JP2008022443A - Contour correction apparatus and video signal processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately perform image contour correction and to also suppress noise. <P>SOLUTION: A contour correction apparatus includes: a first to fourth pixel delay elements 1a-1d for outputting input pixel signal after sequentially delaying them by one pixel; first to fourth inclination measuring units 2a-2d for measuring inclinations of pixels for a delay time for one pixel for each of the pixel delay elements; an intersection measuring unit 3a for measuring the amplitude and time of a point where the inclination of a pixel measured by the inclination measuring unit 2a crosses the inclination of a pixel measured by the inclination measuring unit 2d; an intersection measuring unit 3b for measuring the amplitude and time of a point where the inclination of a pixel measured by the inclination measuring unit 2b crosses the inclination of a pixel measured by the inclination measuring unit 2d; and an intermediate value calculating unit 4 for calculating out a pixel amplitude intermediate value for correcting the amplitude of the pixel outputted from the second pixel delay element 2d based on validity of results measured by the first and second intersection measuring units. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a contour correcting device that corrects a contour portion of an image represented by a digital video signal, and more particularly to a contour correcting device that corrects a contour of a displayed image and reduces noise. .

As a method for enhancing the sharpness and enhancing the sharpness of an image displayed on a video display device, there is contour correction by digital signal processing.
In general, an arbitrary band is extracted from a digitized video signal using a digital filter, and the extracted signal is added to the original signal to emphasize an arbitrary frequency component and contour-corrected video. A signal can be obtained.
However, the conventional contour correction apparatus has a problem that excessive overshoot and undershoot occur in the contour portion.
Further, if the amplitude of the extracted high frequency signal is suppressed in order to reduce overshoot and undershoot, there is a problem that necessary contour correction cannot be performed sufficiently.

As an improvement measure, for example, as disclosed in Japanese Patent Application Laid-Open No. 2002-335424 (Patent Document 1), only a component with an overshoot or undershoot of the contour portion is removed from the extracted high frequency signal, or Some limit correction is performed by adding over the original signal after limiting overshoot and undershoot.
In the contour correction circuit disclosed in Japanese Patent Laid-Open No. 2002-335424, it is possible to perform contour correction by removing overshoot and undershoot of low frequency components, but noise of high frequency components is emphasized. .
In other words, the conventional contour correction apparatus has a problem that not only the contour but also noise which is a high frequency component is emphasized.
JP 2002-335424 A (FIG. 1, summary)

  The present invention has been made to solve the above-described problems, and can correct the contour of an image without overshooting or undershooting, and a high frequency component generated as an isolated point is a contour of an image. It is an object of the present invention to provide a contour correction device that determines noise rather than a portion and can easily suppress (reduce) noise, and a video signal processing device having the contour correction device.

  The contour correction apparatus according to the present invention includes first to fourth pixel delay elements that are connected in series and sequentially output an input pixel signal by delaying by one pixel, and the first to fourth pixel delay elements. 1st to 4th inclination measuring means connected in parallel with each of the first to fourth pixels for measuring the inclination of the pixel representing the change degree of the amplitude of the input pixel and the amplitude of the output pixel with respect to the delay time of one pixel, A first intersection measuring means for measuring the amplitude and time of a point at which the inclination of the pixel measured by the inclination measuring means and the inclination of the pixel measured by the third inclination measuring means intersect, and the second inclination measuring means; By second intersection measuring means for measuring the amplitude and time of a point where the inclination of the pixel to be measured and the inclination of the pixel measured by the fourth inclination measuring means intersect, and by the first and second intersection measuring means The effectiveness of the measurement results In which it has an intermediate value calculating means for calculating and outputting a pixel amplitude intermediate value for correcting the amplitude of pixels output from the second pixel delay elements Zui.

  According to the present invention, since the pixel amplitude is corrected according to a pattern in which the pixel amplitude changes in the time direction (pixel inclination) rather than the contour correction by the digital filter, the image is not accompanied by overshoot or undershoot. In addition, the high-frequency component generated as an isolated point is determined to be noise, not the contour portion of the image, and noise can be easily suppressed.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
In addition, in each figure, the same code | symbol represents that it is the same or equivalent thing.
In the formulas described below, “/” represents “÷” and “*” represents “x”.
Embodiment 1 FIG.
FIG. 1 is a block diagram showing a configuration of a main part of a contour correction apparatus according to Embodiment 1 of the present invention.
As shown in the figure, the contour correction apparatus according to the present embodiment is connected in series to a pixel signal input unit 8 for inputting a pixel signal (for example, a luminance signal of a video signal), and an input pixel signal (luminance signal). ) Are delayed by one pixel and output, the first pixel delay element 1a, the second pixel delay element 1b, the third pixel delay element 1c, and the fourth pixel delay element 1d, 1st to 4th inclination measuring units 2a, 2b, which are connected in parallel with each of the 1st to 4th pixel delay elements 1a, 1b, 1c, 1d to obtain inclinations of adjacent pixels (inclination of pixel amplitude). 2c and 2d.

  Further, a first intersection measurement unit that obtains the amplitude (luminance) of the pixel at the time when the respective inclinations (gradients of the pixel amplitude) of the adjacent pixels measured by the first and third inclination measurement units 2a and 2c intersect. Second intersection measurement for obtaining the amplitude (luminance) of a pixel at a time when 3a and the respective inclinations (gradients of the pixel amplitude) of adjacent pixels measured by the second and fourth inclination measuring units 2b and 2d intersect Unit 3b, an intermediate value calculation unit 4 for determining an intermediate value of the amplitude (luminance) of the pixels obtained by the first intersection measurement unit 3a and the second intersection measurement unit 3b, and a signal obtained by the intermediate unit calculation unit 4 A pixel signal (luminance) output unit 9 that outputs a level is provided.

In FIG. 1, the amplitude of the pixel Sin (for example, the magnitude of the luminance signal) input from the pixel signal input unit 8 is Sin ^# , the amplitude of the pixel Sa delayed by one pixel by the pixel delay element 1a is Sa ^# , and the pixel delay element. 1b, the amplitude of the pixel Sb further delayed by one pixel is Sb ^# , the amplitude of the pixel Sc further delayed by one pixel by the pixel delay element 1c is Sc ^# , and the amplitude of the pixel Sd further delayed by one pixel by the pixel delay element 1d is Sd. ^# .
The “pixel amplitude” is, for example, the magnitude of the luminance signal of the pixel, in other words, the digital value of the digitized luminance signal.

Next, the inclination measuring unit 2a is connected in parallel with the first pixel delay element 1a, and the inclination measuring unit 2a is connected to the amplitude Sin ^# and the pixel Sa of the adjacent pixel Sin via the first pixel delay element 1a. A slope Ka representing the degree of change in the delay time “Xa−Xin” of the amplitude Sa ^# is measured. (See the explanation regarding FIG. 2 and FIG. 3 described later)
The slope Ka is calculated by the following equation.
Ka = (Sa ^# -Sin ^# ) / (Xa-Xin)
Similarly, the inclination measuring unit 2b calculates an inclination Kb that is the difference between the amplitude Sa ^# of the pixel Sa and the amplitude Sb ^{# of the} pixel Sb, and the inclination measuring unit 2c calculates the difference between the amplitude Sb ^# of the pixel Sb and the amplitude Sc ^{# of the} pixel Sc. The inclination measuring unit 2d obtains an inclination Kd that is a difference between the amplitude Sc ^# of the pixel Sc and the amplitude Sd ^{# of the} pixel Sd.

The first intersection measurement unit 3a obtains the time at which the slope Ka and the slope Kc intersect and the amplitude Yac at that time from the slope measurement units 2a and 2c.
Now, assuming that the amplitude of the slope Ka at time X is Ya and the amplitude of the slope Kc at time X is Yc,
Ya = Ka * X + Sin
Yc = Kc * X + Sb-2Kc
It is represented by

In other words, Ya in the above equation is the time of a straight line with a slope Ka representing the degree of change in the delay time “Xa−Xin” of the amplitude Sin ^{# of the} pixel Sin adjacent to the pixel Sain via the pixel delay element 1a and the amplitude Sa ^# of the pixel Sa The amplitude at X. (The same applies to Yc)
The time Xac at the point where the slope Ka and the slope Kc intersect is
Xac = (Sb-Sin-2Kc) / (Ka-Kc)
The amplitude Yac of the intersecting point Xac is Yac = (Ka * Sb-2Ka * Kc-Kc * Sin) / (Ka-Kc)
Similarly, the intersection measuring unit 3b obtains, from the inclination measuring units 2b and 2d, the time Xbd when the respective inclinations Kb and the inclination Kd intersect, and the amplitude Ybd at that time.

FIG. 2 is a diagram for explaining an operation example of the inclination measurement unit and the intersection measurement unit, and shows the relationship between the above-described Sin, Sa, Sb, Sc and Xac, Yac.
2, the horizontal axis is time, Xin is the time when the pixel Sin is input to the pixel delay element 1a, Xa is the time when the pixel Sa is input to the pixel delay element 1b, and Xb is the time when the pixel Sb is the pixel delay element 1c. Xc is the time when the pixel Sc is input to the pixel delay element 1d, and Xd is the time when the pixel Sd is output from the pixel delay element 1d. In addition, the vertical axis in FIG. 2 represents the amplitude (for example, luminance) of the pixel.
That is, since the time of the pixels Sin to Sd is delayed by one pixel each, the amplitude (Sin ^{# to} Sd ^# ) of the pixel in each of the pixels Sin to Sd is also delayed by one pixel. Will be.

First, as shown in FIG. 2A, an inclination Ka representing the degree of change in the delay time “Xa−Xin” between the amplitude Sin ^# of the adjacent pixel Sin and the amplitude Sa ^# of the pixel Sa via the pixel delay element 1a. The amplitude of the point where the amplitude Sb ^# of the adjacent pixel Sb via the pixel delay element 1c and the slope Kc representing the degree of change in the delay time “Xc−Xb” of the amplitude Sc ^# of the pixel Sc intersect is Yac and the slope Ka. And Xac is the time at the point where the slope Kc intersects.

Next, the intermediate value calculation unit 4 obtains an intermediate amplitude Yout between the amplitude Yac and the amplitude Ybd.
At this time, depending on how far the time Xac of the amplitude Yac at the point where the slope Ka and the slope Kc intersect is far from the time Xb of the pixel Sb serving as the base point (that is, the time of the amplitude Sb ^# of the pixel Sb), The intermediate value is obtained by changing the effectiveness of Yac.
Here, the effectiveness of Yac corresponds to a difference in distance between the time Xac of Yac and the time Xb of the pixel Sb, and it is defined that the effectiveness of Yac increases as Yac is closer to Xb.
For example, in the case of FIG. 2A, Xac is between Xc and Xb, and in the case of FIG. 2B, the values of the slope Ka and the slope Kc are close, and the time Xac of the amplitude Yac is as shown in FIG. Compared with, the distance from Xb is increased, and the effectiveness of Yac is reduced.

The following expression is an expression for determining the effectiveness Zac of Yac.
Δtac = | Xac−Xb |
Zac = Δtac * m−n
(Zac is effectiveness, m and n are arbitrary numbers)
Weighting is performed by multiplying the distance Δtac between the time Xac of Yac and the time Xb of the pixel Sb by the coefficient m and subtracting the coefficient n.
Further, the subtraction result is clipped in the range of 0 to 1, and the value Zac obtained by this is set as the validity.
Next, Yac is converted into Yac ′ by the following formula with the effectiveness Zac.
Yac '= Zac * Sb + (1-Zac) * Yac
(Yac 'is the amplitude Yac converted with the effectiveness Zac)
Similarly, Ybd is also converted to Ybd ′ by the effectiveness Zbd of Xb and Xbd.
Next, the amplitude Yout is obtained from the average value of Yac ′ and Ybd ′ converted according to the effectiveness, and is output by the luminance signal output unit 9.

FIG. 3 is a flowchart showing the operation of the contour correction apparatus according to this embodiment described above.
In this embodiment, first, as shown in step S1, pixel amplitudes Sin ^{# to} Sd ^# are input.
Specifically, the inclination measuring unit 2a has pixel amplitudes Sin ^# and Sa ^# , the inclination measuring unit 2b has pixel amplitudes Sa ^# and Sb ^# , and the inclination measuring unit 2c has pixel amplitudes Sb ^# and Sc. ^# , Pixel amplitudes Sc ^# and Sd ^# are input to the inclination measuring unit 2d.
Then, the inclination Ka, the inclination Kb, the inclination Kc, and the inclination Kd are measured. (Step S2)
Next, Xac, Yac, Xbd, and Ybd are obtained by the intersection measuring units 3a and 3b. (Step S3)
Next, the intermediate value calculation unit 4 obtains Yac ′ and Ybd ′ converted according to the effectiveness. (Step S4)
Then, the amplitude Yout is obtained from the average value of Yac ′ and Ybd ′ converted according to the effectiveness. (Step S5)
Finally, the amplitude Yout obtained in step S5 from the intermediate value calculator 4 is output as the amplitude of the correction pixel.

FIG. 4 is a diagram for explaining the operation of the intermediate value calculation unit 4 and shows an example of a signal that is corrected by the contour correction device according to the present embodiment.
FIG. 4A shows that the delay time of the amplitude Yac at the intersection of Ya having the slope Ka and Yc having the slope Kc, and the amplitude Ybd at the intersection of Yb having the slope Kb and Yd having the slope Kd is sufficient for Xb. This is an example in the near case.
At this time, the intermediate value Yout is output, and it is possible to improve the valley-shaped contour of the image by correcting Sb to Yout.
That is, when the amplitude of the pixel gradually changes in a valley shape to Sa ^# , Sb ^# , Sc ^# , the steep change is made by using the corrected intermediate value Yout. The contour of the image can be corrected.

FIG. 4B shows that Yac is converted into Yac ′ by the effectiveness Zac because the amount of change Δt between the time Xac and the time Xb of the amplitude Yac at the intersection of Ya having the slope Ka and Yc having the slope Kc is large. Is done.
At this time, there is no contour to be corrected for Sin, Sa, Sb, Sc, and Sd, and the intermediate value Yout cannot correct the contour of the image.

FIG. 4C shows an example in which the Δb between Ybd delay times Xbd and Xb is large and the Δt between Yac delay times Xac and Xb is small.
At this time, Ybd is converted by the effectiveness Zbd to become Ybd ′, and Sb whose amplitude is isolatedly increased is corrected to Yout, which is an intermediate value between Yac and Ybd ′.
For example, the noise generated at an isolated point as shown in FIG. 4C (that is, the pixel Sb at the isolated point) has no correlation with the left and right pixels, so that the amplitude is suppressed without using a special circuit. Is possible.
That is, when the amplitude of the pixel is large at an isolated point, it is possible to suppress the amplitude by determining that it is impulsive noise instead of the image outline and using the corrected intermediate value Yout. That is, noise can be suppressed.

As described above, the contour correction apparatus according to the present embodiment is connected in series, and delays the input pixel signal by one pixel and sequentially outputs the first to fourth pixel delay elements 1a to 1d.
The first to fourth pixel delay elements 1a to 1d are connected in parallel to each of the first to fourth pixel delay elements 1a to 1d and measure the inclination of the pixel representing the change degree of the amplitude of the input pixel and the amplitude of the output pixel with respect to the delay time for one pixel. 4 tilt measuring means (first to fourth tilt measuring sections 2a to 2d), and the first tilt measuring means (first tilt measuring section 2a) to measure the pixel tilt and the third tilt measuring means ( A first intersection measuring unit (first intersection measuring unit 3a) that measures the amplitude and time of the point at which the pixel inclination measured by the third inclination measuring unit 2c) intersects, and a second inclination measuring unit (first A second intersection that measures the amplitude and time at the point where the inclination of the pixel measured by the second inclination measuring unit 2b) and the inclination of the pixel measured by the fourth inclination measuring means (fourth inclination measuring unit 2d) intersect. Measuring means (second intersection measuring unit 3b), first and second A pixel amplitude intermediate value for correcting the amplitude of the pixel output from the second pixel delay element 1b based on the effectiveness of the measurement result by the intersection measuring means (first and second intersection measuring units 3a and 3b). An intermediate value calculation means (intermediate value calculation unit 4) for calculating and outputting is provided.

As a result, the pixel amplitude is corrected according to the pattern in which the pixel amplitude changes in the time direction (pixel inclination), so that the contour of the image can be appropriately corrected without overshoot or undershoot, and the isolated point Therefore, it is determined that the high-frequency component generated automatically is not an outline portion of the video but noise, and noise can be easily suppressed (reduced).
In addition, since the change pattern is calculated from the inclination of the amplitude of the pixel, it can be realized with a small circuit scale without requiring special pattern matching or a database.

Embodiment 2. FIG.
FIG. 5 is a block diagram illustrating a configuration of a main part of the contour correction apparatus according to the second embodiment.
As shown in FIG. 5, the contour correction apparatus according to the present embodiment is connected in series with a pixel signal input unit 8 that inputs a pixel signal (for example, a luminance signal of a video signal), and delays the input signal by one horizontal period. Are connected in parallel to the first to fourth line delay elements 5a, 5b, 5c, 5d and the first to fourth line delay elements 5a, 5b, 5c, 5d, respectively, and the inclination of the pixels of the adjacent lines is determined. The first intersection for obtaining the amplitude (luminance) of the pixel at the time when the inclinations of the pixels measured by the first to fourth inclination measurement units 2a, 2b, 2c, 2d and the inclination measurement units 2a, 2c intersect, respectively. The measurement unit 3a, the second intersection measurement unit 3b for obtaining the amplitude (luminance) of the pixel at the time when the inclinations of the pixels measured by the inclination measurement units 2b and 2d intersect, the first intersection measurement unit 3a and the first Finding the intersection measuring unit 3b An intermediate value calculation unit 4 for obtaining an intermediate value of the amplitude (brightness) of a pixel, composed of pixel signals (luminance signal) output unit 9 for outputting a signal level obtained by the intermediate unit calculator 4.

In FIG. 5, the amplitude of the pixel Sin input from the pixel signal (luminance signal) input unit 8 is Sin ^# , the amplitude of the pixel Sain delayed by one line by the first line delay element 5a is Sain ^# , The line delay element 5b delays the pixel Sbin by one line, the amplitude of the pixel Sbin is Sbin ^# , the third line delay element 5c delays the pixel Sbin by one line, the amplitude of the pixel Scin is Scin ^# , and the fourth line delay element. The amplitude of the pixel Sdin in which the pixel Scin is delayed by one line by 5d is defined as Sdin ^# .

The operation of the second embodiment, although generally the same as the first embodiment, the amplitude Sin ^# of pixels, the amplitude Sain ^# of pixels line delay, sbin ^#, Scin ^#, contour correction by using the SDIN ^#, noise suppression And the output of the Yout signal is different.
That is, the contour correction apparatus according to the present embodiment uses line delay elements 5a, 5b, 5c, and 5d, respectively, instead of the first to fourth pixel delay elements 1a, 1b, 1c, and 1d in the first embodiment. It is characterized by the fact that
Thereby, the same effect can be obtained with the same circuit both in the horizontal direction and in the vertical direction.

Embodiment 3 FIG.
FIG. 6 is a block diagram illustrating a configuration of a main part of the contour correction apparatus according to the third embodiment.
As shown in the figure, the contour correction device according to the present embodiment includes a pixel signal input unit 8 that inputs a pixel signal (for example, a luminance signal), and pixel delay elements 1a, 1b, 1c, which delay the input signal by one pixel, 1d, a line delay element 5a that delays an input signal by one horizontal period, pixel delay elements 1e, 1f, 1g, and 1h that delay one line delay signal output from the line delay element 5a by one pixel, and a line delay element 5a The line delay element 5b for further delaying the one-line delay signal output from the line delay line by one line, the pixel delay elements 1i, 1j, 1k, 1l for delaying the two-line delay signal output from the line delay element 5b by one pixel, A line delay element 5c for further delaying the two-line delay signal by one line, and pixel delay elements 1m, 1n, 1o for delaying the three-line delay signal from the line delay element 5c by one pixel. p, a line delay element 5d for further delaying the 3-line delay signal from the line delay element 5c by one line, and a pixel delay element 1q, 1r, 1s, 1t for delaying the 4-line delay signal from the line delay element 5d by one pixel. And.

  Furthermore, a correlation detection unit 6 that detects the correlation of each pixel, a pixel selection unit 7 that selects a pixel that intersects a line in the direction with the largest correlation, and an inclination measurement unit that obtains the inclination of the pixel selected by the pixel selection unit 7 2a, 2b, 2c, and 2d and the intersection measurement unit 3a for obtaining the luminance at the time when the inclinations of the pixels measured by the inclination measurement units 2a and 2c intersect, and the inclinations of the pixels measured by the inclination measurement units 2b and 2d. The intersection measuring unit 3b for obtaining the luminance at the intersection time, the intermediate value calculating unit 4 for obtaining the intermediate value of the luminance obtained by the intersection measuring unit 3a and the intersection measuring unit 3b, and the signal level obtained by the intermediate unit calculating unit 4 A luminance signal output unit 9 for outputting is provided.

In FIG. 6, the amplitude of the pixel input from the pixel signal input unit 8 is Sin, Sin delayed by one pixel by the pixel delay element 1a is Sa, Sa is delayed by one pixel by the pixel delay element 1b, Sb, and the pixel delay element 1c is used. Assume that Sb delayed by one pixel is Sc, and Sc delayed by one pixel by the pixel delay element 1d is Sd.
Further, Sin delayed by one line by the line delay element 5a is sain, Sain delayed by one pixel by the pixel delay element 1e, Saa delayed by one pixel by the pixel delay element 1f, Sab, and delayed by one pixel by the pixel delay element 1g. Sab is Sac, and Sac delayed by one pixel by the pixel delay element 1h is Sad.
Further, the Sin delayed by one line by the line delay element 5b is Sbin, the Sbin delayed by one pixel by the pixel delay element 1i is Sba, the Sba delayed by one pixel by the pixel delay element 1j is Sbb, and the pixel delay element 1k is delayed by one pixel. Sbb is Sbc, and Sbc delayed by one pixel by the pixel delay element 11 is Sbd.

Further, Sbin delayed by one line by the line delay element 5c is Scin, Scin delayed by one pixel by the pixel delay element 1m is Sca, Sca delayed by one pixel by the pixel delay element 1n is Scb, and one pixel delay is generated by the pixel delay element 1o. Let Scb be Scc, and Scc delayed by one pixel by the pixel delay element 1p be Scd.
Further, Sdin further delayed by one line by the line delay element 5d is Sdin, Sdin delayed by one pixel by the pixel delay element 1q is Sda, Sda delayed by one pixel by the pixel delay element 1r is Sdb, and one pixel delay is generated by the pixel delay element 1s. Let Sdb be Sdc, and Sdc delayed by one pixel by the pixel delay element 1t be Sdd.

FIG. 7 is a diagram for explaining the operation of the correlation detection unit 6.
The correlation detector 6 detects the correlation between the signals in the directions shown in FIG.
First, in FIG. 7, the correlation of Sbin, Sba, Sbb, Sbc, and Sbd is set as a correlation D1.
When the difference between adjacent pixels from Sbin to Sbd is small, it is determined that the correlation is large, and when the difference between adjacent pixels is large, it is determined that the correlation is small.
Similarly, the correlation between Sb, Sab, Sbb, Scb, and Sdb is D2, and the correlation between Sin, Saa, Sbb, Scc, and Sdd is correlation D3, and the correlation between Sd, Sac, Sbb, Sca, and Sdin is D4.

The pixel selection unit 7 selects a pixel having a correlation that intersects at right angles with the pixel having the highest correlation from the correlations D1, D2, D3, and D4 detected by the correlation detection unit 6.
For example, when it is determined that the correlation D4 is the largest, the pixel selection unit 7 selects pixels Sin, Saa, Sbb, Scc, and Sdd corresponding to the correlation D3 that intersect at right angles.
That is, it is possible to perform contour correction on pixels that intersect at right angles in order to emphasize the contours present in the pixels Sd, Sac, Sbb, Sca, and Sdin corresponding to D4.

In addition, when it is determined that the correlation level is lower than all the correlation levels and all of the correlations are small, the central pixel Sbb is determined as a pixel that exists as an isolated point, and the pixel having the smallest correlation is selected. Is done.
For example, when it is determined that all of the correlations D1, D2, D3, and D4 are small and D1 has the smallest correlation, the pixel selection unit 7 selects the pixels Sbin, Sba, Sbb, Sbc, and Sbd corresponding to the correlation D1. Select.
Accordingly, it is possible to perform more effective pixels as noise suppression instead of contour correction.
The operations of the inclination measuring units 2a, 2b, 2c, 2d, the intersection measuring units 3a, 3b, and the intermediate value measuring unit 4 are substantially the same as those in the first embodiment.

As described above, in the contour correction apparatus according to the present embodiment, the pixel signal input to the pixel signal input unit 8 is further output from the first to fourth pixels that are sequentially output with the pixel signal delayed by one line. The first to fourth line delay elements 5a to 5d, and the first to fourth line delay elements 5a to 5d respectively delay the input pixel signal by one pixel and sequentially output the first to fourth pixel delay elements. (1e-1h, 1i-1l, 1m-1p, 1q-1t) are arranged,
Correlation detection means (correlation detection section 6) for detecting correlations in the horizontal, vertical and diagonal directions of the pixels output from the pixel delay elements corresponding to each line, and correlation detection results by the correlation detection means (correlation detection section 6) Based on the above, pixel selection means (pixel selection unit 7) for selecting a pixel to be determined as noise is provided.

  As a result, in this embodiment, correction is performed by looking at the patterns (that is, correlation detection results) that change in the horizontal direction, vertical direction, and diagonal direction of the video signal (pixel signal). If the pixel to be corrected is noise, noise can be corrected without using a special noise determination circuit or noise reduction circuit. Can be reduced.

Embodiment 4 FIG.
In the first to third embodiments, the case where the input pixel signal is a luminance signal has been described as an example. However, the pixel signal may be an RGB signal that is not a luminance signal.
The operation of the contour correction apparatus according to the present embodiment is substantially the same as that of the first to third embodiments, but performs contour correction on RGB signals and also corrects noise.
That is, in the case of the present embodiment, contour correction and noise suppression (decrease) are not performed according to the amplitude of the luminance signal, but contour correction and noise suppression of the R signal are performed according to the amplitude of the R signal. G signal contour correction and noise suppression are performed according to the signal amplitude, and B signal contour correction and noise suppression are performed according to the B signal amplitude.
Thus, when the pixel signal is an RGB signal, contour correction can be performed for each color signal, and noise can also be suppressed.

It goes without saying that the contour correction apparatus according to the first to fourth embodiments described above may be applied to a video signal processing apparatus for displaying a video signal on a display device.
By using the contour correction device according to Embodiments 1 to 4 as the video signal processing device, a video signal processing device capable of appropriate contour correction and noise reduction can be obtained.

  INDUSTRIAL APPLICABILITY The present invention is useful for realizing a contour correction device that can appropriately correct the contour of an image and can easily suppress noise generated at isolated points.

1 is a block diagram illustrating a configuration of a contour correction apparatus according to a first embodiment. 6 is a diagram for explaining an operation example of an inclination measurement unit and an intersection measurement unit in the contour correction apparatus according to Embodiment 1. FIG. 5 is a flowchart showing an operation of the contour correction apparatus according to the first embodiment. 6 is a diagram for explaining the operation of an intermediate value calculation unit 4 in the contour correction apparatus according to Embodiment 1. FIG. It is a block diagram which shows the structure of the outline correction apparatus by Embodiment 2. It is a block diagram which shows the structure of the outline correction apparatus by Embodiment 3. FIG. 10 is a diagram for explaining the operation of a correlation detection unit in the contour correction apparatus according to the third embodiment.

Explanation of symbols

1a to 1d Pixel delay element 1e to 1t Pixel delay element 2a to 2d Inclination measurement unit 3a, 3b Intersection measurement unit 4 Intermediate value calculation unit 5a to 5d Line delay element 6 Correlation detection unit 7 Pixel selection unit 8 Pixel signal input unit 9 Pixel Signal output section

Claims (5)

First to fourth pixel delay elements connected in series and sequentially outputting an input pixel signal by delaying by one pixel;
First to fourth connected to each of the first to fourth pixel delay elements in parallel to measure the inclination of a pixel representing the change degree of the amplitude of the input pixel and the amplitude of the output pixel with respect to the delay time for one pixel. An inclination measuring means of
First intersection measuring means for measuring amplitude and time of a point at which the inclination of the pixel measured by the first inclination measuring means and the inclination of the pixel measured by the third inclination measuring means intersect;
A second intersection measuring means for measuring the amplitude and time of a point where the inclination of the pixel measured by the second inclination measuring means and the inclination of the pixel measured by the fourth inclination measuring means intersect;
A pixel amplitude intermediate value for correcting the amplitude of the pixel output from the second pixel delay element is calculated and output based on the effectiveness of the measurement result obtained by the first and second intersection measuring means. An outline correction apparatus comprising an intermediate value calculation means.   2. The contour correcting apparatus according to claim 1, wherein a line delay element is used instead of the first to fourth pixel delay elements. The first to fourth line delay elements that sequentially output the input pixel signal by delaying by one line are provided, and each of the first to fourth line delay elements converts the input pixel signal to one pixel. The first to fourth pixel delay elements that sequentially output after being delayed by a minute amount are arranged,
Correlation detection means for detecting correlation in the horizontal direction, vertical direction and diagonal direction of the pixels output from the pixel delay elements corresponding to each line;
The contour correction apparatus according to claim 1, further comprising a pixel selection unit that selects a pixel to be determined as noise based on a correlation detection result by the correlation detection unit.   The contour correction apparatus according to claim 1, wherein the pixel signal is an RGB signal.   A video signal processing apparatus comprising the contour correction apparatus according to claim 1.

Images