JP2002369197A - Image signal transmitting equipment and method, and image signal receiving equipment and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate dot errors and improve picture quality without needing additional data, when an image signal is transmitted effectively by using vicinity image correlation. SOLUTION: Different differential tables are stored in differential vector tables 7a-7d. A trigger producing part 16 produces a trigger on the basis of a scanning signal s and a vector v from a candidate estimating part 8. As the trigger, one or combination of at least two from (1. status around picture elements, 2. status of scanning, 3. the number of times of being a candidate, 4. history of scanning and 5. position of picture element data in a color space) is used. A table selecting part 17 selects one differential vector table by the trigger from the trigger producing part 16. Wile referring to the selected differential vector table, local decoding is performed. Decoded values are supplied to a candidate estimating part 5, which determines transmission picture elements from candidates.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmission apparatus and method for transmitting image data used in a digital television, a digital video recording / reproducing apparatus, and the like, and a receiving apparatus and method.

[0002]

2. Description of the Related Art When transmitting digitized image data for television broadcasting or the like, an encoding process is performed with reference to pixels located around pixels to be transmitted. Such processing makes it possible to perform efficient compression encoding by utilizing the property that a general image not including an edge or the like often has strong autocorrelation in a nearby area. As an order of pixels to be transmitted, a raster scan in which an image is sequentially transmitted from the upper left to the lower right of an image has been conventionally used. In such a transmission method, for example, 8-bit data per pixel including the luminance signal Y and the color difference signals U and V is transmitted. In this case, there is no need to transmit data indicating the address of the pixel.

[0003] In such a transmission method, when the value of a pixel adjacent in the horizontal direction changes sharply, an effect resulting from compression of an image obtained by decoding transmitted data by referring to neighboring pixels is used. Appears. For example, if an image of one vertical "bar" is transmitted using the above-described encoding method, the "bar" may be shaded due to the influence of neighboring pixels referred to during encoding. It becomes an image.

[0004] In order to solve such a problem or to reduce the degree thereof, the present applicant has previously proposed the following transmission method. This transmission method determines a pixel having the highest degree of correlation with the pixel of interest from among untransmitted pixels located in four directions, up, down, left, and right, or oblique to these four directions with respect to the pixel of interest. Then, a difference value between the pixel determined as such and the target pixel is calculated, and data indicating the direction of the target pixel with respect to the pixel determined to have the highest degree of correlation with the target pixel is transmitted together with the calculated difference value. It is something to do.

According to such a transmission method, the conventional NTSC
Compared with the method, much more efficient image transmission becomes possible. However, as a result of a subsequent study, it has been found that a unique error occurs when the above-described transmission method is applied to a moving image. The error is that noise and brightness are different from the surroundings at random positions, resulting in noise. Frame 1
It is an error that you can hardly understand unless you look at it one by one when you look at it one by one, but there is a problem that is noticeable when you watch it continuously as a moving image.

Accordingly, an object of the present invention is to provide a transmission apparatus, a transmission method, a reception apparatus, and a reception apparatus capable of improving an error generated when the transmission method proposed above is applied to a moving image and improving image quality. It is to provide a method.

[0007]

According to a first aspect of the present invention, there is provided an image signal transmitting apparatus for selecting a pixel of interest in an input image signal and positioning the pixel of interest in a predetermined direction. Candidate selection means for selecting one or more pixels as candidate pixels, vector conversion means for converting the pixel value of each candidate pixel into a vector based on a peripheral pixel of the candidate pixel or the pixel of interest, respectively, A transmission pixel determination unit that determines a pixel to be transmitted in the candidate pixel based on a pixel value of the candidate pixel; and The image signal transmission device is characterized in that the conversion method in the vector conversion means is changed according to the following.

[0010] The invention according to claim 11 is an image signal transmission method, wherein a candidate pixel is selected from an input image signal, and one or more pixels located in a predetermined direction are selected as candidate pixels. A vector conversion step of converting a pixel value of each candidate pixel into a vector based on a peripheral pixel or a target pixel of each candidate pixel,
A transmission pixel determination step of determining a pixel to be transmitted in the candidate pixel based on the pixel value of the candidate pixel converted into the vector, and a step of extracting an image transmission state based on the state of the pixel of interest or the transmission method as a feature. The image signal transmission method is characterized in that the conversion method in the vector conversion step is changed according to the characteristics.

According to a twenty-first aspect of the present invention, in an input image signal, a target pixel is selected, one or more pixels located in a predetermined direction are selected as candidate pixels, and each pixel is selected based on a peripheral pixel of the candidate pixel or a target pixel. Then, the pixel value of each candidate pixel is converted into a vector, the pixel to be transmitted is determined in the candidate pixel based on the pixel value of the candidate pixel converted into the vector, and the image based on the state of the pixel of interest or the transmission method is determined. The transmission state is extracted as a feature, and the transmission data is received by an image signal reception device that receives transmission data transmitted from an image signal transmission device configured to change a conversion method in a vector conversion unit according to the characteristic. Receiving means, pixel value generating means for generating a pixel value from the received data, and an output position for determining an output position from the received data Determining means and a means for extracting an image transmission state based on the state of the pixel of interest or the transmission method as a feature, and changing the pixel value generation method in the pixel value generation means according to the characteristic. An image signal receiving device.

According to a twenty-ninth aspect of the present invention, in the input image signal, a pixel of interest is selected, one or more pixels located in a predetermined direction are selected as candidate pixels, and each pixel is selected based on a peripheral pixel of the candidate pixel or a pixel of interest. The pixel value of each candidate pixel is converted into a vector, the pixel to be transmitted is determined in the candidate pixel based on the pixel value of the candidate pixel converted into the vector, and the image transmission is performed based on the state of the pixel of interest or the transmission method. An image signal receiving method for receiving transmission data transmitted from an image signal transmission device, wherein a situation is extracted as a feature, and a conversion method in a vector conversion unit is changed according to the feature. Determining the output position from the received data; and a pixel value generating step of generating a pixel value from the received data. A force position determining step; and a step of extracting as a feature an image transmission state based on the state of the pixel of interest or the transmission method, and changing the pixel value generation method in the pixel value generation step according to the characteristic. Image signal receiving method.

According to the present invention, the conversion method in the vector conversion means is switched according to the state of the pixel of interest or the image transmission state based on the transmission method. Can be eliminated. In addition, the image quality itself is improved, and SN
R can be improved. According to the present invention, there is no need to transmit an additional bit for the switching process. The momentary state of the image is read to form a trigger for switching. Therefore, the present invention has an advantage that the efficiency is not reduced by transmitting the additional bits.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION In order to transmit a digital image signal as efficiently as possible, it is preferable to use the pixel neighborhood correlation of an image. For similar pixels, the efficiency can be considerably improved by sending the difference information. Since the raster scan is used in the NTSC system, the scan method is fixed. On the other hand, in the previously proposed scan, the degree of freedom is given to the scan in order to positively use the image neighborhood correlation. In order to freely scan, it is necessary to transmit address information indicating the position of a pixel in an image together with the pixel information. Even if the address information is added, since the pixel neighborhood correlation is used, the efficiency is improved as a whole. Even so, if the address is transmitted as it is, the amount of information of the address is too large. Therefore, the transmission method proposed above solves this problem by transmitting the relative position from the pixel of interest.

In order to facilitate understanding of the present invention, an image signal transmission device proposed above will be described below with reference to the drawings. FIG. 1 shows a block diagram on the encoder side, and FIG. 2 shows a flowchart of the encoding process.
FIG. 3 shows a block diagram on the decoder side, and FIG. 4 shows a flowchart of the decoding process.

In FIG. 1, reference numerals 1a, 1b, 1c
Is an input terminal to which, for example, a (4: 4: 4) color component signal (YUV) is input. The input component signal is supplied to the vectorization unit 2. In the vectorization unit 2, the input component signal is vectorized as one point in the YUV space. The output of the vectorization unit 2 is written into a frame memory 3 having a size of one screen (this memory 3 is called a vector memory).

A vector memory 3 based on scan information s as a result of a candidate search from a candidate search unit indicated by reference numeral 9
, The direction s and the vector v of the pixel of interest and each candidate are sent to the vector converter 4. The vector conversion unit 4 calculates a difference vector from the target pixel to each candidate vector, and sends the difference vector to the candidate evaluation unit 5.

In the local decoder 6, local decoding is performed with reference to the difference vector table 7, and the decoded value is supplied to the candidate evaluation unit 5, which determines a transmission pixel from the candidates. The multiplexing unit 10 multiplexes the scan direction s of the transmission pixel from the target pixel and the index i of the representative vector, and extracts the transmission data to the output terminal 11. The scan direction s is supplied from the candidate evaluation unit 5 to the scan memory 8. The output of the scan memory 8 is supplied to the candidate search unit 9, and the output of the candidate search unit 9 is provided to the vector memory 3.

Referring to the flowchart shown in FIG.
The flow of the encoding process will be described. In step S1, a pixel to be transmitted first is determined. For example, choose the middle pixel of the image. However, the pixel transmitted first may be at any position. In step S2, the pixel is quantized and transmitted. The transmitted pixel is set as a target pixel (step S
3).

In step S4, the scan memory 8
From the position of the candidate obtained from. In step S5, it is determined whether or not the number of candidates reaches a specified number. If the scanning method is zigzag, the number of four candidates is the specified number. If it is determined in step S5 that the number of candidates does not reach the specified number, candidates are acquired in raster scan order until the number reaches the specified number (step S6).

When the number of candidates reaches the specified number, in step S7, local decoding is performed using the selected table to determine a transmission pixel. Then, in step S8, the scan direction and the index of the difference vector are transmitted. Steps S4, S6 and S7 are processes that require access to the memory. In step S9, attention is paid to the transmitted pixels. Then, step S10
In, it is determined whether all the pixels have been transmitted. If all the pixels have been transmitted, the encoding process ends. If all the pixels have not been transmitted, the process returns to step S4 to transmit the next pixel.

The decoder will be described with reference to FIG. Reference numeral 21 is an input terminal. The input data is supplied to the division unit 22, and the input data is divided into scan information s and index information i. Output position determination unit 23
Receives the scan information s, determines the output position while referring to the scan memory 24, and outputs the output position data a. The vector reconstructing unit 25 reconstructs a vector from the index information i and outputs a vector v.

The vector v is supplied to the component converting section 27, and converts the vector v into a component signal YUV. The component signal is supplied to the frame memory 28. The output position data a is also supplied to the frame memory 28. Then, the pixel value at the output position a is output from the frame memory 28 to the output terminal 29.

Referring to the flowchart shown in FIG.
The flow of the decoding process will be described. In step S21, a signal from the encoder is received as a signal of the first pixel. In step S22, the received pixel data is reconstructed. In step S23, the reconstructed pixel data is output to a specified position. Steps S22 and S23
Is a process that requires access to the memory.

In step S24, a signal from the encoder is received. In step S25, a position to be output in all directions is searched on the scan memory. In the direction in which the output position is not determined, the output position is searched in the raster scan order (step S26). Step S2
At 7, the output position is determined. In step S28, pixel data is reconstructed from the index information. In step S29, the reconstructed pixel data is output to the frame memory. In step S30, it is determined whether all the pixels have been output. If all the pixels have been output, the decoding process ends. If not, the process returns to step S24 to receive the next pixel.

In the conventional raster scan, only pixel information is sent without sending address information by fixing the scan. On the other hand, in the above-described transmission method, the relative position from the pixel of interest is transmitted as a scan signal. At this time, how to jump over the already transmitted pixel is a problem.

Referring to FIG. 5, the jump of the transmitted pixel will be described. In FIG. 5, the target pixel is indicated by a double circle, and a pixel that is a candidate for a transmission pixel in the search is indicated by a light shade. White circles indicate pixels that have not been transmitted, and black circles indicate pixels that have been transmitted. As a search method for searching for a pixel to be transmitted, pixels located in four directions (up, down, left, and right) with respect to a target pixel indicated by a double circle are to be searched.

In FIG. 5A, all the pixels other than the pixel of interest are untransmitted pixels. Therefore, as shown in FIG. 5B, four pixels adjacent in four directions to the pixel of interest are candidates for transmission pixels. . However, as shown in FIG. 5C, when the target pixel is surrounded by the transmitted pixels, the position where the transmitted pixels exist is specified only by the relative relationship of up, down, left, and right. In such a case, pixels that have not been transmitted are skipped and pixels that have not been transmitted are candidates for transmission pixels. That is, as shown in FIG. 5D, two pixels located to the right of the pixel of interest are pixels that have already been transmitted, and the third pixel in the right direction from the pixel of interest that jumps over the two pixels is a candidate for a transmission pixel. . Further, since one pixel in the lower direction of the target pixel has already been transmitted, the second pixel that jumps over the one pixel and is lower than the target pixel is set as a candidate for a transmission pixel.

For this processing, one frame of memory (scan memories 8 and 24) is stored in the encoder and decoder.
, And a flag is set for a transmitted pixel to distinguish a transmitted pixel and an untransmitted pixel. It should be noted that information on whether a pixel has been transmitted (whether it has been transmitted) or not (whether it has not been transmitted) can be shared by the encoder and the decoder, and there is no need to transmit this information.

Next, the scan pattern will be described. In the conventional raster scan, a scan rule is set such that the data is sequentially sent one row at a time from the upper left of the screen. Thus, there is no need to transmit any information about the scan. All bits can be sent as pixel data. However, on the other hand, since the same scanning is performed for any image, the neighborhood correlation of the image is not utilized.

On the other hand, if scanning is to be performed completely randomly, address information must be sent. However, in the case of an image having a standard resolution per pixel using only address information, a data amount of 10 bits is required in both the x and y directions. Since the address information is information that must be transmitted accurately, the data amount cannot be reduced by quantization or the like. As a result, although the efficiency is improved by using the neighborhood correlation, the data amount of the address information is
Pretty heavy.

Therefore, in the proposed transmission system, transmission is made more efficient by giving a degree of freedom to scanning.
"To have freedom" means not completely fixed,
It means not to be free at all. The relative position to the pixel of interest is transmitted based on a predetermined scan pattern.

There are various scan patterns. FIG. 6 shows some examples of scan patterns.
FIG. 6A shows a conventional raster scan for reference. In the case of raster scanning, since the scanning direction is fixed, there is no need to transmit address information. In FIG. 6B, any of the pixels located in four directions of up, down, left, and right with respect to the target pixel is transmitted. In this case, two bits of address information are required for one pixel. The dashed line in the upper left indicates that if no candidate has been obtained, the scan is switched to raster scan.

The scan pattern (zigzag) shown in FIG. 6C is for searching candidates in a zigzag scan for four areas divided around the transmission pixel. As a result, the pixels can be transmitted to the end without a deadlock in scanning. Two bits of address information are required for one pixel. In FIG. 6D, any one of the pixels located in a total of eight directions, that is, up, down, left, right, and oblique to the target pixel is transmitted. In this case, 3-bit address information is required for one pixel. The scan pattern in FIG. 6D selects transmission pixels from among many candidates, so that it is possible to make better use of the neighborhood correlation. However, the address information is increased by one bit as compared with the other scan methods, and the pixels are accordingly increased. There is a problem that data that can be allocated is reduced by one bit.

In the transmission method described above, a candidate is selected based on a certain scan pattern, a difference vector in a difference vector table is applied to the candidate, and an index of the difference vector having the least error is transmitted. . When the number of bits assigned to each pixel is determined, the number of candidates that can be obtained at one time is inversely proportional to the number of difference vectors in the difference vector table. Below 1
An example in which 8 bits are assigned to a pixel is shown.

When eight candidates are obtained at one time as in the scan pattern shown in FIG. 6D described above, three bits are required to specify the candidates, so that five bits are required for the difference vector. And the number of difference vectors is 3
It becomes 2. When acquiring four candidates at a time, as in the scan pattern shown in FIG. 6B or FIG. 6C, two bits are required to specify the candidates, so six bits are used for the difference vector. That is, the number of difference vectors is 64. Further, when only two candidates are obtained at a time only in the horizontal direction, one bit is needed to specify the candidates, so that 7 bits can be used for the difference vector, and the number of difference vectors can be increased. Becomes 128. Further, when the scan pattern is fixed as in the conventional raster scan, all 8 bits can be assigned to the difference vector, so that the number of difference vectors is 256.

As described above, if the number of candidate acquisitions is increased, the number of difference vectors decreases, and if the number of difference vectors is increased, the number of candidate acquisitions decreases. Generally, when many candidates can be obtained, there are many choices, so the number of difference vectors may be small, and vice versa. A well-balanced relationship between the number of candidate acquisitions and the number of difference vectors is considered to change depending on the image transmission status at each time and the tendency of the image itself.

The present invention is an improvement on the previously proposed transmission method. That is, it has been found that a unique error occurs when the transmission method proposed above is applied to a moving image. At random positions, different points of brightness and color from the surroundings are suddenly present, and these become noise. When you look at each frame,
It is an error that can hardly be understood unless you pay attention, but it is noticeable when viewed continuously as a moving image. The present invention improves the image quality by changing a conversion method in a vector conversion unit using information obtained from an image as a trigger, and prevents a dot error from occurring when a moving image is processed. .

A description will now be given of color space difference vector encoding in the above-described transmission method. As shown in FIG. 7, target pixel data and transmission pixel data, which are component signals, are each treated as a vector in a color space, and the difference vector is transmitted. This method can be significantly more efficient than generating differences for each component of the component signal separately. FIG. 8A
As shown in FIG. 8B and FIG. 8B, the set of color space vectors of the YUV component has a stronger correlation than the color space vector of the RGB component, so that the set is small. Further, by forming the color space difference vector, the color space difference vector is collected near (0, 0).

As shown in FIG. 9, a vector representing a set of color space difference vectors is obtained in advance, and a table of difference vectors is created. In the table, each difference vector is distinguished by an index. The difference vector table is shared between the encoder side and the decoder side, and the index is transmitted. If zigzag is used as a scanning method in an 8-bit frame for one pixel, it is possible to have a 6-bit (64) representative vector other than the address. The vectorization unit 2 converts a component signal into a vector v in a color space. However, 64 difference vectors cannot be said to be sufficient, and as described above, when the above-described transmission method is applied to a moving image, the true value is a difference value that cannot be caught by the difference vector in the difference vector table. When it has, there is a problem that an error that occurs in a flickering and dot shape is noticeable.

Therefore, in one embodiment of the present invention, the image quality is improved by switching the difference vector table by a trigger obtained in the course of transmission. If some information can be used as a trigger to switch to the difference vector table that matches the situation at that time, image quality can be improved. If it is possible to react to a situation where a dot-like error occurs, a noticeable error can be eliminated. Further, an increase in the number of difference vector tables means an increase in the number of difference vectors that can be held, so that the image quality itself can be improved.

A trigger used for switching the difference vector table is called a trigger. Normally, when instructing the decoder to perform some kind of switching processing, additional bits are added to data transmitted from the encoder side. However, requiring additional bits is not preferred because it increases redundancy. In the present invention, a trigger for switching the difference vector table from an image to be transmitted without using additional bits is extracted by making good use of the characteristics of the image.

As the trigger, (1. situation around a pixel, 2. situation of a scan, 3. number of candidates, 4. history of a scan, 5. position of a pixel data in a color space) can be considered. Hereinafter, each will be described in order.

As a trigger for the switching process,
“Situation around pixel” can be used. For example, the following factors can be considered as the situation around the pixel.

1. 1. Around the target pixel (a) How many peripheral pixels have been transmitted (b) What percentage of pixels in a peripheral area have been transmitted (c) Activity of peripheral pixels 2. Situation around transmitted pixels (a) How many pixels are transmitted around (b) Percentage of pixels in a certain range around are transmitted (c) Activity of peripheral pixels The linear distance between the target pixel and the transmission pixel.

As an example, the following elements will be used.

1. 1. Around the target pixel (how many neighboring pixels have been transmitted) 2. Situation around transmission pixel (how many pixels are transmitted around) The linear distance between the target pixel and the transmission pixel.

Referring to FIG. 10, each element will be described more specifically. (1. Situation around target pixel)
In FIG. 10A indicates a shaded portion around the target pixel (double circle) in FIG. 10A. Similarly, the surroundings of the transmission pixel in (2. Situation around the transmission pixel) in FIG.
Indicates a shaded portion around the transmission pixel. In both cases, the range around the pixel (3 × 3 = 9 pixels) is set as the range.

FIGS. 11A and 11B show specific examples of the linear distance in (3. Linear distance between target pixel and transmission pixel). Here, the straight line distance represents a straight line distance in an image. In the straight line distance shown in FIG. 11A, the distance between the pixel of interest and the transmission pixel is physically larger than that shown in FIG. 11B. This is determined by determining a certain threshold value, and is used as one of the switching trigger elements.

The "circumstance around the target pixel" will be described. The image has a feature that the correlation in the vicinity of the pixel is strong. As a general property, the closer to the target pixel, the stronger the correlation. Therefore, it is expected that pixels adjacent to the target pixel will have the strongest correlation. However, as shown in FIG. 10C, when all the pixels around the target pixel have already been transmitted, there is a high possibility that a farther pixel, that is, a pixel with lower correlation, has been acquired as a candidate. FIG.
As shown in B, even when pixels around the target pixel are partially transmitted, the same can be said as compared with the case where pixels are not transmitted at all. Using such characteristics, processing is switched depending on how much pixels around the target pixel are transmitted. Thereby, efficiency can be improved.

The "circumstance around the transmission pixel" will be described. When the periphery of the transmission pixel is buried, it is highly likely that the transmission pixel is an isolated point having low correlation with the periphery. It is possible that the pixel is an edge. If the pixel is in a special situation in the image, the processing can be switched from that in the case of transmitting a normal pixel, thereby increasing the efficiency.

The "linear distance between the pixel of interest and the transmission pixel" will be described. In a general image, the correlation near pixels is strong. Therefore, when comparing the case where the distance between the target pixel and the transmission pixel is small as shown in FIG. 11A and the case where the distance is large as shown in FIG. 11B, the correlation in FIG. Considered strong. Therefore, efficiency is improved by providing a threshold value and switching processing according to the linear distance.

Next, "scan status" will be described. In the proposed transmission method, the pixels are transmitted in order from a correlated pixel. As a result, an image that happens to be isolated or a pixel having low correlation remains. If one frame is set to be transmitted until one frame of data is transmitted, such weakly correlated pixels tend to concentrate in the latter half of the scan. Therefore,
Switching to a more appropriate vector conversion table is performed during transmission of one frame of data.

Further, the “number of candidates” will be described. In the proposed transmission method, every time one pixel is transmitted,
The candidate is determined according to the scan pattern while referring to the scan memory, and local decoding is performed to transmit more correlated pixels. At this time, pixels other than the pixels that have been transmitted have been rejected while being selected as candidates. Pixels that have been rejected once are again selected as candidates, and eventually transmitted. However, there are pixels that are selected many times and are not transmitted forever. It is expected that such pixels have low correlation with the surroundings or are likely to be special value pixels in the image. Accordingly, which pixel is selected as a candidate and how many times are stored, and processing is switched based on the selected number. Thereby, efficiency can be improved.

In order to realize a process of switching processes based on the number of candidates, a memory called a candidate memory is prepared in addition to the scan memory. The number of candidates is accumulated in the candidate memory. FIG. 12A to FIG. 12D show how the number of candidates becomes stored in the candidate memory in chronological order. FIG. 12 shows the case of a zigzag scan (FIG. 6C) as a scan pattern.

In FIG. 12A, four untransmitted pixels around the double-circle pixel of interest are selected as candidates. Since this is the first time, a number “1” is added to the candidate pixel. FIG. 12B shows a state in which the candidate pixel on the left side is selected as the transmission pixel. In FIG. 12B, the selected pixel is set as a target pixel, and four surrounding untransmitted pixels are selected as candidates. The number “1” is added to the pixel first selected as a candidate in the state of FIG. 12B,
A number “2” is added to a pixel selected as a candidate in both the state of FIG. 12A and the state of FIG. 12B.
FIG. 12C shows a state in which the candidate pixel immediately below is selected as the transmission pixel in the state of FIG. 12B. Further, FIG.
FIG. 12D shows a state in which the candidate pixel immediately below is selected as the transmission pixel in the state C. As described above, information on the number of times selected as a candidate is stored in the candidate memory for each pixel.

The “scan history” will be described. Information on how scanning has been performed in the past several transmissions is referred to as “scan history”. For example, as shown in FIG. 11A, when a number is assigned to a scan direction, a certain sequence is formed depending on the past scan direction as shown in FIGS. 13B to 13D. This can be used as a scan history to trigger processing switching. As shown in FIG. 13B, in a scan in the horizontal direction only, the sequence becomes (2222).

In the proposed transmission method, scanning is performed while reflecting the characteristics of the image. Therefore, the history of how the scanning is performed includes information indicating the characteristics of each image. Therefore, by using the trigger as a trigger, it is possible to perform a switching process according to the characteristics of each image. Specifically, a method of accumulating past histories and adaptively switching the difference vector table is possible.

The "position of pixel data in the color space" will be described. Pixel data (each component of RGB (three primary colors) is 8-bit data) is placed in a color space. For example, RG
Suppose that it plotted in B space. In the case where the target pixel is (R = 128, G = 128, B = 128) in the RGB space and comes to the position C2 in FIG. Also, the difference value does not take a value equal to or greater than the absolute value 128. This is because RGB does not take a value other than 0 to 255. (R = 255, G = 255, B = 255), and when the target pixel comes to the position of C1 in FIG. 14, the difference value does not become positive. Similarly, (R
= 0, G = 0, B = 0), and the target pixel comes to the position of C3 in FIG. 14, the difference value does not become negative.

As described above, it is possible to narrow down the value that can be taken as the difference value from the pixel to be transmitted next, depending on the position of the target pixel in the color space. By utilizing such a feature, the position of the pixel data in the color space can be used as a trigger for the switching process.

FIG. 15 is a block diagram showing a configuration of an encoder according to one embodiment of the present invention. Parts corresponding to the encoder configuration (see FIG. 1) in the previously proposed transmission method are denoted by the same reference numerals. In one embodiment, a difference vector table 7a, 7b, 7c, 7d for storing a plurality of difference vector tables, a table selector 17 and a trigger generator 16 are provided. The trigger generation unit 16 generates a trigger based on the scan signal s and the vector v from the candidate evaluation unit 5. The table selection unit 17 selects a difference vector table by a trigger from the trigger generation unit 16.

The trigger may be any one of the above-mentioned triggers (1. situation around pixels 2. scan situation 3. number of candidates 4. scan history 5. position of pixel data in color space) Or a combination of two or more.

In FIG. 15, input terminals 1a, 1b, 1
For example, a color component signal (YUV) of (4: 4: 4) is input to c. The input component signal is supplied to the vectorization unit 2. In the vectorization unit 2, the input component signal is vectorized as one point in the YUV space. The output of the vectorization unit 2 is written into a frame memory 3 having a size of one screen (this memory 3 is called a vector memory).

Based on the scan signal s resulting from the candidate search from the candidate search unit 9, the target pixel and the direction s and vector v of each candidate are sent from the vector memory 3 to the vector conversion unit 4. The vector conversion unit 4 calculates a difference vector from the target pixel to each candidate vector, and sends the difference vector to the candidate evaluation unit 5.

In the local decoder 6, the table selection unit 1 in the difference vector tables 7a, 7b, 7c, 7d
The local decoding is performed with reference to the difference vector table selected by 7 and the decoded value is supplied to the candidate evaluation unit 5, and the candidate evaluation unit 5 determines a transmission pixel from the candidates. The multiplexing unit 10 multiplexes the scan direction s of the transmission pixel from the target pixel and the index i of the representative vector, and extracts the transmission data to the output terminal 11.
The direction s is supplied from the candidate evaluation unit 5 to the scan memory 8. The output of the scan memory 8 is supplied to the candidate search unit 9, and the output of the candidate search unit 9 is provided to the vector memory 3.

With reference to the flowchart shown in FIG. 16, the flow of the encoding process according to one embodiment will be described. In step S31, a pixel to be transmitted first is determined. For example, choose the middle pixel of the image. However, the pixel transmitted first may be at any position. Step S32
Then, the pixel is quantized and transmitted. The transmitted pixel is set as a target pixel (step S33).

In step S34, the situation is evaluated based on information from the trigger. In step S35, a difference vector table is selected based on the evaluation result. In step S36, candidates are obtained from the positions of the candidates obtained from the scan memory 8. Step S37
In, it is determined whether the number of candidates reaches a specified number.
If the scanning method is zigzag, the number of four candidates is the specified number. If it is determined in step S37 that the number of candidates does not reach the specified number, candidates are acquired in raster scan order until the number reaches the specified number (step S37).
38).

When the number of candidates reaches the specified number, in step S39, local decoding is performed to determine a transmission pixel. Then, in step S40, the scan direction and the index of the difference vector are transmitted. Steps S32, S36, S38, S39 and S40 are processes that require access to the memory. Step S
At 41, attention is paid to the transmitted pixels. Then, in step S42, it is determined whether all the pixels have been transmitted. If all the pixels have been transmitted, the encoding process ends. If all the pixels have not been transmitted, the process returns to step S34, and the next process is performed.

FIG. 17 is a diagram for explaining a candidate selection method and a process related to local decoding. This processing is performed by the vector conversion unit 4, the local decoder 6, and the candidate evaluation unit 5 in FIG. FIG.
In step 6, the process is performed in step S39. In FIG. 17, a double circle pixel is a target pixel, a circle is a candidate pixel, and a number in the circle is a serial number assigned to the candidate pixel. Each candidate pixel is not necessarily adjacent to the pixel of interest and is obtained from a certain direction in the scan pattern. As shown in FIG. 17A, it is assumed that, for example, four candidate pixels exist for one target pixel.

First, a value obtained by locally decoding the pixel of interest is applied to a difference vector table that is common to the encoding side and the decoding side. As an example, assuming that there are 64 (2 ⁶ ) difference vectors with 6 bits, as shown in FIG. 17B, if the difference vectors are respectively added to the local decode values of the target pixel, 64 color vectors are obtained. Created. These are the transmission pixel values that can currently be created.

Next, as shown in FIG. 17C, the transmittable pixel values that can be created from the candidate and the target pixel are compared, and the transmittable pixel value with the strongest correlation is determined for each candidate based on the comparison result. I do. The transmittable pixel value with the strongest correlation is the optimal approximate value of each candidate.

Finally, as shown in FIG. 17D, a difference value between each candidate and the optimally transmittable pixel is obtained. The candidate having the smallest difference value is determined as the candidate having the smallest error during transmission among the four candidates, and the transmission pixel is determined. The direction of the determined transmission pixel as viewed from the target pixel and the index of the difference vector are multiplexed and transmitted.

A decoder according to an embodiment of the present invention will be described with reference to FIG. The decoder has, for example, four types of difference vector tables 26a, 26b, 26
c, 26d, a table selector 36 and a trigger generator 35 are provided. The trigger generation unit 35 generates a trigger based on the scan signal s and the vector v from the division unit 8. The table selector 36 switches the difference vector table by the trigger.

The input data, which is encoded data, is supplied from the input terminal 21 to the division unit 22, and the input data is divided into scan information s and index information i. The output position determination unit 23 receives the scan information s, determines an output position while referring to the scan memory 24, and outputs output position data a. The vector reconstructing unit 25 reconstructs a vector from the index information i and outputs a vector v. In this case, the table selection unit 36 selects the difference vector table based on the trigger information from the trigger generation unit 35, and the vector reconfiguration unit 25 reconfigures the vector using the selected difference vector table.

The vector v from the vector reconstructing unit 25 is supplied to the component unit 27, and converts the vector v into a component signal YUV. The component signal is supplied to the frame memory 28. Frame memory 2
8 is also supplied with the output position data a. And
The pixel value at the output position a is output from the frame memory 28 to the output terminal 29.

The flow of the decoding process will be described with reference to the flowchart shown in FIG. First step S
At 51, a signal from the encoder is received as a signal for the first pixel. In step S52, the received pixel data is reconstructed. In step S53, the reconstructed pixel data is output to a specified position. Steps S52 and S53, and steps S57, S58,
S60 and S61 are processes that require access to the memory.

In step S54, a signal from the encoder is received. In step S55, the situation is evaluated by the trigger. In step S56, a difference vector table is selected based on the evaluation result. In step S57, a position to be output in all directions is searched on the scan memory. In the direction in which the output position is not determined, the output position is searched in the raster scan order (step S58). In step S59, the output position is determined. In step S60, the pixel data is reconstructed from the index information. In step S61, the reconstructed pixel data is output to the frame memory. Step S
At 62, it is determined whether all pixels have been output. If all the pixels have been output, the decoding process ends. If not, the process returns to step S54 to receive the next pixel.

Next, several examples using more specific examples will be described as trigger generation in one embodiment of the present invention. In the first example, “the situation around the pixel” described with reference to FIGS. 10 and 11 is used as a trigger. FIG. 20 is a block diagram of the encoder of the first example, and FIG. 21 is a flowchart of the encoding process. As shown in FIG.
The same reference numerals are given to the components corresponding to.
In FIG. 21, steps corresponding to those in FIG. 16 are denoted by the same reference numerals.

In the first example, the situation evaluation section 16 around the pixel
a is provided as a trigger generation unit. The situation evaluation unit 16 a generates a control signal for the table selection unit 17 by looking at the contents stored in the scan memory 8.
The difference vector table selected by the table selection unit 17 is supplied to the candidate evaluation unit 5. Other configurations and operations are the same as those of the encoder according to the embodiment, and a description thereof will be omitted.

In FIG. 21, after the number of candidates reaches the specified number, in step S43, the situation around the pixel is evaluated. Based on the result, in step S44, a difference vector table is selected. Local decoding is performed based on the selected difference vector table, and a transmission pixel is determined (step S39). Other processing is the same as the encoding processing in the embodiment, and the description thereof is omitted.

FIG. 22 is a block diagram of the decoder of the first example, and FIG. 23 is a flowchart of the decoding process. Components shown in FIG. 22 corresponding to those in FIG. 18 are denoted by the same reference numerals. 23, steps corresponding to those in FIG. 19 are denoted by the same reference numerals.

On the decoder side, a situation evaluation unit 35a is provided as a trigger generation unit. The situation evaluation unit 35a
Based on the contents stored in the scan memory 24, a control signal for the table selection unit 36 is generated. The difference vector table selected by the table selection unit 36 is supplied to the vector reconstruction unit 25. Other configurations and operations are the same as those of the decoder according to the embodiment, and a description thereof will be omitted.

In FIG. 23, after determining the output position, in step S63, the situation around the pixel is evaluated. Based on the result, in step S64,
Select the difference vector table. The pixel data is reconstructed using the selected difference vector table. Other processing is the same as the decoding processing in the embodiment, and a description thereof will be omitted.

Next, a second example will be described. In the second example, “the number of candidates” described with reference to FIG. 12 is used as a trigger. FIG. 24 is a block diagram of the encoder of the second example, and FIG. 25 is a flowchart of the encoding process. 24, components corresponding to those in FIG. 15 are denoted by the same reference numerals. 25, steps corresponding to those in FIG. 16 are denoted by the same reference numerals.

In the second example, the candidate number evaluation section 16b is provided as a trigger generation section. Candidate count evaluation unit 1
6b generates a control signal for the table selecting unit 17 by looking at the contents stored in the candidate memory 15b.
The candidate memory 15b accumulates the number of times a candidate has been obtained based on the scan signal from the candidate evaluation unit 5. The difference vector table selected by the table selection unit 17 is supplied to the candidate evaluation unit 5. Other configurations and operations are the same as those of the encoder according to the embodiment, and a description thereof will be omitted.

In FIG. 25, after the number of candidates reaches the specified number, in step S45, the number of candidates is evaluated. Based on the result, in step S44, a difference vector table is selected. Local decoding is performed based on the selected difference vector table, and a transmission pixel is determined (step S39). Other processing is the same as the encoding processing in the embodiment, and the description thereof is omitted.

FIG. 26 is a block diagram of the decoder of the second example, and FIG. 27 is a flowchart of the decoding process. 26, components corresponding to those in FIG. 18 are denoted by the same reference numerals. 27, steps corresponding to those in FIG. 19 are denoted by the same reference numerals.

The candidate side evaluation unit 35b is also provided as a trigger generation unit on the decoder side. Candidate count evaluation unit 3
5b generates a control signal for the table selection unit 36 by looking at the contents stored in the candidate memory 34b.
The difference vector table selected by the table selection unit 36 is supplied to the vector reconstruction unit 25. Other configurations and operations are the same as those of the decoder according to the embodiment, and a description thereof will be omitted.

In FIG. 27, after the output position is determined, the number of candidates is evaluated in step S65. Based on the result, in step S64, a difference vector table is selected. The pixel data is reconstructed using the selected difference vector table. Other processing is the same as the decoding processing in the embodiment, and a description thereof will be omitted.

Next, a third example will be described. The third example uses the “scan history” described with reference to FIG. 13 as a trigger. FIG.
FIG. 29 is a block diagram of an encoder of the example, and FIG. 29 is a flowchart of encoding processing. Components shown in FIG. 28 and corresponding to FIG. 15 are denoted by the same reference numerals. 29, steps corresponding to those in FIG. 16 are denoted by the same reference numerals.

In the third example, the scan history evaluation unit 16c
Is provided as a trigger generation unit. The scan history evaluation unit 16c generates a control signal for the table selection unit 17 by looking at the contents stored in the scan history memory 15c. The scan history is stored in the scan history memory 15c based on the scan signal from the candidate evaluation unit 5. The difference vector table selected by the table selection unit 17 is supplied to the candidate evaluation unit 5. Other configurations and operations are the same as those of the encoder according to the embodiment, and a description thereof will be omitted.

In FIG. 29, after the transmitted pixel is set as the target pixel, in step S46, the scan history is evaluated with reference to the scan history memory 15c. Based on the result, in step S44, a difference vector table is selected. Local decoding is performed based on the selected difference vector table, and a transmission pixel is determined (step S39). Other processing is the same as the encoding processing in the embodiment, and the description thereof is omitted.

FIG. 30 is a block diagram of the decoder of the third example, and FIG. 31 is a flowchart of the decoding process. Components shown in FIG. 30 and corresponding to those in FIG. 18 are denoted by the same reference numerals. In FIG. 31, steps corresponding to those in FIG. 19 are given the same reference numerals.

The scan history evaluation unit 35c is also provided on the decoder side.
Is provided as a trigger generation unit. The scan history evaluation unit 35c generates a control signal for the table selection unit 36 by looking at the contents stored in the scan history memory 34c. The difference vector table selected by the table selection unit 36 is supplied to the vector reconstruction unit 25. Other configurations and operations are the same as those of the decoder according to the embodiment, and a description thereof will be omitted.

In FIG. 31, after the pixel data is output to the prescribed position, the scan history is evaluated in step S66. Based on the result, step S6
At 4, a difference vector table is selected. The pixel data is reconstructed using the selected difference vector table. Other processing is the same as the decoding processing in the embodiment, and a description thereof will be omitted.

Next, a fourth example will be described. In the fourth example, the “position of the pixel of interest in the color space” described with reference to FIG. 14 is used as a trigger. FIG. 32 is a block diagram of the encoder of the fourth example, and FIG. 33 is a flowchart of the encoding process. FIG.
And the components corresponding to those in FIG. 15 are denoted by the same reference numerals. 33, steps corresponding to those in FIG. 16 are denoted by the same reference numerals.

In the fourth example, the color space position evaluator 16d is provided as a trigger generator. The color space position evaluation unit 16d generates a control signal for the table selection unit 17 by looking at the contents stored in the color space memory 15d. Vectors are stored in the color space memory 15d based on the vectors from the candidate evaluation unit 5. The difference vector table selected by the table selection unit 17 is supplied to the candidate evaluation unit 5. Other configurations and operations are the same as those of the encoder according to the embodiment, and a description thereof will be omitted.

In FIG. 33, after the number of candidates reaches the specified number, in step S47, the position in the color space is evaluated with reference to the color space memory 15d. Based on the result, in step S44, a difference vector table is selected. Local decoding is performed based on the selected difference vector table, and a transmission pixel is determined (step S39). Other processing is the same as the encoding processing in the embodiment, and the description thereof is omitted.

FIG. 34 is a block diagram of the decoder of the fourth example, and FIG. 35 is a flowchart of the decoding process. 34, components corresponding to those in FIG. 18 are denoted by the same reference numerals. 35, steps corresponding to those in FIG. 19 are denoted by the same reference numerals.

A color space position evaluation section 35d is also provided on the decoder side as a trigger generation section. The color space position evaluating unit 35d generates a control signal for the table selecting unit 36 by looking at the contents stored in the color space memory 34d. The difference vector table selected by the table selection unit 36 is supplied to the vector reconstruction unit 25. Other configurations and operations are the same as those of the decoder according to the embodiment, and a description thereof will be omitted.

In FIG. 35, after the output position is determined, the position in the color space is evaluated in step S67. Based on the result, in step S64, a difference vector table is selected. The pixel data is reconstructed using the selected difference vector table. Other processing is the same as the decoding processing in the embodiment, and a description thereof will be omitted.

The present invention is not limited to the above-described embodiment of the present invention, and various modifications and applications can be made without departing from the gist of the present invention.

[0101]

According to the present invention, the conversion method in the vector conversion means is changed in accordance with the state of the pixel of interest or the image transmission state based on the transmission method. Many errors can be eliminated. In addition, the image quality itself is improved, and SN
R can be improved.

According to the present invention, there is no need to transmit an additional bit for the switching process. The momentary state of the image is read to form a trigger for switching. Therefore, the present invention has an advantage that the efficiency can be prevented from being reduced by transmitting the additional bits.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an encoder in a transmission scheme proposed by the present applicant.

FIG. 2 is a flowchart showing a flow of an encoding process in a transmission method proposed by the applicant of the present application.

FIG. 3 is a block diagram showing a configuration of a decoder in the transmission method proposed by the present applicant.

FIG. 4 is a flowchart showing the flow of a decoding process in the transmission method proposed by the applicant of the present application.

FIG. 5 is a schematic diagram for explaining a method of jumping over already-transmitted pixels in a transmission method proposed by the present applicant earlier.

FIG. 6 is a schematic diagram for explaining a method of searching for a pixel to be transmitted in the transmission method proposed earlier by the present applicant.

FIG. 7 is a schematic diagram for explaining a color space difference vector in a transmission method proposed earlier by the present applicant.

FIG. 8 is a schematic diagram for explaining a color space difference vector in the transmission method proposed earlier by the present applicant.

FIG. 9 is a schematic diagram for explaining a color space difference vector in the transmission method proposed earlier by the present applicant.

FIG. 10 is a schematic diagram illustrating a situation around a pixel according to the present invention.

FIG. 11 is a schematic diagram illustrating a situation around a pixel according to the present invention.

FIG. 12 is a schematic diagram for explaining the number of candidates as candidates in the present invention.

FIG. 13 is a schematic diagram for explaining a scan history according to the present invention.

FIG. 14 is a schematic diagram illustrating the position of pixel data in a color space according to the present invention.

FIG. 15 is a block diagram illustrating a configuration of an encoder according to an embodiment of the present invention.

FIG. 16 is a flowchart illustrating a flow of an encoding process according to an embodiment of the present invention.

FIG. 17 is a schematic diagram used for describing a candidate selection method and local decoding according to an embodiment of the present invention.

FIG. 18 is a block diagram illustrating a configuration of a decoder according to an embodiment of the present invention.

FIG. 19 is a flowchart showing the flow of a decoding process according to an embodiment of the present invention.

FIG. 20 is a block diagram illustrating a configuration of an encoder when a situation around a pixel is used as a trigger according to an embodiment of the present invention.

FIG. 21 is a flowchart illustrating a flow of an encoding process when a situation around a pixel is used as a trigger according to an embodiment of the present invention.

FIG. 22 is a block diagram illustrating a configuration of a decoder when a situation around a pixel is used as a trigger according to an embodiment of the present invention.

FIG. 23 is a flowchart illustrating a flow of a decoding process when a situation around a pixel is used as a trigger according to an embodiment of the present invention.

FIG. 24 is a block diagram illustrating a configuration of an encoder when a candidate count is used as a trigger according to an embodiment of the present invention.

FIG. 25 is a flowchart showing a flow of an encoding process when the number of candidates is used as a trigger according to an embodiment of the present invention.

FIG. 26 is a block diagram illustrating a configuration of a decoder when a candidate count is used as a trigger according to an embodiment of the present invention.

FIG. 27 is a flowchart showing the flow of a decoding process when the number of candidates is used as a trigger according to an embodiment of the present invention.

FIG. 28 is a block diagram illustrating a configuration of an encoder when a scan history is used as a trigger according to an embodiment of the present invention.

FIG. 29 is a flowchart illustrating a flow of an encoding process when a scan history is used as a trigger according to an embodiment of the present invention.

FIG. 30 is a block diagram illustrating a configuration of a decoder when a scan history is used as a trigger according to an embodiment of the present invention.

FIG. 31 is a flowchart illustrating a flow of a decoding process when a scan history is used as a trigger according to an embodiment of the present invention.

FIG. 32 is a block diagram illustrating a configuration of an encoder when a color space position is used as a trigger according to an embodiment of the present invention.

FIG. 33 is a flowchart showing the flow of an encoding process when a color space position is used as a trigger according to an embodiment of the present invention.

FIG. 34 is a block diagram illustrating a configuration of a decoder when a color space position is used as a trigger according to an embodiment of the present invention.

FIG. 35 is a flowchart showing the flow of a decoding process when a color space position is used as a trigger according to an embodiment of the present invention.

[Explanation of symbols]

5: candidate evaluation unit, 8, 24: scan memory,
9: candidate search unit, 17, 36: table selection unit, 7a to 7d, 26a to 26d: difference vector table, 23: output position determination unit, 16, 35 ...
Trigger generator

Continued on the front page (72) Inventor Ken Satoshi Horishi 6-7-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Inside Sony Corporation (72) Inventor Toru Miyake 6-35-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Sony In-house F term (reference) 5C057 AA01 AA06 CC04 EA02 EA07 ED01 EF05 EL01 GG01 GL00 5C059 LA06 MB06 PP16 RC19 RF09 SS02 SS11 UA31 5J064 AA01 AA02 BA13 BB01 BC01 BC02 BC25 BC27 BD02 BD03

Claims (36)

[Claims] 1. An image signal transmission device, comprising: a candidate selecting means for selecting a pixel of interest in an input image signal and selecting one or more pixels located in a predetermined direction as candidate pixels; Vector conversion means for converting a pixel value of the candidate pixel into a vector based on the peripheral pixel or the pixel of interest, and a pixel to be transmitted based on the pixel value of the candidate pixel converted into the vector. A transmission pixel determining unit that determines within a pixel; and a unit that extracts a feature of an image transmission situation based on the situation of the pixel of interest or a transmission method. An image signal transmission device characterized by changing. 2. The image signal transmission device according to claim 1, wherein a vector value of the pixel value of the candidate pixel is represented by a vector representation of a difference value between a pixel of interest and a pixel to be transmitted next. 3. The image signal transmission device according to claim 1, wherein the pixel value itself is represented by a vector representation as a representation of a vector of pixel values of the candidate pixels. 4. The image signal transmission device according to claim 1, wherein the characteristic is a situation around a pixel. 5. The image signal transmission device according to claim 1, wherein the feature is a scan situation. 6. The image signal transmission device according to claim 1, wherein the number of times the feature is a candidate. 7. The image signal transmission device according to claim 1, wherein the feature is a scan history. 8. The image signal transmission device according to claim 1, wherein the characteristic is a position of the pixel data in a color space. 9. The image signal transmitting apparatus according to claim 8, wherein, when the pixel position is located at an end in the color space, conversion is performed based on a vector conversion table including many possible vectors of the candidate pixel. 10. The image signal transmission device according to claim 1, wherein the conversion method is a conversion based on a vector conversion table. 11. A method for transmitting an image signal, comprising: selecting a pixel of interest in an input image signal and selecting one or more pixels located in a predetermined direction as candidate pixels; A vector conversion step of converting the pixel value of the candidate pixel into a vector based on the peripheral pixel of the target pixel or the target pixel, and a pixel to be transmitted based on the pixel value of the candidate pixel converted into the vector. A transmission pixel determination step for determining within the pixel, and a step of extracting as a feature an image transmission state based on the state of the pixel of interest or a transmission method, and according to the characteristic, the conversion method in the vector conversion step An image signal transmission method characterized by changing. 12. The image signal transmission method according to claim 11, wherein a vector value of the pixel value of the candidate pixel is represented by a vector representation of a difference value between a target pixel and a pixel to be transmitted next. 13. The image signal transmission method according to claim 11, wherein a vector of pixel values of the candidate pixels is represented by a vector representation. 14. The image signal transmission method according to claim 11, wherein the feature is a situation around a pixel. 15. The image signal transmission method according to claim 11, wherein the feature is a scan situation. 16. The image signal transmission method according to claim 11, wherein the feature is the number of times that the feature becomes a candidate. 17. The image signal transmission method according to claim 11, wherein the feature is a scan history. 18. The image signal transmission method according to claim 11, wherein the characteristic is a position of the pixel data in a color space. 19. The image signal transmission method according to claim 18, wherein, when the pixel position is at an end in the color space, conversion is performed based on a vector conversion table including many possible vectors of the candidate pixel. 20. The image signal transmission method according to claim 11, wherein the conversion method is conversion based on a vector conversion table. 21. In an input image signal, a pixel of interest is selected, one or more pixels located in a predetermined direction are selected as candidate pixels, and each pixel is selected based on a peripheral pixel of the candidate pixel or the pixel of interest. The pixel value of the candidate pixel is converted into a vector, and a pixel to be transmitted is determined in the candidate pixel based on the pixel value of the candidate pixel converted into the vector, based on the situation or the transmission method of the target pixel. An image signal receiving apparatus that receives transmission data transmitted from an image signal transmitting apparatus configured to extract the image transmission status as a feature and to change the conversion method in the vector converting means according to the feature, comprising: Receiving means for receiving data, pixel value generating means for generating a pixel value from the received data, and outputting from the received data Output position determining means for determining a position; and means for extracting a characteristic of an image transmission state based on the state of the pixel of interest or a transmission method, and generating a pixel value in the pixel value generating means according to the characteristic. An image signal receiving apparatus characterized by changing a method. 22. The image signal receiving apparatus according to claim 21, wherein a vector value of the pixel value of the candidate pixel is represented by a vector representation of a difference value between a target pixel and a pixel to be transmitted next. 23. The image signal receiving device according to claim 21, wherein the pixel value itself is represented by a vector representation as a representation of the pixel value vector of the candidate pixel. 24. The image signal receiving device according to claim 21, wherein the feature is a situation around a pixel. 25. The image signal receiving apparatus according to claim 21, wherein the feature is a scan situation. 26. The image signal receiving apparatus according to claim 21, wherein the feature is the number of times that the feature is a candidate. 27. The image signal receiving apparatus according to claim 21, wherein the feature is a scan history. 28. The image signal receiving device according to claim 21, wherein the characteristic is a position of the pixel data in a color space. 29. In an input image signal, a pixel of interest is selected, one or more pixels located in a predetermined direction are selected as candidate pixels, and each pixel is selected based on a peripheral pixel of the candidate pixel or the pixel of interest. The pixel value of the candidate pixel is converted into a vector, and a pixel to be transmitted is determined in the candidate pixel based on the pixel value of the candidate pixel converted into the vector, based on the situation or the transmission method of the target pixel. An image signal transmission method for extracting transmission data transmitted from an image signal transmission apparatus, wherein the image transmission state is extracted as a feature and the conversion method in the vector conversion means is changed according to the characteristic. A receiving step of receiving data; a pixel value generating step of generating a pixel value from the received data; An output position determining step of determining an output position from a pixel, and a step of extracting the state of the pixel of interest or an image transmission state based on the transmission method as a feature. According to the feature, the pixel in the pixel value generating step An image signal receiving method characterized by changing a value generating method. 30. The image signal receiving method according to claim 29, wherein a vector value of the pixel value of the candidate pixel is represented by a vector representation of a difference value between the pixel of interest and a pixel to be transmitted next. 31. The image signal receiving method according to claim 29, wherein the pixel value itself is represented by a vector representation as a representation of a vector of the pixel values of the candidate pixels. 32. The image signal receiving method according to claim 29, wherein the characteristic is a situation around a pixel. 33. The image signal receiving method according to claim 29, wherein the feature is a scan situation. 34. The image signal receiving method according to claim 29, wherein the feature is the number of times that the feature becomes a candidate. 35. The image signal receiving method according to claim 29, wherein the characteristic is a scan history. 36. The image signal receiving method according to claim 29, wherein the characteristic is a position of the pixel data in a color space.