JPH09200750A - Data transmission method - Google Patents

Abstract

(57) Abstract: Digital image data is compressed at a high compression rate, and errors at the time of restoration are reduced. Fitting is performed for curved surfaces having different pixel level values in a block of a digital image signal. An optimum curved surface is selected by fitting pixel level values to the average value, flat surface, quadric surface, and cubic surface in the block. In the parameter identification units 3, 4, 5, and 6,
The parameters of the curved surface that minimize the sum of squared errors (coefficients of the equation that defines the curved surface) are identified. Then, the minimum sum of squared error when fitting is determined, and the parameter that minimizes the sum of squared error and the selection code indicating the selected curved surface are transmitted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data transmission method in which image data such as a television signal is compressed and transmitted by a high efficiency code.

[0002]

2. Description of the Related Art There is known a high-efficiency coding method for compressing a data amount when transmitting image data having a large amount of information, such as a digital television signal. As one of the high-efficiency encoding methods, one television image is a two-dimensional area (referred to as a block) composed of a plurality of pixels.
There is known a block coding method of dividing the data into blocks and performing coding for each block.

As a conventional block coding method, an average value Av and a standard deviation σ of a plurality of pixel data (luminance values) in a block are obtained, and one bit is assigned to each pixel in the block. (Av + σ) is “1”, (A
A method is known in which the luminance value of (v−σ) is encoded to “0” and the average value Av of each block and the encoded output of each pixel are transmitted.

[0004]

In the block coding method described above, the image restored on the receiving side becomes a set of mosaics corresponding to the size of the block, and the block distortion generated at the boundary between the blocks becomes conspicuous. . Therefore, when it is desired to obtain a high-quality restored image, it is necessary to reduce the size of the block. There is a disadvantage that the compression ratio is reduced by making the block smaller.

Therefore, an object of the present invention is to provide a transmission method capable of obtaining a good restored image at a high compression rate.

Another object of the present invention is to provide a transmission method having a simple configuration for identifying parameters.

[0007]

SUMMARY OF THE INVENTION The present invention is a transmission method for compressing the data amount of input image data for each block composed of a two-dimensional array of a plurality of pixel data and transmitting the compressed data. A first series that minimizes the sum of squared errors when the image data is converted into a data series having an order arranged for each block and the level values of the pixels in the block are fitted to the curved surface of the first order The first parameter, which is the coefficient of the equation defining the curved surface of the order, is identified for each block, and the sum of squares of the error when fitting the level value of the pixel in the block to the curved surface of the second order is set to the minimum. The second parameter, which is a coefficient of the equation defining the curved surface of the second order, is identified for each block, and the second parameter is determined according to the error and the order of the first parameter and the second parameter, respectively. Select the parameter or the second parameter is a data transmission method characterized by transmitting a first parameter or a second parameter selected.

Each block of pixels has the same block size and is defined by a common (x, y) coordinate. The parameter identification unit 3 fits the luminance values of the pixels in the block using a predetermined curved surface of degree 3, for example, to minimize the prediction error at the time of fitting.
Zero parameters are determined for each block. Further, the parameter identification unit 4 obtains six parameters that minimize the sum of squares of the error from the surface of degree 2.

One of the two parameters is selected according to the error and order of each parameter. The selected parameters are transmitted. In the case of a curved surface of degree 3, only ₁₀ parameters a _{1 to} a ₁₀ need to be transmitted, and if the number of pixels in one block is M,
A compression rate of (10 / M) can be realized, and in the case of a curved surface of degree 2, a compression rate of (6 / M) can be realized. Although different depending on each image, on average, the compression ratio can be made higher than in the case where only a curved surface of degree 3 is used. In addition, since the fitting is performed by using a curved surface as compared with the case where the coding is performed by using the binary coding, the restored image quality can be improved. Further, the coordinate data generated based on the coordinates of the block is common to each block, and the identification of the parameters and the formation of the restored image can be easily performed.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. The description of this embodiment is made in the following order. a. Overall configuration b. Parameter identification c. Parameter identification section d. Judgment unit e. Switch circuit and restoration unit

A. Overall Configuration FIG. 1 shows the configuration of a transmission system composed of a transmission side and a reception side. In FIG. 1, reference numeral 1 is an input terminal for a digital television signal. The input digital television signal is supplied to the blocking circuit 2, and the data order is converted from the television scanning order to the block order. In FIG. 2, F indicates one television image. In the television image F, pixel data is generated from the left end to the right end of the line, and in the vertical direction, pixel data of each line is generated from the upper side to the lower side. In the blocking circuit 2, as shown in FIG. 2, the vertical direction of the image F is divided into, for example, five, and the horizontal direction is divided into, for example, five, whereby 1
Block B is formed. Image (luminance) data is generated from the blocking circuit 2 in the order of the numbers assigned to the blocks B. Overlapping the boundaries between adjacent blocks B is effective for preventing block distortion.

FIG. 3 shows one block B. In one block B, (4N + 1) pixels in the horizontal direction and (2N + 1) M pixels in the vertical direction [M = (4N + 1) × (2N
+1)]. The position of each pixel in the block B is represented by x-axis and y-axis coordinates with the center as the origin o. The luminance values of pixels positioned at the left end of the upper corner in the block B and Z _1, the luminance value of pixel data positioned in a horizontal direction from the data and Z _2, Z ₃ ····,
Let the luminance value of the pixel data located at the lower right corner of block B be Z _M. The coordinates of a pixel having a luminance value of Z _i are represented as (xi _, y _i ).

The four parameter identification units 3, 4, using output surfaces of the blocking circuit 2 using curved surfaces of different orders.
5 and 6 and the determination unit 8. The distribution of all the luminance values included in one block can be approximately expressed by the (x, y) coordinates shown in FIG. Parameter identification unit 3
Is for fitting data in a block with a curved surface of degree 3, that is, a cubic curved surface having two peaks. In this case, the predicted luminance value Z _iで is represented by the following expression.

Z _i ＾ = a ₁ x _i ³ + a ₂ x _i ² y _i + a
₃ x _i y _i ² + a ₄ y _i ³ + a ₅ x _i ² + a ₆ x _i y _i
+ A ₇ y _i ² + a ₈ x _i + a ₉ y _i + a ₁₀

In the above equation, (x _i, y _i ) represents the position of the pixel having the luminance value Z _i in the block, and a _{1 to} a ₁₀ are the coefficients in the above equation, and these coefficients are called parameters.

[0016] In the parameter identification unit 3 identifies the parameter a ₁ ~a _10, such as the square sum of the prediction error for the true value Z _i of the predicted value calculated by the above formula Z _i ^ have is minimized.

The parameter identification unit 4 fits the data in the block with a curved surface of degree 2, that is, a quadratic surface having one peak. In this case, the predicted luminance value Z _i
＾ is represented by the following equation.

Z _i ＾ = a ₅ x _i ² + a ₆ x _i y _i + a ₇
y _i ² + a ₈ x _i + a ₉ y _i + a ₁₀

The parameter identifying unit 4 calculates the predicted value Z _{i in the} above equation.
^ Sum of the squares of the error is the smallest of such parameters a ₅ ~
to identify a _10. The parameters a _{5 to} a ₁₀ of each block from the parameter identification unit 4 are supplied to the selection circuit 7 and the determination unit 8.

The parameter identification unit 5 is a curved surface of order 1
That is, the data in the block is
Data, and in this case, the expected brightness
Degree value Z _i＾ is represented by the following equation.

Z _i ^ = a ₈ x _i + a ₉ y _i + a ₁₀

The parameter identifying unit 6 is for fitting the in-block data on a curved surface of degree 0, that is, a plane. In this case, the predicted brightness value Z _i ^ is expressed by the following equation.

Z _i ^ = a ₁₀

These parameter identification sections 3, 4, 5, 6
The parameters from each of the above are supplied to the selection circuit 7 and the determination unit 8. The quantization bit number of the luminance value is 8 bits,
The number of bits of the parameter is 8 bits, for example.

The determination unit 8 obtains the error variance for each block from the error between the predicted brightness value Z _i ^ and the brightness value Z _i that should be obtained using each parameter, reduces the error variance, and reduces the order. Generate additional code for selecting parameters using curved surface. The additional code is 2 bits, and the selecting circuit 7 performs the following selecting operation according to the additional code.

(00) → The parameter a ₁₀ from the parameter identification unit 6 is selected. (01) → parameter a _8, a from the parameter identification unit 5
Select _9, a ₁₀ . (10) → parameters a _5, a from the parameter identification unit 4
_6, ... a ₁₀ is selected. (11) → parameter a _1, a from the parameter identification unit 3
Select _2, a _3, ... a ₁₀ .

The parameter selected by the selection circuit 7 is taken out to the output terminal 9. Further, the additional code is taken out to the output terminal 10. These parameters and additional codes are transmitted.

The parameters received from the input terminal 11 on the receiving side are supplied to the switch circuit 13, and the input terminal 12
The additional code received from is supplied to the control circuit 14. The additional code indicates the order of the curved surface for each block used for parameter identification. The control circuit 14 decodes this additional code and generates a control signal for controlling the switch circuit 13.

The switch circuit 13 outputs a number of parameters corresponding to the degree of the curved surface used on the transmitting side.
In other words, in the case of the cubic surface, 10 parameters are output from the switch circuit 13, and the quadric surface, the linear surface,
In each case of the zero-order curved surface, six, three, and one parameters are output from the switch circuit 13. Switch circuit 1
3 outputs zero data instead of unnecessary parameters.

The parameters from the switch circuit 13 are supplied to the restoration section 15. In the restoration unit 15, the luminance value is restored by the coordinate data of the block B and the received parameter. The restored data from the restoration unit 15 is the scan conversion circuit 1.
6. The scan conversion circuit 16 restores the order of the data to the scan order of the television signal, and the output terminal 17
The data of the restored image can be obtained.

It should be noted that it is not always necessary to use four curved surfaces for parameter identification, but two curved surfaces having different orders may be used.

B. Parameter Identification In this embodiment, M [= (4N +
1) × (2N + 1)] data is compressed into 1 to 10 parameters, and significant compression can be performed.
Parameters a _{1 to} a made by the parameter identification unit 3
The identification of ₁₀ will be described below.

Luminance values Z ₁ to M of the M pixels in the block B
Z _M is expressed by the following matrix calculation when the prediction error is set to e _1, e ₂ ... E _M when a cubic surface of degree 3 is fitted and predicted.

[0034]

[Equation 1]

The above matrix operation can be rewritten as follows.

[Z] = [W] · [a] + [e]

That is, [Z] is an Mth-order vector, and [W]
Is a (M rows and 10 columns) matrix, [a] is a tenth-order vector, and [e] is an Mth-order vector. The parameter [a] that minimizes the sum of squares of the error obtained by the method of least squares is as follows.

[0038] [a] = ([W ^T] (W]) ^-1 [W ^T], [Z]

Coordinates (x _1, x _2, ...
... x _M, y _1, y _2, ... Y _M ) becomes fixed data when the value of N corresponding to the block size is determined. Therefore,
(Section [W ^T] (W]) ^-1 [W ^T] (coordinate data),
Fixed data common to all blocks, eg RO
Can be generated by M.

[0040] ([W ^T] (W]) ^-1 of the matrix, [(10
(Since it is the product of M rows and 10 columns)], (10 rows and 10 columns), and ([W ^T] (W]) rows and M columns) × the matrix ^-1 [W ^T], [(10 rows and 10 (Column) × (10 rows and M columns)], resulting in a matrix of (10 rows and M columns) Therefore, the formula for obtaining the parameter a by the method of least squares is as follows.

[0041]

[Equation 2]

Regarding the fitting of data in a block, an example of fitting a one-dimensional block with a straight line will be described for easy understanding. One block consists of seven pixels on the same line, and as shown in FIG. 4, the x-coordinate is formed with the center pixel as the origin. When the brightness values Z _1, Z _2, ... Z _{7 of} each pixel have the changes shown in FIG. 4, the brightness value in the block is fitted by the straight line (ax + b) shown by the broken line. Therefore, the luminance value of each pixel is represented by the following equation.

[0043]

(Equation 3)

With respect to the above equation, the parameters a and b are identified by the method of least squares. Not a straight line, but a quadratic curve (a
x ² + bx + c) when fitting with a,
It is necessary to identify the three parameters b and c. In the present invention, since it is a two-dimensional block, the data in the block is fitted by a curved surface instead of a straight line or a curved line.

C. Parameter Identification Unit FIG. 5 shows an example of the parameter identification unit 3. The luminance value Z _k from the blocking circuit 2 is supplied to the input terminal 20. This brightness value Z _k has ten multiplication circuits 21, 22,
... is supplied to each of the 30. Multiplier circuits 21, 22,
... 30 is a luminance value Z _k from a ROM (not shown)
Coordinate data W _k (1), W _k (2), ... W _k (10) are supplied.

The output data of each of the multiplication circuits 21, 22, ... 30 is supplied to one input terminal of each of the addition circuits 31, 32 ,. Adder circuits 31, 32, ...
The output data of each of the 40 registers 41, 42, ...
50, and output terminals 51, 52, ...
It is taken out to 60. An integrating circuit is constituted by the pair of the adding circuit and the register. Registers 41, 42, ...
50 is reset every time the calculation of M luminance values of one block is completed.

When M luminance values are supplied to the input terminal 20, the output data obtained at the output terminal 51 is [W ₁ (1) Z
_{1 + W 2 (1) Z} 2 + ··· + W M (1) Z M = a 1 ], and the parameters a ₁ is generated. Similarly, output terminal 5
2, 53, ... 60 respectively, parameters a _2, a _3, ...
・ A ₁₀ is taken out.

The parameter identifying section 4 identifies the six parameters a _5, a _6, ... A ₁₀ and has the configuration shown in FIG. It has a configuration similar to that of the parameter identification unit 3 described above, and a configuration of a unit for generating one parameter includes a multiplication circuit and an integration circuit.

FIG. 7 and FIG. 8 respectively show a parameter identifying section 5 for identifying the three parameters a _8, a _9, a ₁₀ and a parameter identifying section 6 for identifying one parameter a ₁₀ . Each circuit shown in each of FIGS. 6, 7 and 8 is provided with a reference numeral corresponding to FIG.

D. Judgment Unit The judgment unit 8 is a parameter that has a small error variance (σ) among the parameters identified by the parameter identification units 3, 4, 5, and 6 and that the degree of the curved surface used for parameter identification is as low as possible. To judge.

FIG. 9 shows an example of the structure of the judging section 8. In FIG. 9, image data from the blocking circuit 2 is supplied to an input terminal indicated by 61. This image data is used as a square circuit 62, an integrating circuit 73, and nine multiplying circuits 63, 64, 6
5, 66, 67, 68, 69, 70, 71. When the i-th luminance value Z _i of one block is supplied to the input terminal 61, coordinate data (x _i, y _i , x _i ² , x _i y _i , y _i ² , x _i ³ , x _i ²
y _i , x _i y _i ² , y _i ³ ) are supplied.

The output of the squaring circuit 62 is supplied to the integrating circuit 72. The outputs of the multiplying circuits 63 to 71 are integrated by the integrating circuit 7.
4,75,76,77,78,79,80,81,82
Is supplied to. The integrating circuit 72 to which the output of the squaring circuit 62 is supplied, the integrating circuit 73 to which the luminance value Z _i is supplied, and the integrating circuits 74 to 82 to which the above-described multiplication output is supplied correspond to the number of pixels in one block. Perform M times of integration (adding the input and the previous addition output). These integration circuits 7
Outputs 2 to 82 are supplied to arithmetic circuits 83, 84, 85, 86 for selectively obtaining error variance.

The arithmetic circuit 83 calculates the error variance σ ₀ ² when a 0th-order curved surface (flat surface) is used. The arithmetic circuit 84 is 1
Calculate the error variance σ ₁ ² when using a quadric surface (plane). The arithmetic circuit 85 uses the error variance σ when using a quadric surface.
Calculate ₂ ² . The arithmetic circuit 86 calculates the error variance σ ₃ ² when the cubic curved surface is used. These calculated error variances σ ₀ ^{2 to} σ ₃ ² are supplied to the determination circuit 87. The determination circuit 87 generates a control signal at the output terminal 88 such that the selection circuit 7 selects an appropriate parameter according to the error variance and the degree of the curved surface used for parameter identification.

That is, when one of the plurality of parameters is selected, the degree of the curved surface used for the fitting is considered together with the value of the error variance. In general,
Although there is a relationship of (σ ₃ ² <σ ₂ ² <σ ₁ ² <σ ₀ ² ), the compression ratio r has a relationship of (r ₃ > r ₂ > r ₁ > r ₀ ). For an image (block) that changes drastically, the above-described relationship of the magnitude of the error variance is established. In the case of a flat image (block), the error variance has the same relationship, but there is almost no difference between them. When the error variances are almost the same, a parameter obtained by fitting using a low-order curved surface capable of increasing the compression ratio r is selected.

Error variance σ ₀ ² _, σ ₁ ² _, σ ₂ ² _and σ ₃ ²
Each of the calculation methods will be described below. (I) Calculation of σ ₀ ^{2 From} (Z _i ^ = a ₁₀ ), the error variance is given by the following equation, where M is the number of pixels in one block.

Σ ₀ ² = (1 / M) · Σ (Z _i −a ₁₀ ) ² = (1 / M) · Σ (Z _i ) ² −2a ₁₀ (1 / M) Σ (Z
_i ) + a ₁₀ ²

Therefore, while the parameter a ₁₀ is being identified, the sum of squares of the brightness value (Σ (Z _i ) ² ) and the integrated value of the brightness values (Σ (Z _i )) are calculated by the squaring circuit 62 and the integrating circuits 72 and 73. The error variance σ ₀ ² is calculated by the arithmetic circuit 83 according to the above equation using the parameter a ₁₀ calculated and calculated by

Here, by the method of least squares, (δσ ² / δ
a = 0, δ means partial differentiation), (a ₁₀ =
Since it is uniquely determined as (1 / M) · Σ (Z _i )), the error variance σ ₀ ² is given by the following equation.

Σ ₀ ² = (1 / M) · Σ (Z _i ) ² −
[(1 / M) Σ (Z _i )] ²

(Ii) Calculation of σ ₁ ^{2 From} (Z _i ^ = a ₈ x _i + a ₉ y _i + a ₁₀ ), the error variance is given by the following equation.

Σ ₁ ² = (1 / M) · Σ (Z _i −a ₈ x _i −a ₉ y _i −a ₁₀ ) ² = (1 / M) · Σ (Z _i ² ) + a ₁₀ ² + a ₈ ² (1 / M) Σ (x _i ² ) + a ₉ ² (1 / M) Σ (y _i ² ) -2a ₁₀ (1 / M) Σ (Z _i ) -2a ₈ (1 / M) Σ (x _i · Z _i ) −2a ₉ (1 / M) Σ (y _i · Z _i ) + 2a ₁₀ a ₈ (1 / M) Σ (x _i ) + 2a ₁₀ a ₉ (1 / M) Σ (y _i ) + 2a ₈ a ₉ (1 / M) Σ (x _i y _i )

In the above equation, Σ (x _i ² ), Σ (y _i ² ),
Σ (x _i y _i ), Σ (x _i ), Σ (y _i ) are constants uniquely determined from the coordinate data and the number of pixels M, and need not be calculated. Therefore, Σ (Z _i ² ), Σ (Z _i ), Σ (Z _i ·
x _i ), Σ (Z _i · y _i ) identified parameters a _8,
The error variance σ ₁ ² is obtained using a _{9 and} a ₁₀ . Σ (Z _i · x _i ) is obtained by the multiplication circuit 63 and the integration circuit 74, and Σ (Z _i
i · y _i ) is required.

[0063] Of course, also in this case, the parameter a _8, a _9, a
₁₀ is Σ (Z _i ² ), Σ (Z _i ), Σ (Z _i · x _i ), Σ
Since it is obtained from (Z _i · y _i ), σ ₁ ² can be obtained directly from these four integrated results.

(Iii) Calculation of σ ₂ ² (Z _i ^ = a ₅ x _i ² + a ₆ x _i y _i + a ₇ y _i ² + a
_{From 8} x _i + a ₉ x _i + a ₁₀ ), Σ (Z
_i ² ), Σ (Z _i ), Σ (Z _i · x _i ), Σ (Z _i · y _i ),
Σ (Z _i · x _i ² ), Σ (Z _i · x _i y _i ), Σ (Z _i ·
By calculating y _i ² ), the error variance σ ₂ ² when the quadric surface is used is calculated. Σ (Z _i · x _i ² ) is obtained by the multiplication circuit 65 and the integration circuit 76, and Σ (Z _i
X _i y _i ) is obtained by the multiplication circuit 66 and the integration circuit 77, and Σ (Z _i · y _i ² ) is obtained by the multiplication circuit 67 and the integration circuit 78. 7 integrating circuits 72, 7
From the outputs of 3, ... 77, the arithmetic circuit 85 calculates the error variance σ ₂ ² .

(Iv) Calculation of σ ³ (Z _i ^ = a ₁ x _i ³ + a ₂ x _i ² y _i + a ₃ x _i y _i
² + a ₄ y _i ³ + a ₅ x _i ² + a ₆ x _i y _i + a ₇ y _i
² + a ₈ x _i + a ₉ y _i + a ₁₀ )
Σ (Z _i ² ), Σ (Z _i ), Σ (Z _i · x _i ), Σ (Z _i ·
y _i ), Σ (Z _i · x _i ² ), Σ (Z _i · x _i y _i ), Σ (Z
_i · y _i ² ), Σ (Z _i · x _i ³ ), Σ (Z _i · x _i ²
By calculating y _i ), Σ (Z _i · x _i y _i ² ), Σ (Z _i · y _i ³ ), the error variance σ ₃ ² when the cubic surface is used is calculated. Σ (Z _i · x _i ³ ) is obtained by the multiplication circuit 68 and the integration circuit 79. Similarly, Σ (Z _i · x _i
² y _i ), Σ (Z _i · x _i y _i ² ), Σ (Z _i · y _i ³ ) are multiplication circuits 69, 70, 71 and integration circuits 80, 8 respectively.
1, 82 and. Integrating circuits 72, 73, 7
The respective outputs of 4, 75, 79, 80, 81, 82 are supplied to the arithmetic circuit 86, and the error variance σ is calculated by the arithmetic circuit 86.
₃ ² is calculated.

E. Switch circuit and restoration unit Switch circuit 13 and restoration unit 15 provided on the receiving side
Is configured as shown in FIG. 10 as an example. FIG.
In, the parameters (1, 3, 6, or 10 parameter data) received from the input terminals 91, 92, ... 100 are supplied. Switch circuit 13
Are controlled by a control signal formed from the received additional code to select zero data instead of a parameter independent input. The output data of the switch circuit 13 is supplied to the restoration unit 15.

Although not shown, the restoration section 15 is provided with a ROM, from which different coordinate data M
Occurs several times. While this coordinate data is generated M times, one block of the received parameters is supplied from the switch circuit 13, and M restored luminance values are sequentially generated at the output terminal 121.

Parameters a _1, a from the switch circuit 13
_2, ... A ₁₀ are supplied to the multiplication circuits 101, 102, ... 109. For example, the coordinate data [x _i ³ , x _i ² y _i, x _i y _i ² , ... Related to the coordinates (x _i, y _i ) of the i-th pixel.
y _i ] are supplied to the multiplication circuits 101, 102, ... 109, respectively. The outputs of the multiplying circuits 101, 102, ... 109 are added by the adding circuits 112, 113 ,. Therefore, the restored luminance value Z is output to the output terminal 121 derived from the final stage adder circuit 120.
_i ^ is obtained.

[0069]

According to the present invention, a curved surface of a plurality of orders prepared in advance for each block is fitted and well fitted, and a parameter of a curved surface which can increase the compression rate is selected and transmitted. Therefore, according to the present invention, a very large compression rate can be obtained, and the brightness value is expressed by various curved surfaces, so that the image quality is significantly improved as compared with the conventional binary block coding. can do.

Further, according to the present invention, when the parameters are identified, the coordinate data common to the blocks are used. Therefore, by generating the coordinate data by, for example, the ROM, the parameter identifying unit can be easily performed only by the multiplication and addition processes. Can be configured.

[Brief description of drawings]

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

FIG. 2 is a schematic diagram for explaining blocking.

FIG. 3 is a schematic diagram for explaining a block configuration.

FIG. 4 is a schematic diagram for facilitating the understanding of the fitting according to the present invention.

FIG. 5 is a block diagram of a specific example of a parameter identification unit 3 in one embodiment of the present invention.

FIG. 6 is a block diagram of a specific example of a parameter identification unit 4 according to an embodiment of the present invention.

FIG. 7 is a block diagram of a specific example of a parameter identification unit 5 in one embodiment of the present invention.

FIG. 8 is a block diagram of a specific example of a parameter identification unit 6 according to an embodiment of the present invention.

FIG. 9 is a block diagram used to explain a determination unit in one embodiment of the present invention.

FIG. 10 is a block diagram illustrating an example of a restoration unit according to an embodiment of the present invention;

[Explanation of symbols]

2 ... Blocking circuit, 3, 4, 5, 6 ... Parameter identification section, 7 ... Selection circuit, 8 ... Judgment section, 13 ... Switch circuit, 15 ... Restoration section,
16 ... Scan conversion circuit

Claims (1)

[Claims] 1. A data amount of input image data is compressed for each block composed of a two-dimensional array of a plurality of pixel data,
In the transmission method for transmitting the compressed data, the input image data is converted into a data series having an order collected for each block, and a level value of a pixel in the block is converted to a curved surface of a first degree. The first parameter, which is the coefficient of the equation defining the curved surface of the first order that minimizes the sum of squares of the error when fitting is identified for each block, and the level value of the pixel in the block is A second parameter, which is a coefficient of an equation that defines the second-order curved surface that minimizes the sum of squares of errors when fitting to a second-order curved surface, is identified for each block, and the first parameter The first parameter or the second parameter is selected according to the error and the order of the parameter of Data transmission method, wherein the transmission of the parameters or the second parameters.