JP2005102314A - Pixel defect detecting method for solid-state image sensor, and imaging device using the method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect and correct a pixel defect in real time in a pixel defect detecting method for a solid-state image sensor that images a plurality of polarized beams resulting from polarizing image pickup light and outputs a video signal. <P>SOLUTION: In accordance with a result of comparing video signals imaged and outputted at the same image pickup position or nearby image pickup positions for each polarized beam, it is detected whether a pixel defect occurs in the image sensor that images any one of polarized beams. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for detecting pixel defects that occur in a solid-state imaging device used in a television camera, and in particular, when imaging spectral light at the same imaging position of imaging light with a plurality of imaging devices, respectively, the same imaging The present invention relates to a detection method in which it is possible to detect in real time whether or not a pixel defect has occurred in any of the image pickup devices by comparing the video signals picked up and output by the image pickup device at the position.

  As a method for detecting a pixel defect of a solid-state image sensor of a conventional television camera, for example, as described in JP-A-10-98651, an optical image from a subject is applied to a photoelectric conversion surface of a solid-state image sensor. Optical means for irradiating at each first period, and reading means for driving the solid-state image sensor to read out signal charges of the solid-state image sensor at every second period shorter than the first period Storage means for holding, as a reference signal, an output signal charge of the solid-state image sensor when the optical image is not irradiated on the photoelectric conversion surface of the solid-state image sensor due to the difference between the first and second periods. And an output signal charge of the solid-state imaging device when the optical image is irradiated onto the photoelectric conversion surface of the solid-state imaging device as an imaging signal, and means for correcting the imaging signal with the reference signal There are things.

According to this conventional technique, a black reference signal is inserted between video signals using a rotary shutter disk or an electronic shutter, and these signals can be output from the solid-state imaging device. As described in the above publication, in the pixel defect correction circuit, in the reference signal period, a digitized signal as a reference image of the corresponding field is written in the frame memory, and other periods excluding the reference signal period Then, by subtracting the corresponding field image stored in the frame memory from the digitized field signal and outputting it, correction is performed while detecting fixed pattern noise in real time, It is said that the image quality can be improved by eliminating the need to create pixel defect data at the time of manufacture and automatically obtaining an image that has been corrected even if the pixel defect changes during use.
JP-A-10-98651

  However, depending on such a configuration, a pause period occurs in the output signal of the pixel defect correction circuit. Therefore, when a moving image is viewed in real time, it is necessary to obtain video signals at continuous time intervals by passing through a time axis expansion circuit, for example. The timing is adjusted by this time axis expansion circuit. In order to do so, a large circuit configuration must be required, which increases costs. In addition, since the video signal corresponding to the above-described pause period is not captured, the content of the video is interrupted more than necessary before and after that, and the time interval is extended and the time interval is continued regardless of the interruption of the content. However, the video content may become more discontinuous than required, which may cause a problem.

  The object of the present invention is to eliminate the necessity of providing a reference signal period separately from the video signal period as in the above-described prior art, and without using a large-scale circuit such as a pixel defect correction circuit. It is to be able to detect a pixel defect in real time from a captured video signal. In addition, it is possible to control externally whether or not to correct each pixel defect according to the defect signal level of the detected pixel defect, and the image status (existence of accumulation, accumulation time, degree of video gain). The pixel defect correction method in which the defect signal level of the pixel defect and the correction coefficient for correcting the pixel defect are changed according to the above.

  In order to solve the above-described problems, the present invention provides a pixel defect detection method for a solid-state imaging device that images a plurality of spectral lights obtained by spectrally dividing imaging light and outputs a video signal, and the same for each spectral light. Detecting whether or not a pixel defect has occurred in an image sensor that picks up one of the spectral lights according to the result of comparing the video signals picked up and output at the imaging position or in the vicinity of the imaging position It is. The present invention further detects whether or not the pixel defect occurs each time the video signal is output from the image sensor.

  In addition, in order to solve the above-described problem, the present invention provides a pixel defect detection method for a solid-state imaging device that captures a plurality of spectral lights obtained by spectrally dividing imaging light and outputs a video signal, for each spectral light. A difference between a video signal level from a first pixel corresponding to a predetermined imaging position of the imaging light and a video signal level obtained from a second pixel adjacent to or close to the first pixel, and the spectral light The predetermined imaging relating to the spectral light corresponding to the first pixel corresponding to the predetermined imaging position, in particular, a more characteristic difference among the compared differences, by comparing the differences obtained for each The first pixel corresponding to the position is detected as a defective pixel.

  Furthermore, in the process of detecting, among the differences obtained for each spectroscopic light, the difference is the largest deviation with respect to the average value of the differences, and the value of the maximum deviation is smaller than a predetermined value. The first pixel corresponding to the predetermined imaging position related to the spectral light with respect to the difference that is large may be detected as a defective pixel. Further, for each difference obtained for each spectroscopic light, a value obtained by averaging the differences excluding the difference may be used as the average value for obtaining the deviation. Furthermore, the detection process may be repeatedly performed each time a video signal from the first pixel corresponding to each of a plurality of imaging positions of the imaging light is output. Further, the present invention provides an imaging apparatus having a solid-state imaging device that images a plurality of spectral lights obtained by separating imaging light and outputs a video signal, at the same imaging position or a nearby imaging position for each spectral light. Means for comparing image signals that have been picked up and output, and means for detecting whether or not a pixel defect has occurred in an image pickup element that picks up one of the spectral lights according to the comparison result. is there.

  The present invention is also directed to an imaging apparatus having a solid-state imaging device that images a plurality of spectral lights obtained by spectrally separating imaging light and outputs a video signal, for each spectral light, a predetermined imaging position of the imaging light. Means for obtaining a difference between a video signal level from the first pixel corresponding to the first pixel and a video signal level obtained from a second pixel adjacent to or close to the first pixel, and the differences obtained for each spectroscopic light. Means for detecting the first pixel corresponding to the predetermined imaging position as a defective pixel related to the spectral light corresponding to a more characteristic difference among the compared differences. It is. Further, it may be provided with means for correcting the defect signal level of the pixel defect according to the output from the detection means, and outputting a video signal with the defect signal level corrected.

  According to the present invention, it is not necessary to provide a reference signal period separately from the video signal period, and a video signal for each pixel is output without using a large-scale circuit such as a pixel defect correction circuit. It is possible to perform pixel defect detection or pixel defect correction in real time on a pixel-by-pixel basis by repeating sequential processing while sequentially switching the target pixel, eliminating the need to detect the pixel defect position in advance, and pixel defects in the prior art Even when the number of pixels increases, it is possible to save the trouble of detecting the pixel defect position again and writing it in the memory or the like. In addition, it is possible to control externally whether or not to correct each pixel defect according to the defect signal level of the detected pixel defect, and the image status (existence of accumulation, accumulation time, degree of video gain). The defect signal level of the pixel defect and the correction coefficient for correcting the pixel defect can be changed according to, for example, the white defect detection level so that the increase / decrease in the white defect caused by the change in the accumulation time can be corrected appropriately. A correction adapted to the situation of the captured image, such as changing (W) according to the accumulation time, becomes possible.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 is a diagram showing a block configuration example of a television camera according to the present invention. In this figure, 1 is a lens that collects imaging light incident from a subject, 2 is imaging light that has passed through the lens 1, and a plurality of spectral lights such as red, green, and blue (hereinafter R, G, and B, respectively). The prisms 3, 4, and 5 that separate the light of each wavelength in (described in the above) receive the spectral light from the prism 2, and are obtained by photoelectric conversion in accordance with the amount of spectral light received for each pixel. It is a solid-state imaging device (CCD) provided with a plurality of light receiving elements each storing charges.

  6, 7, and 8 input video signals generated by sequentially reading out the charges accumulated in the image sensors 3, 4, and 5 from the light receiving elements, and removing noise components present in the input video signals Then, a correlated double sampling (CDS) circuit that outputs a video signal from which a noise component has been removed by sample-holding only the signal component. 9, 10, and 11 are preamplifier circuits that perform video signal processing such as gain correction and gamma correction on the video signals output from the CDS circuits 6, 7, and 8, and 12, 13, and 14 are preamplifier circuits 9, 10, and 14, respectively. 11 is an A / D converter that converts the processed video signal output from each of the video signals into a digital signal. 15 detects pixel defects for each pixel based on the digital video signals of the respective colors output from the A / D converters 12, 13, and 14, and determines the detected pixel position and the defect signal level of the detected pixel defect. A pixel defect detection circuit 16 for outputting a signal indicating the A / D converters 12, 13, and 16 according to the signal indicating the detected pixel position and the defect signal level of the pixel defect detected and output by the pixel defect detection circuit 15. 14 is a pixel defect correction circuit that corrects pixel defects in the digital video signal output from each of the image signals 14 and outputs the video signals corrected for the pixel defects to the video signal processing circuit 17. In the video signal processing circuit 17, conversion processing of the video signal format and the like are appropriately performed, and the processed video signal is output to a subsequent stage (not shown). 18 is for making the imaging devices 3, 4, 5 and each of the circuits described above operate at a predetermined timing, and for establishing external control between the imaging device and an external device (not shown). CPU circuit.

  Next, operations of the pixel defect detection circuit 15 and the pixel defect correction circuit 16 will be described with reference to FIG. An example will be described in which a white defect occurs in an image due to a pixel defect. Here, since the defect signal level of the defective pixel that becomes a white defect appears as a peak component having a signal level higher than the video signal of the surrounding pixels, the value of the video signal (defect signal) level of the pixel of the white defect is The value is relatively larger than the average value of the video signal levels of the peripheral pixels, and the difference value (difference) obtained by subtracting the average value from the value of the defect signal level is also a large value. On the other hand, for normal pixels that are not pixel-defective, the normal pixels are used in almost all shooting situations, for example, in a situation where a normal imaging device is shot with a field of view comparable to that of a general human visual field. In many cases, the video signal level is almost the same as the video signal level of the surrounding pixels. Therefore, the value of the video signal level of the normal pixel is about the same value as the average value of the video signal level of the peripheral pixels, and the difference obtained by subtracting the average value from the value of the video signal level of the normal pixel. The value is a relatively small value. Further, for example, when a predetermined pixel having an R channel is a white defect pixel, any one of a corresponding G channel pixel and a B channel pixel is similarly positioned at a predetermined imaging position of imaging light corresponding to the predetermined pixel. It is very unlikely that either or both are defective pixels, as are the R channel pixels. That is, at least one pixel of the R, G, and B channel pixels corresponding to the predetermined imaging position of the imaging light may be a pixel defect.

  Therefore, in the example in which the predetermined pixel of the R channel corresponding to the predetermined imaging position of the imaging light is a white defect pixel, the difference between the signal level of the predetermined pixel of the R channel and the average value of the signal levels of surrounding pixels. Is a relatively large value as described above, and at the same time, the value of the difference between the signal level of the G channel pixel and the average signal level of the surrounding pixels corresponding to the predetermined position, and the B channel It can be said that the difference value between the signal level of the pixel and the average value of the signal level of the surrounding pixels is a relatively small value in both the G channel and the B channel. Therefore, by comparing the difference values of the R channel, the G channel, and the B channel described above for the pixel corresponding to the predetermined imaging position of the imaging light, any channel of the R channel, the G channel, and the B channel is compared. It is possible to detect whether or not a pixel defect has occurred in the image sensor.

  Hereinafter, the pixel defect detection method of the present invention will be described. First, the imaging light that has passed through the lens 1 of the imaging apparatus according to the present invention is split by the prism 2 to obtain spectral light of the R, G, and B channels. These spectral lights are received by the solid-state imaging devices 3, 4 and 5 for each channel. In each solid-state imaging device, electric charge is obtained by photoelectric conversion according to the amount of spectral light received for each pixel, the electric charge is accumulated, and further sequentially output as a video signal for each channel. A video signal is generated. These video signals are input to the pixel defect detection circuit 15 via the CDS circuits 6, 7 and 8, preamplifier circuits 9, 10 and 11, and A / D converters 12, 13 and 14.

  The pixel defect detection circuit 15 detects a defective signal portion that is a pixel defect in the input video signal by using signal processing of a procedure as shown in FIG. As the detection procedure, first, pixels of each channel corresponding to an imaging position of interest of imaging light, for example, an imaging position An corresponding to an nth light receiving element among a plurality of light receiving elements arranged in a matrix on a solid-state imaging element. The video signal level value for each channel (hereinafter referred to as the first pixel for each channel) is Rn, Gn, and Bn for each channel. Further, for each channel, a pixel adjacent to or close to the first pixel (hereinafter referred to as a second pixel for each channel) is appropriately selected, and the average value of the video signal levels of the selected second pixel And the average values thus obtained are defined as Rn_av, Gn_av, and Bn_av, respectively (101 in FIG. 1).

Next, for each channel, a difference value (difference) obtained by subtracting the average value of the video signal level of the second pixel obtained from the video signal level of the first pixel is expressed by the following formulas 1, 2, and 3. Respectively, Rw, Gw, and Bw (102 in FIG. 1).
Rw = Rn−Rn_av (Formula 1)
Gw = Gn−Gn_av (Formula 2)
Bw = Bn−Bn_av (Equation 3)
Next, in order to compare the obtained differences Rw, Gw, and Bw, a difference for each channel is obtained by subtracting the average value of the differences of channels other than the channel from the difference of each channel, and these are obtained. The maximum channel deviation value among the channel deviations is defined as Wmax. In the example shown in FIG. 1, the deviation formulas of the respective channels are represented by the following equations 4, 5, and 6 (103 in FIG. 1).
R channel ... | Rw- (Gw + Bw) / 2 | ... (Formula 4)
G channel ... | Gw- (Bw + Rw) / 2 | ... (Formula 5)
B channel ... | Bw- (Rw + Gw) / 2 | ... (Formula 6)
Although not shown, as another formula for the deviation of each channel, the average values of the differences of the three channels R, G, and B are used to obtain the following formulas 7, 8, and 9, which are obtained from these formulas. The maximum value Wmax may be obtained from the obtained values.
R channel ... | Rw- (Rw + Gw + Bw) / 3 | ... (Formula 7)
G channel ... | Gw- (Rw + Gw + Bw) / 3 | ... (Formula 8)
B channel ... | Bw− (Rw + Gw + Bw) / 3 | (Equation 9)
When Wmax is obtained as described above, it is determined whether or not the maximum value Wmax is larger than a predetermined threshold value Wth (104 in FIG. 1). By this determination, it is determined whether or not the maximum value Wmax is a value based on the video signal level of the pixel related to the pixel defect. If the maximum value Wmax is larger than the predetermined threshold value Wth, Is detected as a defective pixel in which a pixel defect corresponding to the predetermined imaging position occurs in the first pixel of the channel related to the deviation Wmax. On the other hand, when the maximum value Wmax is smaller than the predetermined threshold value Wth, it is determined that there is no pixel having a pixel defect corresponding to the predetermined imaging position.

  When a pixel defect is detected by carrying out the pixel defect detection procedure of the present invention described above, a signal (defective pixel position signal) indicating the position of the defective pixel corresponding to the detection result from the pixel defect detection circuit 15. ) Is output and input to the pixel defect correction circuit 16.

The pixel defect correction circuit 16 generates a correction signal using the input defective pixel position signal and the video signal of the pixels around the defective pixel output from any of the A / D converters 12, 13, and 14. . Here, FIG. 3 is a diagram for explaining an example of the correction signal generation method. In the example shown in FIG. 3, when Rn is a defect signal level of a pixel defect, the level Rn ′ of the correction signal is the video signal level Rn−2, Rn−1, Rn + 1, four adjacent pixels. It is generated by obtaining the average value from Rn + 2. Here, the correction signal level and the average value of the video signal levels of the second pixels in the description for obtaining the above-described difference may be obtained by the same formula. In this case, the following formulas 10, 11, and 12 are used. They are calculated.
Rn_av = (Rn−2 + Rn−1 + Rn + 1 + Rn + 2) / 4 (Formula 10)
Gn_av = (Gn-2 + Gn-1 + Gn + 1 + Gn + 2) / 4 (formula 11)
Bn_av = (Bn-2 + Bn-1 + Bn + 1 + Bn + 2) / 4 (formula 12)
It should be noted that the correction signal level and the average value described above are not necessarily the same value, and it goes without saying that various calculation methods and combinations of signals are used as their calculation methods.

  As described above, when Rn is a defect signal level of a pixel defect, as illustrated in 105 of FIG. 1, the correction signal level Rn ′ generated in accordance with the defective pixel position signal, in this example, Rn_av is Instead of the defective signal level Rn, the video signal level of the defective pixel (first pixel) is set, and thereby the pixel defect is corrected.

  Note that the above description has explained the processing procedure of pixel defect detection and pixel defect correction for a predetermined imaging position of imaging light related to a certain pixel of interest. These procedures are operations for outputting a video signal from a solid-state imaging device. Accordingly, each time a video signal for each pixel is output, the sequential processing is repeated while sequentially switching the target pixel, so that pixel defect detection or pixel defect correction can be performed in real time on a pixel basis.

  In the present invention, the operator can set the pixel defect correction by operating the menu screen (not shown) of the imaging apparatus or by an external control signal generated using an external control function. A signal is transmitted to the CPU circuit 18 so that a signal for controlling the circuit is transmitted from the CPU circuit 18 to the pixel defect detection circuit 15 and the pixel defect correction circuit 16. By doing so, it is possible to control from the outside whether or not to correct each pixel defect, or the defect signal of the pixel defect depending on the image situation (existence of accumulation, accumulation time, degree of video gain, etc.) The level and the correction coefficient for correcting the pixel defect can be changed.

The flowchart for demonstrating operation | movement of one Example of this invention. The figure which shows the block structural example of the television camera in connection with this invention. The figure for demonstrating an example of the correction signal generation method.

Explanation of symbols

1: lens, 2: prism, 3, 4, 5: imaging device (CCD), 6, 7, 8: CDS circuit, 9, 10, 11: preamplifier circuit, 12, 13, 14: A / D converter, 15: Pixel defect detection circuit, 16: Pixel defect correction circuit, 17: Video signal processing circuit, 18: CPU circuit.

Claims (3)

  In the pixel defect detection method for a solid-state imaging device that captures a plurality of spectral lights obtained by spectrally separating imaging light and outputs a video signal, a first imaging unit that corresponds to a predetermined imaging position of the imaging light for each spectral light. A difference between a video signal level from a pixel and a video signal level obtained from one or a plurality of second pixels adjacent or close to the first pixel is obtained, and an average of the differences among the differences for each spectroscopic light The first pixel corresponding to the predetermined imaging position related to the spectroscopic light with respect to the difference having the largest deviation with respect to the value and the difference in which the value of the maximum deviation is larger than the predetermined value is defined as the defective pixel. And detecting a pixel defect.   In an imaging apparatus having a solid-state imaging device that images a plurality of spectral lights obtained by spectrally separating imaging light and outputs a video signal, a first pixel corresponding to a predetermined imaging position of the imaging light for each spectral light Means for calculating a difference between the video signal level from the first video signal level and a video signal level obtained from one or a plurality of second pixels adjacent or close to the first pixel, and an average of the differences obtained for each spectroscopic light Means for obtaining a difference having a maximum deviation with respect to the value, and a difference in which the value of the maximum deviation is greater than a predetermined value, and a result obtained from the means according to the predetermined imaging position An image pickup apparatus comprising: means for detecting the first pixel as a defective pixel. 3. The imaging apparatus according to claim 2, further comprising means for correcting a defect signal level of the pixel defect in accordance with an output from the detection means, and outputting a video signal with the defect signal level corrected. An imaging apparatus characterized by the above.