JP2014049780A - Device and method for determining defective pixel - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique by which, even when a defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel, the defective pixel can be determined as a defective pixel.SOLUTION: A defective pixel determination device 10 includes: a first difference calculation part 1511 calculating a first differential value between a first maximum value and a first minimum value among pixel values of a first pixel group neighboring to a center pixel, in a same color layer with the center pixel; a second difference calculation part 1512 calculating a second differential value between a second maximum value and a second minimum value among pixel values of a second pixel group neighboring to the center pixel, in a same color layer with the center pixel; and a determination part 152 determining whether or not the center pixel is a defective pixel on the basis of the relation between the first maximum value or the first minimum value and the pixel value of the center pixel, and the first and second differential values.

Description

  The present invention relates to a defective pixel determination device and a defective pixel determination method.

  Conventionally, there are a static type and a dynamic type as a defect correction technique of a camera sensor. In the static type, an all-black / all-white image is input at the time of shipment to detect a defective pixel, and the position of the defective pixel is stored in a memory. In use, the position of the defective pixel is read from the memory, and the pixel value of the defective pixel is replaced with a value calculated using the peripheral pixels of the defective pixel.

  With this static type, as the number of pixels increases in recent years, a large-capacity memory is required to store the positions of defective pixels, and a defect detection process is required at the time of shipment. Another problem is that defects that appear due to changes in temperature and aging cannot be dealt with. Against this background, there is an increasing demand for dynamic defect correction.

  As a dynamic defect correction technique, for example, there is a technique described in the following patent document.

JP 2009-290653 A JP 2011-135666 A JP 2008-67158 A

  However, the technique disclosed in Patent Document 1 is a general technique for defect correction, but it is a side effect only by looking at the difference value between the average value of the pixel values of the eight pixels near the pixel and the pixel value of the pixel. In reality, it is necessary to add a condition such that all the pixel values of neighboring pixels are separated from the pixel value of the pixel. In such a case, a defective pixel existing near an edge having a value close to that of the defective pixel cannot be determined as a defect, and the defective pixel remains.

  The method disclosed in Patent Document 2 is a defective pixel determination method different from the method disclosed in Patent Document 1, but when the defect determination condition is the maximum pixel value of the pixel in the setting window. However, the defective pixel existing near the edge having a value close to that of the defective pixel cannot be sufficiently determined as a defect, and the defective pixel may remain.

  The technique disclosed in Patent Document 3 is a defective pixel determination technique different from the technique disclosed in Patent Document 1 and the technique disclosed in Patent Document 2, but if the setting is made to reduce the side effects, In some cases, a defective pixel existing in the vicinity of an edge having a close value cannot be sufficiently determined as a defect, and the defective pixel remains.

  Therefore, the present invention makes it possible to determine that a defective pixel is a defective pixel even when the defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel. Is to provide.

  According to an embodiment of the present invention, the first of the first maximum value and the first minimum value among the pixel values of the first pixel group that is the same color layer as the center pixel and is adjacent to the center pixel. A first difference calculation unit that calculates a difference value; a second maximum value and a second minimum value among pixel values of a second pixel group that is the same color layer as the center pixel and is adjacent to the center pixel; A second difference calculation unit for calculating the second difference value, a relationship between the first maximum value or the first minimum value and the pixel value of the central pixel, the first difference value, and the first difference value. And a determination unit that determines whether or not the central pixel is a defective pixel based on the difference value of 2. A defective pixel determination device is provided.

  According to such a configuration, even when a defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel, it is possible to determine that the defective pixel is a defective pixel. In particular, it is possible to determine whether or not the center pixel is a defective pixel even when there is an edge close to the center pixel in the horizontal or vertical direction.

  In the determination unit, the relationship between the first maximum value or the first minimum value and the pixel value of the central pixel satisfies a first condition, and the first difference value satisfies a second condition. If the second difference value satisfies the third condition, the center pixel may be determined to be a defective pixel. According to this configuration, the determination unit can determine that the center pixel is a defective pixel when the first condition, the second condition, and the third condition are satisfied.

  When the pixel value of the central pixel is larger than the first maximum value, the first condition is that a value obtained by subtracting the first maximum value from the pixel value of the central pixel is less than a first threshold value. It may be a condition that is large. According to this configuration, when the pixel value of the center pixel is larger than the first maximum value, the determination unit obtains a value obtained by subtracting the first maximum value from the pixel value of the center pixel. Whether or not the first condition is satisfied can be determined based on whether or not the condition that the threshold value is greater than 1 is satisfied.

  When the pixel value of the central pixel is smaller than the first minimum value, the first condition is that a value obtained by subtracting the pixel value of the central pixel from the first minimum value is less than a first threshold value. It may be a condition that is large. According to this configuration, when the pixel value of the center pixel is smaller than the first minimum value, the determination unit obtains a value obtained by subtracting the pixel value of the center pixel from the first minimum value. Whether or not the first condition is satisfied can be determined based on whether or not the condition that the threshold value is greater than 1 is satisfied.

  The second condition may be a condition that the first difference value is smaller than a second threshold value. According to this configuration, the determination unit can determine whether the second condition is satisfied depending on whether the condition that the first difference value is smaller than the second threshold is satisfied.

  The third condition may be a condition that the second difference value is smaller than a third threshold value. According to such a configuration, the determination unit can determine whether the third condition is satisfied depending on whether the condition that the second difference value is smaller than the third threshold is satisfied.

  Whether or not the central pixel is a defective pixel may be determined for a plurality of patterns obtained by rotating the first pixel group and the second pixel group by a predetermined angle with respect to the central pixel. Good. According to such a configuration, the determination unit can perform defect determination with higher accuracy by performing defect determination on not only one pattern but also on the plurality of patterns.

  According to another embodiment of the present invention, the first maximum value and the first minimum value among the pixel values of the first pixel group adjacent to the center pixel and having the same color layer as that of the center pixel. A step of calculating a first difference value; and a second difference between a second maximum value and a second minimum value among pixel values of a second pixel group adjacent to the center pixel and having the same color layer as the center pixel 2 based on the step of calculating the difference value of 2, the relationship between the first maximum value or the first minimum value and the pixel value of the central pixel, and the first difference value and the second difference value. And determining whether or not the central pixel is a defective pixel.

  According to such a method, even when a defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel, it is possible to determine that the defective pixel is a defective pixel. In particular, it is possible to determine whether or not the center pixel is a defective pixel even when there is an edge close to the center pixel in the horizontal or vertical direction.

  As described above, according to the present invention, even when a defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel, the defective pixel is determined to be a defective pixel. It is possible to provide a technology that makes it possible.

It is a figure which shows the structural example of the four primary color sensor which can be applied to embodiment of this invention. It is a figure which shows the subject which embodiment of this invention tends to solve. It is a figure which shows the function structure of the defective pixel determination apparatus which concerns on embodiment of this invention. It is a figure which shows the detailed structure of a 1st edge vicinity defect determination part. It is a figure which shows the detail of the function which a 1st edge vicinity defect determination part has. It is a figure which shows the detailed structure of a 2nd edge vicinity defect determination part. It is a figure which shows the detail of the function which a 2nd edge vicinity defect determination part has. It is a figure which shows an example of the relationship between the kind of defect determination, and a correction value.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In the present specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

  In the present specification and drawings, a plurality of constituent elements having substantially the same functional configuration may be distinguished by attaching different alphabets after the same reference numeral. However, when it is not necessary to particularly distinguish each of a plurality of constituent elements having substantially the same functional configuration, only the same reference numerals are given.

  First, the types of sensors that can be applied to the embodiment of the present invention will be described. FIG. 1 is a diagram illustrating a configuration example of a four-primary-color sensor 110A that can be applied to the embodiment of the present invention. In the description using FIG. 1, the four primary color sensor 110A will be described as an example of a sensor that can be applied to the embodiment of the present invention. However, as will be described later, the sensor that can be applied to the embodiment of the present invention is It is not limited to the four primary color sensor 110A.

  As shown in FIG. 1, the four primary color sensor 110A is a sensor in which four color filters indicated by A, B, C, and D are arranged in units of 2 × 2 pixels. However, the configuration of the four color filters is not particularly limited as long as each of them detects light in different wavelength regions. For example, the four color filters may have a configuration in which white pixels are added to three primary color pixels of R, G, and B, a configuration in which infrared light pixels are added to the three primary color pixels, or a visible light region. May be capable of detecting light of each wavelength region obtained by equally dividing.

  The types of sensors that can be applied to the embodiment of the present invention have been described above. Subsequently, problems to be solved by the embodiments of the present invention will be described. FIG. 2 is a diagram illustrating a problem to be solved by the embodiment of the present invention. Here, the conventional dynamic defect determination method is based on the condition that the central pixel is isolated in the same color layer 3 × 3. The same color layer corresponds to, for example, the same type of color filter in the four primary color sensor 110A illustrated in FIG. Therefore, if even one pixel in the 3 × 3 pixel group has the same or very close value as the central pixel, the central pixel is not determined to be defective.

  That is, according to the conventional dynamic defect determination method, most defective pixels can be determined and removed centering on the flat portion, but as shown in FIG. 2, the defect correction is performed from the image M1 to the image M2. Then, when an edge having the same level as that of the defective pixel is in the vicinity, the defective pixel (for example, defective pixels R1, R2, etc. shown in FIG. 2) is not determined to be a defective pixel. There was also a problem that it remained without being removed.

  For example, in the example shown in FIG. 2, three pixels existing on the right side of the defective pixel R1 in the same color layer as the defective pixel R1 (more specifically, 1 existing in the upper right side of the defective pixel R1 in the same color layer as the defective pixel R1). The pixel, the one pixel present on the right, and the one pixel present on the lower right have the same pixel value as the defective pixel R1, and the defective pixel R1 is not determined to be a defective pixel. In addition, the upper left pixel of the defective pixel R2 in the same color layer as the defective pixel R2 has the same pixel value as the defective pixel R2, and the defective pixel R2 is not determined to be a defective pixel.

  Therefore, in the embodiment of the present invention, even when a defective pixel exists in the vicinity of an edge having a pixel value close to the pixel value of the defective pixel, the defective pixel is determined with high accuracy as a defective pixel. A technique that makes it possible to do this will be described.

  Subsequently, a functional configuration of the defective pixel determination device 10 according to the embodiment of the present invention will be described. FIG. 3 is a diagram showing a functional configuration of the defective pixel determination device 10 according to the embodiment of the present invention. As shown in FIG. 3, the defective pixel determination apparatus 10 includes a sensor 110, a line memory 120, a correction value calculation unit 130, a dynamic defect determination unit 140, a first edge vicinity defect determination unit 150, and a second edge vicinity defect determination. Unit 160 and selector 170. Hereinafter, the function of each functional block included in the defective pixel determination device 10 will be sequentially described in detail.

  The sensor 110 is configured by an image sensor that forms an image of light from the outside on a light receiving plane of the image sensor, photoelectrically converts the imaged light into a charge amount, and converts the charge amount into an electric signal. The type of the image sensor is not particularly limited, and may be, for example, a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). In the example shown in FIG. 3, the sensor 110 exists inside the defective pixel determination device 10, but the sensor 110 may exist outside the defective pixel determination device 10.

  Here, in the embodiment of the present invention, a four-primary-color sensor 110A in which only one pixel of each color exists in 2 × 2 pixels is adopted as a configuration example of the sensor 110. The configuration of the sensor 110 will be described later. It is not limited to such an example. The RAW signal from the sensor 110 is accumulated in the line memory 120 and read from the correction value calculation unit 130, the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, the second edge vicinity defect determination unit 160, and the selector 170. It is. For example, the RAW signal is read out for every 5 lines as an image signal. In the example shown in FIG. 3, the line memory 120 exists inside the defective pixel determination device 10, but the line memory 120 may exist outside the defective pixel determination device 10.

  The correction value calculation unit 130 calculates a corrected pixel value (hereinafter, also referred to as “correction value”) when the pixel value of the defective pixel is corrected. Here, for example, in the example shown in FIG. 3, the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 exist as functional blocks for performing defect determination. Yes. As described above, when there are a plurality of functional blocks that perform defect determination in the defective pixel determination apparatus 10, the correction value calculation unit 130 calculates correction values corresponding to the respective defect determinations, and calculates the correction values. It may be supplied to the selector 170. A specific example of the correction value will be described later with reference to FIG. In addition, when it is not necessary to correct the defective pixel, the defective pixel determination device 10 does not need to include the correction value calculation unit 130.

  The dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 sequentially select a central pixel from a plurality of pixels constituting an image signal supplied from the line memory 120. Then, it is determined whether or not the central pixel is a defective pixel. The dynamic defect determination unit 140 determines whether or not the center pixel is a defective pixel using a defect determination condition based on conventional dynamic defect determination. Note that the defective pixel determination device 10 may not include the dynamic defect determination unit 140.

  Here, as described above, according to the conventional dynamic defect determination method, when an edge having the same level as the defective pixel exists in the vicinity, the defective pixel is not determined to be a defective pixel. On the other hand, the first edge vicinity defect determination unit 150 and the second edge vicinity defect determination unit 160 determine that the defective pixel is a defective pixel even when the edge at the same level as the defective pixel is in the vicinity. can do.

  Specifically, the first edge vicinity defect determination unit 150 determines whether or not the center pixel is a defective pixel even when an edge close to the center pixel in the horizontal direction or the vertical direction exists. Can do. Further, the second edge vicinity defect determination unit 160 can determine whether or not the center pixel is a defective pixel even when there is an edge near the center pixel in an oblique direction.

  In the example illustrated in FIG. 3, the defective pixel determination apparatus 10 includes both the first edge vicinity defect determination unit 150 and the second edge vicinity defect determination unit 160, but the first edge vicinity defect determination is performed. It is sufficient that at least one of the part 150 and the second edge vicinity defect determination part 160 is provided. Details of the functions of the first edge vicinity defect determination unit 150 and the second edge vicinity defect determination unit 160 will be described later.

  The selector 170 calculates each of the correction values calculated by the correction value calculation unit 130 based on the determination results supplied from the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160. One of the correction values is selected from the correction values. FIG. 8 shows the relationship between the determination results supplied from the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 and the correction value selected by the selector 170. This will be described later with reference.

  For example, the correction value selected by the selector 170 may be output to another device that exists outside the defective pixel determination device 10, and the defective pixel may be corrected to the correction value in the other device. In addition, when it is not necessary to select a correction value (for example, when the dynamic defect determination unit 140 does not exist and only the second edge vicinity defect determination unit 160 exists), the defective pixel determination device 10 May not include the selector 170.

  In addition, as shown in FIG. 3, the correction value calculation unit 130, the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 are provided in parallel, so that the line memory It is not necessary to provide 120 overlappingly, and the memory size of the line memory 120 can be kept small.

  The functional configuration of the defective pixel determination device 10 according to the embodiment of the present invention has been described above.

  Next, details of the first edge vicinity defect determination unit 150 will be described. FIG. 4 is a diagram illustrating a detailed configuration of the first edge vicinity defect determination unit 150. FIG. 5 is a diagram illustrating details of functions of the first edge vicinity defect determination unit 150. As illustrated in FIG. 4, the first edge vicinity defect determination unit 150 includes a first difference calculation unit 1511, a second difference calculation unit 1512, and a determination unit 152.

  In the following description, as shown in FIG. 5, the central pixel is indicated as the central pixel Pc, and the pixel group that is the same color layer as the central pixel Pc and is adjacent to the central pixel Pc is defined as the first pixel group (pixels P1,. The pixel P2, the pixel P4, the pixel P6, and the pixel P7) are divided into the second pixel group (the pixel P3, the pixel P5, and the pixel P8).

  Further, as shown in (Formula 1) of FIG. 5, the first maximum value among the pixel values of the first pixel group is expressed as Pmax = MAX (P1, P2, P4, P6, P7), and FIG. As shown in (Formula 2) of 5, the first minimum value among the pixel values of the first pixel group is represented as Pmin = MIN (P1, P2, P4, P6, P7). Therefore, the first difference value between the first maximum value and the first minimum value is represented as Pmax−Pmin.

  Further, as shown in (Condition 3) in FIG. 5, the second maximum value among the pixel values of the second pixel group is denoted as MAX (P3, P5, P8), and the pixels of the second pixel group The second minimum value among the values is indicated as MIN (P3, P5, P8). Further, as shown in (Condition 3) of FIG. 5, the second difference value between the second maximum value and the second minimum value is expressed as MAX (P3, P5, P8) −MIN (P3, P5, P8).

  Here, the first difference calculation unit 1511 calculates Pmax−Pmin. The second difference calculation unit 1512 calculates MAX (P3, P5, P8) −MIN (P3, P5, P8). The determination unit 152 determines that the center pixel Pc is defective based on the relationship between Pmax or Pmin and the pixel value of the center pixel Pc, and Pmax−Pmin and MAX (P3, P5, P8) −MIN (P3, P5, P8). It is determined whether or not it is a pixel.

  According to such a defective pixel determination method, it is possible to determine whether or not the center pixel Pc is a defective pixel even when there is an edge in the horizontal direction or the vertical direction with respect to the center pixel Pc. . The determination method can be applied to a case where the center pixel Pc has a white defect (Hot Pixel), and can also be applied to a case where the center pixel Pc has a black defect (Cold Pixel).

  FIG. 5 shows a case where a vertical edge is present on the right side of the center pixel Pc. However, when a vertical edge is present on the left side of the center pixel Pc, the horizontal direction is located above the center pixel Pc. This method can also be applied to the case where there is an edge and the case where there is a horizontal edge below the center pixel Pc.

  In order to apply this method, the determination as to whether or not the center pixel Pc is a defective pixel is performed based on the first pixel group (pixel P1, pixel P2, pixel P4, pixel P6, and pixel P6) based on the center pixel Pc. This is performed for a plurality of patterns obtained by rotating the pixel P7) and the second pixel group (pixel P3, pixel P5, and pixel P8) by a predetermined angle. For example, when the predetermined angle is 90 degrees, it is only necessary to determine four patterns obtained by rotating 90 degrees. Thus, by applying the method to a plurality of patterns, it becomes possible to perform defect determination with higher accuracy.

  When the determination with respect to the plurality of patterns is made, the determination unit 152 determines that the center pixel Pc is defective in at least one of the plurality of patterns. What is necessary is just to determine with a pixel. In addition, when the determination unit 152 determines that the central pixel Pc is not a defective pixel in all of the plurality of patterns, the determination unit 152 may determine that the central pixel Pc is not a defective pixel.

  When determination is made for a plurality of patterns obtained by rotating the first pixel group and the second pixel group by a predetermined angle with respect to the center pixel Pc, the first pixel group and the second pixel group are at a predetermined angle. After the rotation, the same determination as in the above method may be performed. Alternatively, the first pixel group and the second pixel group itself are not rotated, but are different from the determination in the above method (for example, a pixel after the target pixel itself is rotated by a predetermined angle with the central pixel Pc as a reference) May be made).

  There are no particular restrictions on when the determination unit 152 determines that the central pixel Pc is a defective pixel. For example, the determination unit 152 has a relationship between Pmax or Pmin and the pixel value of the central pixel Pc. If the condition of 1 is satisfied, Pmax−Pmin satisfies the second condition, and MAX (P3, P5, P8) −MIN (P3, P5, P8) satisfies the third condition, the center pixel You may determine with Pc being a defective pixel.

  For example, as shown in (Formula 3) and (Condition 1) in FIG. 5, when the pixel value of the central pixel Pc is larger than Pmax, the first condition is that Pmax is calculated from the pixel value of the central pixel Pc. The condition may be that the reduced value (Pc−Pmax) is larger than the first threshold (TH1). As shown in (Formula 3) and (Condition 1) in FIG. 5, the first condition is that when the pixel value of the central pixel Pc is smaller than Pmin, the pixel value of the central pixel Pc is changed from Pmin. There may be a condition that the reduced value (Pmin−Pc) is larger than the first threshold (TH1).

  For example, as shown in (Condition 2) of FIG. 5, the second condition may be a condition that Pmax−Pmin is smaller than the second threshold (TH2). For example, as shown in (Condition 3) of FIG. 5, the third condition is that MAX (P3, P5, P8) −MIN (P3, P5, P8) is more than the third threshold (TH3). The condition of being small may be sufficient.

  Note that, for example, the determination unit 152 does not satisfy the first condition described above or the relationship between Pmax or Pmin and the pixel value of the central pixel Pc, or does not satisfy the second condition described above Pmax−Pmin. Alternatively, when MAX (P3, P5, P8) -MIN (P3, P5, P8) does not satisfy the third condition, it may be determined that the center pixel Pc is not a defective pixel.

  The details of the first edge vicinity defect determination unit 150 included in the defective pixel determination apparatus 10 according to the embodiment of the present invention have been described above.

  Next, details of the second edge vicinity defect determination unit 160 will be described. FIG. 6 is a diagram illustrating a detailed configuration of the second edge vicinity defect determination unit 160. FIG. 7 is a diagram illustrating details of functions of the second edge vicinity defect determination unit 160. As shown in FIG. 6, the second edge vicinity defect determination unit 160 includes a difference calculation unit 161 and a determination unit 162.

  In the following description, as shown in FIG. 7, the central pixel is indicated as the central pixel Pc, and the pixel of the four corners from the pixel group adjacent to the central pixel Pc is the same color layer as the central pixel Pc (pixel P1). ) Is removed and a pixel group after removal (pixel P2, pixel P3, pixel P4, pixel P5, pixel P6, pixel P7, and pixel P8) is assumed.

  Further, as shown in (Formula 4) of FIG. 7, the maximum value among the pixel values of the pixel group after the removal is expressed as Pmax = MAX (P2, P3, P4, P5, P6, P7, P8), As shown in (Formula 5) of FIG. 7, the minimum value among the pixel values of the pixel group after the removal is represented as Pmin = MIN (P2, P3, P4, P5, P6, P7, P8). Therefore, the difference value between the maximum value and the minimum value is expressed as Pmax−Pmin.

  Here, the difference calculation unit 161 calculates Pmax−Pmin. The determination unit 162 determines whether or not the center pixel Pc is a defective pixel based on the relationship between Pmax or Pmin and the pixel value of the center pixel Pc and Pmax−Pmin.

  According to such a defective pixel determination method, it is possible to determine whether or not the center pixel Pc is a defective pixel even when there is an edge in the oblique direction with respect to the center pixel Pc. The determination method can be applied to a case where the center pixel Pc has a white defect (Hot Pixel), and can also be applied to a case where the center pixel Pc has a black defect (Cold Pixel).

  FIG. 7 shows a case where an oblique edge exists on the upper left side of the center pixel Pc, but when an oblique edge exists on the upper right side of the center pixel Pc, the lower left side of the center pixel Pc. This method can also be applied to the case where there is an oblique edge and the case where there is an oblique edge on the lower right side of the center pixel Pc.

  In order to apply the method to this case, the determination as to whether or not the center pixel Pc is a defective pixel is performed by determining whether or not the pixel group (P2, P3, P4, P5, P6, P7, This is performed for a plurality of patterns obtained by rotating P8) by a predetermined angle. For example, when the predetermined angle is 90 degrees, it is only necessary to determine four patterns obtained by rotating 90 degrees. Thus, by applying the method to a plurality of patterns, it becomes possible to perform defect determination with higher accuracy.

  When the determination with respect to the plurality of patterns is performed, the determination unit 162 determines that the center pixel Pc is a defective pixel in at least one of the plurality of patterns. What is necessary is just to determine with a pixel. In addition, when the determination unit 162 determines that the central pixel Pc is not a defective pixel in all of the plurality of patterns, the determination unit 162 may determine that the central pixel Pc is not a defective pixel.

  When determination is made for a plurality of patterns obtained by rotating the pixel group after removal by a predetermined angle with respect to the center pixel Pc, the determination in the above method is performed after the pixel group after removal is rotated by a predetermined angle. The same determination may be made. Alternatively, the pixel group itself after removal is not rotated, but is different from the determination in the above method (for example, determination after the target pixel itself is rotated by a predetermined angle with the central pixel Pc as a reference) ) May be made.

  There are no particular restrictions on when the determination unit 162 determines that the central pixel Pc is a defective pixel. For example, the determination unit 162 has a relationship between Pmax or Pmin and the pixel value of the central pixel Pc. When the first condition is satisfied and Pmax−Pmin satisfies the second condition, the center pixel Pc may be determined to be a defective pixel.

  For example, as shown in (Formula 3) and (Condition 1) in FIG. 7, when the pixel value of the central pixel Pc is larger than Pmax, the first condition is that Pmax is calculated from the pixel value of the central pixel Pc. The condition may be that the reduced value (Pc−Pmax) is larger than the first threshold (TH1). Further, as shown in (Formula 3) and (Condition 1) in FIG. 7, when the pixel value of the center pixel Pc is smaller than Pmin, the first condition is that the pixel value of the center pixel Pc is changed from Pmin. There may be a condition that the reduced value (Pmin−Pc) is larger than the first threshold (TH1).

  For example, as shown in (Condition 2) of FIG. 7, the second condition may be a condition that Pmax−Pmin is smaller than the second threshold (TH2). The first threshold value (TH1) shown in FIG. 5 and the first threshold value (TH1) shown in FIG. 7 may be the same or different. Further, the second threshold value (TH2) shown in FIG. 5 and the second threshold value (TH2) shown in FIG. 7 may be the same or different.

  For example, the determination unit 162 determines that the relationship between Pmax or Pmin and the pixel value of the central pixel Pc does not satisfy the first condition described above, or Pmax−Pmin does not satisfy the second condition described above. May determine that the center pixel Pc is not a defective pixel.

  The details of the second edge vicinity defect determination unit 160 included in the defective pixel determination apparatus 10 according to the embodiment of the present invention have been described above.

  Here, as described above, the correction value calculation method by the correction value calculation unit 130 is not particularly limited. Hereinafter, an example of a correction value calculation method performed by the correction value calculation unit 130 will be described. FIG. 8 is a diagram illustrating an example of the relationship between the type of defect determination and the correction value. As a matter of course, the correction value calculation method by the correction value calculation unit 130 is not limited to the example shown in FIG.

  As described above, the pixel value of the pixel determined to be a defective pixel is replaced with the correction value calculated by the correction value calculation unit 130, for example. As shown in FIG. 8, the correction value is determined by the functional block of any one of the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 (that is, the center It may be different depending on whether or not the determination that the pixel Pc is a defective pixel is made.

  For example, when the defect determination is performed by the dynamic defect determination unit 140, an average value of four pixels (pixel P2, pixel P4, pixel P5, and pixel P7) of the upper, lower, left, and right of the center pixel Pc may be used as the correction value. .

  In addition, when the defect determination is performed by the first edge vicinity defect determination unit 150, when the edge exists on the left or right of the center pixel Pc, the average of the upper and lower two pixels (pixel P2 and pixel P7) of the center pixel Pc. The value may be used as a correction value. When the edge exists above or below the center pixel Pc, the average value of the two left and right pixels (pixel P4 and pixel P5) of the center pixel Pc is used as the correction value. Also good.

  Defect determination is performed by the first edge vicinity defect determination unit 150, and the edge exists on the left or right of the center pixel Pc when the edge direction is the horizontal direction. On the other hand, the defect is determined by the first edge vicinity defect determination unit 150 and the edge exists above or below the center pixel Pc when the edge direction is the vertical direction. Therefore, a function block for determining whether the edge direction is the horizontal direction or the vertical direction is provided, and the selector 170 is located on the left or right of the center pixel Pc from the edge direction determined by the function block. It may be determined whether it exists or exists above or below.

  In addition, when the defect determination is performed by the second edge vicinity defect determination unit 160, when the edge exists in any of the upper left, upper right, lower left, and lower right of the center pixel Pc, the upper, lower, left, and right four pixels ( The average value of the pixel P2, the pixel P4, the pixel P5, and the pixel P7) may be used as the correction value.

  In the example shown in FIG. 8, the correction value used is any one of three types: an upper / lower two-pixel average, a left / right two-pixel average, and an upper / lower / left-right four-pixel average. In this case, the correction value calculation unit 130 calculates these three types of correction values and supplies them to the selector 170, which selects the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity. Based on the determination result by each defect determination unit 160, a correction value is selected and output to another apparatus. The selector 170 may output the input pixel value as it is when no defect determination is made.

  Since the determination results from the dynamic defect determination unit 140, the first edge vicinity defect determination unit 150, and the second edge vicinity defect determination unit 160 can be simultaneously supplied to the selector 170, the selector 170 detects a plurality of defect determinations. It is possible that Assuming such a case, the order of referring to the determination results may be determined in advance, and the selector 170 may select a correction value corresponding to the defect detection detected first. For example, as shown in FIG. 8, the order of referring to the determination results is as follows: dynamic defect determination, edge vicinity defect determination 1 (determination result by the first edge vicinity defect determination unit 150), and edge vicinity defect determination 2 (first 2 may be in the order of determination result by the edge vicinity defect determination unit 160, but is not particularly limited.

  Heretofore, an example of the correction value calculation method by the correction value calculation unit 130 has been described.

  According to the embodiment of the present invention, it is possible to determine that a defective pixel near the edge having the same level value as the central pixel is also a defective pixel, and perform correction and removal. Such correction and removal could not be removed by conventional dynamic defect correction. As a result of the preliminary experiment, a high defect removal rate of 99% or more can be achieved by combining the conventional dynamic defect correction and the edge vicinity defect correction in the embodiment of the present invention, and important image information is mistaken as a defect. The side effect of damaging the image by determining and correcting it is also very small.

  Since each of the first edge vicinity defect determination unit 150 and the second edge vicinity defect determination unit 160 can be configured only by size comparison, addition / subtraction, and shift operation, multiplication and division are not required. Arithmetic delay and the like are also kept very small, and the increase in circuit scale is negligible. Furthermore, since the first edge vicinity defect determination unit 150 and the second edge vicinity defect determination unit 160 can use signals of predetermined lines (for example, 5 lines) used in the dynamic defect determination unit 140, the line memory There is no need to add any additional. From the above, the embodiment of the present invention has an advantage of easy mounting.

  The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

  For example, the correction value described in the example shown in FIG. 8 does not need to be used as the correction value, and the range of pixels used for calculating the correction value is expanded to a kernel size larger than 5 × 5, The pixel value of a pixel having a different color from the pixel Pc may be used for calculating the correction value. The main point of the embodiment of the present invention is to provide a method for determining a defect at the same level near the edge.

  In the embodiment of the present invention, the case where the four primary color sensors are used has been mainly described. However, the R, B signals of the three primary color Bayer sensors are present in that one pixel of each color exists in 2 × 2 pixels. This is also the same, and there is no technical barrier to applying the embodiment of the present invention to this. However, in the three primary color Bayer sensor, since the demosaicing is performed using the G signal that is present in 2 × 2 pixels, the resolution of the final image is high and the resolution is not affected as a side effect of defect correction. It is good to do. It should also be noted that the defect determination accuracy may be increased by referring to the G signal.

DESCRIPTION OF SYMBOLS 10 Defect pixel determination apparatus 110 Sensor 120 Line memory 130 Correction value calculation part 140 Dynamic defect determination part 150 1st edge vicinity defect determination part 1511 1st difference calculation part 1512 2nd difference calculation part 152 Determination part 160 2nd Edge vicinity defect determination unit 161 Difference calculation unit 162 Determination unit 170 Selector

Claims (8)

A first difference calculation that calculates a first difference value between a first maximum value and a first minimum value among pixel values of a first pixel group that is the same color layer as the center pixel and is adjacent to the center pixel. And
A second difference for calculating a second difference value between a second maximum value and a second minimum value among pixel values of a second pixel group adjacent to the center pixel and having the same color layer as the center pixel. A calculation unit;
The center pixel is a defective pixel based on the relationship between the first maximum value or the first minimum value and the pixel value of the center pixel, and the first difference value and the second difference value. A determination unit for determining whether or not
A defective pixel determination device comprising: In the determination unit, the relationship between the first maximum value or the first minimum value and the pixel value of the central pixel satisfies a first condition, and the first difference value satisfies a second condition. And when the second difference value satisfies the third condition, the central pixel is determined to be a defective pixel.
The defective pixel determination apparatus according to claim 1. When the pixel value of the central pixel is larger than the first maximum value, the first condition is that a value obtained by subtracting the first maximum value from the pixel value of the central pixel is less than a first threshold value. Is also a condition that
The defective pixel determination apparatus according to claim 2. When the pixel value of the central pixel is smaller than the first minimum value, the first condition is that a value obtained by subtracting the pixel value of the central pixel from the first minimum value is less than a first threshold value. Is also a condition that
The defective pixel determination apparatus according to claim 2. The second condition is a condition that the first difference value is smaller than a second threshold value.
The defective pixel determination apparatus according to claim 2. The third condition is a condition that the second difference value is smaller than a third threshold value.
The defective pixel determination apparatus according to claim 2. Whether or not the central pixel is a defective pixel is determined for a plurality of patterns obtained by rotating the first pixel group and the second pixel group by a predetermined angle with respect to the central pixel.
The defective pixel determination apparatus according to claim 1. Calculating a first difference value between a first maximum value and a first minimum value among pixel values of a first pixel group adjacent to the center pixel and having the same color layer as the center pixel;
Calculating a second difference value between a second maximum value and a second minimum value among pixel values of a second pixel group adjacent to the center pixel in the same color layer as the center pixel;
The center pixel is a defective pixel based on the relationship between the first maximum value or the first minimum value and the pixel value of the center pixel, and the first difference value and the second difference value. Determining whether or not,
A defective pixel determination method.

Images