JP2010045449A - Smear correction circuit, smear correction method, and image device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect only smear component superimposed on an image signal, and remove only the detected smear component efficiently from the image signal. <P>SOLUTION: A smear correction portion 4 uses an image signal and a reference signal supplied from an image sensor. Next, the image sensor has an incidence region where light incident from an optical source is photoelectric converted to generate the image signal and a shading region where the reference signal is generated from dark current produced in a state that light is shaded; the correction amount generation portion removes noise superimposed on the image signal by a predetermined filtering; then the correction amount generation portion subtracts the reference signal from the noise-removed image signal to detect the smear component superimposed on the image signal, and generates a correction signal for eliminating the smear component from the image signal; and a correction amount generation portion 21 is provided. The correction amount subtraction portion 22 outputs an appropriately processed image signal, according to determination as to whether the smear component is superimposed on the image signal. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a smear correction circuit, a smear correction method, and an imaging apparatus that are suitable for application to, for example, correcting smear that occurs in a captured image.

  Conventionally, when a high-brightness subject is imaged using an imaging device such as a CCD (Charge Coupled Devices) sensor, vertical stripes that should not exist originally appear in the vertical direction of the image, and the image quality may deteriorate. Such vertical stripes are called smears. In smear, light is mixed into the vertical transfer path that should have been shielded by intense light irradiation, or the signal charge photoelectrically converted by the photodiode crosses the potential barrier and leaks into the vertical transfer path. It is generated by doing. The smear component that generates smear is noise superimposed on the image signal, and may be superimposed on the image signal generated by the light-shielding region (OPB: optical black) outside the effective pixel region. For this reason, various techniques for removing the influence of smear components have been studied. In the following description, the correction excluding the smear component superimposed on the image signal is referred to as “smear correction”.

Patent Document 1 discloses a technique for correcting smear by subtracting a smear component detected from a light shielding area from an effective pixel.
JP 2006-74658 A

  By the way, the conventional technique for performing smear correction did not detect only the smear component. Therefore, when a lot of noise is superimposed on an image signal or when a plurality of light sources are imaged, a smear-free area is generated. There was also an excessive correction. In other words, by determining that noise other than the smear component superimposed on the image signal is a smear component, excessive smear correction is performed in an area where smear is not generated, and colored vertical streaks appear in the image. There was a thing. Further, when the light source moves (in other words, when the relative position of the image sensor and the light source changes), the timing at which the signal charges are transferred is shifted, so that smear may occur in an oblique direction. Conventionally, since smear correction is performed in the vertical transfer direction of the image sensor, if smear generated in an oblique direction is corrected, an area where smear is not generated is excessively corrected. For this reason, an uncomfortable feeling may occur in the displayed image.

  The present invention has been made in view of such a situation, and an object thereof is to detect only a smear component superimposed on an image signal and efficiently remove only the detected smear component from the image signal.

The present invention provides an image signal supplied from an imaging device having an incident area for photoelectrically converting light incident from a light source to generate an image signal and a light-shielding area for generating a reference signal from a dark current generated in a state where the light beam is shielded. And a reference signal.
Next, the noise superimposed on the image signal is removed by a predetermined filtering process, and the smear component superimposed on the image signal is detected by subtracting the reference signal from the image signal from which the noise has been removed, and the smear component is detected from the image signal. A correction signal for removal is generated.
Then, when the smear component is superimposed on the image signal, the correction signal is subtracted from the image signal to output an image signal from which the smear component has been removed, and when the smear component is not superimposed on the image signal, The signal is output as it is.

  By doing this, it is possible to detect only the smear component superimposed on the image signal and then subtract the correction signal for correcting the smear component from the image signal to output the image signal from which the smear component has been removed. it can.

  According to the present invention, it is possible to reliably detect a smear component by removing noise superimposed on an image signal by a predetermined filter process. Since only the smear component detected from the image signal is removed, the image signal on which the smear component is not superimposed is not excessively corrected. For this reason, only smear is removed from the obtained image, and the image quality can be improved.

  Hereinafter, a first embodiment of the present invention will be described with reference to FIGS. In the present embodiment, an example will be described in which the present invention is applied to an imaging apparatus 10 that appropriately corrects an image signal obtained by imaging a subject by a predetermined process.

FIG. 1 is a block diagram illustrating an internal configuration example of the imaging apparatus 10 of the present example.
The imaging device 10 photoelectrically converts image light formed through an optical system (not shown) to generate R, G, B analog image signals, and R, G, B. Amplifiers 2r, 2g, and 2b for amplifying analog image signals are provided. Also provided are analog / digital converters 3r, 3g, 3b for converting the amplified R, G, B analog image signals into R, G, B digital image signals. Hereinafter, the R, G, B digital image signals are simply referred to as “digital image signals”.

  The imaging apparatus 10 also includes a smear correction unit 4 that corrects noise and smear components superimposed on the digital image signals output from the analog / digital conversion units 3r, 3g, and 3b. Further, a gain adjustment unit 5 that adjusts the gain of the corrected digital image signal and a luminance adjustment unit 6 that adjusts the luminance of the digital image signal whose gain has been adjusted are provided. In addition, a gamma correction unit 7 that gamma-corrects the digital image signal whose luminance has been adjusted, and an output unit 8 that converts the digital image signal into a Y, Cr, Cb digital image signal and outputs the digital image signal.

  The Y, Cr, Cb digital image signal output from the output unit 8 is supplied to a display unit (not shown) including, for example, a liquid crystal display panel and an organic EL display panel, and an image is displayed. The Y, Cr, Cb digital image signal may be supplied to a recording unit (not shown) including, for example, a hard disk drive, a flash memory, and a tape device, and may be recorded in a predetermined format.

Here, the operation of each unit will be described.
The analog image signals output from the image sensors 1r, 1g, and 1b are quantized by the amplifiers 2r, 2g, and 2b and the analog / digital converters 3r, 3g, and 3b, adjusted to appropriate levels, and converted into digital image signals. Is done. The digital image signal is appropriately processed by the smear correction unit 4 and the gain adjustment unit 5, and then the high luminance part is compressed by the luminance adjustment unit 6 to be within a prescribed range. Further, in order to cope with monitor gamma such as CRT (Cathode Ray Tube), the gamma correction unit 7 performs gamma correction, and then the output unit 8 converts the digital image signal into a final output format. Output to the recording unit, the recording unit, and the like.

  By the way, if the smear component is detected from the digital image signal adjusted by the gain adjusting unit 5 in the imaging device 10, for example, it is likely to lead to a false detection that a smear appears in a region where no smear has occurred. For this reason, the smear correction unit 4 performs processing for detecting a smear component from the image signal and processing for smear correction.

FIG. 2 shows a configuration example of the image sensors 1r, 1g, and 1b.
The imaging elements 1r, 1g, and 1b in this example are imaging elements configured with 1125 × 2200 pixels. A horizontal light shielding area 11 (HOPB: Horizontal OPtical Black) is formed at the left end of the imaging elements 1r, 1g, 1b, and a vertical light shielding area (VOPB: Vertical OPtical Black) 35 is formed at the upper end. The horizontal light shielding area 11 and the vertical light shielding area 15 are areas where light rays from the light source are shielded. The direct current level (DC voltage) of the dark current generated by the pixels in the light shielding region in the state where the light beam is shielded is used as a reference signal (also referred to as a black level) used in a black level detection unit 32 described later. .

  In the imaging devices 1r, 1g, and 1b, the light receiving unit 12 that is an area excluding the horizontal light shielding area 11 and the vertical light shielding area 15 functions as an incident area that photoelectrically converts light incident from the light source to generate an image signal. . The image light of the subject formed on the light receiving unit 12 is photoelectrically converted and output as an R, G, B analog image signal.

  When strong light from the light source 13 enters a part of the light receiving unit 12, a smear 14 is generated in the vertical direction of the imaging devices 1r, 1g, and 1b. A black level detection unit 32 (see FIG. 4), which will be described later, detects the signal level of the DC signal generated by the pixels in the light-shielded region 16 in real time as the black level. Since smear does not occur in the region 16, the DC signal generated in the region 16 is used as a reference signal for detecting smear.

FIG. 3 shows an internal configuration example of the smear correction unit 4.
The smear correction unit 4 includes a correction amount generation unit 21 that calculates a correction amount from the digital image signal and removes noise, and a correction amount subtraction unit 22 that subtracts the correction amount calculated by the correction amount generation unit 21 from the digital image signal. .

  The correction amount generator 21 detects a smear component superimposed on the digital image signal by removing noise superimposed on the digital image signal by a predetermined filter process and subtracting the reference signal from the digital image signal from which the noise has been removed. . Then, a correction signal for removing smear components from the digital image signal is generated. The level of the correction signal can be arbitrarily changed by a menu operation or the like performed by the user while viewing an image displayed on a display unit (not shown).

  When the smear component is superimposed on the digital image signal, the correction amount subtraction unit 22 subtracts the correction signal from the digital image signal and outputs a digital image signal from which the smear component has been removed. On the other hand, when the smear component is not superimposed on the digital image signal, the digital image signal is output as it is. Then, the correction amount subtraction unit 22 delivers the digital image signal obtained by subtracting the correction signal to the subsequent gain adjustment unit 5 (see FIG. 1). In this way, the correction amount subtraction unit 22 processes the digital image signal so as not to excessively correct the smear to the area where the smear has not occurred.

FIG. 4 shows an internal configuration example of the correction amount generation unit 21.
The correction amount generation unit 21 includes a smear waveform detection unit 31 that detects a smear component waveform from the digital image signal output from the vertical light-shielding region 15 of the imaging elements 1r, 1g, and 1b. In addition, a black level detection unit 32 that detects the signal level of the digital image signal output from the horizontal light-shielding region 11 (in this example, the region 16) in which the smear component is not superimposed in the light-shielding region as a black level of the reference signal is provided. . In addition, the correction amount generation unit 21 includes a coring processing unit 33 that outputs a correction signal obtained by removing the smear component from the digital image signal in response to detection of a smear component waveform. The coring process performed by the coring processing unit 33 will be described later with reference to FIG.

Here, the processing of the black level detection unit 32 will be described.
While the imaging device 10 is photographing, the smear component of the smear 14 is superimposed on the vertical light shielding area 15 at the time of exposure. For this reason, the target value of the black level cannot be determined only by signals supplied from the pixels in the vertical light shielding region 15. Therefore, the black level detection unit 32 detects the black level in real time by using the signal output from the horizontal light-shielding region 11 (region 16 in this example) as a reference signal. However, since the black level may change depending on the temperature or the like, it is necessary to stabilize the black level by removing the noise component generated in the horizontal light shielding region 11. An IIR filter is used to remove this noise component. The IIR filter is a filter that uses an input signal and a previous output signal by a feedback loop.

  Then, the smear waveform detection unit 31 detects a smear component of the smear 14. Normally, a smear that appears stationary with respect to random noise is a fixed pattern noise that appears at a fixed location, and has high visibility to the user. Therefore, it is necessary to improve the accuracy of detecting only the smear 14. For this reason, the smear waveform detection unit 31 detects a smear component by combining a plurality of types of digital filters.

  The coring processing unit 33 subtracts the black level from the signal level of the digital image signal output from the vertical shading region 15 when the smear waveform detection unit 31 detects the waveform of the smear component. Then, when the signal level of the digital image signal with the black level reduced is within the range from the predetermined threshold value to the predetermined upper limit value, a correction signal is output by removing the smear component from the digital image signal with the black level reduced. To do.

FIG. 5 shows a configuration example of the smear waveform detection unit 31.
The smear waveform detection unit 31 includes a median filter 41 that removes impulse noise. Further, an integration filter 42 that extracts low-frequency components in the vertical direction of the image sensors 1r, 1g, and 1b, and a low-pass filter 43 that extracts low-frequency components that extract the low-frequency components in the horizontal direction of the image sensors 1r, 1g, and 1b. Is provided. The control unit 44 controls operations of the median filter 41, the integration filter 42, and the low-pass filter 43.

The median filter 41 obtains, for example, the levels of three adjacent pixels from the digital image signal obtained from the vertical light shielding region 15, and extracts the median value.
The integration filter 42 includes a line memory 45 that stores digital image signals of a plurality of lines, and averages the signal level of each column of the digital image signals obtained from the vertical light shielding region 15. At this time, the integration filter 42 can extract a low frequency component in the vertical direction from the digital image signal by adding the digital image signals read from the line memory 45. As described above, the integration filter 42 increases the effect of removing the noise component by storing the digital image signal obtained from the vertical light shielding region 15 of the previous frame in the line memory 45.

  The low-pass filter 43 passes the low-frequency component of the digital image signal averaged by the integration filter 42 and further removes the noise component. In this example, the integration filter 42 and the low-pass filter 43 are combined to function as a two-dimensional filter.

  In this way, the smear waveform detection unit 31 can remove random noise generated spatially and temporally from the digital image signal. For this reason, it is possible to detect smear components generated in the vertical light shielding region 15 with high accuracy. The user can switch whether or not to operate the low-pass filter 43 according to the amount of noise.

  The correction amount generation unit 21 obtains the correction amount of the noise component using the VOPB waveform and the black level detected from the digital image signal obtained from the vertical light shielding area 15. Specifically, after subtracting the black level from the VOPB waveform, the coring processing unit 33 performs coring processing in order to further remove noise components. Coring processing can be selected from three types of processing (see FIG. 8 described later) according to the situation.

FIG. 6 shows an internal configuration example of the correction amount subtraction unit 22.
The correction amount subtraction unit 22 includes signal level detection units 51r, 51g, and 51b that detect the signal levels of the digital image signals supplied from the analog / digital conversion units 3r, 3g, and 3b, and a correction gain generation unit that generates a correction gain. 52r, 52g, 52b are provided. Also, a random number generator 57 that generates random numbers, RAMs 55r, 55g, and 55b that store R, G, and B correction signals that correct digital image signals, and RAM controllers 54r, 54g, and 55b that control the RAMs 55r, 55g, and 55b, 54b. In the RAMs 55r, 55g, and 55b, the R, G, and B correction signals supplied from the correction amount generation unit 21 are stored corresponding to the pixels to be corrected.

  The correction amount subtracting unit 22 includes multiplying units 53r, 53g, and 53b that multiply the correction gain generated by the correction gain generating units 52r, 52g, and 52b by the correction signal read from the RAMs 55r, 55g, and 55b. The multiplying units 53r, 53g, and 53b multiply the correction gain supplied from the correction gain generating unit by the correction signal read by the RAM control units 54r, 54g, and 54b to generate a subtraction signal that reduces the signal level of the digital image signal. To do. The correction amount subtraction unit 22 includes a subtraction unit 56 that outputs a digital image signal obtained by subtracting the subtraction signals generated by the multiplication units 53r, 53g, and 53b from the digital image signal.

  The digital image signal is supplied to the signal level detectors 51r, 51g, 51b and the subtractor 56, respectively. The signal level detectors 51r, 51g, 51b detect the signal level of the digital image signal and supply the detected signal level to the correction gain generators 52r, 52g, 52b for each pixel. Then, the correction gain generation units 52r, 52g, and 52b generate a correction gain according to the signal level. The reason for adjusting the correction gain will be described later with reference to FIG.

  On the other hand, the random number generator 57 supplies random numbers to the correction gain generators 52r, 52g, and 52b and the RAM controllers 54r, 54g, and 54b. When the smear component detected from the digital image signal obtained from the vertical shading region 15 is subtracted uniformly in a situation where the amount of noise is very large, due to the influence of noise that could not be removed by the filter processing of the smear waveform detection unit 31. May cause vertical streaks. In order to eliminate the influence of this noise, the correction gain or the pixel is slightly changed using the random number generated by the random number generator 57, thereby causing the vertical stripes to be shaded and making the image noise less visible. it can.

  The random numbers supplied to the correction gain generation units 52r, 52g, and 52b are used to slightly change the correction level for each row. This minute variation is, for example, dithering for each pixel. The random numbers supplied to the RAM control units 54r, 54g, and 54b are used for making minute fluctuations (for example, replacing adjacent pixels). This minute vibration spatially varies the correction value by adding a random number to the read address for reading the correction value from the RAMs 55r, 55g, and 55b. Accordingly, the correction gain generation units 52r, 52g, and 52b change the signal levels detected by the signal level detection units 51r, 51g, and 51b according to the random numbers supplied from the random number generation unit 57, thereby correcting the digital image signal correction gains. Can be generated.

  The RAM 55r stores the R correction signal supplied from the correction amount generation unit 21. The RAM 55g stores the G correction signal supplied from the correction amount generation unit 21. The RAM 55b stores the B correction signal supplied from the correction amount generation unit 21. The RAMs 55r, 55g, and 55b are controlled to store and read the R, G, and B correction signals by the RAM control units 54r, 54g, and 54b, respectively. Thereby, the RAM control units 54r, 54g, and 54b change the pixels corrected by the R, G, and B correction signals stored in the RAMs 55r, 55g, and 55b according to the random numbers supplied from the random number generation unit 57. Can be read.

  The correction gain output from the correction gain generation units 52r, 52g, and 52b and the R, G, and B correction signals that the RAM control units 54r, 54g, and 54b read from the RAMs 55r, 55g, and 55b are the multiplication units 53r, 53g, and 53g, respectively. Multiplied by 53b. Then, the multipliers 53r, 53g, 53b generate a subtraction signal for subtracting the signal level of the input digital image signal. The subtraction signals generated by the multiplication units 53r, 53g, and 53b are supplied to the subtraction unit 56. The subtracting unit 56 outputs a digital image signal obtained by subtracting the subtracted signal from the digital image signal.

FIG. 7 shows an internal configuration example of the subtraction unit 56.
The subtracting unit 56 includes subtracting units 56r, 56g, and 56b that select and output the input digital image signal or the corrected digital image signal. The subtraction units 56r, 56g, and 56b include subtraction signal generation units 61r, 61g, and 61 that generate a digital image signal obtained by subtracting the correction signal from the digital image signal. In addition, threshold determination units 62r, 62g, and 62b that output a digital image signal or a digital image signal obtained by subtracting the R, G, and B correction signals from the digital image signal based on a predetermined threshold are provided.

  The R, G, and B correction signals are supplied from the correction amount generation unit 21 to the subtraction units 56r, 56g, and 56b. The subtraction signal generation units 61r, 61g, and 61b subtract R, G, and B correction signals from the digital image signals input to the subtraction units 56r, 56g, and 56b. Then, the threshold determination units 62r, 62g, and 62b compare the signal level of the digital image signal with a predetermined threshold. When the signal level of the digital image signal is less than a predetermined threshold, the digital image signal is output. On the other hand, when the signal level of the digital image signal exceeds a predetermined threshold value, a digital image signal obtained by subtracting the R, G, and B correction signals from the digital image signal supplied from the subtraction signal generators 61r, 61g, and 61b is output. .

FIG. 8 shows an example of coring processing performed by the coring processing unit 33.
The coring process is a process for efficiently removing noise included in the input digital image signal.
FIG. 8A shows an example of the first coring process.
FIG. 8B shows an example of the second coring process.
FIG. 8C shows an example of the third coring process.

  8A to 8C, the horizontal axis indicates the value of the digital image signal input to the coring processing unit 33, and the vertical axis indicates the value of the digital image signal output by the coring processing unit 33. In addition, a threshold level is set to limit the output of the input digital image signal until a predetermined signal level is reached. When the input digital image signal and the output digital image signal have the same signal level, the relationship between the input and output digital image signals is indicated by a broken line.

  As shown in FIG. 8A, when the level of the input digital image signal exceeds the threshold level, the coring processing unit 33 outputs the digital image signal. At this time, slight noise superimposed on the digital image signal is not output. For this reason, slight noise superimposed on the digital image signal can be removed.

  The levels of the input and output digital image signals are linear and are indicated by straight lines having a predetermined slope. This inclination is variable, and when the user performs a menu operation or the like, the correction amount can be adjusted by changing the inclination. By the way, when the signal level of the detected smear component is saturated, the corrected pixel cannot be correctly restored. Therefore, by providing an upper limit clip so as not to output a smear component above a certain level, the unnaturalness of the pixel after correction is reduced.

  FIG. 8B shows that the digital image signal is output when the level of the input digital image signal is lower than the threshold level. However, the level of the output digital image signal does not become larger than the input digital image signal. For this reason, when the level of the input digital image signal exceeds the threshold level, the level of the output digital image signal increases to the upper limit clip. When the level of the output digital image signal reaches the upper limit clip, the level of the output digital image signal is limited. As a result, smear can be corrected inconspicuously.

  FIG. 8C shows that the level of the output digital image signal matches the broken line when the level of the input digital image signal reaches the threshold level. Thereafter, as in FIG. 8B, the level of the digital image signal to be output rises up to the upper limit value along the broken line. When the level of the output digital image signal reaches the upper limit, the smear component is clipped at a constant value.

  FIG. 9 illustrates an example of processing for adjusting the correction gain performed by the correction gain generation units 52r, 52g, and 52b. In FIG. 9, the horizontal axis indicates the signal level (also referred to as pixel level) for each pixel, and the vertical axis indicates the correction gain.

  If the level of the correction target pixel is close to the saturation level, subtracting the detected smear without adjusting the gain may result in an unnatural color smear after correction. In order to prevent this, processing for adjusting the correction gain is performed. With this adjustment process, the correction gain can be reduced when the level of the pixel to be corrected is large.

  In this example, the pixel level is shown in a range from 0 to 1023. The correction gain of the digital image signal for each pixel corrected by the correction gain generation units 52r, 52g, and 52b is constant until the pixel level reaches the gain adjustment start level. When the pixel level reaches the gain adjustment start level, the correction gain is adjusted to be lowered. Further, when the pixel level reaches the saturation level, the correction gain is adjusted to the lowest value so that the correction gain becomes the lowest value.

  According to the imaging apparatus 10 according to the first embodiment described above, the smear component is reliably obtained by removing noise superimposed on the digital image signals obtained from the imaging elements 1r, 1g, and 1b by a predetermined filtering process. Can be detected. Then, by removing only the smear component detected from the image signal and outputting it, the image signal on which the smear component is not superimposed is not excessively corrected. For this reason, only smear is removed from the obtained image, and the image quality of the image displayed on the display unit can be improved.

  Further, when the smear component waveform is detected from the digital image signal output from the vertical shading region 15 where smear is detected, which is detected by the smear waveform detecting unit 31, the black level is determined from the signal level of the image signal output from the vertical shading region. Reduce. Then, when the signal level of the image signal with the black level reduced is within a range from a predetermined threshold value to a predetermined upper limit value, coring processing for outputting a correction signal is performed. For this reason, a slight noise other than the smear component is not output, and when it is saturated to a level at which the smear component cannot be expressed by intense light, the smear component is limited to a predetermined upper limit value. For this reason, even if the smear component is saturated, the level of the smear component to be corrected is within a certain range, and there is an effect that the image after smear correction does not feel strange.

  The random number generated by the random number generator 57 is supplied to the correction gain generators 52r, 52g, and 52b and the RAM controllers 54r, 54g, and 54b. Thus, by slightly changing the correction gain or pixel, the vertical stripes can be shaded and the image noise can be made invisible. As a result, as a result of smear correction, there is an effect that even if noise such as vertical stripes is slightly superimposed on the digital image signal, the influence of noise can be made difficult to appear.

  The subtracting unit 56 outputs the digital image signal as it is when the value obtained by subtracting the correction signal from the digital image signal is less than a predetermined threshold among the input digital image signals. When the predetermined threshold value is exceeded, an image signal obtained by subtracting the correction signal from the image signal supplied from the subtraction signal generators 61r, 61g, 61b is output. As a result, the image signal from which the smear component has been removed is not excessively corrected, and an image can be displayed in a state where the original image signal is utilized.

  In addition, various noises are usually superimposed on the digital image signal in addition to the smear component. These noises can be removed in advance by various filters provided in the smear waveform detection unit 31. Thus, even a digital image signal on which noise is superimposed, only the smear component can be reliably detected, so that only the detected smear component can be removed from the digital image signal.

Next, a second embodiment according to the present invention will be described with reference to FIGS.
However, in the following description, parts corresponding to those in FIGS. 2 and 4 already described in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.

FIG. 10 shows an internal configuration example of the correction amount generation unit 70 that corrects smear occurring obliquely.
The correction amount generation unit 70 includes a smear waveform detection unit 31, a black level detection unit 32, and a coring processing unit 33. In addition, a smear position registration unit 71 that registers smear position information and a smear inclination detection unit 72 that detects smear inclination and outputs R, G, and B correction signals subjected to smear correction are provided.

  The digital image signal subjected to the coring processing unit by the coring processing unit 33 is supplied to the smear position registration unit 71. The smear position registration unit 71 includes a memory (not shown), and stores the position of the smear 14 generated in the vertical light shielding area 15 as a position in contact with the vertical light shielding area 35 over a plurality of frames. The smear inclination detection unit 72 detects the smear inclination from the smear position information supplied from the smear position registration unit 71 over a plurality of frames. Then, a correction signal for removing a smear component having a slope from the digital image signal is generated.

FIG. 11 shows an example of smear generated obliquely with the moving light source.
FIG. 11A shows an example of smear that occurred in the previous frame.
FIG. 11B shows an example of smear that occurred in the current frame.

  When the relative position of the image sensor and the light source changes (in this example, when the light source 13 where smear occurs moves in the direction of the arrow 80), the smear 81 in the previous frame occurs in an oblique direction rather than a vertical direction. Then, the smear 81 in the current frame is generated in an oblique direction with the smear 81 moving in the direction of the arrow 60. At this time, the smear position registration unit 71 registers position information indicating the smear positions 82 a and 82 b in the vertical light shielding area 35. Then, the smear inclination detection unit 82 detects the inclination of the smear 81 based on the smear position information read from the smear position registration unit 71 and corrects the smear 81.

FIG. 12 shows an example of processing for correcting smear generated obliquely.
FIG. 12A shows an example of smear generated obliquely.
FIG. 12B shows an example in which the smear generated obliquely is excessively corrected.
FIG. 12C shows an example in which smear generated obliquely is appropriately corrected.

  FIG. 12A is the same view as FIG. 11A and shows a state in which smear is generated obliquely as the light source 13 moves in the lateral direction. At this time, the position information indicating the smear positions 82a and 82b is registered in the smear position registration unit 71 as described above.

  In order to correct the smear 81 generated obliquely, the smear inclination detection unit 72 takes a difference between the smear position detected in the vertical light shielding area 15 of the previous frame and the smear position detected by the vertical light shielding area 15 of the current frame. To calculate the difference. Since the vertical length of one frame is fixed, the inclination of the smear 81 can be detected from the relationship with this difference. Further, the smear 81 is corrected based on the inclination. As a result, the smear 81 is corrected in accordance with the surrounding luminance, chromaticity, and the like as in the correction unit 85 shown in FIG. 12B.

  However, when there are a plurality of subjects or when two or more subjects intersect, the smear inclination detection unit 72 may erroneously detect the smear position. When the smear inclination detection unit 72 erroneously detects the smear position, as shown in FIG. 12B, an overcorrection unit 86 that excessively corrects smear occurs. Since the overcorrection unit 86 is an area that does not need to be corrected, the overcorrection unit 86 is more conspicuous than the surrounding luminance and chromaticity as a result of the correction.

  In order to prevent such overcorrection, the subtracting unit 56 in FIG. 8 does not correct the region 87 if the corrected signal is equal to or less than a certain threshold value. Alternatively, a process of reducing the correction gain of the region 87 is performed. By this process, it is possible to suppress an unnatural image due to failure of smear correction.

  According to the correction amount generation unit 70 according to the second embodiment described above, the position information in the vertical light shielding region 15 of the smear in the oblique direction that occurs when the relative positions of the imaging element and the light source change is registered over a plurality of frames. To do. Then, the inclination of the smear is detected based on this position information, and a correction signal for removing this smear can be generated. Therefore, only the smear component of the smear in the oblique direction can be removed from the digital image signal. As a result, there is no effect of overcorrecting an area where smear does not occur, and the image quality can be improved.

  The series of processes in the above-described embodiments can be executed by hardware, but can also be executed by software. When a series of processing is executed by software, it is possible to execute various functions by installing programs that make up the software into a computer built into dedicated hardware, or by installing various programs. For example, a general-purpose personal computer or the like installs and executes a program constituting desired software.

  In addition, a recording medium in which a program code of software that realizes the functions of the above-described embodiments is recorded is supplied to the system or apparatus, and a computer (or a control device such as a CPU) of the system or apparatus is stored in the recording medium. Needless to say, this can also be achieved by reading and executing the program code.

  As a recording medium for supplying the program code in this case, for example, a floppy disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like is used. Can do.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS running on the computer based on the instruction of the program code performs actual processing. In some cases, the functions of the above-described embodiment are realized by performing part or all of the above.

  Furthermore, the present invention is not limited to the above-described embodiment, and it is needless to say that various other configurations can be taken without departing from the gist of the present invention.

1 is a block diagram illustrating an internal configuration example of an imaging device according to a first embodiment of the present invention. It is explanatory drawing which shows the structural example of the image pick-up element in the 1st Embodiment of this invention. It is a block diagram which shows the internal structural example of the smear correction | amendment part in the 1st Embodiment of this invention. It is a block diagram which shows the example of an internal structure of the correction amount production | generation part in the 1st Embodiment of this invention. It is a block diagram which shows the internal structural example of the smear waveform detection part in the 1st Embodiment of this invention. It is a block diagram which shows the example of an internal structure of the correction amount subtraction part in the 1st Embodiment of this invention. It is a block diagram which shows the internal structural example of the subtraction part in the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the coring process part in the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the gain adjustment in the 1st Embodiment of this invention. It is a block diagram which shows the example of an internal structure of the correction amount production | generation part in the 2nd Embodiment of this invention. It is explanatory drawing which shows the example of the smear detection corresponding to the smear which moves in the 2nd Embodiment of this invention. It is explanatory drawing which shows the example of the process which controls the over correction | amendment of smear in the 2nd Embodiment of this invention.

Explanation of symbols

  1r, 1g, 1b ... imaging device, 2r, 2g, 2b ... amplifier, 3r, 3g, 3b ... analog / digital conversion unit, 4 ... smear correction unit, 5 ... gain adjustment unit, 6 ... brightness adjustment unit, 7 ... gamma Correction part, 8 ... Output part, 10 ... Imaging device, 11 ... Horizontal light shielding area, 12 ... Light receiving part, 13 ... Light source, 14 ... Smear component, 21 ... Correction amount generation part, 22 ... Correction amount subtraction part, 31 ... Smear Waveform detection unit, 32 ... black level detection unit, 33 ... coring processing unit, 40 ... filter, 41 ... median filter, 42 ... integral filter, 43 ... low pass filter, 44 ... control unit

Claims (7)

The image signal supplied from an imaging device having an incident region for photoelectrically converting light incident from a light source to generate an image signal, and a light-shielding region for generating a reference signal from a dark current generated in a state where the light ray is shielded. The noise superimposed on the image signal is detected by removing the noise superimposed on the image signal by a predetermined filtering process and subtracting the reference signal from the image signal from which the noise is removed. A correction amount generating unit that generates a correction signal for removing the smear component from the image signal;
When the smear component is superimposed on the image signal, the correction signal is subtracted from the image signal to output an image signal with the smear component removed, and the smear component is superimposed on the image signal. And a correction amount subtraction unit that outputs the image signal as it is when there is no smear correction circuit. The smear correction circuit according to claim 1.
The correction amount generation unit
A smear waveform detection unit that detects a waveform of the smear component from the image signal output by the vertical light shielding region among the light shielding regions;
A black level detection unit that detects a signal level of the image signal output from a horizontal light-shielding region in which the smear component is not superimposed among the light-shielding regions, as a black level of the reference signal;
When the smear waveform detector detects the waveform of the smear component, the signal level of the image signal with the black level reduced by subtracting the black level from the signal level of the image signal output by the vertical shading region A coring processing unit that outputs the correction signal obtained by removing the smear component from the image signal from which the black level has been reduced when the value is within a predetermined upper limit value range from the first threshold value. Correction circuit. The smear correction circuit according to claim 1 or 2,
The correction amount subtraction unit
A signal level detector for detecting a signal level of the image signal;
A random number generator for generating random numbers;
A correction gain generation unit that generates a correction gain of the image signal by changing the signal level detected by the signal level detection unit according to a random number supplied from the random number generation unit;
A storage unit that stores the correction signal supplied from the correction amount generation unit corresponding to the pixel to be corrected;
A read control unit that reads out the correction signal by changing pixels corrected by the correction signal stored in the storage unit according to a random number supplied from the random number generation unit;
A multiplication unit that multiplies the correction gain supplied from the correction gain generation unit by the correction signal read by the read control unit to generate a subtraction signal that reduces the signal level of the image signal;
A smear correction circuit comprising: a subtraction unit that outputs an image signal obtained by subtracting the subtraction signal from the image signal. The smear correction circuit according to claim 3.
The subtraction unit
A subtraction signal generator for generating an image signal obtained by subtracting the correction signal from the image signal;
The signal level of the image signal is compared with a second threshold, and the image signal is output when the signal level of the image signal is less than the second threshold, and the signal level of the image signal is the second threshold A smear correction circuit comprising: a threshold determination unit that outputs an image signal obtained by subtracting the correction signal from the image signal supplied from the subtraction signal generation unit when a threshold value is exceeded. The smear correction circuit according to claim 2,
The correction amount generation unit
A smear position registration unit that registers the position of the smear generated in the light-shielding region as position information over a plurality of frames;
When the relative position of the image sensor and the light source changes, the smear inclination is detected from the digital image signal based on the position information registered in the smear position registration unit and the smear of the smear having the inclination is detected from the digital image signal. A smear inclination detection unit that generates the correction signal for removing a component; and a smear correction circuit. The image signal supplied from an imaging device having an incident region for photoelectrically converting light incident from a light source to generate an image signal, and a light-shielding region for generating a reference signal from a dark current generated in a state where the light ray is shielded. The noise superimposed on the image signal is detected by removing the noise superimposed on the image signal by a predetermined filtering process and subtracting the reference signal from the image signal from which the noise is removed. And generating a correction signal for removing the smear component from the image signal;
When the smear component is superimposed on the image signal, the correction signal is subtracted from the image signal to output an image signal with the smear component removed, and the smear component is superimposed on the image signal. And outputting the image signal as it is when there is no smear correction method. An imaging element having an incident area for photoelectrically converting light incident from a light source to generate an image signal and a light-shielding area for generating a reference signal from a dark current generated in a state where the light ray is shielded;
The noise superimposed on the image signal is removed from the image signal and the reference signal supplied from the image sensor by a predetermined filter process, and the reference signal is subtracted from the image signal from which the noise has been removed. A correction amount generating unit that detects a smear component superimposed on the image signal and generates a correction signal for removing the smear component from the image signal;
When the smear component is superimposed on the image signal, the correction signal is subtracted from the image signal to output an image signal with the smear component removed, and the smear component is superimposed on the image signal. And a correction amount subtraction unit that outputs the image signal as it is when there is no image pickup device.