mode 119.
Can be switched accordingly. Pixel value encoding mode is "1"
In such a case, in order not to be input to the determiner 182,
When "4" is indicated, it is input to the decision unit 182.
Can be switched. Therefore, the pixel value encoding mode 119 is
In the case of “1”, the number of motion vectors is
Then, the shape determination mode 111 of “1” which is the minimum number is
Will be output. On the other hand, pixel value encoding mode 11
When 9 indicates “4”, corresponding to the shape signal 105,
It is determined whether the number of motion vectors is one or four.
And the result of the determination is
It is output as the encryption mode 111. In any case
Also, the shape encoding mode 111
And the mode encoder 122. And the picture
Prime value encoder 120, shape encoder 112, and mode
The operation of the encoder 122 is the same as in Embodiment 18, and
Each encoded signal is output. Since the above operation is performed, the present embodiment
23, the number of pixel value signals is the minimum number.
When encoding using the motion vector of
Signal is always coded using the minimum number of motion vectors
It will be. Increasing the number of motion vectors increases the motion vector
The additional information needed to encode the
In such a case, the shape signal
Increased burden by suppressing the number of motion vectors for encoding
It is to prevent large. Thus, the second embodiment
According to the image coding apparatus of FIG.
And a motion vector count for shape encoding.
A shape encoding mode determination unit 138 including a constant unit 182;
The minimum number of pixel value signal encoding modes
When using the motion vector of
Code mode and the number of encoding modes of the pixel value signal
Is used, the shape mode encoding mode
Is selected by making a judgment about the pixel value signal.
Between the encoding mode 119 and the encoding mode 111 of the shape signal.
Increase the correlation to reduce the number of bits in the mode-coded signal
And increase the number of motion vectors
Suppressing selection increases additional information
Can be suppressed from increasing. Embodiment 24 FIG. Embodiment 24 of the present invention
The image coding apparatus according to
The encoding is performed by switching the modification according to the input signal.
is there. FIG. 29 is a diagram showing an image code according to Embodiment 24 of the present invention.
It is a block diagram which shows the structure of an encoding device. Smell
198 is a quantization step change / non-change of the pixel value coding.
And determines the encoding mode of the pixel value signal.
Whether to change the quantization step or not
Judgment is performed, and the encoding mode 119 of the pixel value signal is output.
You. Reference numeral 138 denotes a shape encoding mode determiner, which is shown in FIG.
138 is the shape encoding mode in the twentieth embodiment.
It corresponds to a determiner. 190 and 196 are switches
Therefore, it is switched according to the output of the determiner 198,
Determine the encoding mode of the signal. 192 is the shape encoding
It is a quantizer step change / non-change determiner that determines the shape signal.
Change the quantization step for the encoding mode
Or not, and determine
The signal mode 193 is output. Other signs
22 is the same as that in FIG.
The same is true. The twenty-fourth embodiment configured as described above
The operation of the image coding apparatus according to the first embodiment will be described. Ma
First, the input image signal 101 corresponds to the image code of the twenty-fourth embodiment.
When input to the decoding device, the blocker 102
Blocking and signal separation are performed as in Embodiment 18.
A pixel value signal and a shape signal are output. Separated pixel value signal
No. 107 changes / non-changes the quantization step of pixel value coding
When input to the determiner 198, the determiner 198 outputs the pixel value signal.
Change the quantization step for signal 107, or
Is not performed, and the result of the determination is
Or pixel value encoding mode indicating either "no change"
119, the pixel value encoder 120 and the mode encoding
To the shaper 122 and the shape coding mode determiner 138
You. In the shape encoding mode determination unit 138, the switch
Switches 190 and 196 are set to the pixel value encoding mode 119.
Can be switched accordingly. If the pixel value encoding mode is
If "update" is indicated, the input is not input to the decision unit 192.
In the case where "change" is indicated, it is input to the decision unit 192.
Can be switched. Therefore, pixel value encoding mode 1
If 19 indicates that the quantization step should not be changed,
The shape determination mode 11 indicating “unchanged” from the integrator 138
1 will be output. On the other hand, pixel value encoding mode
If 119 indicates to change the quantization step,
The quantization step is changed according to the shape signal 105.
The determination unit 192 determines whether or not to perform the determination.
The fixed result is output as the shape encoding mode 111. In each case, the shape encoding mode 11
1 is a shape encoder 112, a mode encoder 122,
Is output to Then, the pixel value encoder 120, the shape code
The operations of the encoder 112 and the mode encoder 122 are implemented.
In this case, each encoded signal is output.
The value of the quantization step is the degree of compression,
Is generally linked to the transmission rate of the
Transmission rate or recording rate of an encoded signal
The transmission rate must be higher than the specified value so that the
If it is large, the quantization step is coarse, and if it is small, the amount
Control for increasing the densification step is performed. Also, quantization
The step value also directly affects the quality of the encoded signal.
Therefore, if the image is such that the pixel value changes sharply,
In this case, image quality degradation in the amplitude direction is difficult to detect visually.
From the above, increase the quantization step to increase the compression ratio
It becomes possible. Quantization according to such a change in pixel value
Changing the steps is also commonly done. Control for Changing the Quantization Step
Indicates that the quantization step has changed.
Additional information is added for each block, and the image data and
Are also encoded. However, changes in the quantization step
For further changes, the pixel value signal and the shape signal are changed simultaneously.
Of the pixel value signal.
If the step is not changed, the quantization step of the shape signal is
Even if restrictions are added so as not to change the
Image quality degradation is slight, while the quantization step
Can be greatly reduced. As described above, the image code of the twenty-fourth embodiment
According to the quantization device, the quantization step of the pixel value encoding is changed /
A non-change decision unit 198 and a quantization step change
Shape encoding mode determination including a change / non-change determination unit 192
And the encoding mode of the pixel value signal.
If the quantization mode is "No change of quantization step",
Set the encoding mode to "No quantization step change", and
When the coding mode of the signal is "change quantization step"
Determines the encoding mode of the shape signal and selects
The encoding mode 119 of the pixel value signal and the shape signal.
Mode coding by increasing the correlation with the coding mode 111 of the signal
It is possible to reduce the number of bits of the signal
Suppress the choice that the quantization step is changed
As a result, additional information can be increased by changing the quantization step.
It is also possible to suppress the number of bits and reduce the number of bits. Note that the image coding of the twenty-second to twenty-fourth embodiments
The device is in accordance with the twentieth embodiment shown in FIG.
Although the configuration is different from that of the nineteenth embodiment shown in FIG.
It is also possible to adopt a configuration according to
Increasing mode correlation and suppressing increase in additional information
By doing so, the number of bits can be reduced. Also,
A configuration according to the eighteenth embodiment shown in FIG.
It is possible to perform encoding suitable for each signal and
Can be reduced. Embodiment 22 to
The coded signal obtained by the image coding device of FIG.
Appropriate decoding can be performed in the image decoding apparatus according to aspect 21.
It is possible. In Embodiments 18 to 21, the input image
The signal is derived from transparency information and shape information in addition to pixel value information.
The pixel value signal, the transmittance signal, and the shape
Although the separation into the state signals is performed, the second embodiment
In 2 to 24, the input image signal is converted into a pixel value signal and a shape signal.
It shall be separated. About this, the embodiment
Even in 22 to 24, the transparency information and the shape information match.
In this case, it is possible to use only the shape information,
On the other hand, if they do not match, the setting of the blocker
Use transparency information as a shape signal or multi-valued signal
Dealing with certain transparency information together with pixel value information,
In this way, the shape signal and the pixel value signal can be obtained.
You. Embodiment 25 FIG. Embodiment 25 of the present invention
Is a two-dimensional image composed of a plurality of pixels.
Inputs image signals and predicts and detects changed pixels.
You. FIG. 30 is a diagram showing image coding according to the twenty-fifth embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of the device. In the figure, 2
01 is an input signal, which is an image code as a binary image signal.
Input to the gasifier. 204c is a first change pixel detector
And the pixel value changes with respect to the input signal 201.
Pixel as the first detected change pixel 205c.
Power. 202a and 202b are memories,
Delayed by temporarily storing the signal, the reference signal 20
Output as 3a and 203b. 204a and 2
Reference numeral 04b denotes a change pixel detector, which includes a reference signal 203a and a reference signal 203a.
And the pixel whose pixel value changes are detected for
And the second and third detection change pixels 203a and 203a
Output as 3b. 206 is a change pixel predictor;
Based on the detected change pixels 203a and 203b, the first
The changed pixel output from the changed pixel detector 204c
Then, a predicted change pixel 207 is output. 208 is a subtractor
Yes, the first changed pixel 205c and the predicted changed pixel 207
Is obtained, the difference is calculated as the prediction error 20
9 is output. 210 is an encoder, which has a prediction error
209, and outputs an encoded signal 211. In the twenty-fifth embodiment configured as described above,
The operation of the image coding apparatus according to the first embodiment will be described. FIG.
Reference numeral 1 denotes encoding of the image by the image encoding apparatus according to Embodiment 25.
It is a figure for explaining the principle of operation. Here is a brief explanation
For the sake of simplicity, it is assumed that each pixel is processed sequentially.
The management procedure will be described. In FIG. 31, from the upper left pixel to the right
Scanning is performed in the lower right direction.
I do. The pixel value of each pixel has a binary value.
It represents a true / false value (binary). Also, here
Encoding is completed in the first and second lines, and three lines
Encoding the eye (the line where the first change pixel is)
And [0136] In the above-described scanning, the changed pixel is not
The first pixel whose prime value changes, which is a coded line
The changed pixel on the (scanning line) is a second changed pixel and a third changed pixel.
And the first changed pixel on the uncoded scan line
Is the first change pixel. Therefore, the second change pixel and the second change pixel
And the first changed pixel is predicted from the three changed pixels.
Difference value between the calculated first changed pixel and the actual first changed pixel
If (prediction error) is calculated, the prediction error is concentrated near 0
Fewer bits using variable length coding etc. because of distribution
It is possible to perform efficient encoding by using numbers. In FIG. 30, first, the input signal 201 is
Input to the device. Normal for image input signal 201
Color signal (pixel value signal), object shape or object
It can be a shape signal representing the ratio of synthesis. Input signal
No. 201 is input to the memories 202a and 202b.
Is temporarily stored. On the other hand, the input signal 201 is the first change image.
The change pixel detector 204 is also input to the elementary detector 204c.
c detects a pixel whose binary pixel value changes. This is
It is a first change pixel of FIG. In FIG. 30, the first
Are input to the subtractor 208. On the other hand, the memory 202a stores the temporarily stored input data.
The signal 201 is delayed by two lines to obtain a reference signal 203a.
And outputs it to the changed pixel detector 204a.
204a detects the second change pixel 205a in FIG.
You. Similarly, the memory 202b stores the temporarily stored input signal.
201 is delayed by one line as a reference signal 203b.
The output to the changed pixel detector 204b is output to the changed pixel detector 20b.
4b detects the third change pixel in FIG. In FIG.
In this case, the changed pixels 205a and 205b are
Input to the measuring instrument 207. [0139] Generally, the image is displayed in the horizontal and vertical directions.
And the first and second changed pixels are substantially straight lines.
Often lined up. The change pixel predictor 206
Prediction based on the input changed pixels based on
The predicted change pixel 207 is output to the subtractor 208. Decrease
The arithmetic unit 208 and the input first change pixel 205c are
By acquiring the difference from the measured pixel 207,
The difference is output to the encoder 210 as a prediction error 209,
Encoder 210 encodes prediction error 209 and encodes
The signal 211 is output. The predicted change pixel and the first detected
The prediction error, which is the difference value with the changed pixel, is concentrated near 0.
If this is coded, it will be closer to 0.
Use variable-length coding that assigns codes with a small number of bits
Can be efficiently encoded with a small number of bits
become. As described above, the image coding apparatus according to the twenty-fifth embodiment is described.
The memory 202a-b and the change pixel detector 204
a to c, a change pixel predictor 207, a subtractor 208,
By providing the encoder 210, the input signal can be delayed.
Based on the changed pixel detected from the reference signal
Predict the changing pixels of the signal and note the error in this prediction.
Encoding to improve coding efficiency.
It becomes possible. Embodiment 26 FIG. Embodiment 26 of the present invention
Is a two-dimensional image composed of a plurality of pixels.
Inputs image signals and predicts and detects changed pixels.
Embodiment 25 is different from Embodiment 25 in how to acquire a changed pixel used for prediction.
The law is different. FIG. 32 shows Embodiment 2 of the present invention.
6 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 6.
You. In the figure, reference numeral 201 denotes an input signal, which is a binary image.
The image signal is input to the image encoding device. 204 is strange
A pixel detector for the input signal 201,
Pixels whose prime value changes are detected and detected as detected change pixels 205.
Output. 216a and 216b are memories,
Delay by temporarily storing the input changed pixels
You. The memory 216a delays the detection change pixel 205
The reference change pixel 217a is stored in the memory 216b by the reference signal 2
17a is delayed and the reference change pixel 217b is output.
Reference numeral 206 denotes a change pixel predictor, which is a reference change pixel 217.
a and 217b to predict a changed pixel,
The pixel 207 is output. Subtractor 208 and encoder 21
0 is the same as in the twenty-fifth embodiment. In the twenty-sixth embodiment configured as described above,
The operation of the image coding apparatus according to the first embodiment will be described. Implementation
An input signal 201 similar to that of the twenty-fifth embodiment is
Is input to the image encoding device of
Therefore, a pixel whose binary pixel value changes is detected,
The changed pixel 205 is output to the memory 216a and the subtractor 208.
Is forced. Detected change pixel 2 input to memory 216a
05 is a reference change pixel 217a after a delay of one line.
To the change pixel predictor 206 and the memory 216b.
Is forced. Reference change pixel 2 input to memory 216b
17a is a reference change pixel 2 after a delay of one line
17b is output to the change pixel predictor 206. reference
The changing pixels 217a and 217b are described in Embodiment 25.
By treating them as the second and third changed pixels,
Pixel prediction unit 206 performs the same prediction as in the twenty-fifth embodiment.
And a predicted change pixel 207 is obtained.
Subsequent processing is the same as in the twenty-fifth embodiment. As described above, the image code of the twenty-sixth embodiment
In the conversion device, the memories 216a and 216b and the changed pixel detector 2
04, a change pixel predictor 207, a subtractor 208,
With the provision of the encoder 210, detection is performed from the input signal.
Reference change image by delaying change pixel in memory
To obtain the input signal based on the reference change pixel.
The change pixel of the signal and sign the error for this prediction
In the same manner as in Embodiment 25,
It is possible to improve the conversion efficiency. Embodiment 27 FIG. Embodiment 27 of the present invention
Is a two-dimensional image composed of a plurality of pixels.
Inputs image signals and predicts and detects changed pixels.
Embodiment 25 is different from Embodiment 25 in how to acquire a changed pixel used for prediction.
The law is different. Image coding according to the twenty-seventh embodiment
The device has the same configuration as the device according to the twenty-fifth embodiment.
FIG. 30 is used for the description. Image according to Embodiment 25
In the image coding apparatus, as described with reference to FIG.
Of one line and two lines immediately before
The change pixel in the scan line is used for prediction,
In the image coding apparatus according to the twenty-seventh embodiment,
The prediction is performed based on the changed pixels in the scanning line.
You. FIG. 33 shows an image coding apparatus according to the twenty-seventh embodiment.
FIG. 7 is a diagram for explaining the principle of the encoding operation by the unit.
For the scan line located at the bottom to perform encoding, 7 lines
And the second detected in the scan line four lines before,
And the first changed pixel is on a straight line based on
Is predicted to exist, the predicted change pixel shown in FIG.
can get. From the predicted change pixel and the input signal,
Using the prediction error with the first changed pixel detected in the line
Information "one pixel to the right of the predicted changed pixel"
By encoding, encoding is performed in the same manner as in Embodiment 25.
Efficiency can be improved. Operation of image coding apparatus according to the twenty-seventh embodiment
Is temporarily stored in the memories 202a and 202b.
Embodiment 25 is the same as Embodiment 25 except that
It will be similar. In addition, the change pixel predictor 206
Then, a changed pixel can be predicted by calculation as follows.
The second changed pixel is the x-th pixel on the m-th line, and the third changed image
To the y-th pixel of the n-th line and the predicted point of the first changed pixel
Assume that the 3rd point is aligned on the straight line with the z-th pixel on the k-th line
Then, the relation x-y: z-y = m-n: k-n holds
Therefore, z−y = (x−y) * (k−n) / (m−n). Therefore, z = y
-(x-y) * (n-k) / (m-n), the first changed pixel is the k-th
The y- (x-y) * (n-k) / (m-n) pixel of the pixel. like this
In the image coding apparatus according to Embodiment 27,
The configuration is the same as that of the image coding apparatus according to the twenty-fifth embodiment.
Delay using the memories 202a and 202b
The same effect can be obtained by changing the time. Embodiment 28 FIG. Embodiment 28 of the Invention
Is a two-dimensional image composed of a plurality of pixels.
Inputs image signals and predicts and detects changed pixels.
Embodiment 25 is different from Embodiment 25 in how to acquire a changed pixel used for prediction.
The law is different. Image coding according to the twenty-eighth embodiment
The device is the same as the device according to the twenty-fifth embodiment shown in FIG.
Configuration for decoding the encoded signal 211
And the encoded and decoded output of the decoder
Only one signal is output to one of the memories. Real
The image coding apparatus according to Embodiment 25 uses FIG.
As described above, one line corresponds to the scan line to be encoded.
Pixel in a scanning line located on two lines
Is used for prediction, but the image of Embodiment 28
In the encoding device, the encoded and decoded
A changed pixel on a scanning line is used for prediction. FIG. 34 shows an image coding apparatus according to the twenty-eighth embodiment.
FIG. 7 is a diagram for explaining the principle of the encoding operation by the unit.
A scan line to be encoded (in the figure, a first change pixel exists)
4 lines above and 3 lines below
And third changed pixels detected in the scan line
It is assumed that the first changed pixel exists on a straight line based on
Then, the predicted change pixel shown in the figure is obtained. This prediction
Changed pixel and detected in the scan line from the input signal
Using the prediction error with the first changed pixel,
By encoding the information “two pixels to the right of the
Thus, as in the case of Embodiments 25 and 27, the coding efficiency is improved.
I can go up. Operation of Image Coding Apparatus in Embodiment 28
Is temporarily stored in the memories 202a and 202b.
The delay time differs depending on the memory, and the encoder 210 outputs
The encoded signal 211 to be decoded is decoded
And the corresponding change pixel detector
Only the point of detecting a changed pixel from the
Different from the form 25. Thus, according to the twenty-eighth embodiment,
An image coding apparatus according to the twenty-fifth embodiment
Path for decoding a coded signal to a reference image
With the added configuration, a similar effect can be obtained. Embodiment 29 FIG. Embodiment 29 of the Present Invention
Is a two-dimensional image composed of a plurality of pixels.
Inputs image signals and predicts and detects changed pixels.
The changed pixel detected in the same manner as in Embodiment 26 is delayed.
Is used for prediction. FIG. 35 shows an embodiment of the present invention.
FIG. 28 is a block diagram showing the configuration of an image encoding device according to aspect 29.
It is. In the figure, 216 and 220 are memories
Yes, it is delayed by temporarily storing the input changed pixels.
Extend. The memory 216 delays the detected change pixel 205.
The delay change pixel 217 is stored in the memory 220 and the prediction error 2
09 is delayed and a delay prediction error 221 is output. 22
2, and 224 are adders, and the adder 222 is a delay unit.
The change pixel 217 and the delay prediction error 221 are added to an adder 224.
Adds the prediction error 209 and the delay prediction error 221
You. Other symbols are the same as those in FIG.
Same as 26. In the twenty-ninth embodiment configured as described above,
The operation of the image encoding apparatus according to the first embodiment will be described. Figure
Numeral 36 denotes encoding by the image encoding apparatus according to the twenty-ninth embodiment.
FIG. 4 is a diagram for explaining the principle of the operation of FIG. Embodiment 26
Now, by delaying the detected first changed pixel, the second,
And a third changed pixel. Against this
In the image coding apparatus according to the twenty-ninth embodiment, the second
The difference between the changed pixel and the third changed pixel is defined as the third changed pixel.
To be added and used as the predicted value of the first changed pixel
It is. As shown, the second change pixel and the third change pixel
The difference from the pixel is “two pixels to the left” and the third changed pixel
By adding the “two pixels to the left”, the encoding is
The scanning line to be performed (in the figure, the line where the first changed pixel exists)
In), the predicted changed pixel in FIG. on the other hand
The first change pixel is detected in the scan line to be encoded.
And the detected first changed pixel and the prediction
By encoding "one pixel to the left" which is the difference from the change pixel,
The same effect as in the twenty-sixth embodiment can be realized. In FIG. 35, the input signal 201 is
Input to the image coding apparatus according to aspect 29,
204 detects the position where the binary pixel value changes
The first changed pixel 205 is subtracted from the memory 216.
And output to the device 208. In the memory 216, one lane
The delay change pixel 217 delayed by IN is shown in FIG.
This is the third change pixel. The delay change pixel 217 is an adder
222, and the second and third
Addition to delay prediction error 221 corresponding to difference of changed pixels
The obtained predicted change pixel 207 is output to the subtractor 208.
Is forced. The subtracter 208 predicts the detected change pixel 204 and
The difference from the measured pixel 207 is output as the prediction error 209.
This prediction error 209 is encoded by the encoder 210.
And the encoded signal 211 is output. Prediction error 2
09 is added to the delay prediction error 22 in the adder 224.
1 is added. Resulting delay prediction error 221
Corresponds to the difference between the second and third changed pixels as described above.
Is temporarily stored in the memory 220.
Therefore, it is delayed and used for the next encoding. That is,
In FIG. 36, the next line (one line below)
In addition, the above-mentioned “two pixels to the left” in the delay error 221
And the above "one pixel to the left" which is the prediction error 209 is added.
That the “3 pixels to the left” is used as the predicted value
Become. As described above, the image code of the twenty-ninth embodiment
In the digitizing device, the memories 216 and 220 and the changed pixel detection
Output unit 204, adders 216 and 220, and subtractor 2
08 and the encoder 210, the input signal
Delay processing for the detected changed pixels and the prediction error
And the addition process, the embodiment 26
Similarly, it is possible to improve the coding efficiency. What
In the image coding apparatuses according to Embodiments 25 to 29,
Input in block units,
It is possible to process by the order. Embodiment 30 FIG. Embodiment 30 of the present invention
The image decoding device according to
Decode the encoded signal output from the device to
To obtain a two-dimensional image signal. Fig. 37
30 shows a configuration of an image coding apparatus according to Embodiment 30 of the present invention.
It is a block diagram. In the figure, 211 is an input signal
And output from the image coding apparatus according to Embodiment 25.
30 is a coded signal (211 in FIG. 30) of a prediction error to be performed.
A decoder 230 decodes the coded signal 211.
Thus, a decoding prediction error 231 is output. 232 is an adder
Yes, the decoding prediction error 231 and the prediction change pixel 207 are added.
The arithmetic processing is performed, and the obtained decoded change pixel 233 is output.
Reference numeral 234 denotes a pixel value generator, which is directly connected to the decoded changed pixel 233.
Pixels located between the previously decoded change pixels are
Pixel value, that is, the pixel value of a pixel that is not a changed pixel
Then, a decoded image signal 235 is generated and output. other
Are the same as those in FIG.
The same is true. Embodiment 30 configured as above
The operation of the image decoding apparatus according to
When the encoded signal 211 is input, the prediction error is encoded.
The input signal 211 is decoded by the decoder 230.
The resulting decoded prediction error 231 is
2 is output. On the other hand, the image signal 235 decoded immediately before
Are input to the memories 202a and 202b, and the
As in state 25, the prediction of the changed pixel is performed and the changed pixel is predicted.
The predicted change pixel 207 is sent from the predictor 206 to the adder 232.
Is output. The adder 232 adds the predicted change pixel 207 to
By adding the input decoding prediction error 231
Thus, the decoding change pixel 233 is obtained, and this is used to generate a pixel value.
Output to the device 234. The pixel value generator 234 calculates the decoding change
Located between pixel 233 and the previously decoded change pixel.
Pixel is a predetermined pixel value, that is, an image that does not become a changed pixel.
A decoded image signal 235 is generated as a raw pixel value.
Output. Thus, the image according to the thirtieth embodiment is
The decoding device comprises memories 202a-b, a changing pixel detector
204a-b, a change pixel predictor 207, and a decoder 2
30, an adder 232, and a pixel value generator 234.
By using the prediction change pixel and the decoding prediction error,
A decoded change pixel is obtained, and the decoded image signal 23 is
5 so that the coded signal according to
Can be decrypted. In the thirtieth embodiment,
Now, the encoded signal by the image encoding apparatus of the twenty-fifth embodiment is
Is decrypted, but according to Embodiments 27 and 2
8 to obtain an encoded signal obtained by the image encoding apparatus.
Can be similarly decoded. Embodiment 31 FIG. Embodiment 31 of the present invention
The image decoding apparatus according to
Decode the encoded signal output from the device to
To obtain a two-dimensional image signal. FIG. 38 shows the present invention.
30 shows a configuration of an image coding apparatus according to Embodiment 31 of the present invention.
It is a block diagram. Referring to FIG.
The arithmetic unit 232 and the pixel value generator 234 are shown in FIG.
This is the same as FIG. 32, and the description is given in the embodiments 30 and 26.
Is the same as The thirty-first embodiment constructed as described above
The operation of the image decoding apparatus according to
When the encoded signal 211 is input, the prediction error is encoded.
The input signal 211 is decoded by the decoder 230.
The resulting decoded prediction error 231 is
2 is output. On the other hand, the decoded changed image decoded immediately before
The element 233 is input to the memory 216a,
In the same manner as described above, the prediction of the changed pixel is
The predicted change pixel 207 is output from the adder 232 to the adder 232.
It is. Subsequent processing is the same as in the thirtieth embodiment.
You. Thus, the image decoding apparatus according to the thirty-first embodiment is described.
The memory 216a-b and the change pixel predictor 207
, A decoder 230, an adder 232, and a pixel value generator
234, the prediction change pixel and the decoding prediction error
Using the difference, a decoded changed pixel is obtained, and
Since the signal image signal 235 is obtained, the code
The decoded signal can be appropriately decoded. Embodiment 32 FIG. Embodiment 32 of the present invention
The image decoding device according to
Decode the encoded signal output from the device to
To obtain a two-dimensional image signal. FIG. 39 shows the present invention
32 shows a configuration of an image coding apparatus according to Embodiment 32 of the present invention.
It is a block diagram. Referring to FIG.
The arithmetic unit 232 and the pixel value generator 234 are shown in FIG.
The description is similar to that of FIG.
Is the same as The thirty-second embodiment constructed as described above
The operation of the image decoding apparatus according to
When the encoded signal 211 is input, the prediction error is encoded.
The input signal 211 is decoded by the decoder 230.
The resulting decoded prediction error 231 is
2 is output. On the other hand, the decoded changed image decoded immediately before
The element 233 is input to the memory 216,
Similarly, the prediction of the changed pixel is performed, and the
The predicted change pixel 207 is output to the adder 232. Or later
Is similar to that of the thirtieth embodiment. As described above, the image according to the thirty-first embodiment is
The decoding device comprises memories 216 and 220 and adder 2
24, 222, and 232, the decoder 230, and the pixel
With the provision of the value generator 234, the prediction change pixel and the decoding
The decoding change pixel is obtained using the signal prediction error and
Therefore, a decoded image signal 235 is obtained according to the twenty-ninth embodiment.
Can be appropriately decoded.
It should be noted that any one of the image coding apparatuses
If encoding was performed in block units in
In the case of the image decoding apparatus according to Embodiments 30 to 32,
Input the coded signal in block units, and
Processing can be performed properly by
You. Embodiment 33 FIG. Embodiment 33 of the present invention
The image encoding device according to
Output by switching the encoding result of the difference or the number of pixels
It is. FIG. 40 shows an image code according to Embodiment 33 of the present invention.
It is a block diagram which shows the structure of an encoding device. Smell
Reference numeral 240 denotes a subtractor, which detects the changed pixel 205
The difference 241 between b and 205c is obtained. 242 is a sign
And encodes the difference 41 to generate an encoded signal 243
Is output. 244 is a comparator, which has a prediction error 209 and
A comparison with a predetermined value is performed, and a switch 24
6 is controlled. Reference numeral 246 denotes a switch.
Either of the coded signals 247 and 243 is
Whether to use the output coded signal 211 of the image coding device
The switching is performed under the control of the comparator 244. Other marks
The symbols are the same as in FIG. 30, and the description is the same as in the twenty-fifth embodiment.
It is. An image coding apparatus according to Embodiment 25 performs prediction
The error was coded, but the prediction error was small.
The encoding is performed on the assumption that
In such a case, the coding efficiency decreases. In such a case,
Rather than encoding the measurement error, the change pixel itself (position) is marked.
Encoding is more efficient. Therefore, this implementation
In the image coding apparatus according to aspect 33, the code of the prediction error
And encoding the number of pixels indicating the position of the changed pixel.
Things. In addition, encoding of the changed pixel itself (position) is performed.
Changes the number of changed pixels, making prediction difficult
Or impossible, making prediction error coding difficult or impossible
Encoding, it is possible to perform encoding.
You. In the thirty-third embodiment configured as described above,
The operation of the image coding apparatus according to the first embodiment will be described. FIG.
Reference numeral 1 denotes encoding of the image by the image encoding apparatus according to Embodiment 33.
It is a figure for explaining the principle of operation. Second and third
The prediction of the first changed pixel from the changed pixel is performed according to the second embodiment.
This is the same as the case of No. 5. In the thirty-third embodiment,
Predetermined values centered on predicted change pixels
Set the corresponding prediction range. Then, the detected first change
Depending on whether or not the pixel is in the prediction range,
The switch is performed, and if it is within the prediction range,
If the difference is not in the prediction range, encode the first changed pixel
You. In the thirty-third embodiment, the third change
Since the pixels are already coded and decoded, the first
To encode the change pixel, a third change pixel and a
The difference in the scanning order from the one changed pixel, that is,
It suffices to encode the number of existing pixels. And
Of the pixels in between, those located in the above prediction range
This is encoded by the prediction error.
It is possible to leave. Therefore, the first changed pixel is
In order to encode, the pixels in the prediction range
That is, it is only necessary to encode the code excluding. For example, the change pixel A in FIG.
Pixel B is outside the prediction range, and
What to do if these points are detected as changed pixels
explain. As described in the twenty-fifth embodiment, the scanning direction is
From upper left to lower right, the first change pixel is 3 * 12 + 6 = 42
Eye, change pixel A is 4 * 12 + 1 = 49th, change pixel B is 4 * 12 + 10
= 58th. Between the third changed pixel and the changed pixel A
Since there is no prediction range, the number of pixels between them is 49-42 =
7 is a change pixel A, that is, information indicating the position of A.
Is encoded. On the other hand, in the case of the change pixel B,
The prediction range is included between the changed pixel 3 and the changed pixel B.
So, excluding 5 pixels in this prediction range,
2-5 = 11 is changed pixel B, that is, information indicating the position of B.
Will be encoded. The input signal 201 is the image of the thirty-third embodiment.
After being input to the encoding device, the memories 202a and 20
From the delay due to 2b, the prediction error 20
9 is performed in the same manner as in the twenty-fifth embodiment.
Signal 210 of the prediction error 209 by the
7 is obtained. In Embodiment 25, this encoded signal is output.
Although the encoded signal was input,
, The encoded signal 247 is output to the switch 246.
Further, the prediction error 209 is calculated by the encoder 242 and the comparator 2
44. On the other hand, when the change pixel detector 204b detects
The third changed pixel 205b and the changed pixel detector 204c
Is subtracted from the first change pixel 105c detected by the subtractor 24.
0, and the difference between them is
The prime number 241 is obtained and output to the encoder 242.
The encoder 242 calculates the difference 241 and the prediction error 2
09 and the number of pixels excluding pixels existing in the prediction range
Obtain coded signal 243 and output it to switch 246
I do. The comparator 244 calculates, for the input prediction error,
It is determined whether it is in the prediction range or not.
In the switch 246, the prediction error coded signal 247 is
Output 211 and if not in prediction range
The switch 246 outputs the pixel number encoded signal 243.
Control is performed by the signal 245 so that the force 211 is set.
U. As described above, the image according to the thirty-third embodiment is
In the encoding device, the image encoding device according to Embodiment 25 is used.
, A subtractor 240 and a pixel number encoder 242
And a configuration including a comparator 244 and a switch 246.
The prediction error is within the specified range
Indicates that the coded signal of the prediction error is out of range.
Since the encoded signal of the prime number is the encoded signal to be output,
If the prediction error is large, the
Coding efficiency can be reduced even when the changing pixel cannot be predicted.
Perform proper encoding to prevent degradation
Becomes possible. Embodiment 34 FIG. Embodiment 34 of the Invention
The image decoding apparatus according to
Decode the encoded signal output from the device to
To obtain a two-dimensional image signal. Fig. 42
34 shows a configuration of an image coding apparatus according to Embodiment 34 of the present invention.
It is a block diagram. In the figure, reference numeral 250 denotes mode decoding.
The prediction error is coded for the input signal
Signal or the position of the changing pixel (number of pixels) is encoded
It is determined whether the signal is an encoded signal and the encoding mode 251 is output.
You. 256 is a pixel number decoder which converts the input signal 255
It decodes and outputs the decoded pixel number 257. 258 is added
A prediction change pixel 205b and the number of decoded pixels 257
And outputs a decoded changed pixel 259. 2
52 and 260 are switches, and the mode decoder 25
Input signal and output corresponding to the encoding mode that outputs 0
Switching of signals is performed. Other symbols are the same as in FIG.
The description is the same as that of the twenty-fifth embodiment. More than
Image decoding apparatus according to the thirty-fourth embodiment
The operation of the device will be described. The encoded signal 211 is
When input, the mode decoder 250 first
Encodes the measurement error or encodes the number of pixels
Are determined, and the result of the determination
The coding mode of “error” or “number of pixels” is output.
The switching between the switches 252 and 260 is controlled. Operation when prediction error is coded
Is the same as in the thirtieth embodiment. On the other hand, the number of pixels
The switch 252
Therefore, the input signal 211 is decoded by the decoder 256.
And the number of pixels, which is the difference between the changed pixels, is decoded,
The decoded pixel number 257 is output to the adder 258. Addition
In the arithmetic unit 258, the number of decoded pixels 257 is
Predicted based on the decoded image signal 235 obtained by
The decoded change pixel 259 is obtained by adding the
Can be In any case, based on the decoded change pixel 261
Thus, the decoded image signal 235 is output as in the thirtieth embodiment.
Is done. As described above, image decoding according to the thirty-fourth embodiment.
The decoding apparatus is based on the image decoding apparatus according to Embodiment 30.
The mode decoder 250, the adder 258, and the
Prime decoder 256 and switches 252 and 260
And the mode decoder 250
The switch 252 and the switch 252 correspond to the encoding mode to be acquired.
And 260 are switched, and appropriate decoding is selectively performed.
Therefore, the encoding coded in Embodiment 33
The signal can be properly decoded. Embodiment 35 FIG. Embodiment 35 of the Present Invention
The image encoding device and the image decoding device according to
Box settings can be changed according to the image signal.
You. Image coding apparatus and image according to Embodiment 35
The decoding apparatus has the same configuration as in Embodiments 33 and 34.
Things. FIG. 43 shows encoding according to the 35th embodiment,
Alternatively, it is a diagram for explaining the principle of the decoding operation. Figure
(a) shows the case where the input image is composed of 8 × 8 pixels, and
FIG. 4 (b) shows a sub-sampled version of the same example, and
The configuration is shown. Subsampled
Is 1/2, while the distance between pixels is doubled.
You. Therefore, if it is subsampled,
The measurement range is defined as a range equivalent to 1/2 of the prediction range of the original one.
Search results in searching for almost the same spatial location.
You. For example, the right subsampled prediction range and
Use the same range of ± 2 pixels as the original one on the left.
Then, it will exceed the number of pixels in one line,
In Embodiments 33 and 34, the mode switching is
Not done properly. On the other hand, as shown
If it is subsampled, reduce the prediction range by half
In that case, mode switching can be performed properly,
Improvement of the coding efficiency according to the embodiment can be realized. As described above, the image according to the thirty-fifth embodiment is
In an encoding device and an image decoding device, a thirty-third embodiment
Encoding apparatus, or an image according to the thirty-fourth embodiment.
In the decoding device, the size of the prediction range is
By changing the size according to the size,
In the case of sample
It is possible to improve the efficiency. Embodiment 36 FIG. Embodiment 36 of the present invention
The image encoding device based on
And encodes a significant area from the image signal.
Extraction and efficient encoding are performed. FIG.
A configuration of an image coding apparatus according to Embodiment 36 of the present invention
FIG. Referring to FIG.
The signal is a two-dimensional shape signal. 402 is a significant area extraction
A significant region extracted from the input shape signal 401
And outputs a significant area signal 403. 404 is a block
The input shape signal 401 is divided into blocks and
The locked shape signal 405 is output. 408 is a switch
And switching is performed in response to the significant area signal 403.
U. Reference numeral 412 denotes a block size changer, which outputs a significant area signal.
Change the size of the block in response to
A further blocked shape signal 413 is output. 418,
And 414 are encoders, each of which has a significant area signal 4
03 and the block shape signal 413
The signals 419 and 415 are output. The thirty-sixth embodiment configured as described above
The operation of the image coding apparatus according to the first embodiment will be described. 2
The input signal 401 which is a dimensional shape signal is used in the third embodiment.
6 and input to the image coding apparatus of FIG.
Is input to the blocking unit 404. Extract significant area
The detector 402 detects the range of the significant region and outputs the significant region signal.
403 is a switch 408, a block size changer 41
2 and output to the encoder 418. FIG. 45 shows an image coding apparatus according to the thirty-sixth embodiment.
FIG. 4 is a diagram for explaining the principle of encoding operation by
You. The shaded area indicates the pixels inside the object,
Pixel that contains a clear image signal, and
A small rectangle, that is, a rectangle shown by a bold line in FIG.
Corresponds to the range. Blocker 405 is the input shape
The signal is blocked and the blocked shape signal 405 is scanned.
Output to switch 408. Here, the switch 408 is
A block is set in the range of the significant area indicated by the significant area signal 403.
The state becomes 0N when the converted shape signal 405 corresponds.
That is, in the case of a region other than the significant region, the block shape signal
Is not coded. When the switch is ON
Changes the block size signal 405 to the block size
Input to the block size changer 412.
The significant region is included in response to the significant region signal 403 input.
Size changed to the smallest block, and the changed shape
The signal 413 is output to the encoder 414,
The encoded signal 415 is the shape signal. On the other hand, significant area
The significant region signal indicating the range of
And an encoded signal 419 is output. in this way,
In the image coding apparatus according to Embodiment 36, the significant area
Including the detector 102 and the block size changer 412
By detecting the range of the significant region,
Block shape signal so that only the part encodes the shape signal
Encode outside of the significant area because the size is changed
And the coding efficiency of the shape signal is improved. Embodiment 37 FIG. Embodiment 37 of the present invention
The image decoding apparatus according to
The encoded signal output from the device is decoded, and the two-dimensional shape signal is decoded.
Issue. FIG. 46 shows Embodiment 37 of the present invention.
FIG. 3 is a block diagram illustrating a configuration of an image decoding device according to the first embodiment.
In the same figure, 419 and 415 of the thirty-sixth embodiment
This is an encoded signal output from the image encoding device. 42
0 is a decoder for the significant area signal, and 422 is a decoder for the shape signal
Which decodes each input signal and
Region signal 421, minimum-blocked decoded shape signal 423
Is output. 430 is a block size changer.
In response to the signal significant area signal 421, the block size is
And outputs the changed decoded block shape signal 431.
Power. Reference numeral 426 denotes a switch, which is a significant area signal 421.
Switching is performed according to. 432 is a deblocker
Yes, the block shape signal 427 is integrated and the decoded shape
The signal 433 is output. This embodiment is constructed as described above.
The operation of the image decoding apparatus according to the above will be described. Mark
The decoded signals 419 and 415 are output from the decoder 42, respectively.
0 and 422 and decoded. Decoder 4
19 converts the decoded significant area signal 421 into a block size
Output to the switch 430 and the switch 426. On the other hand,
The encoder 422 is the smallest block including the range of the significant region.
A certain minimum block shape signal 423 is changed in block size.
Output to the changer 430. Block size changer 430
Is a block based on the input decoded significant area signal.
Change the size to a predetermined size and change the block
The signal is output to the switch 426 as the shape signal 431. S
The switch 426 indicates the range of the significant area indicated by the significant area signal 421.
Turns on only when a signal including
Outside is a value indicating that the value is outside the range of the significant region.
The deblocker 432 outputs the input blocked shape signal.
And the signal indicating the outside of the significant area are integrated to obtain a two-dimensional
The shape signal is output as a decoded signal 433. like this
In the image decoding apparatus according to the thirty-seventh embodiment, the decoding
Modifiers 420 and 422 and a block size changer 43
0, switch 426, and deblocker 432
As a result, the range of the significant area is decoded, and based on that,
By decoding the shape signal, the code
The encoded signal thus encoded can be correctly decoded. Embodiment 38 FIG. Embodiment 38 of the Invention
Image encoding device performs encoding according to the prediction probability.
Thus, good hierarchical coding is realized.
FIG. 47 shows an image coding apparatus according to Embodiment 38 of the present invention.
FIG. 3 is a block diagram showing the configuration of FIG. In FIG.
This is a force image signal. Reference numeral 300 denotes a separator, which is an input image signal.
No. 1 is separated into two image signals 301a and 301b.
Output. 302, 308a, and 308b are encoded
Each of which encodes an input signal and encodes
And output the conversion signal. 330 is a decoder which is an encoded signal
303a, and outputs a decoded image signal 331.
You. Reference numeral 304 denotes a prediction probability calculator, which receives an input image signal.
No. 331 to predict the pixel value of the image signal 301b
Calculate the prediction probability for the prediction, and calculate the probability value 30
5 is output. 306 is a second separator, which is input and
In response to the probability value 305, the image signal
Nos. 307a and 307b are output separately. The thirty-eighth embodiment constructed as described above
The operation of the image coding apparatus according to the first embodiment will be described. Entering
Force signal 1 is input to the image encoding apparatus according to the thirty-eighth embodiment.
First, in the separator, the image signals 301a and 301
01b. Here, the signal 301a is given priority
The selected signal is input to the encoder 302,
The other signal 301b is output to the second separator 306.
FIG. 48 shows a code obtained by the image coding apparatus according to the embodiment 38.
It is a figure for explaining the principle of operation of conversion. Fig. 48 (a)
, The solid circle pixel corresponds to the image signal 301a
A pixel indicated by a broken circle corresponds to the image signal 301b.
FIG. 48 shows a model of a binary image signal.
A circle with no hatching indicates a false value. Mark
The encoder 302 encodes a high-priority image signal 301a.
Then, the obtained encoded signal 303a is used as an encoded output.
At the same time, it also outputs to the decoder 330. [0179] A decoded signal that has been decoded by the decoder 330
Is input to the prediction probability calculator 304. Predicted probability calculator
304 is based on the decoded high-priority image signal
To predict the pixel value of the low priority image signal,
Calculate the probability. In FIG. 48 (a), A is four adjacent directions
Is a false value, and B is a true value in four adjacent directions, whereas C is
Is a true value in two adjacent directions and a false value in two adjacent directions. That
As a result, it is confirmed that A is a false value and B is a true value.
Rate is high, but C is true or false
Also, the probability that the prediction is correct is low. So
Here, C is given priority over A and B shown in FIG.
After the encoding, C is decoded as shown in FIG.
While playing back A and B based on prediction
The quality degradation is small, and the desired gradation encoding is performed.
obtain. Therefore, the prediction probability calculator 304 outputs
Based on the probability value 305, the second separator 306
The probability value 305 is high for the
Is the image signal 307a, and the other is the image signal 307a.
b, and each of the encoders 308a and 3
08b. Each encoder has its own input
The image signal is encoded to generate encoded signals 303b and 303.
Output c. The coded signal output as described above
303a to 303c are transmitted in this order as having the highest priority.
Or, if recorded, a code with a higher priority for decoding
Decoding by decoding in order from the encoded signal
Even if the decoding process is aborted in the process of
A decoded image with little quality degradation can be obtained. This
As described above, in the image coding apparatus according to the thirty-eighth embodiment, the separator
300 and 306, encoders 302, 308a, and
308b, a decoder 330, and a prediction probability calculator 30
4 means that pixels with low prediction probability are preferentially signed.
And hierarchical coding with less image quality degradation
Can be realized without additional information. Embodiment 39 FIG. Embodiment 39 of the Present Invention
The image decoding apparatus according to
It decodes the encoded signal output from the device. Figure
Reference numeral 49 denotes an image decoding apparatus according to Embodiment 39 of the present invention.
FIG. 3 is a block diagram illustrating a configuration. In the figure, 303a
To c are output from the image coding apparatus according to the thirty-eighth embodiment.
A coded signal, of 310, 316a, and 316b.
The decoded signals 311 and 317 are decoded by the decoder.
a and 317b. 320 is a predictor
Yes, the image signal is predicted based on the image signal 311 and the prediction is performed.
The measurement image signal 321 is output. 312 is a prediction probability calculator
And the prediction of the input predicted image signal 331
The probability is calculated and a probability value 313 is output. 322 is Sui
And performs switching in accordance with the probability value 313. The thirty-ninth embodiment configured as described above
The operation of the image decoding apparatus according to the above will be described. Mark
The decoded signals 303a to 303c are supplied to decoders 310 and 31 respectively.
6a and 316b and decoded. Signal 3
03a is decoded, and the decoded image signal 311 is output
Signal, and a prediction probability calculator 312;
It is also input to the predictor 320. The predictor 320 outputs the decoded image signal 311
, The pixel value of the low-priority image signal 321 is predicted.
The prediction probability calculator 312 calculates the prediction of the predicted image signal 321.
Calculate the measurement probability, and determine whether each pixel is
16b to determine which should be decoded. Ma
In addition, the prediction probability calculator 312 calculates the priority order input from the outside.
Referring to the order 309, an encoded signal with a lower priority is transmitted.
Or, it is determined whether or not it is recorded. Transmission or note
If it is determined that the pixel has not been recorded,
Is the image signal 3 predicted as the decoded signal 323
Switch 322 to output 21
Control. For decoded pixels, the switch
The image signal 311, 317a, or 317b is
Is selected, and the decoded signal output from the device is selected.
No. 323. Thus, according to the thirty-ninth embodiment.
In an image decoding apparatus, decoders 310, 316a, and
And 316b, a prediction probability calculator 312, and a predictor 320
And decryption corresponding to the prediction probability and priority
Therefore, the image encoding apparatus according to Embodiment 38
The encoded signal can be decoded properly
You. Embodiment 40 FIG. Embodiment 40 of the present invention
RECORDING MEDIUM FOR IMAGE ENCODING PROGRAM USED BY IMAGE
The program recording medium is implemented on a computer, etc.
Encoding apparatus or image decoding apparatus according to the first to third aspects.
Is realized. FIG. 50 records the program
Floppy (registered trademark) disk as an example of a recording medium
FIG. 51 shows the recorded image encoding process.
FIG. 52 shows the processing procedure of the image decoding program.
It is a flowchart figure shown. Floppy shown in FIG.
The image encoding program shown in FIG.
Ram is a personal computer or workstation
And so on, according to the second embodiment.
An image encoding device is realized. Similarly, the flow chart shown in FIG.
Image decoding shown in FIG. 52 recorded on a oppey disk
The program can be used on personal computers and workstations.
Embodiment 3
Is realized. In this case, the same embodiment
In FIG. 6, the changing pixel decoding process described with reference to FIG.
After that, it is of a type that performs selection by a switch. As described above, the process according to the fortieth embodiment is performed.
The gram recording medium stores an image encoding program or an image decoding program.
Recording of the encryption program
Computer systems such as personal computers
The image encoding device or the image decoding device of the present invention.
It can be realized. Note that, in the present forty embodiment,
Are the image encoding device according to the second embodiment and the image encoding device according to the third embodiment.
A program for realizing an image decoding device is recorded.
However, devices according to other embodiments can be realized similarly.
is there. Also, in the fortieth embodiment, a file is used as a recording medium.
I showed a floppy disk, IC card, CD-RO
Write programs such as M, optical disk, cassette tape, etc.
The present invention can be similarly implemented as long as the medium can be recorded. [0186] The present invention has been described above.ToSuch an image transmission method
According toThe shape information indicating the shape of the object and the
Image signal including pixel value information which is color information of each pixel.
Image transmitting an image coded signal, which is the coded signal of the signal
A transmission method for transmitting a block of a predetermined number of pixels.
To respondBlocked shape information and pixel value information
Encoded identification information that identifies the encoding mode for
Information is included in the above-mentioned image encoded signal and transmitted.Data transmission
Including the steps, the shape information and the pixel value information
The encoding modes corresponding to each are in-screen encoding
Mode or inter-screen coding mode.
CharacterSoShape information and pixel values
Each encoding mode of information is defined as an intra-screen encoding mode.
Whether to use the inter-picture coding mode
Code that can be individually switched between the report and the pixel value information
Decryption processing corresponding to thethe aboveMark for each information
Low bit rate to identify the encryption modeBased on identification information
And can be performed appropriately. [0187]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 1 of the present invention. FIG. 2 is a diagram for explaining an operation principle of the image encoding device according to the first embodiment of the present invention. FIG. 3 is a block diagram illustrating a configuration of an image encoding device according to a second embodiment of the present invention. FIG. 4 is a diagram for explaining an operation principle of an image encoding device according to a second embodiment of the present invention. FIG. 5 is a block diagram illustrating a configuration of an image decoding device according to a third embodiment of the present invention. FIG. 6 is a block diagram illustrating a configuration of another example of an image decoding device according to Embodiment 3 of the present invention. FIG. 7 is a block diagram illustrating a configuration of an image encoding device according to a fourth embodiment of the present invention. FIG. 8 is a diagram for explaining an operation principle of an image encoding device according to a fourth embodiment of the present invention. FIG. 9 is a block diagram illustrating a configuration of an image decoding device according to a fifth embodiment of the present invention. FIG. 10 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 6 of the present invention. FIG. 11 is a block diagram illustrating a configuration of an image decoding device according to a seventh embodiment of the present invention. FIG. 12 is a block diagram showing a configuration of an image encoding device according to Embodiment 8 of the present invention. FIG. 13 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 9 of the present invention. FIG. 14 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 10 of the present invention. FIG. 15 is a block diagram showing a configuration of an image decoding device according to Embodiment 11 of the present invention. FIG. 16 is a block diagram illustrating a configuration of an image decoding device according to a twelfth embodiment of the present invention. FIG. 17 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 13 of the present invention. FIG. 18 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 14 of the present invention. FIG. 19 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 15 of the present invention. FIG. 20 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 16 of the present invention. FIG. 21 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 17 of the present invention. FIG. 22 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 18 of the present invention. FIG. 23 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 19 of the present invention. FIG. 24 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 20 of the present invention. FIG. 25 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 21 of the present invention. FIG. 26 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 22 of the present invention. FIG. 27 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 23 of the present invention. FIG. 28 is a diagram for describing selection of the number of motion vectors in the image encoding device according to Embodiment 23 of the present invention. FIG. 29 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 24 of the present invention. FIG. 30 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 25 of the present invention. FIG. 31 is a diagram for describing an operation principle of an image coding device according to a twenty-fifth embodiment of the present invention. FIG. 32 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 26 of the present invention. FIG. 33 is a diagram for describing an operation principle of an image coding device according to a twenty-seventh embodiment of the present invention. FIG. 34 is a diagram for describing an operation principle of an image encoding device according to a twenty-eighth embodiment of the present invention. FIG. 35 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 29 of the present invention. FIG. 36 is a diagram for describing an operation principle of an image coding device according to a twenty-ninth embodiment of the present invention. FIG. 37 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 30 of the present invention. FIG. 38 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 31 of the present invention. FIG. 39 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 32 of the present invention. FIG. 40 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 33 of the present invention. FIG. 41 is a diagram for describing an operation principle of an image encoding device according to a thirty-third embodiment of the present invention. FIG. 42 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 34 of the present invention. FIG. 43 is a diagram for describing setting of a prediction range in an image encoding device and an image decoding device according to Embodiment 35 of the present invention. FIG. 44 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 36 of the present invention. [Fig. 45] Fig. 45 is a diagram for describing the operation principle of the image coding device according to the embodiment 36 of the present invention. FIG. 46 is a block diagram illustrating a configuration of an image decoding device according to a thirty-seventh embodiment of the present invention. FIG. 47 is a block diagram illustrating a configuration of an image encoding device according to Embodiment 38 of the present invention. FIG. 48 is a diagram for describing an operation principle of an image encoding device according to a thirty-eighth embodiment of the present invention. FIG. 49 is a block diagram illustrating a configuration of an image decoding device according to Embodiment 39 of the present invention. FIG. 50 is a diagram illustrating a floppy disk as an example of a recording medium of an image encoding program and an image decoding program according to Embodiment 40 of the present invention. FIG. 51 is a flowchart showing a processing procedure of an image coding program according to Embodiment 40 of the present invention. FIG. 52 is a flowchart showing a processing procedure of an image decoding program according to Embodiment 40 of the present invention. FIG. 53 is a diagram for describing image shape information in image encoding. [Description of Signs] 1, 1a, 201, 211, 255 Input signal 2, 2a, 2b, 204a, 204b Changed pixel detector 3, 61, 95, 202a, 202b, 216a, 21
6b, 220 memory 4, 4a, 4b change pixel predictor 5, 5a, 5b difference value calculator 6 allowable value 7 difference value rounder 8, 8a, 8b, 65a, 65b, 67a, 67b, 2
10,242,418,414 Encoder 9,9a, 9b, 30a, 30b, 66a, 66b, 6
8a, 68b, 80a, 80b, 82a, 82b, 21
1, 243, 419, 415 coded signal 10 changing pixel decoder 11, 11a, 11b, 75a, 75b, 84a, 84
b, 86 Difference value adders 20, 63a, 63b Motion compensator 21 Mode selectors 22, 32, 33, 50 Switchers 31a, 31b, 81a, 81b, 83a, 83b, 2
30, 310, 316a, 316b, 330 Decoder 34, 85a, 85b, 101, 235 Image signal 40a, 40b Horizontal scanner 41a, 41b Vertical scanner 60a, 107 Pixel value signal 60b, 105 Transparency signal (input shape) Signals) 62a, 62b Motion detectors 64a, 64b, 70 Difference value calculator 90 Motion detection determiner 93 Switching determiner 94 Switcher 102, 404 Blocker 103 Shape signal 110 Shape encoding mode determiner 111, 151 Shape code Encoding mode 112 shape encoder 113 shape encoded signal 114 transmittance encoding mode determiner 115,153 transmittance encoding mode 116 transmittance encoder 117 transmittance encoded signal 118 pixel value encoding mode determiner 119, 155 Pixel value encoding mode 120 Pixel value encoder 121 Pixel value encoded signal 1 2 mode encoder 123 mode encoded signals 130 and 136 transmittance encoding mode determiner 132 pixel value encoding mode determiner 138 shape encoding mode determiner 150 mode decoder 156 shape decoder 158 transparency decoding Unit 160 pixel value decoder 157 shape decoded signal 159 transparency decoded signal 161 pixel value decoded signal 162 inverse blocker 163 decoded image signal 170,176,180,186,190,196 switch 172,178 / Inter-frame coding determiner 173 shape coding mode 182 shape coding motion vector number determiner 183,193 shape signal coding mode 188 pixel value coding motion vector number determiner 192 shape coding quantization step Change / non-change determiner 198 Quantization step change / non-change determiner 20 for pixel value encoding a second detected changed pixel (reference signal) 203b third detected changed pixel (reference signal) 204c first changed pixel detector 205a, 205b changed pixel 205c first detected changed pixel 206 changed pixel predictor 207 predicted change Pixel 208 Subtractor 209 Prediction error 221 Delay prediction error 222, 224, 232 Adder 225 Addition value 231 Decoding prediction error 233 Decoding change pixel 234 Pixel value generator 240 Subtractor 241 Difference 243, 247 Coded signal 244 Comparator 246, 426 Switch 250 Mode decoder 251 Encoding mode 256 Pixel number decoder 257 Decoding pixel number 258 Adder 252, 260, 408 Switch 300 Separator 301a, 301b Image signal 302, 308a, 308b Encoder 303a Encoding Signals 304 and 312 Predicted probability calculator 3 05 Probability value 306 Second separator 307a, 307b Image signal separation 311, 317a, 317b Decoded signal 313 Probability value 320 Predictor 322 Switch 331 Decoded image signal (predicted image signal) 401 Input signal (two-dimensional shape signal) 402 Significant region extractor 403 Significant region signal 405 Blocked shape signal 412 Block size changer 413 Blocked shape signal 420 Significant region signal decoder 422 Shape signal decoder 421 Decoded significant region signal 423 Minimum blocked decoded shape Signal 427 Blocked Shape Signal 430 Block Size Changer 431 Decoded Blocked Shape Signal 432 Deblocker 433 Decoded Shape Signal 1001, 1002, 1004, 1006, 1008
1020, 1022 to 1026, 1029, 1033
1036, 1038 Image coding devices 1003, 1003a, 1005, 1007, 102
1,1030-1032,1034,1037,103
9 Image decoding device

   ────────────────────────────────────────────────── ─── Continuation of front page    (31) No. of priority claim No. 8-210955 (32) Priority date August 9, 1996 (August 8, 1996) (33) Priority claim country Japan (JP)    Application for early examination

Claims (1)

(57) [Claims] [Claim 1] Shape information indicating the shape of an object, and the object
An image transmission method for transmitting an image coded signal that is an encoded signal of an image signal including pixel value information that is color information of each pixel to be formed, wherein the block corresponds to a block including a predetermined number of pixels. Transmitting the encoded identification information for identifying the encoding mode for the encoded shape information and the pixel value information by including the encoded identification information in the image encoded signal. each encoding mode corresponding to each of the intra coding mode and inter marks
An image transmission method, which is one of the encoding modes .