JP2010283720A - Device for correcting aberration - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent an image step after correction from being visually seen by randomly selecting an aberration correction coefficient from a certain range. <P>SOLUTION: This device for correcting aberration is equipped with: a filter processing part 300 for applying predetermined filter processing based on a filter coefficient to respective R and B channels out of RGB video pixel data; and a filter coefficient setting part 200 for temporally randomly selecting and outputting, as a filter coefficient, one of correction filter coefficients X, Y, Z for correcting phase shifting among the respective R, G and B video pixel data caused by color aberration of a lens in a correction area A corresponding to a distance to an objective pixel and two correction areas B, C adjacent to the correction area A by using, as inputs, a distance to an object pixel becoming an object of the filter processing or relative coordinates, the diaphragm, focal length and focus position of an optical lens by concentrically dividing a video signal into a plurality of correction areas from the position thereof corresponding to the optical center of the optical lens. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a signal processing apparatus and a signal processing method for video signals, and more particularly to an aberration correction apparatus that reduces chromatic dispersion when used for correcting chromatic aberration by an optical lens.

  The refractive index of light varies depending on the wavelength. For this reason, even in visible light, for example, red light and blue light have different focal lengths even if the same lens is used. This phenomenon is called chromatic dispersion. As a result, there is a difference between the size and position of an image connected by red light and the size and position of an image connected by blue light. This is called chromatic aberration of the lens. Since chromatic aberration is caused by a difference in refractive index of light, it is known that the chromatic aberration increases in the peripheral portion away from the optical axis of the optical lens having a large incident angle, and the phase shift between the colors increases. As a means for correcting such chromatic aberration, for example, there is one described in Patent Document 1. FIG. 1 shows a conventional image processing apparatus described in Patent Document 1.

  In FIG. 1, image information is read by a CCD sensor 101 via an optical system. The obtained RGB color image signals are input to the A / D converter 102. The A / D converter 102 converts RGB color video signals, which are analog signals, into digital RGB color video data. The obtained color video data is input to the interpolation processing units 103R and 103B for the R and B video pixel data. The processing unit is not provided for G video pixel data. For the G video pixel data with high visibility, the input data is directly used as output data without performing interpolation processing. Then, for each of the R and B video pixel data, the phase shift is corrected based on the G video pixel data.

  Each of the interpolation processing units 103R and 103B outputs five types of interpolation data using the interpolation coefficients given from the interpolation coefficient setting unit 104. As the interpolation coefficient, the main scanning effective pixel area is divided into n concentric circles centered on the optical center (center of image data), and a predetermined interpolation coefficient set is selected for each area. These interpolation data Q, Q0, Q1, Q2, and Q3 are respectively input to the saturation value generation units 105A, 105B, 105C, 105D, and 105E, and the saturation value W = MAX (R, G, B) −MIN (R, G, B). The saturation values W output from the saturation value generation units 105A, 105B, 105C, 105D, and 105E are input to W0, W1, W2, W3, and W4 of the data selection units 106R and 106B, respectively. Outputs Q0, Q1, Q, Q2, and Q3 of the interpolation processing unit 103R are input to data input terminals A0, A1, A2, A3, and A4 of the data selection unit 106R. The outputs Q0, Q1, Q, Q2, and Q3 of the interpolation processing unit 103B are input to the data input terminals A0, A1, A2, A3, and A4 of the data selection unit 106B. Then, the data selection units 106R and 106B select the data input corresponding to the control signal having the smallest value among the control signals input to the five selection control terminals W0 to W4, and output from the output terminal Y.

JP 2000-236450 A

  In the chromatic aberration correction circuit described above, the main scanning effective pixel area is divided into n as shown in FIG. 3, and chromatic aberration correction is performed by performing a filter operation using a predetermined interpolation coefficient for each divided area. In FIG. 3, a solid line square represents an imaging region, and a dotted circle represents a boundary between regions using the same interpolation coefficient in aberration correction. For example, in the line segment LM, an area from the point L to the point N is corrected using the coefficient X (area A), and an area from the point N to the point O is corrected using the coefficient Y (area). B) In the region from the point O to the point M, aberration correction is performed using the coefficient Z (region C). An ideal interpolation filter for correcting chromatic aberration is a filter that shifts only the phase without changing the signal level. That is, only the phase characteristic changes, and the frequency characteristic is constant regardless of the frequency. However, it is not possible to keep the frequency characteristics constant regardless of the frequency with a filter using a finite number of coefficients. For this reason, the filter determined for each divided area has frequency characteristics.

  Further, since the chromatic aberration appears more conspicuously in the periphery of the lens, the amount of aberration is larger in the region B than in the region A and in the region C than in the region B. Therefore, different values are set for the coefficients X, Y, and Z, respectively. Therefore, the frequency characteristics of the filters applied to the regions A, B, and C are different, and a signal level step occurs in the image after interpolation near the boundary between the regions. If the interpolation filter is applied by dividing the main scanning effective pixel area by the number of effective pixels, the correction coefficient can be changed in units of one dot in practice, and as a result, the image step becomes difficult to understand. When the number is in the hundreds or thousands, the circuit scale increases, which is not realistic.

  Further, the above-described image step is noticeably generated particularly when the intensity of light entering the lens is small, that is, in a dark place. This is because when the light input is small, that is, when the electric signal level after photoelectric conversion is small, the influence of noise generated in the circuit becomes relatively dominant. This noise is caused by the circuit configuration and naturally occurs even before the aberration correction circuit. Therefore, when this aberration is corrected, it appears as noise with different characteristics in each divided area, for example, the regions A, B, and C. An image step is formed.

  Also, when imaging in the dark using an optical lens as described above, the input signal gain is often increased electrically in post-processing, and in that case the input signal and noise are equally gained. Therefore, the image level difference becomes more conspicuous.

  In view of the above problems, an object of the present invention is to provide an aberration correction apparatus that reduces an image level difference after aberration correction.

  In order to solve the above problems, the present invention divides an image area into a plurality of areas concentrically from a position corresponding to the optical center of the optical lens, and a filter processing unit for correcting chromatic aberration. A coefficient setting unit that sets the filter coefficient for the aberration correction optimized for each area with the characteristics of the input as an input, and the filter coefficient selected by this coefficient setting unit is a single value that is optimized by the area Rather than selecting, it selects at random in time from values optimized in the area and areas adjacent to the area. Alternatively, by selecting which coefficient to select according to an arbitrary probability function, the boundary where the filter coefficient is switched is made difficult to understand, and the influence of the image step in chromatic aberration correction is reduced.

  According to the chromatic aberration correction apparatus of the present invention, when an image is divided into a plurality of areas and a chromatic aberration correction is performed by setting a unique filter coefficient for each area, an output image due to a difference in filter frequency characteristics for each area The image step can be reduced. This is particularly useful in imaging using an optical lens in an environment where noise is conspicuous, such as in a dark place.

Block diagram of a conventional chromatic aberration correction circuit 1 is a block diagram of a chromatic aberration correction circuit according to a first embodiment of the present invention. The figure which shows the relationship of the aberration correction area | region with respect to an imaging area | region The figure which shows the relationship between a counter, a moving radius, and a deflection angle in 1650x720 pixels Diagram showing the relationship between lens parameters and aberration correction The figure which shows the relationship of the aberration correction area | region on line segment LM

  Hereinafter, embodiments will be described with reference to the drawings.

(Embodiment 1)
FIG. 2 is a block diagram of the aberration correction apparatus according to the first embodiment. The aberration correction apparatus according to the present embodiment roughly includes a filter coefficient setting unit 200 and a filter processing unit 300. Detailed contents are described below. In FIG. 2, image information is read by the CCD sensor 101 through the optical system. The obtained RGB color image signals are input to the A / D converter 102. As in the conventional example, the RGB video data output from the A / D converter 102 does not perform interpolation for the highly visible G video pixel data, and uses the input data as output data as it is. Then, for each of the R and B video pixel data, the phase shift is corrected based on the G video pixel data. However, the G video pixel data is output after being delayed by the processing time of the R and B video pixel data. This delay is performed by the delay unit 303.

  The pixel position counter 201 outputs horizontal and vertical coordinates when the origin is a position corresponding to the optical center (lens center) of the video pixel data. For example, FIG. 4 shows coordinates in the case where video data having effective pixels of horizontal 1650 pixels × vertical 720 pixels is handled. The top left vertex in FIG. 4 is the coordinates (−850, −360), and the counter starts from here and counts the coordinates. For example, the count value of the point A is x in the horizontal direction and y in the vertical direction, which are orthogonal coordinates with the center of the screen as the origin. In order to correctly link the count value and the video pixel data, it can be easily realized by resetting an image sensor such as a CCD or CMOS at the same timing as the timing of photoelectric conversion.

  The coordinates output by the pixel position counter 201 are orthogonal coordinates. However, since the calculation becomes complicated to use them as they are in the correction calculation, conversion to polar coordinates is performed.

  This is the radius calculation unit 202 and the declination calculation unit 203. The radius calculation unit receives the orthogonal coordinates of the pixel position as an input, outputs the radius of the pixel position, and the declination calculation unit 203 calculates the orthogonal coordinates of the pixel position. The deviation angle of the pixel position is output as an input.

  The radius vector calculation unit 202 obtains the distance from the center. This is represented by z at point A in FIG. 4. If the input coordinates to the radial calculation unit 202 are horizontal x and vertical y, the output z is

It is represented by

The declination calculator 203 calculates the clockwise angle from the horizontal axis and outputs sine and cosine values. The declination is represented by θ at point A in FIG. Assuming that the input coordinates to the deflection angle calculation unit 203 are horizontal x (right direction is positive) and vertical y (down direction is positive), the output is sin θ = y / z.
cos θ = x / z
It is represented by

  Using these z, sin θ, and cos θ, the horizontal aberration correction amount and the vertical aberration correction amount are determined.

  Specifically, as the absolute value of the aberration correction amount, data measured in advance based on the lens characteristics is stored in the storage medium 204, and the absolute value of the aberration correction amount read out from the storage medium 204, and z, sin θ. , Cos θ, the correction amount in the vertical direction and the correction amount in the horizontal direction are calculated for the pixel to be subjected to aberration correction. The following describes how the measured data is stored.

  The amount of aberration correction varies depending on the aperture, focal length, and focus position of the lens. Therefore, a three-dimensional lens parameter space as shown in FIG. 5 is considered. In a certain lens state, the generated aberration characteristic corresponds to one point in this space. Therefore, for each axis of the aperture, focal length, and focus position, set the parameter range to cover and the parameter value that divides this range into multiple sections, and the parameter values for the aperture, focal length, and focus position intersect Chromatic aberration data is measured and stored. Since the capacity of the storage medium 204 is limited, the chromatic aberration data holds data measured by roughly dividing the parameters of the aperture, focal length, and focus position to some extent, and data of arbitrary parameter values is stored at adjacent points. Complemented from chromatic aberration data.

  Of course, the aberration correction amount varies depending on whether the image pixel data to be corrected is an R signal or a B signal, and also varies depending on the distance from the origin of the image data. Therefore, the chromatic aberration data is the R signal at a location where the distance from the origin to the end is divided to some extent, such as r / 10, 2r / 10, 3r / 10... Where r is the distance from the origin to the end. It means a data string of chromatic aberration amount and a data string of chromatic aberration amount of B signal. The reason why the discrete data string is held in this way is to prevent the limit of the capacity of the storage medium 204 and the increase in the calculation amount, which causes the image step described in the above problem.

  The chromatic aberration data stored as described above is first read from the storage medium 204 in accordance with the three lens states of the aperture, focal length, and focus position. The read chromatic aberration data is input to the aberration correction absolute amount determination unit 205. In the aberration correction absolute amount determination unit 205, according to the distance z from the center, for example, if 2r / 10 ≦ z <3r / 10, three R, corresponding to the positions r / 10, 2r / 10, and 3r / 10 One of the B chromatic aberration data is selected at random, and the selected correction amount is output as the absolute value of the correction amount of the R and B signals in this pixel.

  By selecting the correction amount at random as described above, it becomes difficult to understand the boundary where the chromatic aberration data is switched, and it is difficult to understand the image step. Further, since chromatic aberration data at relatively close positions is selected at random, if the division unit from the center to the end of the lens is made somewhat fine, the correction of chromatic aberration, which is the original purpose, is hardly affected.

  The absolute values of the correction amounts of the R and B signals are input to the vertical / horizontal correction amount determination unit 206, respectively. The absolute value of the input correction amount is multiplied by sin θ to obtain the correction amounts in the vertical direction of the R and B signals, which are output as the R vertical filter coefficient and the B vertical filter coefficient, respectively. Similarly, the correction amount in the horizontal direction of the R and B signals is calculated by multiplying the absolute value of the input correction amount by cos θ and output as an R horizontal filter coefficient and a B horizontal filter coefficient.

  The R video pixel data is input to the vertical filter 301R. The vertical filter 301 </ b> R also has an R vertical filter coefficient as input, performs vertical filter processing on the R video pixel data according to the correction amount, and outputs the result. The output of the R signal vertical filter 301R is input to the R signal horizontal filter 302R. The horizontal filter 302R also has an R horizontal filter coefficient as input, and performs a filtering process in the horizontal direction on the R video pixel data in accordance with the correction amount and outputs it.

  The B video pixel data is input to the vertical filter 301B. The vertical filter 301 </ b> B also has a B vertical filter coefficient as input, and performs vertical filter processing on the B video pixel data according to the correction amount and outputs it. The output of the B signal vertical filter 301B is input to the B signal horizontal filter 302B. The horizontal filter 302R also has a B horizontal filter coefficient as an input, performs horizontal filtering on the B video pixel data according to the correction amount, and outputs the result.

  With the above configuration, the chromatic aberration correction apparatus according to the present embodiment divides an image into a plurality of areas, and sets chromatic aberration correction by setting a unique filter coefficient for each area. The image level difference of the output image due to the difference can be reduced.

(Embodiment 2)
This embodiment is different from the first embodiment in the aberration correction absolute amount determination unit 205. In the first embodiment, the aberration correction absolute amount determination unit 205 selects one from the three R and B chromatic aberration data completely at random, but in this embodiment, the coefficient selection has a certain degree of regularity. . FIG. 6 shows the line segment LM of FIG. 3 cut out. For example, in the region using the coefficient Y on the line segment LM, the line segment is divided into three, and in the region P, the coefficients Y and X are selected with a probability of 50%. In region Q, coefficient Y is selected with probability A, coefficient X is selected with probability B, and coefficient Z is selected with probability C. In the region R, each of the coefficients Y and Z is selected with a probability of 50%.

  By selecting the coefficients in this way, it is possible to make the image level difference in the boundary area of the divided area less noticeable.

  However, the present invention is not limited to the above example. For example, the region in the divided area may be further divided, or the probability that the coefficient is selected may be set to a value different from the above.

  According to the aberration correction device of the present invention, in a circuit that divides an image into a plurality of areas and sets a unique filter coefficient for each area to perform chromatic aberration correction, output based on a difference in filter coefficient for each area The image level difference of the image can be reduced. In particular, it is also useful for a video signal processing circuit using an optical lens in an environment where noise is conspicuous, such as in a dark place.

101 CCD
102 A / D conversion 103R Interpolation processing unit (R channel)
103B Interpolation processing unit (B channel)
104 Interpolation coefficient setting unit 105A Saturation value generation unit 105B Saturation value generation unit 105C Saturation value generation unit 105D Saturation value generation unit 105E Saturation value generation unit 106R Data selection unit (R channel)
106B Data selection part (B channel)
DESCRIPTION OF SYMBOLS 200 Filter coefficient setting part 201 Pixel position counter 202 Radial radius calculation part 203 Declination angle calculation part 204 Storage medium 205 Aberration correction absolute amount determination part 206 Vertical / horizontal correction amount determination part 300 Filter processing part 301R R Vertical filter 302R R Horizontal filter 301B B vertical filter 302B B horizontal filter 303 delay unit

Claims (2)

An aberration correction apparatus that detects light from an optical lens and corrects a video signal output from a photoelectric conversion element,
A filter processing unit that performs a predetermined filter process on the R and B channels of the video signal based on a filter coefficient;
As the filter coefficient, the video signal is divided into a plurality of correction areas concentrically from a position corresponding to the optical center of the optical lens, and the distance or relative from the center to the target pixel to be filtered In the correction area A corresponding to the distance to the target pixel, and the two correction areas B and C adjacent to the correction area A, using the coordinates and the aperture, focal length, and focus position of the optical lens as inputs. A filter coefficient setting unit that selects one of correction filter coefficients X, Y, and Z for correcting a phase shift between R, G, and B video signals caused by chromatic aberration of the lens, and outputs the selected filter coefficient to the filter processing unit When,
An aberration correction apparatus comprising: An aberration correction apparatus that detects light from an optical lens and corrects a video signal output from a photoelectric conversion element,
A filter processing unit that performs a predetermined filter process on the R and B channels of the video signal based on a filter coefficient;
The distance or relative coordinate from the position corresponding to the optical center of the optical lens to the target pixel to be filtered, and the aperture, focal length, and focus position of the optical lens, and the input to the target pixel A correction filter for correcting a phase shift between R, G, and B video signals caused by chromatic aberration of a lens in a correction area A corresponding to a distance and two correction areas B and C adjacent to the correction area A Which of the coefficients X, Y, and Z is applied to the target pixel, the closer the input target pixel position coordinate is to the correction area B, the higher the probability of selecting the filter coefficient Y, and the target pixel position coordinate is corrected. A filter coefficient setting unit that increases the probability of selecting the filter coefficient Z closer to the area C and outputs the filter coefficient to the filter processing unit;
An aberration correction apparatus comprising: