JP2007017989A - Device for converting pixel number and method for converting pixel number - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for converting pixel number that decreases bleeding in characters or the like after converting numbers of pixels and enhances visibility. <P>SOLUTION: When the position of pixels is identical before and after conversion, a zero-order interpolating means 11 determines the pixel value before conversion as a pixel value after conversion. When the position of pixels is not identical before and after conversion, a pixel value calculating means 12 calculates an average of a linear interpolation value obtained by interpolation of nearby pixels and nearby pixel values. Thus, the device 10 for converting pixel number converts the number of pixels in image signals. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a pixel number conversion device and a pixel number conversion method, and more particularly to a pixel number conversion device that converts the number of pixels of an image signal and a pixel number conversion method that converts the number of pixels of an image signal.

  When displaying an image signal of various display pixel numbers displayed on a personal computer screen to a fixed pixel number such as a liquid crystal display, it is necessary to perform pixel number conversion.

  For example, if the personal computer screen is VGA (Video Graphics Array), the number of pixels is 640 × 480, and if the liquid crystal display is SVGA (Super Video Graphics Array), the number of pixels is 800 × 600. Convert and display.

  In the conventional pixel number conversion, neighboring pixels are generally output (zero-order interpolation) or linear interpolation (linear interpolation).

  However, with conventional zero-order interpolation and linear interpolation, if there is a line such as a character in a computer image at a pixel where the position before and after conversion is not the same, the line becomes thicker or the brightness decreases and the character There was a problem that the brightness and whiteness of the line were reduced.

  FIG. 10 is a diagram showing pixel values when pixel number conversion is performed by zero-time interpolation. The pixel number conversion from VGA to SVGA is shown. Pixels having the same conversion position, such as the pixel P1 before conversion and the pixel p1a after conversion, have the same pixel value 100, and there is no problem.

  However, for pixels whose conversion positions do not match, for example, the pixel P7 is spread over two pixels like pixels p8a and p9a. In such a case, problems such as the thickness of the ruled line differ depending on the location on the screen, or the thickness of the character line changes.

  FIG. 11 is a diagram showing pixel values when the number of pixels is converted by linear interpolation. The pixel number conversion from VGA to SVGA is shown. Further, the conversion formula to the pixels p2b and p8b is as follows. In addition, the symbol in a type | formula shows the pixel value of each pixel.

p2b = P1 * 0.2 + P2 * 0.8 (1a)
p8b = P6 * 0.4 + P7 * 0.6 (1b)
The pixels having the same conversion position, such as the pixel P1 before conversion and the pixel p1b after conversion, have the same pixel value 100, and there is no problem.

  However, the pixel p2b adjacent to the pixel p1b is interpolated by the expression (1a) between the pixel P1 and the pixel P2, and becomes the pixel value 20. This causes a problem that on the screen, white protrudes from adjacent pixels and lines blur.

  In the case of the pixel p8b, when the pixel P6 and the pixel P7 are interpolated by the expression (1b), the pixel value 60 is obtained. The same applies to the pixel p9b. This causes problems such as dark lines on the screen, and ruled lines and character lines blurring or thinning depending on the location.

  The present invention has been made in view of the above points, and an object of the present invention is to provide a pixel number conversion device that reduces blurring of characters after the pixel number conversion and improves visibility.

  Another object of the present invention is to provide a pixel number conversion method with improved visibility by reducing blurring of characters after the pixel number conversion.

  The pixel number conversion device according to one aspect of the present invention is a pixel number conversion device that converts the number of pixels of an input image signal, wherein the image signal is delayed by one pixel by a clock; A second delay unit that is directly connected to one delay unit and delays the output from the first delay unit by one pixel by the clock; the output from the first delay unit; and the second delay unit Multiplication means for multiplying a pair of interpolation coefficients by the output from each other, addition means for adding the results obtained by multiplying the pair of interpolation coefficients, outputs from the first delay means, and Selecting means for selecting one of the outputs from the adding means, and the selecting means selects the output from the first delay means when the pre-conversion and post-conversion pixel positions are the same. By doing so, before conversion When zero-order interpolation is performed using the prime value as the pixel value after conversion, and the position of the pixel before conversion is different from that after conversion, the pixel value of two pixels before conversion adjacent to the position of the pixel after conversion is linearly interpolated. An average value obtained from the larger one of the pair of linear interpolation coefficients and 1 and an average value obtained from the smaller one of the pair of linear interpolation coefficients and 0 to the pair of interpolation coefficients The selecting means selects the output of the adding means obtained by adding the results obtained by multiplying the output from the first delay means and the output from the second delay means, respectively. Thus, the number of pixels is converted.

  The multiplication means may be an interpolation filter.

  The pixel number conversion method according to one aspect of the present invention is a pixel number conversion method for converting the number of pixels of an image signal. The image signal is delayed by one pixel by a clock, and an output delayed by one pixel by the clock When the pixel position is further delayed by one pixel and the pixel positions before and after the conversion are the same, the zeroth-order pixel value after the conversion is selected by selecting the output delayed by the one pixel. When interpolation is performed and the pixel position before conversion is different from that after conversion, the pixel value of two pixels before conversion adjacent to the position of the pixel after conversion is out of a pair of linear interpolation coefficients for linear interpolation An output delayed by one pixel, with the average value obtained from the larger one and 1 and the average value obtained from the smaller one of the pair of linear interpolation coefficients and 0 as a pair of interpolation coefficients. And output further delayed by one pixel. And multiplying the multiplication results obtained by multiplying each of the pair of interpolation coefficients and selecting an output obtained by adding the multiplication results to each other to convert the number of pixels. .

  In the present invention, the image signal is delayed by one pixel by the clock, and the output delayed by one pixel by the clock is further delayed by one pixel. When the positions of the pixels before conversion and after conversion are the same, zero-order interpolation is performed using the pixel value before conversion as the pixel value after conversion by selecting the output delayed by one pixel. When the position of the pixel before conversion is different from that after conversion, the larger one of a pair of linear interpolation coefficients for linearly interpolating the pixel values of two pixels before conversion adjacent to the position of the pixel after conversion and 1 And the average value obtained from the smaller one of the pair of linear interpolation coefficients and 0 as the pair of interpolation coefficients, and the output delayed by one pixel and the further 1 The number of pixels is converted by multiplying the outputs delayed by pixels, adding the multiplication results obtained by multiplying the pair of interpolation coefficients, and selecting the output obtained by adding the multiplication results. Is done.

  According to the present invention, it is possible to reduce blurring of characters after the conversion of the number of pixels and improve visibility. Further, according to the present invention, it is possible to reduce blurring of characters after the conversion of the number of pixels and improve visibility.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a principle diagram of a pixel number conversion apparatus according to the present invention. The pixel number conversion device 10 converts the number of pixels of the image signal. In the following description, it is assumed that the 4: 5 pixel number conversion from the VGA screen to the SVGA screen is performed.

  If the position of the pixel before conversion and after conversion is the same, the zero-order interpolation unit 11 sets the pixel value before conversion as the pixel value after conversion. In the figure, since the position of the pixel PEL1 before conversion and the position of the pixel PEL1a after conversion are the same, the pixel value is 100 for both.

  The pixel value calculation means 12 calculates the average of the linear interpolation value interpolated from the neighboring pixels and the neighboring pixel values when the pixel positions before and after the conversion do not match, To do.

  Since the number of pixels is converted to 4: 5, pixels whose pixel positions are shifted before and after conversion appear. In the figure, it is assumed that the positions of the pixel PEL2 and the pixel PEL2a are shifted.

  In such a case, the pixel value calculation unit 12 calculates and converts the pixel value of the pixel PEL2a to 8 from the pixel value 100 of the pixel PEL1 and the pixel value 0 of the pixel PEL2 using a conversion equation. Details will be described later.

  Next, a first embodiment of the pixel number conversion apparatus 10 of the present invention will be described. In the first embodiment, the pixel value calculation means 12 performs linear interpolation by linear interpolation, and calculates and converts the average of the linear interpolation value and neighboring pixel values.

  FIG. 2 is a diagram illustrating pixel values when the first embodiment is applied. Pixel p2c and pixel p8c are converted by the pixel value calculation means 12 as shown in the following equation. For simplicity of explanation, only the conversion of the pixel p2c and the pixel p8c is shown, and the symbol in the formula indicates the pixel value of each pixel.

p2c = (P2 + (P1 * 0.2 + P2 * 0.8)) / 2 (2a)
p8c = (P7 + (P6 * 0.4 + P7 * 0.6)) / 2 (2b)
In the present invention, the neighboring pixel values are P2 and P7 in the above equation, and the linear interpolation values (here, the linear interpolation values) are (P1 * 0.2 + P2 * 0.8) and (P6 * 0.4 + P7 * 0.6). ).

  Pixels having the same conversion position, such as the pixel P1 before conversion and the pixel p1c after conversion, have the same pixel value 100. On the other hand, in the converted pixel p2c, a value of 20 (equal to the value of the pixel p2b in FIG. 11) linearly interpolated with the pixels P1 and P2 is further averaged with the neighboring pixel P1, and becomes 10.

  Therefore, it is possible to reduce the bleeding of the white protruding line to the adjacent pixel caused by the pixel number conversion by the present invention. In addition, for a pixel whose conversion position after conversion does not match, such as the pixel P7, the linearly interpolated value 60 (equal to the value of the pixel p8b in FIG. 11) is further averaged to 80 with the neighboring pixel P7. Become.

  Therefore, it is possible to reduce inconveniences such as dark lines caused by the pixel number conversion. The same applies to the pixel p9c. Further, even in the case where the pixel expands to two pixels like the pixel P7 in FIG. 10, since the average of the linear interpolation and the neighboring pixels is 80 from the pixel value 100, it becomes thin or dark. Reduce the problem of becoming a line.

  Next, a circuit configuration for performing linear interpolation will be described. FIG. 3 is a diagram showing a circuit configuration for performing linear interpolation. IC1 to IC3 are D flip-flops, IC4 to IC7 are adders, and IC8 is a quinary counter. F1 to F8 are interpolation filters, SW is a switch, and S1 to S5 are switch contacts.

  The pixel line before conversion is input to the D terminal of IC1. The Q terminal of IC1 and the D terminal of IC2 are connected. A clock CK on the VGA side is input to the clock terminals of IC1 and IC2.

  The Q terminal of IC1 is connected to the S1 terminal. Further, the outputs of F1 (filter coefficient: 0.2) and F2 (filter coefficient: 0.8) are added by IC4, and the output of IC4 is connected to the S2 terminal.

  The outputs of F3 (filter coefficient: 0.4) and F4 (filter coefficient: 0.6) are added by IC5, and the output of IC5 is connected to the S3 terminal. The outputs of F5 (filter coefficient: 0.6) and F6 (filter coefficient: 0.4) are added by IC6, and the output of IC6 is connected to the S4 terminal.

  The outputs of F7 (filter coefficient: 0.8) and F8 (filter coefficient: 0.2) are added by IC7, and the output of IC7 is connected to the S5 terminal. The SW is connected to the D terminal of IC3. The clock terminals of IC3 and IC8 are connected to CK × 5/4 which is a clock on the SVGA side. The Q terminal of IC3 outputs the converted pixel line.

  Next, a first embodiment will be described. FIG. 4 is a diagram showing the configuration of the first embodiment. The circuit configuration of the first embodiment is the same as that of FIG. 3, and only the filter coefficient is different.

  FIG. 5 is a diagram for explaining the operation of the circuit configuration of the first embodiment shown in FIG. The pixel values of the pre-conversion VGA pixels P1 to P8 are A to H, and the post-conversion SVGA pixels p1c to p9c are ai.

  Moreover, the conversion formula 1 of the conventional linear interpolation corresponding to the circuit of FIG. 3 and the conversion formula 2 of the present invention corresponding to the circuit of FIG. 4 are shown. First, when the count value of the quinary counter IC8 in FIG. 4 is 1, the switch SW is connected to the S1 terminal, and the pixel value of the pixel p1c is a (= A).

  When the count value of the quinary counter IC8 is 2, the switch SW is connected to the S2 terminal, and the pixel value of the pixel p2c is b (= 0.1A + 0.9B). When the count value of the quinary counter IC8 is 3, the switch SW is connected to the S3 terminal, and the pixel value of the pixel p3c is c (= 0.2B + 0.8C).

  When the count value of the quinary counter IC8 is 4, the switch SW is connected to the S4 terminal, and the pixel value of the pixel p4c is d (= 0.8C + 0.2D). When the count value of the quinary counter IC8 is 5, the switch SW is connected to the S5 terminal, and the pixel value of the pixel p5c is e (= 0.9D + 0.1E).

  When the count value of the quinary counter IC8 returns to 1, the switch SW is connected to the S1 terminal, and the pixel value of the pixel p6c is f (= E). The same applies thereafter. Next, a second embodiment of the pixel number conversion apparatus 10 of the present invention will be described. In the second embodiment, the pixel value calculation means 12 performs linear interpolation with a cosine roll-off low-pass filter, and calculates and converts the average of the linear interpolation value and the neighboring pixel value.

  FIG. 6 is a diagram illustrating frequency characteristics of the cosine roll-off low-pass filter. The vertical axis represents gain and the horizontal axis represents frequency. FIG. 7 is a correspondence table of filter coefficients between the linear interpolation and the cosine roll-off low-pass filter. The correspondence between the coefficients used in the first embodiment and the coefficients of the cosine roll-off low-pass filter is shown, and will be described below with the coefficients shown in Table 3.

  FIG. 8 is a diagram illustrating pixel values when the second embodiment is applied. The pixel p2d and the pixel p8d are converted by the pixel value calculation unit 12 as follows. In order to simplify the explanation, only the conversion of the pixel p2d and the pixel p8d is shown, and the symbol in the formula indicates the pixel value of each pixel.

p2d = (P2 + (P1 * 0.15 + P2 * 0.85)) / 2 (3a)
p8d = (P7 + (P6 * 0.37 + P7 * 0.63)) / 2 (3b)
In the present invention, neighboring pixel values are P2 and P7 in the above equation, and linear interpolation values (here, cosine roll-off low-pass filter interpolation values) are (P1 * 0.15 + P2 * 0.85) and (P6 * 0.37 + P7). * 0.63).

  Pixels having the same conversion position, such as the pixel P1 before conversion and the pixel p1d after conversion, have the same pixel value 100. On the other hand, in the converted pixel p2d, the value of 15 interpolated by the cosine roll-off low-pass filter with the pixels P1 and P2 is further averaged with the neighboring pixel P2, and becomes about half.

  Therefore, compared with the first embodiment, it is possible to further reduce the bleeding of the white protrusion line to the adjacent pixel. For a pixel whose conversion position after conversion does not match, such as the pixel P7, the value 63 interpolated by the cosine roll-off low-pass filter is further averaged to 82 with the neighboring pixel P7.

  Therefore, it is possible to further reduce inconveniences such as dark lines as compared with the first embodiment. Further, even when the pixel spreads to two pixels as in the pixel P7 in FIG. 10, it is the same to alleviate the problem of thinning or dark lines.

  As described above, in the pixel number conversion device 10 of the present invention, when the positions of the pre-conversion and post-conversion pixels do not match, the average of the linear interpolation value interpolated from the neighboring pixels and the neighboring pixel values is calculated. The number of pixels is converted as the pixel value after calculation and conversion.

  This reduces the problem that the thickness of the ruled lines and text lines changes, blurs, or becomes thin depending on the location when the lines such as text on the computer image are in pixels where the positions before and after conversion are not the same. It becomes possible to do.

  Note that the pixel value calculation unit 12 may calculate the average from an arbitrary mixture ratio of the neighboring pixel value and the linear interpolation value. For example, the ratio of neighboring pixel values is increased on a screen with many characters such as a personal computer screen, and the ratio of linear interpolation values is increased on a screen with many natural images such as general TV images. The ratio of the mixing ratio is automatically set from the user's operation or from the nature of the image.

Next, the pixel number conversion method of the present invention will be described. FIG. 9 is a flowchart showing the processing procedure of the pixel number conversion method of the present invention.
[S1] It is determined whether or not the positions of the pixels before conversion and after conversion are the same. If they match, go to step S2, otherwise go to step S3.
[S2] Zero-order interpolation is performed using the pixel value before conversion as the pixel value after conversion.
[S3] The average of the linear interpolation value interpolated from the neighboring pixels and the neighboring pixel value is calculated, and this calculated value is used as the converted pixel value.

  As described above, the pixel number conversion method of the present invention calculates the average of the linear interpolation value interpolated from the neighboring pixels and the neighboring pixel values when the positions of the pixels before and after the conversion do not match. Thus, the number of pixels is converted as the converted pixel value. As a result, it is possible to reduce the bleeding of characters after the conversion of the number of pixels and to improve the visibility.

1 is a principle diagram of a pixel number conversion apparatus of the present invention. It is a figure which shows the pixel value at the time of applying 1st Embodiment. It is a figure which shows the circuit structure which performs line interpolation. It is a figure which shows the structure of 1st Embodiment. FIG. 5 is a diagram for explaining the operation of the circuit configuration of the first embodiment shown in FIG. 4. It is a figure which shows the frequency characteristic of a cosine roll-off low-pass filter. 6 is a correspondence table of filter coefficients between linear interpolation and a cosine roll-off low-pass filter. It is a figure which shows the pixel value at the time of applying 2nd Embodiment. It is a flowchart which shows the process sequence of the pixel number conversion method of this invention. It is a figure which shows a pixel value at the time of performing pixel number conversion by midnight interpolation. It is a figure which shows the pixel value at the time of performing pixel number conversion by linear interpolation.

Explanation of symbols

  10 pixel number conversion device, 11 zero-order interpolation means, 12 pixel value calculation means, PEL1, PEL2 pixel before conversion, PEL1a, PEL2a pixel after conversion

Claims (3)

In a pixel number conversion device that converts the number of pixels of an input image signal,
First delay means for delaying the image signal by one pixel by a clock;
A second delay unit connected directly to the first delay unit and delaying an output from the first delay unit by one pixel by the clock;
Multiplying means for multiplying the output from the first delay means and the output from the second delay means by a pair of interpolation coefficients,
Adding means for adding the results obtained by multiplying the pair of interpolation coefficients;
Selecting means for selecting one of the output from the first delay means and the output from the addition means,
When the position of the pixel before conversion and the pixel after conversion is the same, the selection unit selects the output from the first delay unit so that the pixel value before conversion becomes the pixel value after conversion. If the pixel positions before conversion and after conversion are different, the larger of the pair of linear interpolation coefficients for linearly interpolating the pixel values of the two pixels before conversion adjacent to the pixel positions after conversion And the average value obtained from the smaller one of the pair of linear interpolation coefficients and the average value obtained from 0 as the pair of interpolation coefficients, and the output from the first delay means and The number of pixels is converted by the selection means selecting the output of the addition means obtained by adding the results obtained by multiplying the outputs from the second delay means, respectively. Pixel number conversion device. The pixel number conversion apparatus according to claim 1, wherein the multiplication unit is an interpolation filter. In a pixel number conversion method for converting the number of pixels of an image signal,
The image signal is delayed by one pixel by a clock,
The output delayed by one pixel by the clock is further delayed by one pixel,
When the pixel positions before conversion and after conversion are the same, by selecting the output delayed by one pixel, zero-order interpolation is performed with the pixel value before conversion as the pixel value after conversion,
When the position of the pixel before conversion is different from that after conversion, the larger one of a pair of linear interpolation coefficients for linearly interpolating the pixel values of two pixels before conversion adjacent to the position of the pixel after conversion and 1 And the average value obtained from the smaller one of the pair of linear interpolation coefficients and 0 as the pair of interpolation coefficients, and the output delayed by one pixel and the further 1 Multiply each output delayed by pixels,
Add the multiplication results obtained by multiplying the pair of interpolation coefficients,
A pixel number conversion method, wherein pixel number conversion is performed by selecting an output obtained by adding the multiplication results.

Images