JP4429348B2 - IMAGING DEVICE AND IMAGING DEVICE CONTROL METHOD - Google Patents

Description

  The present invention relates to an imaging apparatus that captures a subject image.

  There are imaging devices such as a digital camera that records and reproduces still images and moving images captured by a solid-state imaging device such as a CCD or CMOS sensor electronically and magnetically on various media.

  In order to reduce the price and improve the performance of these image pickup devices, an image pickup device as described in US Pat. No. 5,965,875 focusing on the fact that the light absorption coefficient of a semiconductor substrate used in the image pickup device varies depending on the wavelength of light ( (See FIG. 6).

  This patent describes the principle and pixel circuit of a three-layer photodiode structure imaging device in which a photodiode is formed in a triple well structure. According to this patent, photodiodes are sequentially diffused from the surface of a p-type silicon substrate, and an n-type layer, a p-type layer, and an n-type layer are formed deeply in this order, so that a pn junction diode is formed at the depth of silicon in the same pixel. Three layers are formed in the direction. Light entering the diode from the surface side penetrates deeper as the wavelength is longer. Since the incident wavelength and the absorption coefficient show values specific to silicon, optical signals in different wavelength bands can be detected by separately detecting current from the three-layer diode.

  In the three-layer photodiode, the depth of the pn junction is designed so as to cover the wavelength band of visible light. By calculating the three signals, a signal obtained by color-separating the subject image into three colors of R, G, and B can be obtained.

By using such an image sensor, it is possible to reduce a color filter for color separation of a subject image, an optical low-pass filter for preventing moire due to spatial sampling, and the like. Further, each color obtained by color-separating the subject image has an advantage that the sensitivity centroids coincide with each other and color moiré is difficult to occur.
US Pat. No. 5,965,875

  However, in an image sensor that performs color separation in the depth direction of a semiconductor, if there is noise that is not constant in the depth direction of the semiconductor, such as dark current, even in the same pixel, it differs for each color. Noise and deteriorates the image.

  In the imaging device as described above, two photodiodes adjacent in the vertical direction among the three photodiodes corresponding to the respective colors of the same pixel are capacitively coupled to each other through a pn junction. Further, the capacitance of the photodiode changes as it is accumulated in the charge photodiode generated by the photoelectric conversion. Therefore, the potential of a certain layer photodiode is also affected by the amount of charge accumulated in the photodiodes of other layers. Further, when the uppermost photodiode is saturated, the excess electrons in the diode overcome the potential barrier formed by the second p-type layer from the top and flow into the n-type region of the lowermost photodiode. To do. Therefore, when there are white flaws or the like due to local crystal defects in the semiconductor, the color signal without a defect is also affected and the image is deteriorated.

  An object of the present invention is to provide a solid-state imaging device and a control method thereof that can reduce image quality degradation due to dark current noise, white scratches, and the like of a solid-state imaging device that performs color separation in the depth direction of a semiconductor.

  In order to solve the above-described problems, the present invention relates to an imaging region in which a plurality of pixels each having a plurality of photoelectric conversion units are arranged in the depth direction of a semiconductor, detection means for detecting defective pixels in the imaging region, and the defect When performing pixel correction, the pixel has correction means for correcting each of the signals from the plurality of photoelectric conversion units in the depth direction in the same pixel, and the detection means is one photoelectric sensor in the same pixel. An image pickup characterized in that, when it is detected that correction is required for a signal from a conversion unit, the correction unit also corrects a signal from another photoelectric conversion unit in the same pixel. A device is provided.

  Similarly, in order to solve the above-described problem, the present invention relates to a detection step of detecting a defective pixel in an imaging region in which a plurality of pixels having a plurality of photoelectric conversion units are arranged in a semiconductor depth direction, and correction of the defective pixel A correction step of correcting each of the signals from the plurality of photoelectric conversion units in the depth direction in the same pixel, and from the one photoelectric conversion unit in the same pixel in the detection step When it is detected that the signal needs to be corrected, in the correction step, the signal from other photoelectric conversion units in the same pixel is also corrected. A method is provided.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to obtain a favorable image in the imaging device which has the imaging area which arranged the pixel which has several photoelectric conversion parts in the depth direction of the semiconductor.

(First embodiment)
FIG. 1 shows the configuration of an imaging apparatus according to the first embodiment of the present invention.

  In FIG. 1, 1 is a photographing optical system, 2 is a shutter, 3 is an image sensor, 4 is an amplifier circuit, 5 is an A / D converter, 6 is a front signal processing unit, 7 is a color separation switch, 8 is a changeover switch, 9 Is an image memory, 10 is a subtractor, 11 is a flaw data memory, 12 is a flaw correction processing unit, 13 is a post signal processing unit, 14 is a recording unit, and 15 is a display unit.

  A first embodiment of the present invention will be described with reference to FIG.

  The image pickup device 3 is an image pickup device having an image pickup region in which a plurality of pixels having photodiodes as a plurality of photoelectric conversion units are arranged in the depth direction of a semiconductor as shown in FIG. The flaw data memory 11 stores flaw data (positions of defective pixels) of the image sensor 3 detected in advance. For example, the scratch data is obtained by detecting the position of a pixel that exceeds a certain threshold from an image captured in a light-shielded state.

  The subject image formed by the photographing optical system 1 is converted into an electric signal by the image pickup device 3 via the shutter 2 that is controlled and driven so as to obtain an appropriate exposure amount. In the amplifier circuit 4, the signal is amplified or level-shifted so as to be optimal for the input range of the A / D converter 5. The analog signal is digitized by the A / D converter 5. The previous signal processing unit 6 performs black level adjustment such as clamping on the digitized image data. At this time, each of the three change-over switches 8 is connected to the image memory side. The color separation switch 7 is driven to store the image signal output from the image sensor 3 in an image memory corresponding to each color. That is, the signal from the photodiode at the shallowest position in the depth direction is sent to the image memory R, and the signal from the photodiode at the shallowest position in the depth direction is sent to the image memory G in the depth direction. A signal from the photodiode at the deepest position is stored in the image memory B. Thus, the image data of the subject image subjected to color separation is stored in the image memory 9.

  Thereafter, shooting is performed for the same accumulation time as shooting of the subject in a state where the shutter 2 is closed, that is, in a state where the image sensor 3 is shielded from light. At this time, the three change-over switches 8 are connected in the opposite direction to the image memory 9, that is, on the subtractor 10 side. Further, the image memory 9 outputs image data of the subject image at the same position as the shielded image data to the subtracter 10. As a result, the subtracter 10 outputs image data of the subject image in which unevenness and minute scratches due to the dark current of the image sensor are corrected. These corrections are performed independently for each color separated by the image sensor 3. That is, the dark current signal from the photodiode at the shallowest position in the depth direction is subjected to differential processing with the signal stored in the image memory R, and the dark current signal from the photodiode at the shallowest position in the depth direction. The current signal is subjected to differential processing with the signal stored in the image memory G, and the dark current signal from the photodiode at the deepest position in the depth direction is subjected to differential processing with the signal stored in the image memory B. .

  As described above, in the imaging device of the present embodiment, the imaging region in which a plurality of pixels having at least the first, second, and third photoelectric conversion units are arranged in the semiconductor depth direction, and the imaging region is shielded from light The difference processing between the signal obtained by the first photoelectric conversion unit and the signal obtained by the first photoelectric conversion unit by capturing a subject image in the imaging region, and the first region while the imaging region is shielded from light. Difference processing between a signal obtained by the second photoelectric conversion unit and a signal obtained by the second photoelectric conversion unit by capturing a subject image in the imaging region, and the third photoelectric conversion in a state where the imaging region is shielded from light. Difference processing between the signal obtained by the conversion unit and the signal obtained by the third photoelectric conversion unit by capturing a subject image in the imaging region is performed. 9 is an image memory and 10 is a first subtractor composed of a subtractor. Each of which has a correction means. It can be corrected, even if there is a different noise.

  The scratch correction processing unit 12 performs scratch correction processing based on the scratch data stored in the scratch data memory 11 in advance. The defect-corrected image data is subjected to color processing and the like by the post-signal processing unit 11 and then sent to the recording unit 12 and the display unit 13 for recording or display.

  The scratch correction process will be described with reference to FIGS. 2, 3, and 4. FIG. 2 shows the vicinity of the m-th column and the n-th row of the R color of the image signal of the subject image that has been color-separated and stored in the image memory 9. In this example, the R color of the pixel in the m-th column and the n-th row is detected as a scratch in advance and stored in the scratch data memory 11. In the flaw correction processing unit 12, the flaw data memory 11 reads flaw position information, and the image data at the flaw position is replaced with the same color image data of the surrounding pixels and interpolated. In the case of FIG. 2, since the R color of the pixel in the m-th column and the n-th row is scratched, for example, the R color image data Ra of the pixel in the (m−1) -th row and the n-th row and the R color of the pixel in the m + 1-th row and the n-th row The image data Rc is used, and the average value is set as the R color image data of the pixel in the m-th column and the n-th row.

  3 shows the vicinity of the m-th column and the n-th row of the image signal of the subject image of the G color in FIG. 4B. The pixels Gb and Bb in the m-th column and the n-th row in FIGS. 3 and 4 are not detected as scratches in advance. However, in the present invention, when other colors of the same pixel are flawed as shown in FIG. 2, flaw correction is similarly performed on image data of all colors. Accordingly, the G color image data Gb is replaced with an average value of Ga and Gc. Similarly, the B-color image data Bb is replaced with an average value of Ba and Bc. In this way, when an image signal of any one color of the same pixel is detected as a scratch in advance, a similar defect correction process is performed on the image signal of all the colors, so that there is no defect in the structure. It is possible to prevent image deterioration caused by affecting the color signal.

  As described above, in the imaging apparatus of this embodiment, when correcting an imaging region in which a plurality of pixels having a plurality of photoelectric conversion units are arranged in the depth direction of a semiconductor and a defective pixel, the depth within the same pixel is corrected. Since the scratch data memory 11 for correcting each of the signals from the plurality of photoelectric conversion units in the vertical direction and the second correction means including the scratch correction processing unit are provided, it is possible to surely perform the scratch correction.

  In the embodiment described above, the preceding stage of the image memory 9 is configured to send signals in time series, but in parallel to the image memory R, the image memory G, and the image memory B from the image sensor. 4 may be configured to provide an amplifier circuit, 5 an A / D converter, and 6 a pre-signal processing unit in parallel. In this case, the image sensor has a configuration in which signals from three photoelectric conversion units in the depth direction of the semiconductor of the same pixel are read in parallel.

(Second Embodiment)
FIG. 5 shows a configuration of an imaging apparatus according to the second embodiment of the present invention. In FIG. 5, the same components as those in FIG. Reference numeral 16 denotes a flaw detection circuit, and reference numeral 17 denotes a flaw data mixing circuit.

  A second embodiment of the present invention will be described with reference to FIG. The image sensor 3 is an image sensor having a three-layer photodiode structure shown in the conventional example. The flaw detection circuit 16 detects a flaw caused by a dark current from image data when the shutter 2 is closed, that is, in a state where the image sensor 3 is shielded from light, when the light is shielded. In this method, for example, if the ratio or difference between the image data value of the detection target pixel and the average value of the image data of the surrounding pixels is equal to or greater than a specified value, it is determined as a scratch and is output as scratch data.

  Scratch data detected for each color is synthesized by the scratch data mixing circuit into data in which the pixel position is scratch even if any one color of the same pixel is scratch.

  The subject image formed by the photographing optical system 1 is converted into an electrical signal by the image sensor 3 via the shutter 2 that is controlled and driven so as to have an appropriate exposure amount. In the amplifier circuit 4, the signal is amplified or level-shifted so as to be optimal for the input range of the A / D converter 5. The analog signal is digitized by the A / D converter 5. The front signal processing unit 5 performs black level adjustment such as clamping on the digitized image data. At this time, each of the three selector switches 8 is connected to the image memory 9 side. The color separation switch 7 is driven to store the image signal output from the image sensor 3 in the image memory 9 corresponding to each color. Thus, the image data of the subject image subjected to color separation is stored in the image memory 9. Thereafter, shooting is performed for the same accumulation time as shooting of the subject in a state where the shutter 2 is closed, that is, in a state where the image sensor 3 is shielded from light. At this time, the three change-over switches 8 are connected in the opposite direction to the image memory 9, that is, on the subtractor 10 side. At the same time, the shaded image data is input to the scratch detection circuit 16 for each color. The flaw detection circuit 16 performs the flaw detection described above and outputs flaw data for each color. Further, the image memory 9 outputs image data of the subject image at the same position as the shielded image data to the subtracter 10. As a result, the subtracter 10 outputs image data of the subject image in which unevenness and minute scratches due to the dark current of the image sensor are corrected. In addition, since these corrections are performed independently for each color separated by the image sensor 3, correction can be performed even when there is a different noise in each color. The flaw correction processing unit 12 performs flaw correction processing based on flaw data after the flaw data output from the three flaw detection circuits 16 are combined by the flaw data mixing circuit 17. The defect-corrected image data is subjected to color processing and the like by the post-signal processing unit 11 and then sent to the recording unit 12 and the display unit 13 for recording or display. Since the defect data mixing circuit 17 combines the defect data of all the colors and the correction is performed by the defect correction processing unit 12, the image deterioration due to the influence on the color signal having a defect-free depth due to its structure. Can be prevented. Further, it is not necessary to prepare scratch data in advance, and scratch correction can be performed.

  As described above, the imaging device according to the present embodiment has an imaging region in which a plurality of pixels each having a plurality of photoelectric conversion units are arranged in the depth direction of a semiconductor, and a depth within the same pixel when correcting defective pixels. In order to have an image pickup apparatus having a second correction unit including a flaw detection circuit 16, a flaw data mixing circuit 17, and a flaw correction processing unit 12 for correcting each of the signals from the plurality of photoelectric conversion units in the vertical direction. Scratch correction can be reliably performed.

  In the embodiment described above, the preceding stage of the image memory 9 is configured to send signals in time series, but in parallel to the image memory R, the image memory G, and the image memory B from the image sensor. 4 may be configured to provide an amplifier circuit, 5 an A / D converter, and 6 a pre-signal processing unit in parallel. In this case, the image sensor has a configuration in which signals from three photoelectric conversion units in the depth direction of the semiconductor of the same pixel are read in parallel. In addition, the flaw data is detected after photographing the subject image, but may be performed when the control time of the imaging device is sufficient, for example, after the imaging device is turned on or before photographing the subject.

  In the first and second embodiments described above, the number of color separations of the image sensor that performs color separation in the depth direction of the semiconductor is three colors of R, G, and B. However, the number of colors is two or more. There may be. That is, the imaging device has a configuration having at least first and second photoelectric conversion units in the depth direction of the semiconductor.

  In the first and second embodiments, a circuit having both a circuit for correcting dark current (first correction unit) and a circuit for correcting defective pixels (second correction unit) is shown. It may be configured to have either one of the circuits.

It is a block diagram of the imaging device of the 1st Embodiment of this invention. It is a figure explaining the damage correction of R color. It is a figure explaining the defect correction of G color. It is a figure explaining the defect correction of B color. It is a block diagram of the imaging device of the 2nd Embodiment of this invention. It is a figure explaining the pixel which has a photodiode in the depth direction of a semiconductor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image pick-up optical system 2 Shutter 3 Image pick-up element 4 Amplifying circuit 5 A / D converter 6 Front signal processing part 7 Color separation switch 8 Changeover switch 9 Image signal memory 10 Subtractor 11 Scratch data memory 12 Scratch correction processing part 13 Post signal processing Section 14 Recording section 15 Display section 16 Scratch detection circuit 17 Scratch data mixing circuit

Claims (2)

An imaging region in which a plurality of pixels having a plurality of photoelectric conversion units are arranged in the depth direction of the semiconductor;
Detecting means for detecting defective pixels in the imaging region;
A correction unit that corrects each of the signals from the plurality of photoelectric conversion units in the depth direction in the same pixel when correcting the defective pixel;
When the detection unit detects that the signal from one photoelectric conversion unit in the same pixel needs to be corrected, the correction unit detects the signal from the other photoelectric conversion unit in the same pixel. Even if it corrects, the imaging device characterized by the above-mentioned. A detection step of detecting defective pixels in an imaging region in which a plurality of pixels having a plurality of photoelectric conversion units are arranged in the depth direction of the semiconductor;
A correction step of correcting each of the signals from the plurality of photoelectric conversion units in the depth direction in the same pixel when correcting the defective pixel;
When it is detected in the detection step that correction is required for a signal from one photoelectric conversion unit in the same pixel, in the correction step, a signal from another photoelectric conversion unit in the same pixel is detected. Even if it corrects, the control method of the imaging device characterized by the above-mentioned.

Images