JP3096619B2 - Video signal processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a video signal processing apparatus for subjecting a video signal output from a solid-state image sensor to contour correction of a subject image, that is, aperture processing.

[0002]

2. Description of the Related Art When color imaging is performed using a single image sensor, a stripe or mosaic color filter is mounted on a light receiving portion of the image sensor to associate each pixel of the light receiving portion with a predetermined color component. doing. A video signal output from such an image sensor corresponds to a configuration of a color filter, and video information representing each color component is repeated in a predetermined order. In the process of signal processing for this video signal, usually,
The processing of the luminance information and the processing of the color information are performed independently. In the processing of the luminance information, a video signal representing each color component is synthesized at a predetermined ratio to generate a luminance signal. In the processing of the color information, a color signal representing a desired color component is generated by matrix processing. . In these signal processing, analog signal processing has been conventionally used. However, in the case of a recent multimedia-capable imaging device, in order to be able to supply video information as digital data to a computer device or the like, an output side is used. Digital signal processing has been adopted.

FIG. 7 is a block diagram showing a configuration of a conventional image pickup apparatus. The CCD solid-state imaging device 1 has a plurality of light receiving pixels arranged in a matrix and a shift register associated with each light receiving pixel, and accumulates information charges generated by photoelectric conversion in each light receiving pixel. Further, by mounting the color filter, each light receiving pixel is associated with a predetermined color component.
The driver 2 generates a multi-phase clock pulse in response to the horizontal synchronization signal and the vertical synchronization signal, and supplies the multi-phase clock pulse to the CCD 1. The timing control circuit 3 includes a counter that counts a reference clock, and various synchronization signals (horizontal synchronization signals,
Vertical synchronization signal, etc.). As a result, the CCD 1 is pulse-driven at a predetermined timing, and information charges accumulated in each light receiving pixel are sequentially transferred and output. Then, on the output side of the shift register, the amount of the information charge is converted into a voltage value, and video information corresponding to each light receiving pixel is continuously output in units of one row. The analog signal processing unit 4 captures the output voltage of the CCD 1, performs processing such as sampling and gain control, and generates a video signal according to a predetermined format. The A / D conversion circuit 5 performs analog / digital conversion of the video signal supplied from the analog signal processing unit 4 for each pixel, and generates video data. Digital signal processing unit 6
Captures video data generated by the A / D conversion circuit 5, generates luminance data and color data through predetermined matrix processing, and then performs processing such as contour correction and gamma correction on the luminance data. For example, as shown in FIG. 8, when a mosaic type color filter including yellow (Ye), cyan (Cy), green (G), and white (W) is employed, each color component is averaged to obtain luminance data Y. The red component data R is generated from the difference between W and Ye or the difference between Cy and G, and the blue component data B is generated from the difference between W and Cy or the difference between Ye and G. Then, the brightness data Y is subjected to a so-called aperture process for performing contour correction of a video so as to emphasize a component of a specific frequency band, and further, an input signal level and a luminous brightness on a side for displaying a video signal are provided. Gamma correction is performed to correct the nonlinearity of. For each of the color component data R and B, color difference data U and V are generated from the difference from the luminance data Y, and output as video data together with the luminance data Y.

[0004]

In the gamma correction process for the luminance data Y, as shown in FIG. 9, there is a tendency that low-level luminance data is amplified and high-level luminance data is suppressed. For this reason, if the luminance data subjected to the aperture processing is subjected to gamma correction processing, noise generated in the aperture processing is likely to be amplified in a low luminance part,
As a result, the S / N ratio of the low-luminance portion will be degraded. And
In the high luminance portion, the aperture component is suppressed together with the luminance data, so that the contour correction becomes insufficient, and as a result, the image quality of the reproduced screen is deteriorated.

Accordingly, it is an object of the present invention to perform appropriate aperture processing while performing gamma correction processing on luminance data.

[0006]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and is characterized in that a plurality of light-receiving pixels are arranged in a matrix in a solid-state imaging device. In a video signal processing device for generating luminance data representing the luminance of a video based on a video signal output in units of one row corresponding to information charges stored in pixels, the video signal output from the solid-state imaging device is An A / D conversion circuit for converting each pixel into digital data to generate video data corresponding to each light receiving pixel; and sequentially taking in the continuous video data and removing a predetermined high-frequency component to obtain first luminance data A first filter circuit that sequentially captures the video data and selectively extracts frequency components in a specific band to generate aperture data; and a second filter circuit that generates aperture data. A correction circuit for non-linearly converting the first luminance data obtained from the first filter circuit to generate second luminance data; and obtaining the second luminance data obtained from the correction circuit from the second filter circuit. An aperture adding circuit that adds the obtained aperture data to generate third luminance data.

Thus, aperture data for contour correction is generated from the luminance data before the non-linear conversion, and the aperture data is added to the luminance data after the non-linear conversion processing. Therefore, the nonlinear conversion process can be performed only on the luminance data without performing the nonlinear conversion on the aperture data itself.

[0008]

FIG. 1 is a block diagram showing the configuration of a video signal processing apparatus according to the present invention. The analog signal processing unit 10 receives an output of a CCD solid-state imaging device driven at a predetermined cycle, and includes a correlated double sampling circuit (CDS) 11 and an automatic gain control circuit (AGC) 12.
It is composed of The correlated double sampling circuit 11 receives the output of the CCD in which the reference potential and the signal potential are repeated at a predetermined clock cycle, samples the reference potential portion and the signal potential portion, and extracts the potential difference as a video signal. As a result, a video signal is obtained in which a potential level corresponding to the amount of information charges stored in each light receiving pixel of the CCD is maintained for one clock period. Automatic gain control circuit 1
The control unit 2 gives a gain to the video signal output from the correlated double sampling circuit 11 according to, for example, an average level in one vertical scanning period, and controls the average level in each vertical scanning period to be substantially uniform. I do. That is, the gain is variably set based on the integral value of the video signal output from the correlated double sampling circuit 12 for each vertical scanning period, and feedback control is performed so that the integral value falls within a predetermined range. . As a result, a video signal whose average level is almost uniformly set for each screen is output.

The A / D conversion circuit 13 converts the video signal output from the analog signal processing unit 10 in synchronization with the output operation of the CCD, that is, the sampling operation of the correlated double sampling circuit 11 of the analog signal processing unit 10. Analog / digital conversion is performed sequentially to generate video data. As a result, video data Y0 corresponding to each light receiving pixel of the CCD is generated.

The digital signal processing section 14 comprises a luminance data generating circuit 15, a gamma correction circuit 16, an aperture data generating circuit 17, and an aperture adding circuit 18. The luminance data generation circuit 15 is a digital filter having a low-pass characteristic.
By extracting the following frequency components, the luminance data Y1
Generate The gamma correction circuit 16 is a non-linear conversion circuit having a predetermined conversion table, and its conversion characteristics are shown in FIG.
Are set to follow a gamma curve as shown in FIG. As a result, luminance data Y2 that has been gamma-corrected corresponding to the luminance data Y1 is generated. Note that, in addition to using the conversion table, the gamma correction circuit shown in FIG. 9 can be approximated by combining a plurality of linear conversions. Aperture data generation circuit 17
Is a digital filter having a band-pass characteristic for extracting a frequency component in a specific band.
From 0 to 1 of the sampling frequency of the A / D conversion circuit 13
Aperture data A is generated by extracting the 周波 数 frequency component as the center. The aperture addition circuit 18 is an addition circuit, and adds the aperture data generation circuit 1 to the luminance data Y2 obtained from the gamma correction circuit 16.
The luminance data Y3 is generated by adding the aperture data A obtained from. In the luminance data Y3, since the aperture component has not been subjected to the gamma correction processing, the outline is not unnecessarily emphasized in the portion representing the low luminance image.

The operation timing of each unit of the analog signal processing unit 10 and the digital signal processing unit 14 together with the operation timing of the A / D conversion circuit 13 is controlled by a timing control circuit (not shown) for determining the drive timing of the CCD. The supplied timing clocks synchronize with each other. FIG. 2 is a block diagram showing an example of the luminance data generation circuit 15, which shows a digital filter for three pixels in the horizontal direction.

First to third latches 21, 22, 23
Are connected in series to form a shift register, and store successive video data Y0 (n + 1), Y0 (n) and Y0 (n-1). The multiplier 24 multiplies the video data Y0 (n) latched by the second latch 22 by a coefficient “2”, and supplies the multiplication result to the adder 25. The adder 25 adds the video data Y0 (n + 1) latched by the first latch 21 and the video data Y0 (n-1) latched by the third latch 23 to the multiplication result obtained from the multiplier 24. And the result of the addition is divided by a divider 2
6 Divider 26 divides the addition result provided from adder 25 by divisor “4”, and outputs the division result as luminance data Y1 (n). Thus, the luminance data Y1 (n)
Is given by Y1 (n) = (Y0 (n + 1) + 2Y0 (n) + Y0 (n-1)) / 4.

FIG. 3 is a block diagram showing another example of the luminance data generation circuit 15, which shows a digital low-pass filter for three pixels in the vertical direction. The first to third line memories 31, 32, and 33 each have a capacity to store video data for one row, and are connected in series with each other to sequentially store video data Y0 in row units. That is, the read output of the first line memory 31 is connected to the write input of the second line memory 32, and the read output of the second line memory 32 is connected to the write input of the third line memory 33, Video data Y0 read from the first and second line memories 31 and 32 is written to the second and third line memories 32 and 33 as they are. As a result, three consecutive rows (m + 1) are stored in the first to third line memories 31, 32, and 33.
Rows, m rows, m-1 rows) of video data Y0 (m + 1), Y0 (m),
Y0 (m-1) is stored. The multiplier 34 multiplies the video data Y0 (m) read from the second line memory 32 by a multiplier “2”, and supplies the multiplication result to the adder 35. The adder 35 adds the video data Y0 read from the first line memory 31 to the multiplication result given from the multiplier 24.
(m + 1) is added to the video data Y0 (m-1) read from the third line memory 33, and the result of the addition is supplied to the divider 36. Divider 36 divides the addition result provided from adder 35 by divisor “4”, and outputs the multiplication result as luminance data Y
Output as 1 (m). Thus, the luminance data Y1 (m)
Is given by Y1 (m) = (Y0 (m + 1) + 2Y0 (m) + Y0 (m-1)) / 4.

The frequency characteristics of these digital low-pass filters are as shown in FIG. 4, and a frequency component of 1/2 of the sampling frequency fs is removed. Note that the sampling frequency fs is the sampling frequency of the A / D conversion circuit 13 and coincides with the sampling frequency of the correlated double sampling circuit 11 and the frequency of the output operation of the CCD. In addition, a frequency component equal to or less than の of the sampling frequency fs has been removed in advance by providing an optical low-pass filter such as a diffraction grating in an optical path of light applied to the CCD solid-state imaging device. It does not appear on the output side.

Here, in order to actually obtain the luminance data Y1, it is necessary to construct a two-dimensional filter in the horizontal direction and the vertical direction. The generation circuit 15 is configured.
For example, the output of the vertical digital filter shown in FIG. 3 may be given to the input of the horizontal digital filter shown in FIG. As a result, the luminance data Y1 (m, n) is generated from the video data of a total of 9 pixels of 3 pixels in the horizontal direction and 3 pixels in the vertical direction. In this case, the luminance data Y
1 (m, n) is Y1 (m, n) = (Y0 (m + 1, n + 1) + 2Y0 (m + 1, n) + Y0 (m + 1, m-1) + 2Y0 (m, n + 1) + 4Y0 (m, n) + 2Y0 (m, n-1) + Y0 (m-1, n + 1) + 2Y0 (m-1, n) + Y0 (m-1, n-1)) / 16 Will be. For example, in the color filter as shown in FIG. 8, for the light receiving pixels in the second row and second column, Y1 (2,2) = (W + 2G + W + 2Ye + 4Cy + 2Ye + G + 2W + G) / 4 = Ye + Cy + G + W = 2R + 4G + 2B. Such an operation has the same result in all light receiving pixels, and luminance information composed of the same component can be commonly obtained from all light receiving pixels.

FIG. 4 is a block diagram showing an example of the aperture data generating circuit 17, which shows a digital bandpass filter for five pixels in the horizontal direction. First to fifth
Of latches 41, 42, 43, 44, and 45 are connected in series to form a shift register, and the continuous video data Y0
(n + 2), Y0 (n + 1), Y0 (n), Y0 (n-1) and Y0 (n-2) are stored. The multiplier 46 multiplies the video data Y0 (n) latched by the third latch 43 by the multiplier “2”, and supplies the multiplication result to the subtractor 47. The subtractor 47 is a multiplier 2
4 is subtracted from the video data Y0 (n + 2) latched by the first latch 41 and the video data Y0 (m-2) latched by the fifth latch 45, and is added. The result is supplied to a divider 48. The divider 48 includes a subtractor 47
Is divided by the divisor “4” and output as aperture data A (n). Thus, the aperture data A (n) is given by A (n) = (-Y0 (n + 2) + 2Y0 (n) -Y0 (n-2)) / 4. The frequency characteristic of such a digital bandpass filter is as shown in FIG. 6, and a frequency component of 1/4 of the sampling frequency fs is mainly extracted.

The aperture data generation circuit 17
For, a digital bandpass filter may be configured in the vertical direction. In that case, five line memories are connected in series, and video data Y0 (m-2), Y0 (m-1), Y0 (m), Y0 (m +
1), Y0 (m + 2) is stored, and the aperture data A (m) may be obtained using a multiplier, a subtractor, and a divider in the same manner as in FIG. In this case, the aperture data A
(m) is given by A (m) = (− Y0 (m + 2) + 2Y0 (m) −Y0 (m−2)) / 4.

Then, by combining a horizontal digital bandpass filter and a vertical digital bandpass filter, the aperture data A (m, n) is obtained from a total of 25 pixels of 5 pixels in the horizontal direction and 5 pixels in the vertical direction. )
Can be obtained. In this case, the aperture data A (m, n) is represented by A (m, n) = (Y0 (m + 2, n + 2) -2Y0 (m + 2, n) + Y0 (m + 2, n-2) -2Y0 (m, n + 2) + 4Y0 (m, n) -2Y0 (m, n-2) + Y0 (m-2, n + 2) -2Y0 (m-2, n) + Y0 (m-2, n -2)) / 4.

In the above embodiment, the case where the number of taps of the low-pass filter for obtaining the luminance data is 3 and the number of taps of the band-pass filter for obtaining the aperture data is 5 is exemplified. The number of taps may be 5 or more or 7 or more. The coefficient by which each video data is multiplied determines the characteristics of each filter, and can be set arbitrarily according to the desired characteristics.

[0020]

According to the present invention, the gamma correction is not performed on the aperture data generated by the band-pass filter, and the gamma correction is performed only on the luminance data generated by the low-pass filter. . Then, by adding aperture data not subjected to gamma correction to luminance data subjected to gamma correction, uniform aperture processing is performed on all video data regardless of the luminance of the subject. Therefore, the outline of the video is not excessively emphasized in the low luminance portion, and the outline is further emphasized in the high luminance portion, so that more effective outline correction can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a video signal processing device according to the present invention.

FIG. 2 is a block diagram illustrating a configuration of a horizontal digital filter that generates luminance data.

FIG. 3 is a block diagram illustrating a configuration of a vertical digital filter that generates luminance data.

FIG. 4 is a diagram illustrating frequency characteristics of the digital filters of FIGS. 2 and 3;

FIG. 5 is a block diagram illustrating a configuration of a horizontal digital filter that generates aperture data.

FIG. 6 is a diagram illustrating frequency characteristics of a digital filter.

FIG. 7 is a block diagram illustrating a configuration of a conventional solid-state imaging device.

FIG. 8 is a plan view illustrating a configuration of a color filter mounted on the solid-state imaging device.

FIG. 9 is a diagram illustrating input / output characteristics of gamma correction.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 CCD solid-state imaging device 2 driver 3 timing generation circuit 4, 10 analog signal processing unit 5, 13 A / D conversion circuit 6, 14 digital signal processing unit 11 correlated double sampling circuit (CDS) 12 automatic gain control circuit (AGC) Reference Signs List 15 luminance data generation circuit 16 gamma correction circuit 17 aperture data generation circuit 18 aperture addition circuit 21 to 23, 41 to 45 latch 24, 34, 46 multiplier 25, 35, 47 adder 26, 36, 48 divider

Claims (1)

(57) [Claims] 1. A luminance representing a luminance of an image based on a video signal output from a solid-state imaging device in which a plurality of light receiving pixels are arranged in a matrix corresponding to information charges accumulated in each light receiving pixel in units of one row. A video signal processing device for generating data; an A / D conversion circuit for converting a video signal output from the solid-state imaging device into digital data for each pixel to generate video data corresponding to each light receiving pixel; A first filter circuit that sequentially captures the video data and removes a predetermined high-frequency component to generate first luminance data; and sequentially captures the continuous video data and selectively extracts a frequency component of a specific band. A second filter circuit for generating aperture data, and a correction for nonlinearly converting the first luminance data obtained from the first filter circuit to generate second luminance data. And an aperture addition circuit for adding aperture data obtained from the second filter circuit to second brightness data obtained from the correction circuit to generate third brightness data. Video signal processing device.