JP2013115675A - Imaging apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the event that a defect cannot be corrected and a defective pixel correction is made even in the absence of a defective pixel because of a long distance to a pixel of an identical color and a different weighting factor onto an adjacent pixel brightness in a color imaging apparatus having a plurality of color filters disposed therein.SOLUTION: In a pixel defect detection algorithm, pixel defect determination processing with respect to pixels of an identical color (for example, a target pixel is red [R]) is performed as a first determination processing (GS1). Further, pixel defect determination processing with respect to pixels of a different color (pixels of a color other than the color used in the first determination processing) is performed as a second determination processing (GS2). Furthermore, final determination processing (GS3) is performed as a final pixel defect determination on the basis of the results of the first determination processing and the second determination processing.

Description

  The present invention relates to an imaging apparatus.

  As a background art of this technical field, there is Patent Document 1. According to this publication, “the defect detection circuit 24 detects a defect using the luminance level of the image sensor 13 when the lens 11 is out of focus, and the address of the detected defective pixel is held in the system controller 26. The defect correction is performed at the address position ”(refer to the summary).

JP-A-2005-328134

  In image sensors, there are pixel defects that occur during the manufacturing process and pixel defects that occur due to changes over time due to radiation, such as temperature changes and cosmic rays. Pixel defects have so-called “white defects” and defects that increase the brightness of the defect location. There is a so-called “black defect” in which the brightness of the portion is lowered.

  Since the pixel defect that occurs in the manufacturing process is a fixed position, shipping is performed after performing defect correction that corrects the output of the defective pixel with a value obtained by interpolating the output of the neighboring pixel at the time of factory shipment.

  However, there is a problem that the image quality is impaired because interpolation cannot be performed for pixel defects of the image sensor caused by temperature change or change with time.

  As an example for solving this problem, in Patent Document 1, pixel defect detection is performed based on an output signal of a solid-state imaging device when a lens is in an out-of-focus state, so that it is possible to deal with defective pixels that appear with time. Is described.

  However, in Patent Document 1, a white defect is determined to be a white defect when a difference between adjacent pixel signals is larger than a preset difference width and the difference is positive, and a black defect is determined as an adjacent pixel. When the difference from the above signal is larger than the difference width set in advance and the difference is negative, it is determined that there is a black defect.

  For this reason, a monochrome imaging device can correctly determine, but in a color imaging device in which a plurality of color filters are arranged, the distance from a pixel of the same color is increased, or the weighting factor for the luminance of the adjacent pixel is Due to the difference, there is a problem that the fixed threshold cannot correct the defect, or the pixel is corrected as a defective pixel even though it is not a defective pixel. The present invention solves the above-described problems and provides an imaging apparatus capable of detecting a pixel result and correcting a pixel defect with higher accuracy even in a color imaging apparatus.

Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.
(1) image pickup means for picking up an image of a subject; and defect determination means for determining a defect of a target pixel using an image signal from the image pickup means, wherein the defect determination means has a signal level of the target pixel First determination processing means for comparing a signal level of a comparison target pixel around the target pixel with the same color as the target pixel and determining a defect, and a signal level of the target pixel is different from the target pixel A second determination processing unit that compares a signal level of a comparison target pixel around the target pixel with a color to determine a defect, a defect determination result by the first determination processing unit, and a second determination processing unit An image pickup apparatus comprising: a third determination processing unit that determines whether or not a pixel defect is generated using a defect determination result.
(2) In the imaging device according to (1), the image signal is a primary color Bayer pixel array, and the second determination processing unit sets Gb as a peripheral pixel of a different color when the target pixel is Gr. An image pickup apparatus characterized by determining a pixel defect by comparing signal levels and determining a pixel defect by comparing the signal level as Gr for surrounding pixels of different colors when the target pixel is Gb is there.
(3) The imaging apparatus according to (1), wherein the image signal is a primary color Bayer pixel array, and the second determination processing unit is adjacent to the target pixel when the target pixel is R or B. The average value or weighted sum of the signal levels of the first group of pixels of Gr and Gb is calculated, and the average value or weighted sum of the signal levels of the second group of pixels of Gr and Gb that are not adjacent to the target pixel is calculated. And determining a pixel defect by comparing the average value or weighted addition value of the signal levels of the pixels of the first group with the average value or weighted addition value of the second group. It is.

  The center pixel in the range of N * M pixels (N and M are odd numbers) in the horizontal and vertical directions is the target pixel (defect determination pixel), and the comparison target pixel is the surrounding pixel of the same color as the target pixel The imaging apparatus is characterized by determining whether the pixel of interest is defective using only the same color.

  The center pixel in the range of N * M pixels (N and M are odd numbers) in the horizontal and vertical directions is the target pixel (defect determination pixel), the comparison target pixel is green, and the distance from the target pixel The image pickup apparatus is characterized in that the green group is divided by the above, an average value is calculated, and whether or not the pixel of interest is defective is determined based on a difference in the average value for each group.

  The center pixel in the range of N * M pixels (N and M are odd numbers) in the horizontal and vertical directions is the target pixel (defect determination pixel), and the comparison target pixel is the surrounding pixel of the same color as the target pixel The difference between the target pixel and the comparison target pixel is excluded from the comparison target pixel that is smaller than the threshold value, and the average value of the comparison target pixel and the absolute maximum and minimum values of the difference between the target pixel and the comparison target pixel are calculated. An interpolation value is calculated with a value between the average value of the target pixel and the comparison target pixel based on the average value of the comparison target pixel and the maximum absolute value of the difference minus the minimum value. If the target pixel is defective, the interpolation value Is output as the value of the target pixel.

  As described above, according to the present invention, the threshold value can be switched or determined in consideration of the weighting factor with respect to the luminance of the pixel determined to be defective, and the threshold value setting and determination method can be changed according to the white balance state. Even in a color imaging apparatus, it becomes possible to detect pixel defects with higher accuracy, and to perform defect correction with higher accuracy.

  In addition, according to the present invention, since the defect is determined using only the same color, it is possible to provide an imaging device capable of detecting a pixel defect with higher accuracy even when, for example, the white balance is shifted. it can.

  In addition, according to the present invention, since defect determination is performed using an output of a pixel that is close to the defect determination pixel, an imaging device that can correctly detect a pixel defect even in a subject having a higher frequency component is provided. Can do.

  In addition, according to the present invention, a distinctive defect and an inconspicuous defect are separated, a correction is strengthened for a conspicuous defect, and a correction is weakened for an inconspicuous defect, thereby obtaining a large correction effect and greatly degrading an image. It is possible to provide an image pickup apparatus that can perform both of the non-existence and the defect correction that balances the two.

  According to the present invention, it is possible to provide an imaging apparatus capable of highly accurate pixel defect detection and pixel defect correction even in a color imaging apparatus.

It is a figure showing an example of 1 composition of an imaging device concerning the present invention. It is a figure which shows the 1st structural example of a defect detection and correction circuit. FIG. 6 is a diagram showing an operation flow of the comparator 201. 3 is a diagram illustrating an example of a pixel for supplementary explanation of a threshold selection circuit 203. FIG. 12 is a diagram illustrating another example of a pixel for supplementary explanation of the threshold selection circuit 203. FIG. It is a figure which shows the 2nd structural example of a defect detection and correction circuit. It is a figure which shows the structural example of a 1st defect detection circuit. It is a supplementary explanatory diagram of the first defect detection circuit. It is a figure which shows the other structural example of a defect detection circuit. It is a figure which shows the processing algorithm suitable for the defective pixel detection which concerns on this invention. It is a figure which shows the structural example of a 2nd defect detection circuit. It is a figure which shows the structural example of the 3rd defect detection circuit.

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

  In the present embodiment, a video camera that is one form of the imaging apparatus will be described as an example. The configuration example of the first embodiment of the imaging apparatus according to the present invention includes an imaging lens 101, an imaging element 102, an A / D conversion circuit 103, a defect detection / correction circuit 104, a signal, as shown in FIG. The processing circuit 105 and the timing generator 106 are used as appropriate. Hereinafter, defect detection / defect correction by the imaging apparatus will be described. First, light incident from a subject via the imaging lens 101 is irradiated to the image sensor 102, and a subject image is formed. The image sensor 102 is scanned in the horizontal and vertical directions by the drive pulse from the timing generator 106, captures a subject image, and generates an electrical signal. This electric signal is converted into a digital signal by the A / D conversion circuit 103 and input to the defect detection / correction circuit 104. In the defect detection / correction circuit 104, the threshold value for detecting a defect (for white defect / for black defect) is appropriately switched according to the color information from the timing generator 106, and the difference between the input signal and neighboring pixels and the threshold value (for white defect / black). When it is determined as “normal” due to the relationship (for defects), an input signal is output to the signal processing circuit 105, and when it is determined as “defective”, a signal interpolated from neighboring pixels is output to the signal processing circuit 105. . The signal processing circuit 105 performs various camera signal processing such as noise removal and gamma correction, converts the signal into a signal such as a TV signal, and outputs the signal.

  Next, details of the first configuration example of the defect detection / correction circuit 104 will be described with reference to FIG. As shown in FIG. 2, the first configuration example of the defect detection / correction circuit 104 is configured using a comparator 201, a data latch 202, and a threshold selection circuit 203 as appropriate. An input signal to the defect detection / correction circuit 104 is supplied to the comparator 201 and the data latch 202. The data latch 202 holds an input signal one pixel before and outputs a signal one pixel before the input pixel to the sequential comparator 201. On the other hand, color information for an input pixel is input to the threshold selection circuit 203. In the threshold selection circuit 203, thresholds (for white defects / for black defects) for each color are set in advance, and thresholds (for white defects / for black defects) corresponding to the type of input color information are selected, Output to the comparator 201.

  Here, the operation of the comparator 201 will be supplemented with reference to FIG. The comparator 201 determines whether the signal amount of the input pixel is larger than the sum of the signal amount of the white defect threshold value and the data latch 202 (STEP 1), and the signal amount of the input pixel is equal to the white defect threshold value and the data latch 202. When the sum is larger than the sum of the signal amounts, a “white defect” is determined, and the signal amount of the data latch 202 corresponding to the signal interpolated from the neighboring pixels is output to the signal processing circuit 105 (STEP 2). When the signal amount of the input pixel is smaller than the sum of the signal amount of the white defect threshold value and the data latch 202, the signal amount of the input pixel is more than the signal amount obtained by subtracting the black defect threshold value from the signal amount of the data latch 202. (STEP 3). If the signal amount of the input pixel is smaller than the signal amount obtained by subtracting the black defect threshold from the signal amount of the data latch 202, it is determined as “black defect”, and interpolation is performed from neighboring pixels. The signal amount of the data latch 202 corresponding to the received signal is output to the signal processing circuit 105 (same as STEP 2). If it is not determined as the “white defect” or “black defect”, it is determined as “normal” and an input signal is output to the signal processing circuit 105 (STEP 4). Note that the order of STEP1 and STEP3 may be reversed.

Further, the operation of the threshold selection circuit 203 will be supplemented with reference to FIG. FIG. 4 shows an example of a color image pickup device using primary color filters. For example, in an NTSC TV image pickup device, a luminance signal (Y) is input by inputting red (R), green (G), and blue (B), and Y = It is generated with the formula 0.29 * R + 0.60 * G + 0.11 * B. Focusing on the horizontal * vertical 2 * 2 pixels surrounded by a thick frame in FIG. 4, there are two Gs, so the weighting factor for the luminance signal (Y) is
R: G: B = 0.29: 0.30: 0.11 (Equation 1)
It has become.

  Here, considering the case where each of the red (R), green (G), and blue (B) pixels has a defect, the luminance signal (Y) is affected by the defect at the ratio of the above equation (1). Therefore, the pixel ratio can be detected with high accuracy by adjusting the ratio of the threshold to the ratio of the above formula (1). That is, in the threshold selection circuit 203, threshold values (for white defects / for black defects) for each color are set in advance approximately in the ratio of the above expression (1), and the type of color information of the pixel that is determined to be a defect (input) It turns out that it is good to switch according to.

As another example, in an imaging apparatus using a complementary color filter as shown in FIG. 5, a plurality of ways of assembling a signal processing matrix can be considered. For example, Y = K1 * Ye + K2 * Cy + K3 * Mg + If it is generated by the K4 * G formula,
Ye: Cy: Mg: G = K1: K2: K3: K4 (Formula 2)
In the same way as the above formula (1), it is preferable to set a threshold value (for white defect / for black defect) in advance for each color in the ratio of the above formula (2). It turns out that it is good to set similarly.

  As described above, in this execution penalty, the threshold value for determining whether or not a defect is detected is switched in accordance with the color information of the pixel whose defect is detected in consideration of the degree of influence of the color information of the pixel whose defect is detected on the luminance signal. Also in the apparatus, it is possible to provide an imaging apparatus capable of detecting pixel defects and correcting pixel defects with high accuracy. In addition, this embodiment is particularly effective in an imaging apparatus such as a surveillance camera system in which an energization is continued for a long period of time and defective pixels appear due to temperature changes or temporal changes.

  In the present embodiment, an example has been described in which the data latch 202 is used to detect and correct the defect using the information of the immediately preceding pixel as a neighboring pixel. However, for example, the pixel one line before using a line memory. Defect detection and defect correction may be performed using this information, and using a feed (frame) memory, etc., left and right, up and down one line, diagonal upper right, diagonal lower right, diagonal upper left, diagonal lower left, etc. Defect detection and defect correction may be performed using peripheral pixel information, and various changes are possible.

  As a second embodiment of the imaging apparatus according to the present invention, an example in which the defect detection / correction circuit is different will be described. FIG. 6 is a diagram illustrating a second configuration example of the defect detection / correction circuit. 2 that perform the same operations as those in the first configuration example of the defect detection / correction circuit shown in FIG.

  In the imaging apparatus, since the color temperature of the light source to be photographed varies, so-called auto white balance control that automatically controls the color gain so that white can be photographed as white is widely adopted. For example, as the color temperature increases, the red (R) gain is increased and the blue (B) gain is decreased. Conversely, the lower the color temperature, the smaller the red (R) gain and the larger the blue (B) gain. In addition, the green (G) gain may be controlled under a light source such as a fluorescent lamp.

  When the defect detection / correction circuit 104-2 is arranged before the auto white balance control, as described above, the threshold value (for white defect) set for each color is changed in order to change the gain of each color. (For black defects), it is necessary to consider the gain of each color.

  In the threshold selection circuit 303, the gain of each color (R, G, B) from an auto white balance control unit (not shown) is input, and the threshold (for white defect / black defect) for each color set in advance is input. A threshold value (for white defect / for black defect) corresponding to the type of the input color information is selected as a value obtained by multiplying the input gain of each color, and is output to the comparator 201.

  In this embodiment, even when the light source to be photographed is changed, the degree of influence of the color information suitable for the light source on the luminance signal is taken into consideration, so that an image pickup apparatus capable of detecting pixel defects and correcting pixel defects with higher accuracy. Can be provided.

  As a third embodiment of the imaging apparatus according to the present invention, an example using the first defect detection circuit shown in FIG. 7 will be described below. First, a processing algorithm suitable for detecting defective pixels according to the present invention will be described with reference to FIG. As shown in the figure, in this pixel defect detection algorithm, the pixel defect determination process is performed in the first determination process (GS1) with the same color pixel, and further different colors (other than the colors used in the first process). The pixel defect determination process is performed as the second determination process (GS2) for the color pixel), and the final determination process (GS3) is performed as the final pixel defect determination from the results of the first determination process and the second determination process. There are features to do.

  FIG. 7 is a diagram illustrating a configuration example of a first defect detection circuit that performs pixel defect determination processing using pixels of the same color shown in STEP 1 of FIG. For example, in the primary color filter as shown in FIG. 4, for green (G), green (G) in the red (R) row is Gr, and green (G) in the blue (B) row is Gb. It is described. In addition, a range of 5 * 5 pixels (hereinafter referred to as a scope) in the horizontal direction and the vertical direction is taken as an example.

  In FIG. 7, the center pixel of the scope is a pixel of interest (a pixel for defect determination) and is represented by coordinates (0, 0). The coordinates are assigned based on the target pixel, with the left side being negative in the horizontal direction, the right side being positive, the upper side being negative in the vertical direction, and the lower side being positive. The target pixel moves according to the scanning of the image sensor. The comparison target pixels are eight surrounding pixels of the same color as the target pixel. That is, coordinates (-2, -2), (0, -2), (+ 2, -2), (-2,0), (+2,0), (-2, + 2), (0, +2) and (+2, +2) are the target pixels. Here, as shown in FIG. 7, a case where the target pixel is red (R) will be described as an example.

The signal of the target pixel, the output of the pixel to be compared, and the difference threshold 709 are respectively added to the comparators 701 to 708. The difference threshold 709 is a threshold for distinguishing white defects, black defects, and normal pixels, and is set in advance for each color.
The comparators 701 to 708 respectively compare (A) | the target pixel-comparison target pixel | and the difference threshold 701, and (B) the target pixel-comparison target pixel sign determination result. Output two results. The symbol || is an absolute value calculation. In the comparators 701 to 708, true (1) is obtained when | target pixel−comparison target pixel | ≧ difference threshold 701, and false (0) is otherwise obtained. Output. Also, true (1) is output when the calculation result of the target pixel-comparison target pixel is positive or 0, and false (0) is output when it is negative.

  In the comparator 701, the comparison target pixel coordinate (−2, −2), in the comparator 702 the comparison target pixel coordinate (0, −2), in the comparator 703, the comparison target pixel coordinate (+ 2, −2), and the comparator 704 Is a comparison target pixel coordinate (−2, 0), the comparator 705 is a comparison target pixel coordinate (+2, 0), a comparator 706 is a comparison target pixel coordinate (+2, −2), and a comparator 707 is a comparison target pixel. The comparison is performed with the coordinates (0, + 2) and the comparator 708 as the comparison target pixel coordinates (+ 2, + 2). The determination results of the comparators 701 to 708 are input to the adder 710 and the sign match determination 713, respectively. The adder 710 counts the number of pixels to be compared in which | target pixel−comparison target pixel | ≧ difference threshold value 701. The count result of the adder 710 and the threshold value 711 of the number are input to the comparator 712. The comparator 712 is true (1) if the count result of the adder 710 is equal to or greater than the threshold value 711 of the number. Outputs false (0). On the other hand, in the code coincidence determination 713, true (1) is output when the signs of the target pixel and the comparison target pixel all coincide, and false (0) is output when they do not coincide. The result of the comparator 712 and the result of the sign match determination 713 are ANDed by AND 714. If the result of the comparator 712 and the result of the sign match determination 713 are both true (1), true (1), otherwise Outputs false (0) as the determination result. As a result of the above, the determination result is obtained when | the pixel of interest−comparison target pixel | ≧ the threshold value 701 of the difference is equal to or greater than the threshold value 711 of the number and all the codes of the pixel of interest−comparison target pixel match. True (1) = “defective”, otherwise false (0) = “defective”.

  In addition, when the target pixel is two pixels below, above, below, left, and right of the effective pixel range of the image sensor, a process of invalidating the comparators (some of 701 to 708) where there is no comparison target pixel is It may be added to the code match determination 713. In this case, the addition result may be corrected by the adder 710 at a ratio corresponding to the number of invalid pixels, or the number threshold 711 may be corrected for determination. Also, the setting value of the difference threshold 709 may change for blue (B) and green (Gr, Gb), but the operation is the same as that for red (R). Further, in the present embodiment, the pixels having a distance of 2 and 2√2 are described using the same difference threshold 709, but different thresholds may be used for the pixels having a distance of 2 and 2√2. .

  In this embodiment, since comparison and determination are performed using only the same color, for example, it is possible to provide an imaging apparatus that can detect pixel defects with higher accuracy even when the white balance is shifted.

  In addition, although it demonstrated using the primary color filter, it is clear that the same effect can be acquired also in an image pick-up element of a complementary color filter and another color filter arrangement | positioning. In addition, although the scope has a range of 5 * 5 pixels in the horizontal and vertical directions, it is clear that the same effect can be obtained with scopes of other sizes.

  As a fourth embodiment of the imaging apparatus according to the present invention, in addition to the determination result by the first defect detection circuit shown in FIG. 7, the second defect detection circuit shown in FIG. 11 and the third defect shown in FIG. An example using a detection circuit will be described below.

  In the detection in the third embodiment, since the defect detection is performed using the comparison target pixel having a distance of 2 and 2√2 with respect to the target pixel, for example, an RGB Bayer pixel array is used as the input image signal. As shown in FIG. 8, there is a problem that it is impossible to distinguish between an example of a white defect (solid line) and an example of a subject having a high frequency component (dotted line). (Conversely, the same applies to the case of a black defect.) For this reason, in this embodiment, in addition to the determination result of the embodiment 3, the green color between the target pixel and the comparison target pixel of the embodiment 3 (G: By performing defect determination using Gr or Gb), pixel defect detection with higher accuracy is obtained.

  For example, as shown in FIG. 9, in addition to the determination result of the first defect detection circuit 901 according to the third embodiment, a second defect detection circuit 902 (FIG. 11) that performs the processing of STEP 3 shown in FIG. The determination result of the third defect detection circuit 903 (FIG. 12) that performs the processing of STEP4 shown in FIG. 10 is used to determine between the target pixel and the comparison target pixel in the third embodiment. In the defect determination 904 in which the processing of STEP 5 and STEP 6 shown in FIG. 10 is performed, color information for the pixel of interest is input, and when it is green (G: Gr or Gb), the first defect detection circuit 901 and the second defect detection circuit 901 The AND of the defect detection circuit 902 is output as a determination result. When the target pixel is red (R) or blue (B), the AND of the first defect detection circuit 901 and the third defect detection circuit 903 is determined. Output as a result. According to this pixel defect determination method, it is possible to distinguish between an example of a white defect (solid line) shown in FIG. 8 and an example of a subject with a high frequency component (dotted line), and erroneously determines a subject with a high frequency component as a defect. Can prevent the negative effects.

  Also, for example, as shown in FIG. 10, when the determination result of the first defect detection circuit is “no defect” in STEP 1, the determination result is ended as “no defect” in STEP 6, and in the case of “defect” If the target pixel is green (G: Gr or Gb), the process branches to STEP3, and if the target pixel is red (R) or blue (B), the process branches to STEP4. In STEP 3, if the determination result of the second defect detection circuit is “defect”, the determination result is ended in STEP 5 = defect exists, and if “defect is not present”, the determination result is ended in STEP 6 = defect is not present. In STEP4, if the determination result of the third defect detection circuit is “defect”, the determination result is ended in STEP5 = defect exists, and if it is “no defect”, the determination result is ended in STEP6 = defect is not present. The same result as 9 is obtained.

  Next, the operation of the second defect detection circuit 902 will be described with reference to FIG. 11, and the operation of the third defect detection circuit 903 will be described with reference to FIG.

  FIG. 11 shows a configuration example of the second defect detection circuit. In the third embodiment, the scope is equivalent to a scope of 3 * 3 pixels in the horizontal direction and the vertical direction, and the comparison target pixel is (− 1, -1), (+ 1, -1), (-1, + 1), (+ 1, + 1). The distance between the target pixel and the comparison target pixel is √2. If the target pixel is Gr, the comparison target pixel is Gb. Conversely, if the target pixel is Gb, the comparison target pixel is Gr.

  In the comparator 1101, comparison target pixel coordinates (-1, -1). In the comparator 1102, comparison target pixel coordinates (+1, -1). In the comparator 1103, comparison target pixel coordinates (-1, + 1). The comparator 1104 performs a comparison operation using the pixel coordinates for comparison (+ 1, + 1). The difference threshold 1105, the adder 1106, the number threshold 1107, the comparator 1108, the sign match determination 1109, and the AND 1110 are the difference threshold 709, the adder 710, the number threshold 711, and the comparator 712, respectively, in the third embodiment. Since it operates in the same manner as the code coincidence determination 713 and the AND 714, the description thereof is omitted. Since the color is fixed to green (G: Gr or Gb) at the difference threshold 1105, input of color information at the difference threshold 709 is omitted. Further, the numbers of inputs of the comparator 1108 and the code match determination 1109 are different, and the number threshold 1107 is approximately half the value of the number threshold 711.

  As described above, the determination result in the configuration example of the second defect detection circuit is as follows: | target pixel-comparison target pixel | ≧ difference threshold value 1105 The number of pixels is equal to or greater than the threshold value 1107 and the target pixel-comparison The judgment result is true (1) when all the signs of the target pixels match, and false (0) otherwise. According to this pixel defect determination method, since Gr and Gb are color components having the same or very similar spectral sensitivity, it is possible to compare only by the difference between pixels by regarding them as the same color. Therefore, since it can be realized with a simple circuit, low cost and low power consumption can be realized. Further, since the distribution of high-frequency signals is compared using pixels closer to the target pixel, the white defect example (solid line) shown in FIG. 8 is distinguished from the subject with a high frequency component (dotted line). It is possible to prevent the adverse effect of erroneously determining a subject having a high frequency component as a defect.

  FIG. 12 is a configuration example of the third defect detection circuit. For example, the coordinates (0, -1), (-1,0), (+1,0), (0, + 1), which are separated from the target pixel by the distance = 1 as the comparison target pixel as the first group, Coordinates (-1, -2), (+1, -2), (-2, -1), (+2, -1) (-2, +2) away from the pixel of interest by distance = √5 as two groups It is divided into two groups: 1), (+2, +1), (-1, +2), (+1, +2). The target pixel is red (R) or blue (B), and the first group and the second group are green (G: Gr or Gb). The average value calculation circuit 1201 calculates and outputs the average of the four pixels in the first group. The average value calculation circuit 1202 calculates and outputs the average of the 8 pixels in the second group.

  The comparator 1204 makes a determination of | average value of the first group−average value of the second group | ≦ difference threshold 1203, and | average value of the first group−average value of the second group | ≦ difference threshold 1203. In this case, true (1) = “defective” is output, and if not, false (0) = “defective” is output as the determination result. According to this pixel defect determination method, a comparator is used to compare after obtaining an average value of the first group and the second group among the G signals distributed around the R or B pixel as the target pixel. It can be configured by one, and low cost and low power consumption can be realized. In addition, since the distribution of high-frequency signals is compared using G pixels distributed around the pixel of interest, the white defect example (solid line) shown in FIG. 8 and the example of a subject with a high frequency component (dotted line) are compared. This makes it possible to discriminate and prevent the adverse effect of erroneously determining a subject having a high frequency component as a defect. Although the average value is obtained and compared for the first group and the second group, it is obvious that the same effect can be obtained even by the method of comparing the signals after weighted addition. In that case, a divider can be omitted from the method for obtaining the average value.

  In the second defect detection circuit 902 and the third defect detection circuit 903 described above, for example, when the difference threshold value 1105 and the difference threshold value 1203 are set as shown in FIG. 8, the white defect example (solid line) and frequency in FIG. It can be distinguished from an example of a subject with a high component (dotted line).

  In the present embodiment, as in the third embodiment, since only the same color is used for comparison and determination, for example, even when the white balance is shifted, it is possible to detect a pixel defect with higher accuracy. To provide an image pickup apparatus that can correctly detect a pixel defect even in a subject having a high frequency component because a defect is determined using an output of a pixel that is close to the target pixel. Can do.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  In addition, each of the above-described configurations may be configured such that some or all of them are configured by hardware, or are implemented by executing a program by a processor. Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

DESCRIPTION OF SYMBOLS 101 Image pick-up lens 102 Image pick-up element 103 A / D conversion circuit 104 Defect detection / correction circuit 105 Signal processing circuit 106 Timing generator 201 Comparator 202 Data latch 203 Threshold selection circuit 303 Threshold selection circuit 701-708 Comparator 709 Difference threshold 710 Addition 711 Number threshold 712 Comparator 713 Sign match determination 714 AND
901 First defect detection circuit 902 Second defect detection circuit 903 Third defect detection circuit 904 Defect determination 1101 to 1104 Comparator 1105 Difference threshold 1106 Adder 1107 Number threshold 1108 Comparator 1109 Sign match determination 1110 AND
1201, 1202 Average value calculation circuit 1203 Difference threshold 1204 Comparator

Claims (5)

Imaging means for imaging a subject;
Defect determination means for performing defect determination of a pixel of interest using an image signal from the imaging means;
Have
The defect determination means includes
First determination processing means for comparing the signal level of the target pixel with the signal level of the comparison target pixel that is the same color as the target pixel and is around the target pixel;
Second determination processing means for comparing the signal level of the target pixel with a signal level of a comparison target pixel around the target pixel in a color different from that of the target pixel;
Third determination processing means for determining whether or not there is a pixel defect using the defect determination result by the first determination processing means and the defect determination result by the second determination processing means;
An imaging device comprising: The imaging apparatus according to claim 1,
The imaging apparatus according to claim 1, wherein the first determination processing unit determines a defect using a different threshold set in advance for each color of the target pixel. The imaging apparatus according to claim 2,
The image signal is a primary color Bayer pixel array;
The second determination processing means includes
If the pixel of interest is Gr, the surrounding pixel of different color is Gb and the signal level is compared to determine the pixel defect. If the pixel of interest is Gb, the surrounding pixel of different color is Gr and the signal level is compared. To determine pixel defects
An imaging device characterized by the above. The imaging apparatus according to claim 2,
The image signal is a primary color Bayer pixel array;
When the target pixel is R or B, the second determination processing unit calculates an average value or a weighted addition value of the signal levels of the first group pixels of Gr and Gb adjacent to the target pixel, and An average value or a weighted addition value of signal levels of pixels of the second group of Gr and Gb that are not adjacent to each other is calculated, and an average value or weighted addition value of the signal levels of the pixels of the first group and the average value of the second group Determining pixel defects by comparing values or weighted sums;
An imaging device characterized by the above. In the imaging device in any one of Claims 1 thru | or 4,
The third determination processing means determines a pixel defect when the results of the first determination processing means and the second determination processing means are both determined to be pixel defects, and the third determination processing means and the first determination processing means If it is determined that at least one of the results of the two determination processing means is not a pixel defect, determining that it is not a pixel defect;
An imaging device characterized by the above.

Images