JP2841314B2 - Data transmission method and transmission device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention
Image data such as a signal is compressed by a high-efficiency code and transmitted.
Data transmission methodAnd transmission equipmentAbout. [0002] 2. Description of the Related Art Image data having a large amount of information such as digital data
When transmitting a television signal, the data amount is compressed.
A highly efficient coding method is known. High efficiency coding method
One method is to convert one television image into multiple
Two-dimensional area composed of elements (referred to as block)
Block coding method that divides the data into blocks and codes each block
The law is known. [0003] Conventional block coding methods include block
Average value Av of a plurality of pixel data (brightness values) in the
The standard deviation σ is calculated and one pixel is calculated for each pixel in the block.
And the luminance value of (Av + σ) is set to “1”, (Av + σ)
v-σ) is coded to “0”, and the
The method of transmitting the average value Av and the encoded output of each pixel is known.
Have been. [0004] SUMMARY OF THE INVENTION The above block coding
The method is that the image restored on the receiving side is
A set of corresponding mosaics, starting at the boundaries between blocks
The resulting block distortion becomes conspicuous. Therefore, the restored image
When you want to get a good image, the size of the block
Need to be smaller. Making blocks smaller
Therefore, there was a disadvantage that the compression ratio was lowered. Accordingly, an object of the present invention is to provide a high compression ratio
Transmission method that can obtain good restored imagesAnd transmission
apparatusIs to provide. Another object of the present invention is to identify parameters.
Configuration to provide a simple transmission method.
You. [0007] [Means for Solving the Problems]Claim 1The invention
Input for each block composed of a two-dimensional array of pixel data
Compress the amount of force image data and compress the compressed data
In the transmission method of transmitting, block the input image data
The data is converted into a data sequence having the
The level values of the pixels in the block are checked against the first-order surface.
1st which minimizes the sum of squares of the error when performing the fitting
The first parameter, which is the coefficient of the equation that defines the surface of order
Data is identified for each block, and the level values of the pixels in the block are
To the surface of the second order
Defines a second-order surface that minimizes the sum of squares of the error
Identify the second parameter, which is the coefficient of the equation, for each block
And each of the first parameter and the second parameterMistake
Difference valueAnd degreeAnd based on the comparison resultThe first
Select a parameter or a second parameter and select
The first or second parameter
The data transmission method is characterized in that:Also,
As described above, according to the invention, the input image data is
Data volume and transmit the compressed data.
Transmission device. Each block of a plurality of pixels is equal
With a block size and common (x, y) coordinates
Stipulated. In the parameter identification unit 3, the image in the block
The elementary luminance value is fitted to a predetermined surface of degree 3, for example.
1 that minimizes the prediction error during fitting
Zero parameters are determined for each block. Also,
The parameter identification unit 4 uses the curved surface of order 2 to calculate the error itself.
Six parameters that minimize the sum of squares are obtained. [0009] Between these two parameters, each parameter
One parameter is selected according to the error and order of the
You. The selected parameters are transmitted. Degree 3
In the case of a curved surface, ten parameters a₁~ A_TenOnly
If the number of pixels in one block is M,
(10 / M), and a degree 2
For curved surfaces, a compression ratio of (6 / M) can be achieved.
it can. Depends on each image, but on average,
Higher compression ratio than when only the curved surface 3 is used
it can. Also, compared to coding with binary,
The restoration quality should be good
Can be. Furthermore, it is generated based on the coordinates of the block.
The coordinate data is common to each block, and the
Identification and formation of a restored image can be easily performed. [0010] DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below.
This will be described with reference to the drawings. The description of this embodiment is as follows:
Is made according to the order of. 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 a configuration of a transmission system including a transmission side and a reception side.
In FIG. 1, 1 is a digital television signal.
Signal input terminal. Input digital television signal
Is supplied to the blocking circuit 2 and the order of the data is
It is converted from the order of John scanning to the order of blocks. Figure
In 2, F indicates one television image. Te
In the revision image F, from the left end of the line to the right end
Pixel data is generated, and in the vertical direction,
As a result, pixel data of each line is generated. Blocking
In the circuit 2, as shown in FIG.
For example, by dividing into five and dividing the horizontal direction into, for example, five,
Block B is formed. The order of the numbers attached to each block B
First, image (luminance) data is generated from the blocking circuit 2
You. The boundaries between adjacent blocks B overlap
It is effective to prevent block distortion. FIG. 3 shows one block B and one block B.
B has (4N + 1) pixels in the horizontal direction and
(2N + 1) M [M = (4N + 1) × (2N
+1)]. Each pixel in this block B
Is determined by the coordinates on the x-axis and y-axis with the center as the origin o.
Is represented. Located in the upper left corner of block B
The brightness value of the pixel₁And from this data
The luminance value of the pixel data located at_Two, Z_Three····age,
Pixel data located at the lower right corner of block B
The luminance value of_MAnd Z_iCoordinates of a pixel with a luminance value of
Is (x_i,y_i). The output signals of the blocking circuit 2 are different from each other.
Parameter identification sections 3, 4, using curved surfaces of different orders
5 and 6 and the determination unit 8. Included in one block
The distribution of all luminance values is represented by the (x, y) coordinates shown in FIG.
It can be approximately expressed. Parameter identification unit 3
Is a cubic surface of degree 3, ie, a cubic surface with two peaks.
This is for fitting the data in the lock, in this case
, The predicted luminance value Z_i＾ is represented by the following equation. Z_i＾ = a₁x_i ^Three+ A_Twox_i ^Twoy_i+ A
_Threex_iy_i ^Two+ A_Foury_i ^Three+ A_Fivex_i ^Two+ A₆x_iy_i
+ A₇y_i ^Two+ A₈x_i+ A₉y_i+ A_Ten In the above equation, (x_i,y_i) Is the luminance value Z_iPixel
Indicates the position in the block of₁~ A_TenIs the coefficient of the above equation
And such a coefficient is referred to as a parameter. The parameter identification unit 3 calculates
Predicted value Z to be issued_iTrue value Z of ＾_iPrediction error for
Parameter a that minimizes the sum of squares of₁~ A_TenThe same
Set. The parameter identification unit 4 generates a surface of degree 2, that is,
Filter data in a block on a quadric surface with one mountain
In this case, the predicted luminance value Z_i
＾ is represented by the following equation. Z_i＾ = a_Fivex_i ^Two+ A₆x_iy_i+ A₇
y_i ^Two+ A₈x_i+ A₉y_i+ A_Ten The parameter identification section 4 calculates the predicted value Z_i
Parameter a that minimizes the sum of squares of the error of ＾_Five~
a_TenIs identified. Each block from the parameter identification unit 4
Parameter a_Five~ A_TenAre supplied to the selection circuit 7 and the determination unit 8.
Be paid. The parameter identification unit 5 calculates the degree 1 surface
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＾ = a₈x_i+ A₉y_i+ A_Ten The parameter identification unit 6 calculates the degree 0 surface
That is, the data in the block is fitted with a plane
In this case, the predicted luminance value Z_i＾ is represented by the following formula
Becomes Z_i＾ = a_Ten These parameter identification sections 3, 4, 5, 6
Are supplied to the selection circuit 7 and the determination unit 8.
Be paid. The number of quantization bits of the luminance value is set to 8 bits,
The number of bits of the parameter is, for example, 8 bits. The judgment section 8 is obtained by using each parameter.
Expected luminance value Z to be obtained_i＾ and luminance value Z_iError from
Find the error variance for each
In order to select parameters using a low-order surface
Generates an additional code. This additional code is 2 bits
And the selection circuit 7 performs the following according to the additional code.
Perform a selection operation. (00) → Parameter from the parameter identification unit 6
Meter a_TenSelect (01) → Parameter a from the parameter identification unit 5_8,a
_9,a_TenSelect (10) → Parameter a from parameter identification unit 4_Five,a
_6,... a_TenSelect (11) → Parameter a from the parameter identification unit 3_1,a
_2,a_3,... a_TenSelect The parameter selected by the selection circuit 7 is
It is taken out to the output terminal 9. Also, the additional code is
The child 10 takes it out. These parameters and additional commands
Mode is transmitted. The packet received from the input terminal 11 on the receiving side is
The parameter is supplied to the switch circuit 13 and the input terminal 12
Is supplied to the control circuit 14.
You. Additional code is used for each block used for parameter identification.
Shows the order of the curved surface. The control circuit 14
Control signal for decoding the code and controlling the switch circuit 13
Occurs. From the switch circuit 13, the signal used on the transmission side is used.
The number of parameters corresponding to the degree of the curved surface is output.
In other words, in the case of a cubic surface, ten parameters
The output from the switch circuit 13 is a secondary surface, a primary surface,
For each of the 0th order surfaces, 6, 3, and 1 parameters
Is output from the switch circuit 13. Switch circuit 1
3 outputs zero data instead of unnecessary parameters
You. The parameters from the switch circuit 13 are restored
It is supplied to the unit 15. In the restoration unit 15, the seat of the block B
Luminance value is restored based on target data and received parameters
Is done. The restoration data from the restoration unit 15 is used for the scan conversion circuit 1
6. Data order by scan conversion circuit 16
Is returned to the scanning order of the television signal, and the output terminal 17
The data of the restored image is obtained. The surface used for parameter identification is
Therefore, it is not always necessary to use four
Instead, both may be used. B. Parameter identification In this embodiment, M [= (4N +
1) × (2N + 1)] data is 1 to 10 packets.
It is compressed into parameters and can perform significant compression.
Parameter a made by the parameter identification unit 3₁~ A
_TenIs described below. The luminance value Z of the M pixels in the block B₁~
Z_MPredicts by fitting a cubic surface of degree 3
The prediction error is e_1,e_Two.... e_MThen
Indicated by matrix operations. [0034] (Equation 1) The above matrix operation can be rewritten as follows:
Can be [Z] = [W] · [a] + [e] That is, [Z] is an M-order vector, [W]
Is a matrix of (M rows and 10 columns), and [a] is a 10th-order vector
And [e] are M-order vectors. The least squares method
Parameter that minimizes the sum of squares of the error found by
[A] is given by the following equation. [0038] [A] = ([W^T] (W))^-1[W^T] ・ [Z] The coordinates (x_1,x_2,・ ・
... x_M,y_1,y_2,.... y_M) Corresponds to the block size
Once the value of N is determined, the data becomes fixed data. Therefore,
([W^T] (W))^-1[W^T] (Coordinate data)
It becomes fixed data common to all blocks, for example, RO
M. ([W^T] (W))^-1The matrix of [(10
Row M columns) × (M rows 10 columns)], so (10 rows
10 rows) and ([W^T] (W))^-1[W^T] Matrix
Is the product of [(10 rows and 10 columns) × (10 rows and M columns)]
From (10 rows and M columns). Therefore, the parameter
The equation for obtaining the data a by the least squares method is as follows:
become. [0041] (Equation 2) About the fitting of the data in the block
For easy understanding, one-dimensional blocks are
An example of setting will be described. One block is the same
It consists of seven pixels on the line, as shown in FIG.
An x coordinate is formed with the pixel of the heart as the origin. Brightness of each pixel
Degree value Z_1,Z_2,... Z₇Has the changes shown in FIG.
And a straight line (ax + b) indicated by a broken line,
The luminance value is fitted. Therefore, the brightness of each pixel
The value is represented by the following equation. [0043] (Equation 3) With respect to the above equation, parameters are calculated by the least square method.
Data a and b are identified. Not a straight line but a quadratic curve (a
x^Two+ Bx + c), a,
It is necessary to identify the three parameters b and c
You. In the present invention, since it is a two-dimensional block, a straight line or
Data in blocks using curved surfaces instead of curves.
It is something to dig. C. Parameter identification section FIG. 5 shows an example of the parameter identification unit 3. Show at 20
The luminance value Z from the blocking circuit 2 is input to the input terminal._kIs supplied
It is. This luminance value Z_kAre ten multiplication circuits 21, 22,
.. Are supplied to each of. Multiplication circuits 21, 22,
.., 30 has a luminance value Z from a ROM (not shown)._k
Coordinate data W_k(1), W_k(2) ・ ・ ・ ・ ・ ・ W_k(10) supplied
Is done. Each of the multipliers 21, 22,... 30,
Is one of the adders 31, 32,... 40
Is supplied to the input terminal of. Adder circuits 31, 32, ...
The output data of each of the registers 40 are registers 41, 42,.
, And output terminals 51, 52,.
It is taken out to 60. Addition circuit and register pair
To form an integrating circuit. Registers 41, 42, ...
50 is each time the calculation of M luminance values of one block is completed.
Is reset to The M luminance values are supplied to the input terminal 20
And the output data obtained at the output terminal 51 is [W₁(1) Z
₁+ W_Two(1) Z_Two+ ... + W_M(1) Z_M= A₁]
Parameter a₁Is generated. Similarly, output terminal 5
The parameter a is assigned to each of 2, 53,._2,a_3,・ ・
・ A_TenIs taken out. The parameter identification section 4 has six parameters.
a_Five,a_6,..A_TenThe configuration shown in FIG.
Have been. It has the same configuration as the parameter identification unit 3 described above.
And the configuration of the unit for generating one parameter is
It consists of an arithmetic circuit and an integrating circuit. Three parameters a_8,a_9,a_TenIdentify
Parameter identification unit 5 and one parameter a_TenIdentify
The parameter identification unit 6 is shown in FIGS. 7 and 8, respectively. Figure
6, FIG. 7 and FIG.
Corresponding reference numerals have been applied. D. Judgment unit The determination unit 8 determines whether each of the parameter identification units 3, 4, 5, and 6 has
Smaller error variance (σ) among the more identified parameters
At the same time, the order of the surface used for parameter identification is
Determine the lowest possible parameter. FIG. 9 shows an example of the configuration of the determination unit 8. Figure
In 9, the input circuit indicated by 61 is connected to the blocking circuit 2
These image data are supplied. This image data is squared
The path 62, the integrating circuit 73, and the nine multiplying circuits 63, 64, 6
5, 66, 67, 68, 69, 70, 71
You. The i-th luminance value Z of one block is input to the input terminal 61._iBut
When supplied, the coordinate data is stored in each of the multiplication circuits 63 to 71.
TA (x_i,y_i, X_i ^Two, x_iy_i, Y_i ^Two , X_i ^Three, x_i ^Two
y_i, x_iy_i ^Two , Y_i ^Three ) Is supplied. The output of the squaring circuit 62 is supplied to the integrating circuit 72.
Is done. The output of each of the multiplication circuits 63 to 71 is
4,75,76,77,78,79,80,81,82
Supplied to The number of times the output of the squaring circuit 62 is supplied
Path 72, brightness value Z_iIs supplied to the integrating circuit 73 and above
The multiplying circuits 74 to 82 to which the multiplied outputs described above are supplied include one block.
M multiplications (input and previous additions) corresponding to the number of pixels in the lock
(Adding the calculated force) operation. These integrating circuits 7
An operation for selectively obtaining the error variance of the outputs of 2-82
The circuits 83, 84, 85, 86 are supplied. The arithmetic circuit 83 uses a zero-order curved surface (plane).
Error variance σ₀ ^TwoIs calculated. The arithmetic circuit 84
Error variance σ when using a secondary curved surface (plane)₁ ^TwoCalculate
You. The arithmetic circuit 85 calculates the error variance σ when using a quadratic surface.
_Two ^TwoIs calculated. The arithmetic circuit 86 uses a cubic surface
Error variance σ_Three ^TwoIs calculated. These determined errors
Variance σ₀ ^Two~ Σ_Three ^TwoIs supplied to the determination circuit 87. Judgment
The circuit 87 calculates the error variance and the surface used for parameter identification.
The selection circuit 7 selects an appropriate parameter according to the order.
Such a control signal is generated at the output terminal 88. That is, one of a plurality of parameters is selected.
Option, along with the error variance value, is used for fitting.
The order of the used curved surface is also considered. In general,
(Σ_Three ^Two<Σ_Two ^Two<Σ₁ ^Two<Σ₀ ^Two), But the pressure
The reduction ratio r is (r_Three> R_Two> R₁> R₀) Has a relationship. Strange
For images (blocks) that are severely transformed, the above error
Although the relationship of the size of the scatter is established, a flat image (block
G), the error variance has a similar relationship, but the difference
Almost disappears. If the error variances are similar, the compression ratio
Fitting using a low-order surface that can increase r
Then, the obtained parameter is selected. Error variance σ₀ ^Two _,σ₁ ^Two _,σ_Two ^Two _,σ_Three ^Two
Each calculation method will be described below. (I) σ₀ ^TwoCalculation of (Z_i＾ = a_Ten), The error variance is M within one block
The number of pixels is given by the following equation. [0056] σ₀ ^Two= (1 / M) Σ (Z_i-A_Ten)^Two = (1 / M) Σ (Z_i)^Two-2a_Ten(1 / M) Σ (Z
_i) + A_Ten ^Two Therefore, the parameter a_TenWhile identifying
From the sum of squares of the luminance value_i)^Two ) And integrated luminance value
(Σ (Z_i)) Into the squaring circuit 62 and the integrating circuits 72 and 73.
The calculated and identified parameter a_TenUsing
According to the equation, the error variance σ₀ ^TwoIs obtained by the arithmetic circuit 83
Can be Here, by the least square method, (δσ^Two/ Δ
When a = 0 and δ mean partial differentiation), (a_Ten=
(1 / M) · Σ (Z_i))
Variance σ₀ ^TwoIs represented by the following equation. Σ₀ ^Two= (1 / M) Σ (Z_i)^Two−
[(1 / M) Σ (Z_i)]^Two (Ii) σ₁ ^TwoCalculation of (Z_i＾ = a₈x_i+ A₉y_i+ A_Ten), Error variance
Is represented by the following equation. [0061] σ₁ ^Two= (1 / M) Σ (Z_i-A₈x_i-A₉y_i-A_Ten)^Two = (1 / M) Σ (Z_i ^Two) + A_Ten ^Two+ A₈ ^Two(1 / M) Σ (x_i ^Two) + A₉ ^Two(1 / M) Σ (y_i ^Two) -2a_Ten(1 / M) Σ (Z_i) -2a₈(1 / M) Σ (x_i・ Z_i) -2a₉(1 / M) Σ (y_i・ Z_i) + 2a_Tena₈(1 / M) Σ (x_i) + 2a_Tena₉(1 / M) Σ (y_i) + 2a₈a₉(1 / M) Σ (x_iy_i) In the above equation, Σ (x_i ^Two), Σ (y_i ^Two),
Σ (x_iy_i), Σ (x_i), Σ (y_i) Is the coordinate data
It is a constant that is uniquely determined from the number of pixels M and must be determined.
No need. Therefore, Σ (Z_i ^Two), Σ (Z_i), Σ (Z_i・
x_i), Σ (Z_i・ Y_i) And the parameter a identified_8,
a_9,a_TenAnd the error variance σ₁ ^TwoIs required. Multiplication times
By the path 63 and the integrating circuit 74, Σ (Z_i・ X_i)
乗 算 (Z)
_i・ Y_i) Is required. Of course, also in this case, the parameter a_8,a_9,a
_TenIs Σ (Z_i ^Two), Σ (Z_i), Σ (Z_i・ X_i), Σ
(Z_i・ Y_i)), These four integration results
From the result σ₁ ^TwoIs obtained directly. (Iii) σ_Two ^TwoCalculation of (Z_i＾ = a_Fivex_i ^Two+ A₆x_iy_i+ A₇y_i ^Two+ A
₈x_i+ A₉x_i+ A_Ten ), Σ (Z
_i ^Two), Σ (Z_i), Σ (Z_i・ X_i), Σ (Z_i・ Y_i),
Σ (Z_i・ X_i ^Two), Σ (Z_i・ X_iy_i), Σ (Z_i・
y_i ^Two), The error in using a quadratic surface
Difference variance σ_Two ^TwoIs calculated. Σ (Z_i・ X_i ^Two) Is the power
Calculated by the arithmetic circuit 65 and the integrating circuit 76, Σ (Z_i
・ X_iy_i) Is calculated by the multiplication circuit 66 and the accumulation circuit 77.
Required, Σ (Z_i・ Y_i ^Two) Indicates the multiplication circuit 67 and the product
It is obtained by the arithmetic circuit 78. 7 integrating circuits 72, 7
From the outputs of 3,... 77, the arithmetic circuit 85 calculates the error variance
σ_Two ^TwoIs calculated. (Iv) σ^ThreeCalculation of (Z_i＾ = a₁x_i ^Three+ A_Twox_i ^Twoy_i+ A_Threex_iy_i
^Two+ A_Foury_i ^Three+ A_Fivex_i ^Two+ A₆x_iy_i+ A₇y_i
^Two+ A₈x_i+ A₉y_i+ A_Ten), As above,
Σ (Z_i ^Two), Σ (Z_i), Σ (Z_i・ X_i), Σ (Z_i・
y_i), Σ (Z_i・ X_i ^Two), Σ (Z_i・ X_iy_i), Σ (Z
_i・ Y_i ^Two), Σ (Z_i・ X_i ^Three), Σ (Z_i・ X_i ^Two
y_i), Σ (Z_i・ X_iy_i ^Two), Σ (Z_i・ Y_i ^Three)
Error variance σ when using a cubic surface_Three ^TwoBut
Is calculated. Σ (Z_i・ X_i ^Three) Indicates the multiplication circuit 68 and the product
It is obtained by the arithmetic circuit 79. Similarly, Σ (Z_i・ X_i
^Two y_i), Σ (Z_i・ X_iy_i ^Two ), Σ (Z_i・ Y_i ^Three)husband of
Are multiplication circuits 69, 70, 71 and integration circuits 80, 8
1, 82. Integration circuits 72, 73, 7
4,75,79,80,81,82 output is calculated
The error variance σ is supplied to the
_Three ^TwoIs calculated. E. Switch circuit and restoration unit Switch circuit 13 and restoration unit 15 provided on reception side
Is configured as shown in FIG. 10 as an example. FIG.
, Input terminals 91, 92,... 100
Parameters (1, 3, 6, or 10
Is supplied. Switch circuit 13
Is controlled by a control signal formed from the received additional code.
Select zero data in place of inputs unrelated to parameters.
Is controlled to be selected. Output of this switch circuit 13
The data is supplied to the restoration unit 15. Although not shown, the restoring unit 15 is provided with a ROM.
The different coordinate data from this ROM is M
Occurs several times. While this coordinate data is generated M times,
One block of received parameters from the switch circuit 13
Are supplied to the output terminal 121, and M restored
Luminance values are sequentially generated. Parameter a from switch circuit 13_1,a
_2,... a_TenAre supplied to the multiplication circuits 101, 102,.
Be paid. For example, the coordinates (x_i,y_i) And Seki
Continuous coordinate data [x_i ^Three, x_i ^Two y_i,x_iy_i ^Two, ...
y_i] To the multiplication circuits 101, 102,...
Supplied. Of the multiplication circuits 101, 102,.
The respective outputs are sent to the addition circuits 112, 113,.
Therefore, they are added. Therefore, the last addition circuit 120
The output terminal 121 derived from the calculated luminance value Z
_i＾ is obtained. [0069] According to the present invention, each block is prepared in advance.
Perform fitting of curved surfaces of multiple orders
Surface parameters that can increase compression ratio
A meter is selected and transmitted. Therefore, this
According to the clarification, a very high compression ratio can be obtained,
In addition, the brightness value is expressed by various curved surfaces, and the conventional binary
Significantly improved image quality compared to block coding
it can. In the present invention, the parameters are identified.
Use common coordinate data between blocks when
So, for example, to generate coordinate data by ROM
Parameter identification with only multiplication and addition processing
Can be configured.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a 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 understanding of a 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; [Description of Signs] 2 ... Blocking circuit, 3, 4, 5, 6 ... Parameter identification unit, 7 ... Selection circuit, 8 ... Judgment unit, 13 ... Switch circuit, 15 ..Restoration unit,
16 ... Scan conversion circuit

Claims (1)

(57) [Claims] In a transmission method of compressing the data amount of input image data for each block configured of a two-dimensional array of a plurality of pixel data and transmitting the compressed data, the order in which the input image data is grouped for each block Is converted to a data sequence having the following formula: where the level value of the pixel in the block is fitted to the first-order surface, and the sum of the squares of the errors when the level value of the first-order surface is minimized, A first parameter which is a coefficient is identified for each of the blocks, and the second order in which the sum of squares of an error when the level value of a pixel in the block is fitted to a surface of a second order is minimized. A second parameter, which is a coefficient of an equation that defines the curved surface of, is identified for each block, and each of the first parameter and the second parameter is
Based on the result of comparing the value indicating the error of
And selecting the first parameter or the second parameter, and selecting the selected first parameter or the second parameter.
A data transmission method characterized by transmitting the following parameters: 2. Block consisting of a two-dimensional array of multiple pixel data
The amount of input image data is compressed for each
In the transmission device for transmitting the input data, the input image data is arranged in the
Means for converting into a data series having a number, and converting the level values of the pixels in the block into a first-order surface
And the sum of squares of the error when fitting
Is the coefficient of the equation defining the surface of the first order
Means for identifying one parameter for each of the blocks, and a level value of a pixel in the block for a second-order curved surface.
And the sum of squares of the error when fitting
Is the coefficient of the equation defining the curved surface of the second order
Means for identifying two parameters for each of the blocks, and each of the first parameter and the second parameter.
Based on the result of comparing the value indicating the error of
The first parameter or the second parameter
Means for selecting a parameter, the selected first parameter or the second parameter.
Means for transmitting the parameters of
Data transmission equipment.