JP2912540B2 - Spatial filter circuit - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to, for example, performing spatial filtering on a digital video signal by using effective image data and data outside an image area where there is no originally valid image data. when, to a spatial filter circuit to create a pseudo data outside the image area.

[0002]

2. Description of the Related Art Conventionally, techniques in such a field include:
For example, there is one described in the following literature. Literature; TTC Standard Vol. V, 3rd volume, "Higher Layer Protocol Coding Method", published by the Telegraph and Telephone Technical Committee, p. 323
-324 first conventional example Figure 2 shows the spatial filtering method of the first conventional example
This shows filter characteristics in a case where a filter process is performed on one image described in the above document. This filter is a spatial filter which is divided one image area of 3 × 3, basically 1 / 4,1
It is composed of a combination of non-recursive filters having coefficients of / 2 and 1/4 in the horizontal and vertical directions. In the processing of the boundary part of the effective screen area ( this is called an image edge), one of the taps of the filter goes out of the effective area screen, and the value of the part outside the effective screen without the original data is used. Calculation. Therefore, at the four corners A , upper and lower ends B , and left and right ends C of the image shown in FIG.
The filter is replaced by a filter having a coefficient of 0, and the internal region D is processed by a non-recursive filter having coefficients of 1/4, 1/2, and 1/4. In this manner, different processing is performed at the four corners A, the upper and lower ends B, the left and right ends C, and the internal area D of the screen, so that the calculation is not performed using the value of the part outside the effective screen without the original data. ing.

[0003] Figure 3 is a configuration <br/> diagram illustrating an example of a spatial filter circuit for implementing the filtering process of the first conventional example of FIG. In the space filter circuit is used, the number and the input terminal 1 for inputting digital image data IN, an input terminal 2 for inputting an enable signal ENA indicating the valid interval of the data
You are. The input terminal 1, delay and 3-tap horizontal filter 3 for filtering in the horizontal direction has a coefficient of 1 / 4,1 / 2,1 / 4, the time duration signals required for processing in the 3-tap horizontal filter 3 A time adjusting circuit 4 for connection is connected. On the output side of the 3-tap horizontal filter 3 and the output of the time adjustment circuit 4, a selector 5 for switching the characteristics of the filter is connected. On the output side of the selector 5, 1 /
And 3-tap vertical filter 6 for filtering in the vertical direction has a coefficient of 4,1 / 2,1 / 4, and a combined circuit 7 time for delaying the time duration signals required for processing a three-tap vertical filter 6 of that It is connected. 3 to the output side of the combined circuit 7-tap vertical filter 6 and the time, El selector 8 switches the characteristic of the filter is connected. Selector 8
Is connected to an output terminal 9. The enable signal input terminal 2, to detect the image edge, toggle signal S10a for switching the characteristics of the filter, the edge detection circuit 10 which outputs a S10b are connected. The output side of the edge detection circuit 10, the selector 5 and the selector 8 is connected. S3, 3 tap output signal of the horizontal filter 3, the time combined output signal of the circuit 4 S4, S6 is the output signal of the 3-tap vertical filter 6, S7 the output signal of the combined circuit 7 time,及 Beauty OUT is the output of the selector 8 Signal.

Next, the operation of the spatial filter circuit shown in FIG. 3 will be described. Digital image data IN is input to the input terminal 1 for each scanning line and sent to the 3-tap horizontal filter 3 and the time alignment circuit 4. 3-tap horizontal filter 3
In, to filter by a factor of 1 / 4,1 / 2,1 / 4, and outputs a signal S3 to the selector 5. The time alignment circuit 4 delays the input data by the time required by the three-tap horizontal filter 3 and outputs it to the selector 5 as a signal S4. That is, the same processing as that performed by performing the filter processing with the coefficients of 0, 1, 0 is performed. Enable signal ENA is input terminal 2
Once entered, the four corners illustrated in Figure 2 based on the edge detection circuit 10, the enable <br/> Le signal ENA A, upper and lower ends B, left and right ends C, and to distinguish the internal region D, the switching signal
S10a and S10b are output to selectors 5 and 8, respectively . The selector 5 selects either the signal S3 or the signal S4 based on the signal S10a, and
And to force out the mating circuit 7 time. In the 3-tap vertical filter 6 performs spatial filter processing by a factor of 1 / 4,1 / 2,1 / 4, and outputs a signal S6 to the selector 8. The time alignment circuit 7 outputs the input data to the selector 8 as a signal S7 with a delay of the time required for processing by the 3-tap vertical filter 7. The selector 8 selects one of the signal S6 and the signal S7 to have groups <br/> Dzu the signal S10b, and outputs a signal OUT at the output terminal 9. Thus, by changing the calculation contents of the filter for the existent parts and other areas of the data outside the screen area have performed a spatial filter processing. In addition, a method of replacing a value of a portion where no data exists outside a screen area with a fixed value is generally performed. Hereinafter, this method will be described.

[0005] second prior art Figure 4 (A), (B) , (C) is an edge processing explanatory diagram for explaining a spatial filtering method of the second conventional example. When creating an output data using FIG yo urchin five inputs (A), the output Y (0) valid input data X (0) with the valid values of the image edges ~X (2) Three ,
Pseudo data X (-
2) and X (-1). Seven in the case of Figure 4 to create an output data using input (B), the output of the image edges and four valid input data with a valid value, pseudo Day without data blanking area TA3
Obtained from pieces. FIG. 4 in which output data is created using seven inputs and the output is located only one pixel from the image edge
In the case ( C), the output of the image edge is obtained from five pieces of valid input data having valid values and two pieces of pseudo data of a blanking area. As an example, in the case of FIG. 4 (A), the case will be described of outputting the average of the five data represented by the following formula (1). Y (n) = {X (n−2) + X (n−1) + X (n) + X (n + 1) + X (n + 2)} / 5 (1) The value of the pseudo data in the blanking area is how to replace the fixed value will be described for the case of image data of 8 bits. The image data takes values from 0 to 255, and the image data X (0) to X (5) change by a linear function, and are respectively 60, 80, 100, 120, 140, 1
60. At this time, there is a second method of using the first method of the value of the blanking region 16 is a black value, the data X of the image edge to the value of the blanking region (0). In the first method, the output data is Y (0) = 72 , Y
(1) = 75.2 , Y (2) = 100 , Y (3) = 12
It becomes 0. In the second method, the output data is Y (0) = 7
2 , Y (1) = 84 , Y (2) = 100 , Y (3) = 1
It will be 20.

[0006] Figure 5 is a block diagram of a spatial filter circuit for implementing a spatial filtering method of the second conventional example. This spatial filter circuit is provided with digital image data I
Input terminal 20 for inputting the N, and the enable signal ENA
Is input. The input terminal 20 has a delay flip-flop (hereinafter, referred to as DFF) 2
Data input pin 2 and DFF23 are connected.
A DFF 24 is connected to the output side of the DFF 22, and a DFF 25 is connected to the output side. DFF2
A selector 26 is connected to the output side of 5, and a 5-pixel average value circuit 27 is connected to the output side. An edge detection circuit 28 is connected to the input terminal 21, and a DFF 23 and a selector 26 are connected to the output side. S23 is the output signal of DFF23, S25 is the DFF
25 the output signal of the output signal of the selector 26 S26, S
28a and S28b are control signals of the edge detection circuit 28. FIG. 6 is a diagram for explaining the operation of the spatial filter circuit of FIG. 5 , and the operation will be described with reference to FIG. The edge detection circuit 28 detects the position of the edge image X (0) based on the enable signal ENA and controls the control signal S28a.
Is output to the DFF 23, and the control signal S28b is output to the selector 2
6 is output. The DFF 23 latches the edge image X (0) and sends the edge image X (0) to the selector 26 as a signal S2.
Output as 3. DFF digital image data IN
The delay is made by three clocks CLK at 22 , 24 and 25. The selector 26, signal based on the control signal S28b S2
5 and S23, and the digital image data X (0), X (0), X (1), X (2), X (3) ,
Outputs ... order signals S26 to 5 pixel averaging circuit 27. The five-pixel average value circuit 27 calculates an average value of the input five pixels according to the equation (1), and outputs an output signal OUT to an output terminal 29.

[0007] The third conventional example will be explained a third conventional example spatial filtering method may varying only computations in image edge portion that is described in the literature. In the third conventional example, the data Y (0), which is an image edge, is processed by averaging three pieces of data represented by the following equation (2), and only one pixel is located within the image edge. For the data Y (1) , 4 represented by the following equation (3)
The processing is performed by averaging the pieces of data. Y (0) = {X (0) + X (1) + X (2)} / 3 (2) Y (1) = {X (0) + X (1) + X (2) + X (3)} / 4 (3) Substituting the values of the portion without image data into the equations (2) and (3) under the same conditions as when replacing the values with fixed values, the output is Y (0) = 80 , Y (1) ) = 90 , Y (2) = 100 ,
Y (3) = 120. Figure 7 is a spatial filter circuit for implementing a spatial filtering method of the third conventional example
It is a block diagram. This spatial filter circuit has an input terminal 30 for inputting digital image data IN , and an input terminal 31 for inputting an enable signal ENA. The input terminal 30, three pixels averaging circuit 32,4 pixel average value circuit 33 and the 5 pixel averaging circuit 34 is connected. 3
A selector 35 is connected to the output side of the pixel average value circuit 32, the 4-pixel average value circuit 33, and the 5-pixel average value circuit 34, and an output terminal 37 is connected to the output side. Edge detection circuit 36 is connected to the input terminal 31,
Selector 35 is connected to its output side.

FIG. 8 is a diagram for explaining the operation of the spatial filter circuit of FIG. 7, and the operation will be described with reference to FIG. The edge detection circuit 36 detects the position of the image data X (0) and X (1) based on the enable signal ENA, and outputs a signal S36a及beauty S 36b to the selector 35. In the three-pixel average value circuit 32, 3
Pixel averaging is performed , and output data Ya (n) (n = 0, 1, 1)
, ...) Are output to the selector 35 as a signal S32.
The 4-pixel averaging circuit 33 performs 4-pixel averaging based on Equation (3), and outputs data Yb (n) (n = 0, 1, 2, 2, 3).
..) Are output to the selector 35 as a signal S33. The 5-pixel average value circuit 34 performs 5-pixel averaging based on the following equation (4), and outputs the output data Yc (n) (n = 0, 1, 2,.
..) Are output to the selector 35 as a signal S34. Yc (n) = {X ( n-2) + X (n-1) + X (n) + X (n + 1) + X (n + 2)} / 5 ··· (4) In the selector 35, the control signal S36a及beauty S 36b select one of the signals S32, S33, S34 on the basis of the
And output data Ya (0), Yb (1), Yc (2),
Yc (3),... Are output to the output terminal 37 as a signal OUT.

[0009]

However, the conventional spatial filter circuit has the following problems (1) and (2) . (1) The problem that the output image becomes discontinuous When the input data X (n) changes by a linear function and the output is obtained by the average of five input pixels , the input data X (n) is expressed by the following equation (5). represented, when it is substituted into equation (1), and the input and output data as shown in the following equation (6) becomes equal <br/> properly. X (n) = an + b (5) Y (n) = {X (n−2) + X (n−1) + X (n) + X (n + 1) + X (n + 2)} / 5 = {a (n) −2) + b + a (n−1) + b + a (n) + b + a (n + 1) + b + a (n + 2) + b｝ / 5 = an + b = X (n) However, the first and second prior art examples As shown in the second method and the third conventional example, at the image edge, input data and output data are different and discontinuous with other portions, which causes a problem that image quality is deteriorated. (2) In the challenge third conventional example of the circuit scale, as shown in FIGS. 7 and 8, by changing the processing procedure by the image edge portion, circuitry for performing different spatial filtering is required independently However, there is a problem that the scale of the circuit is increased. The present invention, examples of the prior problems that art had, the output data of the image edges cause image quality deterioration becomes the other portions and the discontinuity, the circuit for changing the procedure in the portion of the image edge is independently solutions for terms of scale of the circuit becomes larger because it requires Te
It is to provide a fixed spatial filter circuit.

[0010]

The present invention solves the above problems.
To determine the valid image data and image area in the image area.
Created in an area where no valid image data exists
Sky that performs spatial filtering based on pseudo data
A pixel on a boundary of the image area
With the image edge consisting of
The image located symmetrically with the pseudo data creation area to be formed
A first method for selecting and holding data of a paired pixel region of a region
Selection holding means, the pair pixel area and the pseudo data
Of the pixel area of the image edge located at the center of the
A second selection holding means for selecting and holding data values,
The image of the image edge held by the second selection holding means
Calculating means for doubling the value of the data in the elementary region;
From the output of the stage held by the first selection holding means
Subtracts the data and creates the pseudo data in the pseudo data creation area.
Subtracting means for creating similar data;
The pseudo data created and the valid data in the image area
A spatial filter that performs a spatial filter operation based on image data
And a filter calculating means.

[0011]

According to the present invention, the spatial filter times as above
Since it is configured the road, first selection holding means, with respect to an image edge comprising pixels on the boundary of the image area, the image area of the pair of the pseudo data generation pixel area and symmetrical which is a unit of creation of pseudo data The data of the pixel area is selected and held. Second selection holding means holds the selected data value of a pixel of the image edge to be located on the straight line connecting the data and pseudo data operation Naruryo zone pair pixel region. The calculating means is
The value of the image edge data held by the second selection holding unit is doubled . Subtraction means for subtracting the data more held in the first selection holding means from the output of the calculation means. The spatial filter operation means is created by the subtraction means.
Calculating the spatial filter based on the effective image data of pseudo data and image area. Therefore, the above problem can be solved.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, a spatial filter processing method used in an embodiment of the present invention will be described. Examples of the present invention, for example,
In a spatial filter processing method that handles digital image data, a fixed value is determined during image edge processing in which output data is obtained using data in an effective screen area and pseudo data values in an area where data is not originally present. The value obtained by performing the process given by the following equation ( 7 ) on the effective screen data is used as the pseudo data of the area where there is no original data that does not have it. X (−n) = X (0) − {X (n) −X (0)} = 2X (0) −X (n) ( 7 ) where n: valid pixels in the image area Horizontal address 0: Horizontal address of image edge X (n); Data of pixel whose horizontal address is n -n; Position symmetrically with pixel whose horizontal address is n around image edge horizontal address Figure 9 of the blanking region is the pseudo data generation target region is an explanatory diagram showing the creation of a pseudo data to the blanking areas where there is no effective image data. The horizontal axis n in the figure is the horizontal address of the pixel, and the vertical axis X
(N) is the value of the pixel. The value of the pseudo data X (-n) of the blanking area represented by the equation ( 7 ) is obtained by setting the effective area X = n symmetrically located with the blanking area X = -n around the image edge X = 0. Pseudo data X (-n) in the area so that the average value of the data X (n) of the blanking area and the pseudo data X (-n) in the blanking area is equal to the image edge data X (0). It is obtained from the data.
As shown in FIG. 9, the difference between the pseudo data X (−n) obtained by Expression ( 7 ) and the image edge data X (0), the effective image data X (n) and the image edge Data X
The difference from (0) matches with the sign reversed. Therefore, the pseudo data changes smoothly and continuously. Therefore, when the 5-pixel averaging process represented by Expression (1) is performed using the pseudo data X (-n) of the blanking region represented by Expression ( 7 ), the input data changes with a linear function. in this case, the input value X (n) and the output value Y (n) and although Ri equal Ku Do <br/>, can be eliminated unnatural screen, it is possible to suppress deterioration in image quality.

FIG. 10 shows an embodiment of the present invention and a conventional space filter.
FIG. 9 is a diagram illustrating an example of data of a filter processing method, in which input data changes according to a linear function, and an example of a case where a 5-pixel averaging process represented by Expression (1) is performed. Input data X in the figure
(N) is as described in the second conventional example, the output data Y (n) is the first method of the second conventional example, the second method is the second method of the second conventional example, It is the output respectively obtained by the method of the third conventional example and the method of the embodiment of the present invention. As shown in FIG. 10, if the input data X (n) is determined by the average of the five pixels is changed by a linear function, of the present invention
In the spatial filtering method used in the embodiment , the input data
Data X (n) and the output data Y (n) is equal, the first
Method, the second method and the third conventional method, input data
It can be seen that the data X (n) and the output data Y (n) are different. Me other, the spatial filter processing method for use in embodiments of the present invention, it is possible to eliminate unnaturalness of the screen, it is possible to suppress the deterioration of image quality. Figure 1 is a block diagram showing the spatial filter circuit according to an embodiment of the present invention in order to realize the spatial filtering method. This spatial filter circuit has an input terminal 4 for inputting digital image data IN.
0, and the enable signal ENA indicating the valid interval of the data
Is input. The input terminal 40 is connected to a pre-processing circuit 50 that performs an operation represented by Expression ( 7 ) and generates pseudo data in a blanking area. The output side of the pre-processing circuit 50, filter circuit 71 as a spatial filter operation means to the 5 pixel averaging operation represented by the formula (1) is connected. Input terminal 4
1 is connected to an edge detection circuit 72 that detects an image edge or the like and outputs first to fourth control signals S72a to S72d to the preprocessing circuit 50 . The output side of the filter circuit 71, an output terminal 73 for outputting an output signal OUT is connected.

[0014] Figure 11 is a structure of the pre-processing circuit 50 in FIG. 1
It is a formed view. The input terminal 40 of the pre-processing circuit 50 is connected to the data input terminals of the DFFs 51, 52 , 53 and the selector 54. The DFF 53 and the selector 54 constitute a first selection holding unit , and the DFF 52
That make up the selection holding means. The output side of the DFF 51 is connected to the data input terminal of the DFF 55,
Data input terminal of the D FF 56 is connected to the output side of 55, and is further connected to a data input terminal of the D FF 57 on the output side of the DFF56. The output side of the DFF57 selector 58 is connected, the output terminal 61 is connected to the selector 58. The output side of the DFF52, 2-fold circuit 59 is connected as the operation manual stage by the bit shift. Subtractor 60 as reduced calculation stage to the output side of the double circuit 59 is connected, and further connected selector 58 at its output. The output side of the DFF53, selector 5
4 are connected. Output from the edge detection circuit 72.
Out of the control signals S72a to S72d, the first control signal S72a indicating the image edge data X (0) is a DFF
It is output to a clock input terminal 52. Second control signal S7 indicating data X (1) one pixel inside the image edge
2b is outputted to the clock input terminal of the DFF53.
The third showing the data X (2) two pixels inside the image edge
Is output to the selector 54. The fourth control signal S72d indicating X = 1 one pixel inside the edge of the image indicating the pseudo data X (−2) and X (−1) and X = 3 three pixels inside the edge of the image , It is output to the selector 58. DFF51, each <br/> clock input terminals of 55, 56, 57, and the edge detection circuit 72, the clock signal CLK is input. S53 is the output of DFF53
Force signal, the output signal of the selector 54 S54, the S57 D
The output signal of the FF 57, S58 is the output signal of the selector 58, S59 is the output signal of the doubling circuit 59, and S60 is the output signal of the subtractor 60.

FIG. 12 is a diagram for explaining the operation of the spatial filter circuit of FIG. 1 showing an embodiment of the present invention . The operation will be described with reference to FIG. DFF51, 52,53 of
Each data input terminal and the selector 54 have an input terminal 4
Digital image data IN is input from 0. DFF
At 51, the digital image data IN is latched at the rise of the clock signal CLK and output to the DFF 55.
The image edge data X (0) is first input to the DFF 55 , the doubling circuit 59, and the selector 54. On the other hand, to the edge detection circuit 72, an enable signal ENA and a clock signal CLK representing a valid section of the digital image data IN at a high level (hereinafter, referred to as “ H ” ) are input.
In the edge detection circuit 72, the enable signal ENA is
In rising after becoming "H" of the first clock signal CLK "H", low level at the falling edge of the clock signal CLK (hereinafter, "L" hereinafter) by the image edge data X (0) Control signal S72 indicating
a is output to the clock input terminal of the DFF 52. DFF
At 52, the data X (0) of the image edge is latched and stored at the rise of the first control signal S72a, and the data X (0) is output to the doubling circuit 59. Double circuit 59
Then, the data X (0) of the image edge is doubled and the signal S5
As 9, data 2X (0) is output to the subtractor 60.

In the edge detection circuit 72, the second clock signal CL after the enable signal ENA becomes " H"
"H" at the rising of K, by the "L" in next falling rising of the clock signal CLK, and the second control signal S72b indicating one inner data X a (1) from the image edges of DFF53 clock Output to the input terminal.
The DFF 53 latches and stores the data X (1) immediately inside the image edge at the rising edge of the second control signal S72b, and outputs the data X (1) to the selector 54 as the signal S53. Further, in the edge detection circuit 72, the data X (2) which is two inward from the image edge is set by setting " H" at the third rising of the clock signal CLK after the enable signal ENA has changed to " H".
Is output to the selector 54. In the selector 54, the third control signal S72c is
When " H", the output signal S53 of the DFF 53 is selected, and when the third control signal S72c is " L", the digital image data IN is selected, and the image data X (0), X (1),
The signal S54 is output to the subtracter 60 in the order of X (2), X (1),..., X (1). In the subtracter 60, a doubler circuit 59
From the output data 2X (0) of the selector 54 to the output signal S of the selector 54.
54 is subtracted to obtain the data X (−) represented by the equation ( 7 ).
The signal S60 is output to the selector 58 in the order of 1), X (-2), X (-1),..., X (-1). The signal S60 output from the subtractor 60 has zero time required for processing by the doubling circuit 59 and the subtractor 60, so that the output of the pseudo data X (-2) is the input of X (2). Ends at the same time as the end of.

On the other hand, the digital image data IN is DF
F51, the clock signal CLK by 55, 56, 57
Is 4 cycles delay, the image data X (0) as its DFF57 or Racing No. S57, X (1), X (2), is output ... sequentially selector 58. Since the processing time of the doubling circuit 59 and the subtractor 60 is zero, at the time when the image data X (0) is output from the signal 57, the subtractor 6
From 0, pseudo data X (-2) and X (-1) are sequentially output to the selector 58. Furthermore, the edge detection circuit 72, pseudo data X (-2) and the fourth control signal to the "H" the section to the start of the edge of the signal S57 the image data X (0) of the signal S60 S72d is output to the selector 58. In the selector 58, when the fourth control signal S72d is “ H”, the output signal S
54, when the fourth control signal S72d is at “ L”, the DFF5
7 is selected and the output terminal 61 is selected.
X (-2), X (-1), X (0), X
(1), and outputs a signal S58 in the order of. In this way, the pseudo data X (−2),
X (-1) and image data X (0), X (1), X
Are sequentially output from the output terminal 61 to the filter circuit 71 . In the filter circuit 71 , the average value of the five pixels is obtained by the equation (1), and the output signal OU is output to the output terminal 73.
Output T. As described above, the filter circuit 72 can perform continuous processing without changing the processing content of the spatial filter regardless of the processing of the edge portion.

As described above, this embodiment has the following advantages. The pseudo data outside the area is set so that the average of two data X (n) and X (-n) located at a position separated from each other by the distance n from the image edge has the same value as the data X (0) of the image edge. by obtaining data X a (-n) from the data in the region, eliminating unnaturalness of the screen, it is possible to improve the image quality by spatial filtering, the process of image edge by using the data of its The five-pixel average filter processing can be performed continuously without distinction. The present invention is not limited to the above-described embodiment, and various modifications are possible. For example, there are the following modifications. (1) In the real施例, the time required for the double circuit 59 and the subtractor 60 treated as zero, DF F5 1,55,5
Output of digital image data IN by 4
Although the configuration is such that the period is delayed, the doubling circuit 59 and the subtractor 6
When considering the time required for processing 0, the number of DFFs may be increased as appropriate. (2) In the real施例has been described horizontal processing, it can be applied to vertical processing. Also ,
It can be applied to horizontal and vertical union Align. In this case, it is necessary to store image edge data using a line memory. (3) In the filter circuit 71, even without a 5 pixel average as shown in formula (1) may be any processing as long as using the data of 5 pixels. (4) In the real施例, although an example of using the data of 5 pixels, change the control signal S72a~S72d and DFF5
Changes in the number of 1,55,56,57, data other than 5 pixel data, it is possible to apply the present invention even when using the data of, for example, seven pixels.

[0019]

As described in detail above, according to the present invention,
If so, the first selection holding means, the second selection holding means,
Operation means, subtraction means, and spatial filter operation means.
Therefore, it is possible to eliminate the unnaturalness of the image and improve the image quality by the spatial filter processing.

[Brief description of the drawings]

FIG. 1 shows a configuration of a spatial filter circuit according to an embodiment of the present invention.
FIG .

FIG. 2 is a diagram showing a first conventional example of a spatial filter processing method.

FIG. 3 is a configuration diagram showing an example of a spatial filter circuit for realizing the first conventional example of FIG . 2 ;

FIG. 4 illustrates a spatial filter processing method according to a second embodiment.
An edge processing illustration of order.

FIG. 5 is a configuration diagram of a spatial filter circuit for realizing a second conventional example.

6 is a diagram for explaining the operation of the spatial filter circuit of FIG.

FIG. 7 is a configuration diagram of a spatial filter circuit for realizing a third conventional example.

FIG. 8 is a diagram for explaining an operation of the spatial filter circuit of FIG. 7;

FIG. 9 is an explanatory diagram of pseudo data creation according to the embodiment of the present invention.

[10] Example and spatial filtering side of the conventional invention
It is a figure which shows the data example of a method .

FIG. 11 is a configuration diagram of a preprocessing circuit 50 in FIG. 1;

FIG. 12 is a diagram illustrating an operation of the spatial filter circuit of FIG. 1 showing the embodiment of the present invention .

[Explanation of symbols]

Reference Signs List 50 Preprocessing circuit 51, 52, 53, 55 , 56 , 57 DFF 54 , 58 selector 59 Doubler circuit 60 Subtractor 71 Filter circuit 72 Edge detection circuit

──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-60-218171 (JP, A) JP-A-63-266982 (JP, A) JP-A-60-63589 (JP, A) JP-A-4- 42376 (JP, A) JP-A-60-218180 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06T 5/20 H04N 5/14

Claims (1)

(57) [Claims] 1. The method according to claim 1 , wherein the effective image data in the image area and the image area
Created in an area where no valid image data exists
Sky that performs spatial filtering based on pseudo data
In the inter-filter circuit, an image edge composed of pixels on the boundary of the image area is defined as an axis
The pseudo data creation area of the pseudo data creation target
Data of a pair pixel area of the image area located symmetrically with the area
First selection and holding means for selecting and holding data, and a center at the center of the paired pixel area and the pseudo data creation area.
Select the data value of the pixel area of the image edge to be placed
A second selection and holding unit for storing the image edges held by the second selection and holding unit.
Calculating means for doubling the value of the data in the pixel area, and the first selection and holding means from the output of the calculating means
Subtract the retained data to create the pseudo data creation area.
Subtracting means for generating the pseudo-data range, pseudo data to the image created by said subtraction means
Spatial filter based on valid image data in image area
A spatial filter circuit comprising: a spatial filter calculating means for calculating .