JP2013207673A - Video signal processing device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a video signal processing device in which the number of pixels of a video signal is increased, a contour of an image can be corrected to be sharp, blurs or notches of the contour can be reduced further and a smoothly connected contour line can be obtained.SOLUTION: A contour inclination direction determining section 1 determines an inclination direction of a contour in an image of a video signal S1. A pixel interpolating section 2 generates interpolation pixel data using a plurality of original pixel data items in the video signal S1 in accordance with a result of determining the inclination direction of the contour, and generates a video signal S2 in which the number of pixels is increased. A contour correcting section 3 includes a high frequency signal component generating part in which property to generate a high frequency signal component is made different in accordance with the result of determining the inclination direction of the contour. The contour correcting section 3 corrects the contour by adding the high frequency signal component to the video signal S2 and outputs the contour-corrected video signal as a video signal S3.

Description

  The present invention relates to a video signal processing apparatus and method for increasing the number of pixels of a video signal and abruptly correcting an image outline.

  A pixel interpolation circuit that increases the number of pixels of a video signal may be used to enlarge an image or convert interlaced scanning to sequential scanning. The pixel interpolation in the horizontal direction will be described with reference to FIG. In FIGS. 23A to 23C, arrows indicate pixels. FIG. 23A shows an original signal before interpolation, and FIG. 23B shows an interpolated signal obtained by interpolating pixels based on the original signal of FIG. 23A to double the number of pixels. The original signal shown in FIG. 23A is linearly interpolated as B1 = A1, B2 = 0.5 × A1 + 0.5 × A2, B3 = A2, B4 = 0.5 × A2 + 0.5 × A3. Thus, the signal shown in FIG.

  This linear interpolation is realized by processing using an interpolation filter described below. FIG. 23C shows the same waveform as FIG. 23A, and the black circle is a zero sample inserted at the interpolation point. In order to double the number of pixels of the original signal in FIG. 23A, first, zero samples are inserted at the interpolation points as shown in FIG. 23C to convert the sampling rate to double. When the signal shown in FIG. 23C with the sampling rate doubled is subjected to filtering by a 3-tap interpolation filter with a tap gain of (1/2, 1, 1/2), the signal shown in FIG. The signal shown in FIG.

  The frequency characteristics in the pixel interpolation process shown in FIG. 23 will be described with reference to FIG. The sampling rate of the signal shown in FIG. 23A is fs, and the frequency characteristic is FIG. 24A. FIG. 24B shows frequency characteristics of a 3-tap interpolation filter with a sampling rate of 2 fs with a tap gain of (1/2, 1, 1/2). The frequency characteristics of the signal of FIG. 23B interpolated by the interpolation filter having the frequency characteristics of FIG. 24B are as shown in FIG.

  As is clear from FIG. 24C, the high frequency component of the interpolated signal in which the number of pixels is doubled decreases. This is because the low-pass characteristic of the interpolation filter has a gentle attenuation characteristic. Therefore, the interpolated signal in FIG. 23B is a blurred video signal.

  The two-dimensional pixel interpolation in the horizontal direction and the vertical direction will be described with reference to FIG. FIG. 25A shows the original signal before interpolation, and FIG. 25B shows that the number of pixels is doubled by interpolating pixels in the horizontal and vertical directions based on the original signal in FIG. This is a signal after interpolation. In FIGS. 25A and 25B, hatched circles indicate original pixels, and unhatched circles indicate interpolation pixels. FIG. 25B shows coefficients for generating interpolation pixels.

Even when the original signal in FIG. 25A is converted into a signal in which the number of pixels is doubled in each of the horizontal direction and the vertical direction as shown in FIG. 25B, zero is applied to the interpolation point in the horizontal direction and the vertical direction. Insert the sample. And
1/4 1/2 1/4
1/2 1 1/2
1/4 1/2 1/4
The signal shown in FIG. 25B is obtained by performing filtering using an interpolation filter having a tap gain.

  Also in the case of two-dimensional pixel interpolation, since the low-pass characteristic of the interpolation filter has a gentle attenuation characteristic, the interpolated signal in FIG. 25B is a blurred video signal with the high-frequency component being lowered.

  Thus, the video signal in which the number of pixels is increased by the pixel interpolation circuit becomes a blurred video signal due to a decrease in the high frequency characteristics of the low-pass characteristics of the interpolation filter. In order to reduce the deterioration of the high-frequency characteristics, an interpolation filter with a high-order tap is necessary. However, if a high-order tap interpolation filter is used, the gap between the pass band and the stop band of the low-pass characteristic becomes steep and causes a problem that ringing occurs. Ringing is more likely to occur as the distance between the pass band and the stop band of the low-pass characteristic becomes steeper.

  Therefore, it is difficult to improve the blur of the video signal with only the pixel interpolation circuit, and it is necessary to perform a so-called enhancement process for adding a high frequency component to the video signal. Patent Document 1 describes a video signal processing apparatus configured to perform contour enhancement processing on a video signal, generate interpolation pixels based on the video signal subjected to contour enhancement processing, and increase the number of pixels. ing. Patent Document 2 describes a video signal processing device configured to increase the number of pixels by a pixel interpolation circuit and to correct a contour portion by a contour correction circuit.

JP 2010-171624 A Japanese Patent Laid-Open No. 11-346320

  Combining the pixel interpolation circuit and the contour correction circuit that performs enhancement processing improves the blurring of the video signal to some extent. However, the blur and jaggedness of the contour portion of the image are easily noticeable, and it is required to further reduce the blur and jaggedness of the contour portion.

  In order to meet such a demand, the present invention can increase the number of pixels of the video signal and sharply correct the contour of the image, and can further reduce the blur and jaggedness of the contour, An object of the present invention is to provide a video signal processing apparatus and method capable of obtaining smoothly connected contour lines.

  In order to solve the above-described problems of the related art, the present invention provides a contour inclination direction determination unit (1) for determining the inclination direction of the contour portion in the image of the input first video signal (S1), and the inclination The plurality of original pixel data in the first video signal is selected according to the direction determination result, the interpolation pixel data is generated using the plurality of original pixel data, and the number of pixels included in the first video signal A pixel interpolation unit (2) that generates a second video signal (S2) with an increased number of pixels, and based on the second video signal, a high characteristic with different characteristics according to the determination result of the tilt direction. A high-frequency signal component generation unit (328) that generates a high-frequency signal component, and adding the high-frequency signal component to the second video signal, thereby creating a contour in the image of the second video signal Third video signal after correcting To provide a video signal processing apparatus characterized by comprising a contour correction portion for outputting as S3) and (3).

  In the above video signal processing device, the pixel interpolation unit has a 2-tap interpolation filter when original pixel data for generating the interpolation pixel data exists in a direction along the inclination of the contour portion. It is preferable to generate the interpolated pixel data using.

  In the video signal processing device, the high-frequency signal component generation unit is a two-dimensional high-pass filter, and the contour correction unit is a tap gain in the two-dimensional high-pass filter according to a determination result of the inclination direction of the contour unit. Is preferably switched.

  In the video signal processing apparatus, the contour correction unit switches a band characteristic in the two-dimensional high-pass filter so as to have a characteristic of extracting the high-frequency signal component in a direction orthogonal to the inclination direction of the contour part. It is preferable.

  In the video signal processing device, the contour correction unit detects an upper limit value of a pixel having a level greater than the central level from an area of at least 25 pixels centered on the target pixel in the second video signal. (326), a lower limit detection unit (327) that detects, as a lower limit value, a pixel having a level lower than the central level from an area of at least 25 pixels centered on the target pixel, and the high frequency signal component in the target pixel It is preferable to include an adding unit (331) for adding, and an amplitude limiting unit (332, 333) for limiting the amplitude of the signal level of the output of the adding unit between the upper limit value and the lower limit value.

  In addition, in order to solve the above-described problems of the related art, the present invention determines the inclination direction of the contour portion in the image of the input first video signal (S1), and according to the determination result of the inclination direction. A plurality of original pixel data in the first video signal is selected, interpolation pixel data is generated using the plurality of original pixel data, and the number of pixels is increased from the number of pixels of the first video signal. The second video signal (S2) is generated, and based on the second video signal, a different high frequency signal component is generated according to the determination result of the tilt direction, and the high frequency signal component is Provided is a video signal processing method characterized in that, by adding to a second video signal, a contour portion in the image of the second video signal is corrected and output as a third video signal (S3).

  In the above video signal processing method, when there is original pixel data for generating the interpolation pixel data in a direction along the inclination of the contour portion, the interpolation pixel is used by using a 2-tap interpolation filter. It is preferable to generate data.

  In the above video signal processing method, it is preferable that the high-frequency signal component is generated using a two-dimensional high-pass filter, and the tap gain in the two-dimensional high-pass filter is switched according to the determination result of the inclination direction of the contour portion. .

  In the video signal processing method described above, it is preferable to switch the band characteristics in the two-dimensional high-pass filter so that the high-frequency signal component is extracted in a direction orthogonal to the inclination direction of the contour portion.

  In the video signal processing method, a pixel having a level higher than a central level is detected as an upper limit value from an area of at least 25 pixels centered on the target pixel in the second video signal, and at least about the target pixel. A signal of a signal in which a pixel having a level lower than the center level is detected as a lower limit value from a 25-pixel region, the high-frequency signal component is added to the target pixel, and the high-frequency signal component is added to the target pixel It is preferable to limit the amplitude between the upper limit value and the lower limit value.

  According to the video signal processing apparatus and method of the present invention, the number of pixels of the video signal can be increased and the contour of the image can be corrected sharply, and the blur and jaggedness of the contour can be further reduced. It is possible to obtain a smoothly connected contour line.

It is a block diagram which shows one Embodiment of the video signal processing apparatus of this invention. It is a figure which shows the example of the inclination direction of the outline part determined by the outline inclination direction determination part 1 in FIG. It is a figure which shows the image pattern corresponded to each outline inclination direction shown in FIG. It is a block diagram which shows the specific structural example of the outline inclination direction determination part 1 in FIG. It is a figure which shows the several pixel of the horizontal direction and the perpendicular direction which the outline inclination direction determination part 1 shown in FIG. 4 produces | generates. It is a figure which shows the example of the template which the template circuits 131-138 in FIG. 4 have. It is a block diagram which shows the specific structural example of the pixel interpolation part 2 in FIG. It is a block diagram which shows the specific structural example of the rearrangement part 221 in FIG. It is a figure for demonstrating the positional relationship of the interpolation pixel which the pixel interpolation part 2 shown in FIG. 7 produces | generates, and a reference | standard pixel. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 217 in FIG. 7 producing | generating the interpolation pixel Pi1 according to the outline inclination direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 218 in FIG. 7 producing | generating the interpolation pixel Pi2 according to the outline inclination direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 218 in FIG. 7 producing | generating the interpolation pixel Pi2 according to the outline inclination direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 219 in FIG. 7 producing | generating the interpolation pixel Pi3 according to the outline inclination direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 219 in FIG. 7 producing | generating the interpolation pixel Pi3 according to the outline inclination direction determination signal Scd. It is a block diagram which shows the specific structural example of the outline correction | amendment part 3 in FIG. It is a wave form diagram for demonstrating operation | movement of the outline correction | amendment part 3 shown in FIG. FIG. 16 is a diagram illustrating a first example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 18 is a characteristic diagram showing the spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the first example shown in FIG. 17. FIG. 16 is a diagram illustrating a second example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 20 is a characteristic diagram showing spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the second example shown in FIG. 19. FIG. 16 is a diagram illustrating a third example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 22 is a characteristic diagram showing spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the third example shown in FIG. 21. It is a figure for demonstrating the pixel interpolation of a horizontal direction. It is a figure for demonstrating the frequency characteristic in the process of the pixel interpolation shown in FIG. It is a figure for demonstrating two-dimensional pixel interpolation.

  Hereinafter, an embodiment of a video signal processing apparatus and method according to the present invention will be described with reference to the accompanying drawings. In FIG. 1, the video signal processing apparatus according to an embodiment includes a contour inclination direction determination unit 1, a pixel interpolation unit 2, and a contour correction unit 3. The video signal S1 input to the video signal processing apparatus is supplied to the contour inclination direction determination unit 1 and the pixel interpolation unit 2. The contour inclination direction determination unit 1 determines the inclination direction of the contour portion of the image of the video signal S1. An example of a specific configuration of the contour inclination direction determination unit 1 and a method of determining the inclination direction of the contour portion will be described later.

  An example of the inclination direction of the contour determined by the contour inclination direction determination unit 1 will be described with reference to FIG. In FIG. 2, white circles indicate original pixels, broken lines extending in the horizontal direction indicate the lines of the video signal S1, and broken lines extending in the vertical direction indicate the horizontal position of the video signal S1. FIG. 2 shows a range of N-1 line to N + 2 line in the vertical direction and M-1 dot to M + 2 dot in the horizontal direction. A black circle not located on the horizontal and vertical broken lines indicates one of the interpolation pixels generated by the pixel interpolation unit 2.

  As shown in FIG. 2, the contour inclination direction determination unit 1 determines which of the inclination directions A to H is the inclination direction of the contour part. The eight directions of the inclination directions A to H are merely examples, and are not limited to the eight directions. If the inclination direction of the contour portion cannot be determined, it may be determined that “there is no inclination direction”. In the present embodiment, the contour inclination direction determination unit 1 determines eight types of inclination directions A to H and nine types of “no inclination direction”.

  The inclination directions A to H of the contour portion correspond to image patterns as shown in FIGS. 3A to 3H, white circles indicate white or high luminance pixels, and black circles indicate black or low luminance pixels. An image pattern that does not apply to any of FIGS. 3A to 3H is defined as “no tilt direction”. The contour tilt direction determination unit 1 generates a contour tilt direction determination signal Scd indicating eight types of tilt directions A to H and nine types of determination results “no tilt direction”. The contour inclination direction determination signal Scd is supplied to the pixel interpolation unit 2 and the contour correction unit 3.

  Here, a specific configuration of the contour inclination direction determination unit 1 and an example of a method for determining the inclination direction of the contour portion will be described with reference to FIGS. In FIG. 4, the pixel data constituting the video signal S1 is sequentially input to the input terminal 101. Pixel data is sequentially input to the delay elements 102 to 106 in one horizontal period, and is delayed by one horizontal period. The delay elements 102 to 106 can be configured by line memories. Pixel data input to the input terminal 101 is sequentially input to D flip-flops 107 to 109 which are 1-clock delay elements, and is delayed by one clock.

  The pixel data delayed by one horizontal period output from the delay element 102 is sequentially input to the D flip-flops 110 to 113 and delayed by one clock. The two horizontal period delayed pixel data output from the delay element 103 is sequentially input to the D flip-flops 114 to 118 and delayed by one clock. Pixel data with a delay of 3 horizontal periods output from the delay element 104 is sequentially input to the D flip-flops 119 to 123 and delayed by one clock.

  The four horizontal period pixel data output from the delay element 105 are sequentially input to the D flip-flops 124 to 127 and delayed by one clock. Pixel data with a delay of 5 horizontal periods output from the delay element 106 is sequentially input to the D flip-flops 128 to 130 and delayed by one clock.

  Pixel data output from each of the D flip-flops 108 to 130 is input to the template circuits 131 to 138. Template circuits 131 to 138 have different templates. The templates included in the template circuits 131 to 138 are referred to as templates 1 to 8.

  The template circuits 131 to 138 receive pixel data of 24 pixels distributed in the horizontal direction and the vertical direction shown in FIG. In FIG. 5, pixel data of 24 pixels is distributed in a range of N-2 line to N + 3 line and M-2 dot to M + 3 dot. As can be seen from the configuration shown in FIG. 4, the template circuits 131 to 138 include two pixels of M dots and M + 1 dots for the N-2 line and the N + 3 line, and M elements for the N-1 line and the N + 2 line. Pixel data of 4 pixels from -1 dot to M + 2 dots, and 6 pixels from M-2 dots to M + 3 dots are input to the N line and the N + 1 line.

  The template circuits 131 to 138 have templates 1 to 8 that are aligned with the inclination directions of the contour portions described with reference to FIG. 3, and the degree of coincidence between the pixel data of 24 pixels in FIG. A coincidence determination value indicating the degree of coincidence is calculated and output.

  6A to 6H show examples of templates 1 to 8 included in the template circuits 131 to 138, respectively. Templates 1 to 8 are templates corresponding to the inclination directions A to H shown in FIG. The template circuits 131 to 138 multiply the pixel values of the input pixel data of 24 pixels by the numerical values indicated by the templates 1 to 8 and output the sum of the multiplied values as the coincidence determination value.

  For example, the inclination direction of the contour portion is the inclination direction A, and the pixel values of the M-2 dot, M-1 dot, and M dot row in FIG. 5 are 100, M + 1 dot, M + 2 dot, and M + 3 dot. When the pixel value of the column is 10, the matching degree determination values output from the template circuits 131 to 138 are as follows.

  In the template circuit 131, 100 × 5 × 12 + 10 × (−5) × 12 = 5400, and in the template circuit 132, 100 × 5 × 10 + 100 × (−5) × 2 + 10 × 5 × 2 + 10 × (-5) × 10 = 3600. In the template circuit 133, 100 × 6 × 8 + 100 × (−6) × 2 + 10 × 6 × 2 + 10 × (−6) × 8 = 3240, and in the template circuit 134, 100 × 5 × 8 + 100 × (-5) × 4 + 10 × 5 × 4 + 10 × (-5) × 8 = 1800.

  In the template circuit 135, 100 × 5 × 6 + 100 × (−5) × 6 + 10 × 5 × 6 + 10 × (−5) × 6 = 0, and in the template circuit 136, 100 × 5 × 4 + 100 × (-5) × 8 + 10 × 5 × 8 + 10 × (-5) × 4 = −1800. In the template circuit 137, 100 × 6 × 2 + 100 × (−6) × 8 + 10 × 6 × 8 + 10 × (−6) × 2 = −3240, and in the template circuit 138, 100 × 5 × 2 + 100 * (-5) * 10 + 10 * 5 * 10 + 10 * (-5) * 2 = -3600.

  As described above, the coincidence degree determination values calculated by the template circuits 131 to 138 are input to the direction determination unit 139. The direction determining unit 139 compares the magnitudes of the coincidence determination values from the input template circuits 131 to 138, and determines the template direction set in the template circuit that outputs the coincidence determination value indicating the largest value. It determines that it is the inclination direction of an outline part. In the above example, it can be seen that the coincidence degree determination value output from the template circuit 131 is the largest at 5400 and is in the tilt direction A.

  The direction determination unit 139 can determine which of the inclination directions A to H is the inclination direction of the contour part based on the matching degree determination values from the template circuits 131 to 138. The direction determining unit 139 determines that there is no clear contour and determines “no tilt direction” when the coincidence determination values output from the template circuits 131 to 138 are all smaller than a predetermined value. To do.

  As described above, the direction determination unit 139 generates the contour inclination direction determination signal Scd indicating the nine determination results of the eight directions of the inclination directions A to H and “no inclination direction”, and outputs it from the output terminal 140. . The contour inclination direction determination signal Scd is supplied to the pixel interpolation unit 2 and the contour correction unit 3.

  In the configuration example shown in FIG. 4, one template circuit per direction is used to determine the inclination direction of the contour portion, but a plurality of template circuits having different templates in one direction may be used. The specific configuration of the contour inclination direction determination unit 1 and the method for determining the inclination direction of the contour portion described with reference to FIGS. 4 to 6 are merely examples, and an arbitrary inclination direction determination method can be employed. However, the configuration in which the tilt direction is determined using the template is preferable because the tilt direction can be easily and accurately determined.

  The pixel interpolation unit 2 interpolates the pixels of the video signal S1 based on the contour inclination direction determination signal Scd and increases the number of pixels. The pixel interpolation unit 2 is configured as shown in FIG. 7 as an example of a specific configuration. A specific configuration and operation of the pixel interpolation unit 2 will be described with reference to FIG.

  In FIG. 7, the pixel data constituting the video signal S1 is sequentially input to the input terminal 201. Pixel data is sequentially input to the delay elements 202 to 204 for one horizontal period, and is delayed by one horizontal period. The delay elements 202 to 204 can be configured by line memories. Pixel data input to the input terminal 201 is sequentially input to D flip-flops 205 to 207 which are 1-clock delay elements, and is delayed by one clock. Pixel data with a delay of one horizontal period output from the delay element 202 is sequentially input to the D flip-flops 208 to 210 and delayed by one clock.

  The two horizontal period delayed pixel data output from the delay element 203 is sequentially input to the D flip-flops 211 to 213 and delayed by one clock. The three horizontal period pixel data output from the delay element 204 are sequentially input to the D flip-flops 214 to 216 and delayed by one clock.

  Pixel data input from the input terminal 201, pixel data output from each of the delay elements 202 to 204, and pixel data output from each of the D flip-flops 205 to 216 are input to the interpolation calculation units 217 to 219. . A total of 16 pixel data of 4 pixels in the horizontal direction and 4 pixels in the vertical direction are input to the interpolation calculation units 217 to 219. In the present embodiment, pixel data output from the D flip-flop 208 is set as a target pixel. The input terminal 220 receives a contour inclination direction determination signal Scd.

  As shown in FIG. 7, the pixel data input to the input terminal 201 is P1, and the pixel data output from the D flip-flops 205 to 207 are P2 to P4. The pixel data output from the delay element 202 is P5, and the pixel data output from the D flip-flops 208 to 210 is P6 to P8. Pixel data P6 is pixel data of the target pixel. The pixel data output from the delay element 203 is P9, and the pixel data output from the D flip-flops 211 to 213 is P10 to P12. The pixel data output from the delay element 204 is P13, and the pixel data output from the D flip-flops 214 to 216 is P14 to P16.

  The pixel data P1 to P16 are pixel data (original pixel data) of original pixels indicated by white circles in FIG. The target pixel data P6 is original pixel data of 2N + 2 lines and 2M + 2 dots. The interpolation calculation unit 217 generates interpolation pixel data Pi1 at the position of 2N + 1 lines and 2M + 1 dots. The interpolation calculation unit 218 generates interpolation pixel data Pi2 at the position of 2N + 1 lines and 2M + 2 dots. The interpolation calculation unit 219 generates interpolation pixel data Pi3 at the position of 2N + 2 lines and 2M + 1 dots. The interpolation calculation units 217 to 219 differ in the position and tap gain of the original pixel data used when generating the interpolation pixel data Pi1 to Pi3 according to the determination result of the inclination direction of the contour indicated by the contour inclination direction determination signal Scd. Make it.

  The interpolation calculation unit 217 generates the interpolated pixel data Pi1 by one of the interpolation filters shown in FIGS. 9A to 9C according to the determination result indicated by the contour inclination direction determination signal Scd.

  When the contour inclination direction determination signal Scd indicates the directions F, G, and H, the interpolation calculation unit 217 performs (1 / 2,1 / 1 /) on the pixel data P6 and P11 as shown in FIG. Filtering by the 2-tap interpolation filter of 2) is performed to generate interpolated pixel data Pi1. When the contour inclination direction determination signal Scd indicates the directions B, C, and D, the interpolation calculation unit 217 has a tap gain of (1/2) for the pixel data P7 and P10 as shown in FIG. 9B. , 1/2) is subjected to filtering by a 2-tap interpolation filter to generate interpolated pixel data Pi1.

When the contour inclination direction determination signal Scd indicates the directions A and E, and “no inclination direction”, the interpolation calculation unit 217 applies the pixel data P1 to P16 to the pixel data P1 to P16 as shown in FIG.
1/64 -5/64 -5/64 1/64
-5/64 25/64 25/64 -5/64
-5/64 25/64 25/64 -5/64
1/64 -5/64 -5/64 1/64
Interpolated pixel data Pi1 is generated by performing filtering using the 16-tap interpolation filter.

  The interpolation calculation unit 218 performs interpolation pixel data Pi2 using one of the interpolation filters shown in FIGS. 10A and 10B and FIGS. 11A and 11B according to the determination result indicated by the contour inclination direction determination signal Scd. Is generated.

  When the contour inclination direction determination signal Scd indicates the directions F, G, and H, the interpolation calculation unit 218 applies (1/2, 1/1) to the pixel data P5 and P11 as shown in FIG. Filtering by the 2-tap interpolation filter of 2) is performed to generate interpolated pixel data Pi2. When the contour inclination direction determination signal Scd indicates the directions B, C, and D, the interpolation calculation unit 218 has a tap gain of (1/2) for the pixel data P7 and P9 as shown in FIG. , 1/2) is subjected to filtering by a 2-tap interpolation filter to generate interpolated pixel data Pi2.

  When the contour tilt direction determination signal Scd indicates the direction E, “no tilt direction”, the interpolation calculation unit 218 taps the pixel data P2, P6, P10, and P14 as shown in FIG. Filtering is performed by a 4-tap interpolation filter having a gain of (-1/8, 5/8, 5/8, -1/8) to generate interpolation pixel data Pi2. When the contour inclination direction determination signal Scd indicates the direction A, the interpolation calculation unit 218 has a tap gain of (1/2, 1/2) with respect to the pixel data P6, P10 as shown in FIG. ) To generate interpolated pixel data Pi2.

  When the contour inclination direction determination signal Scd indicates the directions F, G, and H, the interpolation calculation unit 219 performs (1/2, 1/1) on the pixel data P2 and P11 as shown in FIG. Filtering by the 2-tap interpolation filter of 2) is performed to generate interpolated pixel data Pi3. When the contour inclination direction determination signal Scd indicates directions B, C, and D, the interpolation calculation unit 219 has a tap gain of (1/2) with respect to the pixel data P3 and P10 as shown in FIG. , 1/2) is subjected to filtering by a 2-tap interpolation filter to generate interpolated pixel data Pi3.

  When the contour inclination direction determination signal Scd indicates the direction A, “no inclination direction”, the interpolation calculation unit 219 has a tap gain of (−−) for the pixel data P5 to P8 as shown in FIG. 1/8, 5/8, 5/8, -1/8) is filtered by a 4-tap interpolation filter to generate interpolation pixel data Pi3. When the contour inclination direction determination signal Scd indicates the direction E, the interpolation calculation unit 219 performs (1/2, 1/2) 2 on pixel data P6 and P7 as shown in FIG. Interpolated pixel data Pi3 is generated by performing filtering using a tap interpolation filter.

  As described above, the interpolation pixel data Pi1 to Pi3 and the target pixel data P6 generated by the interpolation calculation units 217 to 219 are input to the rearrangement unit 221. The configuration and operation of the rearrangement unit 221 will be described with reference to FIG. As shown in FIG. 14, the rearrangement unit 221 includes line buffers 2215 to 2218, switches 2219, 2220 and 2222, and toggle circuits 2221 and 2223.

  Interpolated pixel data Pi1 to Pi3 are input to line buffers 2215 to 2217 via input terminals 2211 to 2213. The target pixel data P6 is input to the line buffer 2218 via the input terminal 2214. The line buffers 2215 to 2217 respectively hold interpolation pixel data Pi1 to Pi3 for one line. The line buffer 2218 holds the target pixel data P6 for one line.

  The toggle circuit 2221 is a circuit whose value repeatedly changes from “0” to “1” and from “1” to “0” in a predetermined cycle. As shown in FIG. 8, the toggle circuit 2221 outputs “0” when the original pixel is located, 2M-2 dots, 2M dots, 2M + 2 dots, 2M + 4 dots,. The interpolation pixel to be generated should be generated: 2M-1 dot, 2M + 1 dot, 2M + 3 dot,. The line buffers 2215 to 2217 read out the respective data with a read clock four times the write clock in which the interpolated pixel data Pi1 to Pi3 and the target pixel data P6 are written.

  Interpolated pixel data Pi1 and Pi2 read from the line buffers 2215 and 2216 are input to the switch 2219. The interpolated pixel data Pi3 and the target pixel data P6 read from the line buffers 2217 and 2218 are input to the switch 2220. The switch 2219 selects the interpolation pixel data Pi1 output from the line buffer 2215 when the toggle circuit 2221 outputs “1”, and the interpolation pixel output from the line buffer 2216 when the toggle circuit 2221 outputs “0”. Data Pi2 is selected. The switch 2220 selects the interpolated pixel data Pi3 output from the line buffer 2217 when the toggle circuit 2221 outputs “1”, and the target pixel output from the line buffer 2218 when the toggle circuit 2221 outputs “0”. Data P6 is selected.

  The interpolation pixel data Pi1 and Pi2 selectively output from the switch 2219 and the interpolation pixel data Pi3 and target pixel data P6 selectively output from the switch 2220 are input to the switch 2222.

  The toggle circuit 2223 is a circuit whose value repeatedly changes from “0” to “1” and from “1” to “0” in a predetermined cycle. As shown in FIG. 8, the toggle circuit 2223 outputs “0” in the 2N-2 line, 2N line, 2N + 2 line, 2N + 4 line,. "1" is output for the 2N-1 line, 2N + 1 line, 2N + 3 line,. The switch 2222 selects the interpolation pixel data Pi1 or the interpolation pixel data Pi2 output from the switch 2219 when the toggle circuit 2223 outputs “1”, and is output from the switch 2220 when the toggle circuit 2223 outputs “0”. The interpolated pixel data Pi3 or the target pixel data P6 is selected and output from the output terminal 222.

  Through the above operation, the pixel interpolation unit 2 shown in FIG. 1 outputs the video signal S2 in which the number of pixels is doubled in the horizontal direction and the vertical direction.

  In the present embodiment, the contour tilt direction determination unit 1 determines the tilt direction of the contour portion of the image, and according to the nine types of determination results of the eight tilt directions A to H of the contour portion and “no tilt direction”. Interpolated pixel data is generated by an interpolation filter having the number of taps. When the interpolation pixel data Pi1 is generated in the directions B to D and F to H, the interpolation pixel data Pi3 is generated in the directions B to E and F to H when the interpolation pixel data Pi2 is generated in the directions A to D and F to H. When generating, a 2-tap interpolation filter is used.

  That is, in this embodiment, when there is a pixel in a direction along the inclination of the contour portion, a 2-tap interpolation filter is used. Therefore, ringing hardly occurs in the pixel interpolation performed by the pixel interpolation unit 2 described with reference to FIGS.

  In FIG. 1, the video signal S <b> 2 output from the pixel interpolation unit 2 is input to the contour correction unit 3. The contour correcting unit 3 corrects the contour portion of the video signal S2 to be steep in accordance with the contour tilt direction determination signal Scd. A specific configuration and operation of the contour correction unit 3 will be described with reference to FIG.

  In FIG. 15, pixel data constituting the video signal S2 is sequentially input to the input terminal 301. Pixel data is sequentially input to the delay elements 302 to 305 in one horizontal period, and is delayed by one horizontal period. The delay elements 302 to 305 can be configured by line memories. Pixel data input to the input terminal 301 is sequentially input to D flip-flops 306 to 309 which are one-clock delay elements, and is delayed by one clock. The pixel data delayed by one horizontal period output from the delay element 302 is sequentially input to the D flip-flops 310 to 313 and delayed by one clock.

  Pixel data with a delay of two horizontal periods output from the delay element 303 is sequentially input to the D flip-flops 314 to 317 and delayed by one clock. The three horizontal period delayed pixel data output from the delay element 304 is sequentially input to the D flip-flops 318 to 321 and delayed by one clock. The 4-horizontal period delayed pixel data output from the delay element 305 is sequentially input to the D flip-flops 322 to 325 and delayed by one clock.

  The pixel data input to the input terminal 301, the pixel data output from each of the delay elements 302 to 305, and the pixel data output from each of the D flip-flops 306 to 325 include an upper limit detector 326, a lower limit detector 327, Input to a two-dimensional high-pass filter (HPF) 328.

  As shown in FIG. 15, the pixel data input to the input terminal 301 is P01, the pixel data output from the D flip-flops 306 to 309 is P02 to P05, the pixel data output from the delay element 302 is P06, The pixel data output from the D flip-flops 310 to 313 are P07 to P10.

  The pixel data output from the delay element 303 is P11, the pixel data output from the D flip-flops 314 to 317 is P12 to P15, the pixel data output from the delay element 304 is P16, and the D flip-flops 318 to 321 are used. Each output pixel data is set to P17 to P20. The pixel data output from the delay element 305 is P21, and the pixel data output from the D flip-flops 322 to 325 is P22 to P25.

  The upper limit detection unit 326, the lower limit detection unit 327, and the two-dimensional HPF 328 receive pixel data of a total of 25 pixels including 5 pixels in the horizontal direction and 5 pixels in the vertical direction centered on the pixel data P13 output from the D flip-flop 315. It will be. In the contour correcting unit 3, the pixel data P13 is pixel data of the target pixel. Pixel data P13 is input to adder 331 as signal S315. It is assumed that the signal S315 has a waveform in which the pixel data P13 changes as time passes as shown in FIG.

  A black circle constituting the signal S315 is the pixel data P13. The signal S315 includes noises N1 and N2. In FIGS. 16A to 16G, only the horizontal waveform of each signal is shown for simplification.

  The upper limit detection unit 326 ranks the input pixel data of 25 pixels centered on the pixel data P13 that is the target pixel from 1 to 25 in order of decreasing level (or increasing order), and the pixels of the 25 pixels For example, pixel data having a level that is the fourth largest from the maximum level is detected instead of the pixel data having the maximum level. The pixel data detected by the upper limit detector 326 is input to the minimum value detector 333 as a signal S326 having a waveform as shown in FIG.

  The upper limit detection unit 326 detects pixel data of a level larger than the central level except for the pixel data of the maximum level among the input pixel data. Preferably, the upper limit detection unit 326 equally divides the input pixel data from the minimum level to the maximum level into a small level portion, an intermediate level portion, and a large level portion, and the maximum level pixel in the large level portion. Pixel data at a level excluding data is detected. Therefore, in the configuration of FIG. 15, the upper limit detection unit 326 may detect the second highest level pixel data, the third highest level pixel data, and the fifth highest level pixel data.

  The reason why the upper limit detection unit 326 detects pixel data of a level larger than the central level except for the pixel data of the maximum level is to remove the noise N2 included in the signal S315. What level of pixel data the upper limit detection unit 326 detects may be selected in accordance with the relationship between the noise removal effect and the contour correction effect.

  The lower limit detection unit 327 ranks the input pixel data of 25 pixels centered on the pixel data P13 that is the target pixel from 1 to 25 in order of increasing level (or increasing order), and the pixels of the 25 pixels For example, pixel data having a level that is the fourth smallest from the minimum level is detected instead of the pixel data having the minimum level. The pixel data detected by the lower limit detection unit 327 is input to the maximum value detection unit 332 as a signal S327 having a waveform as shown in FIG.

  The lower limit detection unit 327 detects pixel data of a level higher than the central level except for pixel data of the maximum level among the input pixel data. Desirably, the lower limit detection unit 327 equally divides the input pixel data from the minimum level to the maximum level into a small level portion, an intermediate level portion, and a large level portion, and the minimum level pixel in the small level portion. Pixel data at a level excluding data is detected. Therefore, in the configuration of FIG. 15, the lower limit detection unit 327 may detect the second lowest level pixel data, the third lowest level pixel data, and the fifth lowest level pixel data.

  The reason why the lower limit detection unit 327 detects pixel data of a level smaller than the center level except for pixel data of the minimum level is to remove noise N1 included in the signal S315. What level of pixel data the lower limit detection unit 327 detects may be selected based on the relationship between the noise removal effect and the contour correction effect.

  When the video signal S2 input to the contour correcting unit 3 does not contain noise, the upper limit detecting unit 326 may detect the maximum level and the lower limit detecting unit 327 may detect the minimum level. Even if the video signal S2 includes noise, if the noise is not removed, the maximum level may be detected by the upper limit detection unit 326 and the minimum level may be detected by the lower limit detection unit 327.

  The two-dimensional HPF 328 is a 25 tap filter. The two-dimensional HPF 328 generates a high frequency signal component to be added to the signal S315 by filtering with the tap gain as described later using the input pixel data of 25 pixels. At this time, the two-dimensional HPF 328 is configured to switch the tap gain according to the contour inclination direction determination signal Scd input from the input terminal 329. It is preferable that the two-dimensional HPF 328 switches the band characteristics of the filter so that the high-frequency signal component is extracted in a direction orthogonal to the inclination direction of the contour portion indicated by the contour inclination direction determination signal Scd.

  For example, the two-dimensional HPF 328 extracts a high frequency signal component so that the horizontal high frequency is obtained in the direction A shown in FIG. 3A and the vertical high frequency is obtained in the direction E shown in FIG. In the case of the direction C shown in FIG. 3C, the two-dimensional HPF 328 has a tap gain shown in FIG. 17 and a spatial frequency characteristic as shown in FIG. In the case of the direction G shown in FIG. 3G, the two-dimensional HPF 328 has a tap gain shown in FIG. 19 and a spatial frequency characteristic as shown in FIG. In the cases other than the directions A and G, the two-dimensional HPF 328 has a tap gain shown in FIG. 21 and a spatial frequency characteristic as shown in FIG.

  In this embodiment, the tap gain and the spatial frequency characteristic are switched in three directions other than the direction A, the direction G, the direction A, and the direction G, but the tap gain and the spatial frequency characteristic may be switched more finely. Different tap gain and spatial frequency characteristics may be used for all nine types of directions A to H and “no tilt direction”. It is preferable that all nine types have different tap gains and spatial frequency characteristics because the configuration of the two-dimensional HPF 328 is complicated, but tap gains and spatial frequency characteristics corresponding to each of the nine types can be obtained.

  As described above, the two-dimensional HPF 328 generates a high-frequency signal component by switching the tap gain according to the contour inclination direction determination signal Scd, so that the contour portion is steep according to the direction of each contour portion. An optimum high frequency signal component can be obtained. The two-dimensional HPF 328 operates as a high frequency signal component generation unit.

  The high frequency signal component output from the two-dimensional HPF 328 is input to the multiplier 330. If the gain coefficient of the multiplier 330 is 2, for example, the multiplier 330 doubles the input high frequency signal component and outputs a signal S330 having a waveform as shown in FIG. The signal S330 is input to an adder 331 that is an addition unit. The adder 331 adds the input signal S315 and the signal S330, and outputs a signal S331 having a waveform as shown in FIG. The signal S331 is a signal obtained by adding a high-frequency signal component to the pixel data P13 of the target pixel that is sequentially input as time passes. The signal S331 is input to the maximum value detection unit 332.

  The maximum value detection unit 332 selects the larger one of the signal S327 output from the lower limit detection unit 327 and the signal S331 output from the adder 331, and the signal having a waveform as shown in FIG. S332 is output. Undershoot is added to the signal S331 output from the adder 331 as shown in FIG. By selecting the larger one of the signal S327 and the signal S331 by the maximum value detection unit 332, the signal S332 becomes a signal from which the undershoot portion has been removed as shown in FIG. The signal S332 is input to the minimum value detection unit 333.

The minimum value detection unit 333 selects the smaller one of the signal S326 and the signal S332,
A signal S333 having a waveform as shown in FIG. Overshoot is added to the signal S332 as shown in FIG. By selecting the smaller one of the signal S326 and the signal S332 by the minimum value detection unit 333, the signal S333 becomes a signal from which the overshoot portion is removed as shown in FIG. The signal S333 is output from the output terminal 334 as the video signal S3 output from the contour correcting unit 3 in FIG.

  The maximum value detection unit 332 and the minimum value detection unit 333 operate as an amplitude limiting unit that limits the amplitude of the signal S331 between the upper limit value by the upper limit detection unit 326 and the lower limit value by the lower limit detection unit 327.

  With the above operation, the video signal S3 (S333) output from the contour correcting unit 3 has a steeper slope near the center of the tilt than the signal S315, and the contour is corrected without adding a shoot component. Note that the inclination near the center of the inclination in the contour portion of the signal S333 can be freely set by the value of the gain coefficient of the multiplier 330.

  The upper limit detection unit 326 detects pixels having a level greater than the central level except for the maximum level pixel among the input pixels, and the lower limit detection unit 327 determines the center of the input pixels except for the minimum level pixel. Since pixels having a level lower than the level are detected, even if noise is mixed in the input signal, the noise is completely removed. In the present embodiment, since the contour correcting unit 3 having the above configuration is provided, a video signal processing apparatus having excellent noise resistance characteristics can be obtained.

  In this embodiment, the two-dimensional HPF 328, which is a high-frequency signal component generation unit, is configured to switch the band characteristics of the filter in accordance with the inclination direction of the contour indicated by the contour inclination direction determination signal Scd. Compared with the conventional video signal processing apparatus and method, blurring and jaggedness of the contour portion can be further reduced, and a smoothly connected contour line can be obtained.

  The present invention is not limited to the embodiment described above, and various modifications can be made without departing from the scope of the present invention. In the contour correction unit 3 shown in FIG. 15, the region of the two-dimensional HPF 328 for generating a high frequency signal is 25 taps in the horizontal direction 5 and the vertical direction 5, and the region by the upper limit detection unit 326 and the lower limit detection unit 327 is also horizontal. Although the two areas are set to be the same as 25 pixels in the direction 5 and the vertical direction 5, they are not necessarily the same.

  When the contour portion is corrected both in the horizontal direction and in the vertical direction, the upper limit detection unit 326 performs centering from an area of at least 5 pixels in the horizontal direction and 5 pixels in the vertical direction centered on the target pixel. The lower level detection unit 327 may detect a pixel having a level lower than the central level as a lower limit value from an area of at least 25 pixels centered on the target pixel.

  In the present embodiment, as a preferred configuration, the contour correction unit 3 that performs contour correction without adding a shoot component is taken as an example, but a contour correction unit that adds a shoot component and corrects the contour may be used. A high-frequency signal component generating unit that generates a high-frequency signal component to be added to pixel data of the pixel of interest is provided with a contour correction unit that corrects the contour, and the high-frequency is determined according to the inclination direction of the contour of the image. What is necessary is just to switch the band characteristic of the filter which produces | generates a signal component.

DESCRIPTION OF SYMBOLS 1 Contour inclination direction determination part 2 Pixel interpolation part 3 Contour correction part 326 Upper limit detection part 327 Lower limit detection part 328 Two-dimensional high-pass filter (high frequency signal component generation part)
331 Adder (additional part)
332 Maximum value detector (amplitude limiter)
333 Minimum value detection unit (amplitude limiting unit)

It is a block diagram which shows one Embodiment of the video signal processing apparatus of this invention. It is a figure which shows the example of the inclination direction of the outline part determined by the outline inclination direction determination part 1 in FIG. It is a figure which shows the image pattern corresponded to each outline inclination direction shown in FIG. It is a block diagram which shows the specific structural example of the outline inclination direction determination part 1 in FIG. It is a figure which shows the several pixel of the horizontal direction and the perpendicular direction which the outline inclination direction determination part 1 shown in FIG. 4 produces | generates. It is a figure which shows the example of the template which the template circuits 131-138 in FIG. 4 have. It is a block diagram which shows the specific structural example of the pixel interpolation part 2 in FIG. It is a figure for demonstrating the positional relationship of the interpolation pixel which the pixel interpolation part 2 shown in FIG. 7 produces | generates, and a reference | standard pixel . It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 217 in FIG. 7 producing | generating the interpolation pixel Pi1 according to the outline inclination direction determination signal Scd . FIG. 8 is a diagram illustrating an example of tap gain when the interpolation calculation unit 218 in FIG. 7 generates an interpolation pixel Pi 2 in accordance with the contour inclination direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 218 in FIG. 7 producing | generating the interpolation pixel Pi2 according to the outline inclination direction determination signal Scd. Is a diagram illustrating an example of a tap gain in generating the interpolated pixel Pi 3 interpolation operation unit 21 9 in Figure 7 is in accordance with the contour tilt direction determination signal Scd. It is a figure which shows the example of the tap gain at the time of the interpolation calculating part 219 in FIG. 7 producing | generating the interpolation pixel Pi3 according to the outline inclination direction determination signal Scd. It is a block diagram which shows the specific structural example of the rearrangement part 221 in FIG. It is a block diagram which shows the specific structural example of the outline correction | amendment part 3 in FIG. It is a wave form diagram for demonstrating operation | movement of the outline correction | amendment part 3 shown in FIG. FIG. 16 is a diagram illustrating a first example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 18 is a characteristic diagram showing the spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the first example shown in FIG. 17. FIG. 16 is a diagram illustrating a second example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 20 is a characteristic diagram showing spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the second example shown in FIG. 19. FIG. 16 is a diagram illustrating a third example of tap gain when the two-dimensional high-pass filter 328 in FIG. 15 generates a high-frequency signal component. FIG. 22 is a characteristic diagram showing spatial frequency characteristics when the tap gain of the two-dimensional high-pass filter 328 in FIG. 15 is the third example shown in FIG. 21. It is a figure for demonstrating the pixel interpolation of a horizontal direction. It is a figure for demonstrating the frequency characteristic in the process of the pixel interpolation shown in FIG. It is a figure for demonstrating two-dimensional pixel interpolation.

Claims (10)

A contour inclination direction determination unit that determines the inclination direction of the contour portion in the image of the input first video signal;
A plurality of original pixel data in the first video signal is selected according to the determination result of the tilt direction, interpolation pixel data is generated using the plurality of original pixel data, and the first video signal has A pixel interpolation unit that generates a second video signal in which the number of pixels is increased from the number of pixels;
A high-frequency signal component generating unit configured to generate a high-frequency signal component having different characteristics according to the determination result of the tilt direction based on the second video signal; An outline correction unit that corrects the outline in the image of the second video signal and outputs it as a third video signal by adding to the video signal of 2;
A video signal processing apparatus comprising:   When the original pixel data for generating the interpolation pixel data exists in a direction along the inclination of the contour portion, the pixel interpolation unit converts the interpolation pixel data using a 2-tap interpolation filter. The video signal processing apparatus according to claim 1, wherein the video signal processing apparatus generates the video signal processing apparatus. The high-frequency signal component generation unit is a two-dimensional high-pass filter;
The video signal processing apparatus according to claim 1, wherein the contour correction unit switches a tap gain in the two-dimensional high-pass filter according to a determination result of a tilt direction of the contour portion.   The contour correction unit switches a band characteristic in the two-dimensional high-pass filter so as to have a characteristic of extracting the high-frequency signal component in a direction orthogonal to the inclination direction of the contour part. The video signal processing apparatus described. The contour correction unit
An upper limit detector for detecting, as an upper limit, a pixel having a level greater than the central level from an area of at least 25 pixels centered on the target pixel in the second video signal;
A lower limit detection unit for detecting, as a lower limit value, a pixel having a level smaller than the central level from an area of at least 25 pixels centered on the target pixel;
An adding unit for adding the high frequency signal component to the target pixel;
An amplitude limiting unit that limits the signal level of the output of the additional unit between the upper limit value and the lower limit value;
5. The video signal processing apparatus according to claim 1, wherein the video signal processing apparatus includes: Determining the inclination direction of the contour in the image of the input first video signal;
A plurality of original pixel data in the first video signal is selected according to the determination result of the tilt direction, interpolation pixel data is generated using the plurality of original pixel data, and the first video signal has Generating a second video signal having a larger number of pixels than the number of pixels;
Based on the second video signal, a different high-frequency signal component is generated according to the determination result of the tilt direction,
Video signal processing characterized in that the high frequency signal component is added to the second video signal to correct a contour in the image of the second video signal and output as a third video signal. Method.   When the original pixel data for generating the interpolation pixel data exists in a direction along the inclination of the contour portion, the interpolation pixel data is generated using a 2-tap interpolation filter. The video signal processing method according to claim 6. Generating the high-frequency signal component using a two-dimensional high-pass filter;
The video signal processing method according to claim 6 or 7, wherein a tap gain in the two-dimensional high-pass filter is switched in accordance with a determination result of an inclination direction of the contour portion.   9. The video signal processing method according to claim 8, wherein band characteristics in the two-dimensional high-pass filter are switched so that the high-frequency signal component is extracted in a direction orthogonal to the inclination direction of the contour portion. . Detecting a pixel having a level greater than the central level as an upper limit value from an area of at least 25 pixels centered on the target pixel in the second video signal;
From a region of at least 25 pixels centered on the target pixel, a pixel having a level smaller than the central level is detected as a lower limit value,
Adding the high frequency signal component to the pixel of interest;
10. The signal level of a signal obtained by adding the high frequency signal component to the target pixel is amplitude-limited between the upper limit value and the lower limit value. 10. Video signal processing method.

Images