JP4146186B2 - Solid-state imaging device and defect correction method thereof - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
  The present inventionThe solid-state imaging device according to the present invention relates to a solid-state imaging device and a defect correction method thereof.The light receiving elementMultiple light receiving elementsIs a solid-state imaging device arranged in a two-dimensional arrayDigital camera usingAnd the present inventionPertaining toSolid imagingapparatusDefectsThe correction method is that the photosensitive area is divided into multiple photosensitive areas.DetectDetected light receiving elementDefecteasilyIt relates to a correction method.
[0002]
[Prior art]
In general, flaw correction for defective pixels included in a solid-state imaging device is performed by interpolation processing using pixel data obtained by storing the positions of defective pixels in advance and obtaining pixels adjacent to the defective pixels. With this interpolation processing, for example, if the gray level boundary of the image is immediately before the defective pixel, the processing result may be noticeable in the reproduced image.
[0003]
Therefore, before performing the interpolation process, the signal level of the pixel data in the vicinity of the defective pixel is compared, and the discrimination for patterning the direction in which the boundary exists is performed, and the pixel used for flaw correction is determined based on the discrimination result is doing. Thus, complicated preprocessing is required for flaw correction.
[0004]
However, this flaw correction is only performed on a pixel at a pre-stored position. For example, the flaw correction occurs after the flaw or aging that occurs when the solid-state imaging device is exposed to a high temperature during operation. Cannot handle scratches.
[0005]
In such a case, a flaw detection circuit and a flaw correction circuit that perform appropriate flaw correction even on a complicated boundary are proposed in Japanese Patent Laid-Open No. 6-319082. The flaw detection circuit captures the target pixel and surrounding pixels by the pixel fetch circuit, and obtains a correlation between the target pixel and surrounding pixels by the flaw determination circuit, and if there is at least one correlation with the target pixel, the target pixel Is detected as normal, and other than this, it is determined that there is a defect, and scratch detection not based on memory is performed. In addition, the flaw correction circuit includes a pixel capturing circuit and an arithmetic unit that calculates an average of the peripheral pixels, and replaces the output of the arithmetic unit as an interpolation output.
[0006]
[Problems to be solved by the invention]
By the way, the flaw detection circuit and flaw correction circuit disclosed in Japanese Patent Laid-Open No. 6-319082 deal with acquired defects, so that complex data processing is always performed by taking in peripheral pixels for the target pixel, and flaw detection is performed. Since this is performed, it takes more time than the case of storing the defect position in advance. In particular, the flaw correction circuit performs an interpolation process using pixel data of peripheral pixels for the target pixel. In the interpolation processing, there are respective intervals between the target pixel and the peripheral pixels, but interpolation processing is performed in consideration of the number of pixels to be used without considering the interval. As a result, an error with respect to the level of the target pixel becomes large.
[0007]
  The present invention eliminates such disadvantages of the prior art,When correcting for pixel defectsComplicatedcorrectionSolid-state imaging that eliminates processing and reduces correction errors for target pixelsEquipment and itsAn object is to provide a defect correction method.
[0008]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, the present invention converts the incident light into a signal charge in a photosensitive region that receives the incident light, and the light receiving elements that accumulate the signal charge in the formed potential well are arranged in a two-dimensional array. In this solid-state imaging device, a barrier for partitioning the potential well is formed in the light receiving device, and charge reading means for reading the signal charge accumulated in each potential well formed in the photosensitive region divided by the barrier is independent. It is characterized by being arranged.
[0009]
The solid-state imaging device of the present invention prevents the supply of dark current caused by a defect by partitioning the potential well of each light receiving device with a barrier, and charges reading means arranged adjacent to the partitioned potential well By separately reading out the signal charges accumulated through each of the two, the signal charges can be obtained from the region while corresponding to one piece of positional information as one light receiving element. Thereby, one position information is assigned to each area. For example, even if a defect occurs in one of the areas, signal charges obtained from other normal areas can have the same position information.
[0010]
In order to solve the above-mentioned problems, the present invention detects a defect in a solid-state imaging device in which light receiving elements that convert incident light into signal charges are arranged in a two-dimensional array in a photosensitive region that receives incident light. In the defect detection method, a potential well region in which signal charges are accumulated is partitioned, and a solid-state imaging device that independently reads the signal charges accumulated in the partitioned regions is used to capture an image of uniform brightness, It is determined whether or not there is imaging data obtained by this imaging within a preset threshold range, and the position information of the light receiving element in which the defect is detected by this determination and the information on the defective area in this light receiving element are stored. 1 process is included.
[0011]
The defect detection method for a solid-state imaging device according to the present invention determines whether there is imaging data within a threshold range when an image of uniform brightness is captured, and not only the position information of the light receiving element but also the information on the defect area. Can also be stored, and the normal area and the abnormal (defect) area can be distinguished to promote use from the normal area.
[0012]
Furthermore, in order to solve the above-described problems, the present invention responds to defects in a solid-state imaging device in which light receiving elements that convert incident light into signal charges in a photosensitive region that receives incident light are arranged in a two-dimensional array. In a defect correction method for performing correction, an area of a potential well that accumulates signal charges is partitioned, and a solid-state imaging device that independently reads the signal charges accumulated in the partitioned areas is used to capture an image of a scene and receive light. A signal charge accumulated from each region of the element is read out, and the first step of storing the imaging data obtained by this imaging, and among the stored imaging data, it is obtained in advance by detecting a defect related to the solid-state imaging element. Using imaging data obtained from a normal area with a defective light receiving element based on the position information of the defect and the information on the defective area, the area of this normal area is the entire photosensitive area. Characterized in that it comprises a second step of converting.
[0013]
According to the defect correction method for a solid-state imaging device of the present invention, imaging data of each region is stored, imaging data of a normal region is extracted from a light receiving element including a defect based on the defect position information and the defect region information, By correcting the imaging data by converting it to the entire photosensitive area of the light receiving element, it is possible to accurately obtain the imaging data originally obtained by the light receiving element despite the occurrence of a defect.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Next, an embodiment of a solid-state imaging device according to the present invention will be described in detail with reference to the accompanying drawings.
[0015]
In this embodiment, the solid-state imaging device of the present invention is applied to the light receiving device 10. The illustration and description of parts not directly related to the present invention are omitted. In the following description, the signal is indicated by the reference number of the connecting line in which it appears.
[0016]
The light receiving element 10 is a photodiode and performs photoelectric conversion. The photodiode shown in this example is n⁺Has a pn structure. When viewed from above, the light receiving element 10 has a photosensitive region 12 formed in an octagon as shown in FIG. 1A, and the photosensitive region 12 is divided into photosensitive regions 12a and 12b. The photosensitive area 12a occupies 3/4 of the total area of the photosensitive area 12, and the photosensitive area 12b occupies 1/4 of the total area. This division is performed through the center of the photosensitive region 12. In the division of the photosensitive region 12, an oxide film is formed as a potential barrier 12c that partitions the photosensitive regions 12a and 12b. In this embodiment, the oxide film is made of SiO.₂Is used.
[0017]
The light receiving element 10 is formed with transfer gates 12d and 12e (black circle symbols). The transfer gates 12d and 12e correspond to the photosensitive areas 12a and 12b, and are formed so as to independently read out signal charges accumulated in a vertical transfer register (not shown).
[0018]
FIG. 1B shows a cross section of the light receiving element 10 cut along the breaking line Ib-Ib. The light receiving element 10 includes a p-channel layer 16 formed on a silicon substrate 14 and an n-channel⁺Layer 18 is embedded. n⁺The upper surface 16a of the layer 18 is formed according to the photosensitive region 12, and n⁺A p + thin film 20 is formed on the layer 18. The potential barrier 12c described above is formed vertically from the silicon substrate 14 to the thin film 20.
[0019]
On the light receiving element 10, an internal lens 22 is formed so that incident light also enters the periphery of the photosensitive region 12 to increase the photoelectric conversion efficiency. The internal lens 22 flattens the incident light side. A transparent protective film 24 is provided between the inner lens 22 and the surface 20a formed by the thin film 20. On the flattened inner lens 22, a color filter 26 for color-separating incident light is formed, and an on-chip microlens 28 is formed on the incident light side. The on-chip microlens 28 has a role of condensing incident light on the photosensitive region 12 and a filter function that makes the spatial frequency of the incident light equal to or lower than the Nyquist frequency.
[0020]
In this way, the light receiving element 10 divides the photosensitive region 12 into the regions 12a and 12b by forming the potential barrier 12c in the photosensitive region 12, and even if, for example, a lattice defect occurs in one of the regions 12a and 12b, the light receiving element 10 corresponds to the amount of light. Even if the conversion to the signal charge cannot be performed accurately, the influence of the dark current generated in the defective region can be prevented and the adverse effect on the normal region can be avoided. The defect is not limited to a lattice defect, and is effective when a defect such as dust that blocks incident light on the photosensitive region occurs.
[0021]
The division of the photosensitive area 12 is not limited to 3: 1 as shown in FIG. 1, and the photosensitive area on the vertical transfer register (not shown) side may be equally divided into a plurality. Thereby, the area | region corresponding to abnormality generation can be increased.
[0022]
Also, as shown in FIGS. 2 (a) and 2 (b), transfer gates 12d and 12e for reading the signal charges accumulated in the respective regions of the light receiving element 10 to the vertical transfer register may be disposed in a substantially opposed positional relationship. . The transfer gate is indicated by a black circle symbol. Even if data is read out to the vertical transfer register at the same time, the vertical transfer register is driven so as to form independent packets so that the signal charges are not mixed, and the signal charge in each region is read out.
[0023]
In the light receiving element 10 of FIGS. 1 and 2 (a), a potential barrier 12c is formed in the photosensitive region 12 as when the clock hand indicates 3 o'clock. Further, in the light receiving element 10 of FIG. 2B, a potential barrier 12c is formed so that the right half of the light receiving element 10 is divided into four equal parts. In this case, the light receiving element 10 is formed with transfer gates 120e, 122e, 124e, 126e on the right half.
[0024]
In consideration of the position where the transfer gate is formed so that the signal charge can be read out independently while satisfying this region division condition, if the signal charge can be read out, the potential barrier 12c formed in the light receiving element 10 with respect to the center line They may be formed symmetrically. That is, formation of a potential barrier 12c that divides the photosensitive region 12 of FIG. 2A into four equal parts and in the case of FIG. 2B, formation of a potential barrier 12c that also divides the left half photosensitive region of the light receiving element 10 into four equal parts. May be performed.
[0025]
Next, a digital camera 30 using a solid-state image sensor on which the light receiving element 10 is formed will be described with reference to FIG. The digital camera 30 includes an optical lens system 32, an aperture adjustment mechanism 34, an imaging unit 36, a preprocessing unit 38, a buffer 40, a signal processing unit 42, an operation unit 44, a system control unit 46, a defect data memory 48, and a timing signal generation. Part 50, driver 52, monitor 54 and storage 56. The optical lens system 32 includes an automatic focus (AF) adjustment mechanism that automatically adjusts the arrangement of the optical lenses to adjust the subject to a focused positional relationship. Each of the mechanisms is provided with a motor for moving the optical lens to the position described above. These mechanisms operate in response to a drive signal 52a supplied from the driver 52 to each motor.
[0026]
The diaphragm adjusting mechanism 34 is an AE (Automatic Exposure) adjusting mechanism that adjusts the amount of incident light, although not specifically shown, and rotates the ring portion in accordance with a drive signal 52b from the driver 52. The ring part partially overlaps the blades to form a round shape of the iris, and forms the iris so that the incident light beam passes therethrough. In this way, the aperture adjustment mechanism 34 changes the aperture of the iris. The aperture adjustment mechanism 34 may be incorporated in the optical lens system 32 as a mechanical shutter as a lens shutter.
[0027]
The imaging unit 36 uses a solid-state imaging element 36a in which the light receiving elements 10 are arranged in a two-dimensional array. As described above, the color filter 26 and the on-chip microlens 28 are sequentially formed on the light receiving element 10 in the solid-state imaging device 36a toward the incident light side. The color filter 26 is, for example, a filter in which three primary color RGB color filter segments are arranged in a predetermined positional relationship with individual light-sensitive elements (photosensitive cells) 10 on a one-to-one basis. Therefore, the color filter depends on the arrangement of the light receiving elements 10.
[0028]
The solid-state imaging device 36a includes a charge coupled device (CCD) and a metal oxide semiconductor (MOS) type. In the present embodiment, a CCD is used for the solid-state image sensor 36a. The CCD has the characteristics shown in FIG. That is, the light receiving element 10 shown in FIG. 1 (a) is formed by dividing the photosensitive region 12 into two photosensitive regions 12a and 12b, and has a structure for reading out independently from each of the photosensitive regions 12a and 12b. . The sensor sensitivities of the photosensitive regions 12a and 12b are the same, only the areas of the photosensitive regions are different.
[0029]
A driving signal 52c is supplied from the driver 52 to the solid-state imaging device 36a. The solid-state imaging device 36a is supplied with a drive signal 52c according to the operation mode. Of the operation modes, signal charges generated by photoelectric conversion from each of the light receiving elements 10 at the time of exposure are field-shifted to the vertical transfer register via the readout gate according to the defect detection mode, the imaging mode, and the like.
[0030]
However, in the above-described field shift signal charge readout, either the photosensitive region 12a or the photosensitive region 12b is selectively read out of the photosensitive regions 12 included in the light receiving elements 10 arranged in an array. After this readout, the signal charge accumulated in the other photosensitive area is read out. In the vertical transfer register, if the signal charges in the photosensitive areas 12a and 12b are not mixed even if they are simultaneously read, simultaneous reading is possible.
[0031]
The signal charge of the vertical transfer register is shifted horizontally by shifting the signal charge transferred to the horizontal transfer register arranged in the direction orthogonal to the vertical transfer register, that is, in the horizontal direction. The horizontally transferred signal charge is Q / V converted into a voltage signal using a floating diffusion amplifier (FDA) as an output amplifier. The imaging unit 36 outputs the Q / V converted analog signal 36b to the preprocessing unit 38.
[0032]
The preprocessing unit 38 includes a correlated double sampling (CDS) circuit, a gain control amplifier (GCA), and an analog-to-digital converter (A / D converter) for noise removal. (Not shown). The preprocessing unit 38 converts all of the imaging data obtained by performing noise removal, waveform shaping, and digitization to the supplied analog signal 36b as image data 38a through the system bus 100 and the data bus 40a, and the buffer 40. Output to.
[0033]
The buffer 40 is a non-volatile memory, and is also a temporary memory in that it stores image data 38a read for each area for one image. A control signal 46a from the system control unit 46 is supplied to the buffer 40 as a control signal 40b, and writing / reading control for the image data 38a is performed in response to the control signal 40b. The image data 38a stored in the buffer 40 should have a data capacity so as to store not only the image data of each area obtained for one light receiving element 10, but also the image data collected as the entire photosensitive area 12. .
[0034]
As will be further described in the subsequent defect detection, the former image data is used to determine which region is defective, and the latter image data is used to determine the defect of the light receiving element 10. Finally, the buffer 40 stores image data that does not include a defect after the defect correction is performed on one supplied image. The buffer 40 outputs image data 40a for one image to the signal processing unit 42 via the system bus 100 and the data bus 42a.
[0035]
The signal processing unit 42 includes a signal generation circuit (not shown), a memory, a gamma correction circuit, an evaluation value calculation unit, a pixel interpolation processing circuit, a matrix processing circuit, and a compression / decompression processing circuit. The memory may be shared with the buffer 40. The signal processing unit 42 is supplied with a control signal 46a from the system control unit 46 via the system bus 100 as a control signal 46b via a control line. A signal generation circuit (not shown) of the signal processing unit 42 operates in response to the control signal 46b. The signal generation circuit has a PLL (Phase Locked Loop) circuit that can generate a plurality of frequencies. The signal generation circuit multiplies the oscillation frequency of the source as a reference clock to generate a plurality of types of clock signals 42b, and outputs them as a clock signal 42c to the system control unit 46 and the timing signal generation unit 50 via the system bus 100. .
[0036]
Further, the signal processing unit 42 is supplied with a timing signal from the timing signal generation unit 50 (not shown). This timing signal includes a horizontal synchronization signal HD, a vertical synchronization signal VD, and operation clocks of respective units described later.
[0037]
Image data 40a is input to the memory via the system bus 100 and a data bus 42a for inputting and outputting signals, and is temporarily stored. In particular, the memory is supplied with image data obtained from, for example, signal charges thinned out in the vertical direction by 1/4 in the moving image shooting mode. The memory may also perform thinning readout in the horizontal direction in order to increase the aspect ratio and signal readout speed at the time of readout. Also in this case, image data is read so as not to destroy the original color arrangement pattern. The memory performs thinning in the horizontal direction.
[0038]
The memory supplies the stored image data to the gamma correction circuit. In addition, as described above, it is preferable to use a non-volatile memory as the memory when reading data repeatedly. The gamma correction circuit includes a lookup table for gamma correction, for example. The gamma correction circuit performs gamma correction on the image data supplied as one of the pre-processes in the image processing using the table data. The gamma correction circuit supplies gamma-corrected image data to an evaluation value calculation unit and a pixel interpolation processing circuit (not shown).
[0039]
The evaluation value calculation unit includes an arithmetic circuit that performs aperture value / shutter speed, white balance (WB) adjustment, gradation correction, and the like. The evaluation value calculation unit calculates not only scene information but also appropriate parameters by arithmetic processing based on the supplied image data in the above-described circuit. These calculation results are supplied to the system control unit 46 as parameters via the data bus 42a, the system bus 100, and the data bus 46c.
[0040]
The evaluation value calculation unit is not limited to the signal processing unit 42, and may be provided in the system control unit 46. In this case, the signal processing unit supplies the gamma-corrected image data to the system control unit.
[0041]
The pixel interpolation processing circuit has a function of calculating and interpolating pixel data. Since the imaging unit 36 uses the single-plate color filter 26, colors other than the actual color filter segment colors cannot be obtained from the imaging device. Therefore, the pixel interpolation processing circuit generates pixel data of an unobtainable color by interpolation in the still image shooting mode. The pixel interpolation processing circuit supplies plain image data to the matrix processing circuit.
[0042]
Note that the pixel interpolation processing circuit may include a function of widening the generated pixel data. Further, as described above, when the so-called honeycomb type solid-state imaging device 36a is used in the imaging unit 36, the pixel interpolation processing circuit uses the gamma-corrected image data to actually detect the position where the pixel exists (actual Pixel data of the three primary colors RGB is generated by interpolation processing at a position (virtual pixel) where no pixel) or pixel exists.
[0043]
The matrix processing circuit generates luminance data Y and color data Cb, Cr using the image data supplied from the pixel interpolation processing circuit and predetermined coefficients. The generated image data is supplied to a compression / decompression processing circuit.
[0044]
The compression / decompression processing circuit converts the image data (Y / C) supplied in the still image mode to MPEG (Joint Photographic coding Experts Group) and the image data (Y / C) supplied in the movie (movie) mode to MPEG ( Moving Picture coding Experts Group) -1, MPEG-2, etc. The compression / decompression processing circuit sends the compressed image data to the storage 56 via the data bus 42a, the system bus 100, and the data bus 56a and records them. The compression / decompression processing unit supplies the image data recorded in the storage 56 via the data bus 56a, the system bus 100, and the data bus 42a, and performs decompression processing. This decompression process is the reverse process of the compression process.
[0045]
Further, the signal processing unit 42 performs RGB conversion on the generated image data and the image data (Y / C) expanded along with reproduction, and this RGB converted image data 42a is transferred to the system bus 100 and the data bus 54a. To the monitor 54. The monitor 54 is controlled by a display controller (not shown), and the image data 54a is displayed as an image on the display device.
[0046]
The signal processing unit 42 may include an external I / F circuit (not shown) when inputting / outputting image data to / from an external device. Examples of external I / F circuits include PIO (Programmed Input / Output), UART (Universal Asynchronous Receiver / Transmitter), USB (Universal Serial Bus), IEEE1394 standard (the Institute of Electrical and Electronics) Engineers: Interfaces based on the American Institute of Electrical and Electronics Engineers).
[0047]
Although not shown, the operation unit 44 includes a mode selection switch and a release shutter button. The mode selection switch has a function of selecting which one of the defect detection mode and the imaging mode to use. The mode selection switch outputs the selected mode signal 44a to the system control unit 46. The release shutter button is a button having a two-stage stroke, and a trigger signal for setting the digital camera 30 to the preliminary imaging stage (S1) with the first stroke and the main imaging stage (S2) with the second stroke. 44b is output to the system control unit 46. In addition, the operation unit 44 may be provided with a zoom selection switch and a cross button, and may have a function of selecting conditions displayed on the liquid crystal display panel.
[0048]
The system control unit 46 is a microcomputer or CPU (Central Processing Unit) that controls a general-purpose part of the entire camera and a part that performs digital processing. The system control unit 46 sets the digital camera 30 to the defect detection mode or the imaging mode according to the supplied mode signal 44a. Further, the system control unit 46 generates the control signal 46a so as to perform the through image display for displaying the object scene image on the monitor 54 regardless of the mode.
[0049]
Then, the system control unit 46 receives the mode signal 44a to be set and the trigger signal 44b that notifies the imaging timing from the release shutter button, and also generates the control signal 46a accompanying the imaging recording. The control signal 46a generated in this way is supplied as control signals 46d and 46e to the signal processing unit 42, the timing signal generating unit 50, and the driver 52 via the system bus 100, respectively.
[0050]
The system control unit 46 also generates a control signal 46a that takes into account control for performing line interpolation and signal generation circuit in the signal processing unit 42 and signal processing. Although not shown, the system control unit 46 also performs read / write control in the preprocessing unit 38 and the storage 56.
[0051]
In the present embodiment, the system control unit 46 includes a defect detection function unit 460 and a defect correction function unit 462. The defect detection function unit 460 reads the image data 40a supplied from the buffer 40, detects the defect by comparing with the threshold value indicating the allowable range set for each of the light receiving elements 10, and detects the defect by this comparison determination. Has a function of detecting defects in the photosensitive regions 12a and 12b. The defect detection function unit 460 writes the position information and area information of the light receiving element 10 in which the defect is detected as a set to the defect data memory 48.
[0052]
In the information to be written, it is preferable to store a coefficient for converting a normal area into a total photosensitive area in the light receiving element 10 in which a defect is detected. This is effective when performing defect correction.
[0053]
The defect correction function unit 462 combines the process of combining the image data 40a with the photosensitive region 12 of the light receiving element 10 and the pixel data corresponding to each set of the defect data memory 48 using all data from the normal region. It has a function of performing area conversion processing of the photosensitive region. However, when the defect of the light receiving element 10 extends to the whole, the defect correction function unit 462 corrects the defect by the addition averaging process using the pixel data in the vicinity of the light receiving element to be defective as before.
[0054]
The defect data memory 48 uses a nonvolatile memory such as an EEPROM (Electrically Erasable and Programmable Read-Only Memory). When the defect detection mode is selected, the defect data memory 48 transfers the stored data to the buffer 40 or the storage 56, for example, and saves it. After the transfer, the defective data memory 48 erases the memory contents under the control of the system control unit 46. The defect data memory 48 writes a data set related to the defective light receiving element under the control of the system control unit 46. At this time, the system control unit 46 performs the write control while comparing the saved data set and the data set in which the defect is detected in consideration of the order of the read positions. Thereby, the defect data memory 48 can provide defect data according to the order of the supplied image data 40a.
[0055]
The timing signal generation unit 50 generates a timing signal according to the control signal 46d supplied from the system control unit 46 with reference to the clock signal 42c supplied from the signal processing unit 42. The timing signal includes a vertical synchronization signal, a horizontal synchronization signal, a field shift pulse, a vertical transfer signal, a horizontal transfer signal, and the like. The timing signal generation unit 50 supplies the generated timing signal 50a to the driver 52, the preprocessing unit 38, the signal processing unit 42 and the system control unit 46 (not shown), respectively, according to the operation.
[0056]
The driver 52 has a drive circuit that generates drive signals 52a, 52b, and 52c based on the supplied timing signal 50a and control signal 46e. The driver 52 supplies drive signals 52a and 52b to the optical lens system 32 and the aperture adjustment mechanism 34 based on the control signal 46e, respectively, to perform AF adjustment and AE adjustment. The driver 52 supplies the drive signal 52c generated based on the timing signal 50a to the solid-state image sensor 36a, and the signal charge obtained by photoelectric conversion in each light receiving element 10 (photosensitive regions 12a, 12b) The signal charge accumulated in at least one photosensitive area is read out to the vertical transfer register, the read signal charge is transferred vertically, the signal charge is line-shifted and supplied to the horizontal transfer register, and then transferred horizontally. Yes.
[0057]
Image data 42a is supplied from the signal processing unit 42 to the monitor 54 via the system bus 100 and the data bus 54a. As the monitor 54, a liquid crystal monitor is generally used. The liquid crystal monitor is provided with a liquid crystal display controller. The liquid crystal controller performs switching control by arranging liquid crystal molecules and applying voltage based on the image data 42a. By this control, the liquid crystal monitor displays an image. Needless to say, the monitor 54 is not limited to a liquid crystal monitor, and can be sufficiently used as long as the display device is small in size, can confirm images, and can suppress power consumption.
[0058]
The storage 56 records the image data supplied from the signal processing unit 42 via the data bus 42a, the system bus 100, and the data bus 56a using a semiconductor memory or the like as a recording medium. As the recording medium, an optical disk, a magneto-optical disk, or the like may be used. The storage 56 writes / reads data using a recording / reproducing head by combining a pickup suitable for each recording medium, a pickup and a magnetic head. Data writing / reading is performed under the control of the system control unit 46 although not shown.
[0059]
Next, the operation of the digital camera 30 will be described with reference to FIGS. After the power is turned on, initial setting is performed (step S10). Thereafter, a shooting mode is selected (step S12). The mode is selected and set by the operation unit 44. In this embodiment, there are a defect detection mode, a shooting mode, a reproduction display mode, and the like, and any one mode is selected. After this setting, the process proceeds to the determination of whether or not the defect detection mode is set (step S14).
[0060]
Next, it is determined whether or not the selected mode is defect detection (step S14). If the defect detection mode is set (YES), the process proceeds to the defect detection process (subroutine SUB1). If a mode other than this is set (NO), the process proceeds to a determination of whether or not it is a shooting mode (to step S16). The defect detection process is performed on the image data 40a supplied from the buffer 40 by the defect detection function unit 460 of the system control unit 46. Detailed explanation will be given later. The defect detection function unit 460 supplies information on defects detected by the control of the system control unit 46 to the defect data memory 48.
[0061]
After the defect detection processing, the processing procedure is returned to mode selection (step S12). This is because the defect detection may be performed when the user desires. Then, mode selection is performed, and the processing after the selection setting described above is repeated.
[0062]
It is determined whether or not the imaging mode is set (step S16). In the photographing mode (YES), the process proceeds to reading out signal charges for each region from the solid-state imaging device 36a (to step S18). Here, although not shown, a series of settings relating to shooting is performed until step S18, and the imaging timing is supplied from the operation unit 44 to the system control unit 46, and exposure is performed. If a mode other than the shooting mode is selected and set (NO), the process proceeds to the processing procedure of the corresponding mode. In this embodiment, for example, the process proceeds to the reproduction display mode.
[0063]
After the exposure, the digital camera 30 drives the imaging unit 36 and reads the signal charges accumulated separately from the regions of the light receiving element 10 to a vertical transfer register (not shown) (step S18). The signal charge is converted into an analog signal by Q / V conversion and supplied to the preprocessing unit 38. In the preprocessing unit, image data 38 a obtained by subjecting the supplied signal to digital conversion processing is supplied to the buffer 40 via the system bus 100.
[0064]
Next, pixel data corresponding to the photosensitive region 12 in the normal light receiving element 10 is generated, and defect correction processing is performed on the defective light receiving element (step S20). The generation of the pixel data means that the pixel data corresponding to the signal charges obtained in the respective photosensitive regions 12a and 12b in the light receiving element 10 is combined into pixel data normally obtained from the light receiving element 10. In the defect correction, the position information of the defective light receiving element, the information on the defect area, and the coefficient stored in advance in the defect data memory 48 are read out, and the process is performed on the pixel data of the target light receiving element.
[0065]
The defect correction is performed by multiplying the pixel data obtained from the normal light-sensitive area of the defective light-receiving element by a coefficient and converting it into pixel data obtained from the entire light-sensitive area 12 in the light-receiving element. Therefore, the coefficient used for conversion is the ratio of the area of the entire photosensitive region to the area of the normal photosensitive region. Thus, by converting the pixel data of the defective light receiving element, it is possible to suppress errors that occur due to the averaging process.
[0066]
The buffer 40 finally stores the image data 40a subjected to the defect correction and is read out to the signal processing unit 42. In the signal processing unit 42, the supplied image data 40a is subjected to gamma correction, WB adjustment, gradation correction, pixel interpolation processing, matrix processing, and compression / decompression processing (step S22).
[0067]
Next, the signal-processed image is written into the storage 56 under the control of the system control unit 46 (step S24). By this writing, recording to the storage 56 is performed. Thereafter, it is determined whether or not to end the imaging mode (step S26). The termination determination condition includes a storage capacity of the storage 56, a battery capacity (not shown), a forced termination from the user, and the like. If the determination condition is not satisfied (NO), the above-described shooting procedure is repeated in the shooting mode. If the determination condition is satisfied (YES), the operation of the digital camera 30 is stopped and the process ends.
[0068]
Next, the defect detection process will be described. A field having a uniform luminance is imaged, and then, as shown in FIG. 5, the light-receiving element 10 is driven so that the transfer gate opens at the same timing without distinction of the region, and a vertical (not shown) Read to transfer register (substep SS10). The signal charge read for each light receiving element 10 is used as pixel data, and each pixel data for one image is stored in the buffer 40.
[0069]
Next, defect determination for each of the light receiving elements 10 is performed (substep SS12). The buffer 40 reads the image data 40a stored in the defect detection function unit 460. In the defect detection function unit 460, an allowable threshold range is set in advance. The defect detection function unit 460 determines that there is a white defect when the pixel data of the image data 40a shows a level that is greater than or equal to the allowable upper limit. In addition, the defect detection function unit 460 determines that there is a black defect when the pixel data of the image data 40a shows a level that is less than or equal to the allowable lower limit. Therefore, it is determined that pixel data whose pixel data is within the allowable threshold range is normal. Here, the positions of the light receiving elements 10 determined to be white and black are temporarily stored in the buffer 40.
[0070]
Next, after the same image is exposed again, the signal charge is read out to a vertical transfer register (not shown) by distinguishing each region (substep SS14). The signal charges read for each region are stored in the buffer 40 as pixel data. The buffer 40 stores pixel data obtained by dividing the number of light receiving elements by the number of pixel regions. The signal charge read from the light receiving element cannot be driven to read and read only the defective light receiving element in units of one line. Therefore, the signal charge is read separately for each region.
[0071]
  Multiplying the coefficient for the pixel data of each area is converted into the pixel data of the entire photosensitive area 12 (substep SS16). As described above, the coefficient is the ratio of the area of the entire photosensitive region to the area of the region. In the case of FIG. 1, the coefficient for converting the photosensitive region 12a to the total photosensitive region 12 is 4/3 because it is (the area of the total photosensitive region) / (the area of the photosensitive region 12a). AlsoThere was an abnormalityIn photosensitive area 12bPhotosensitive area 12a soThe conversion factor is1/3It is.
[0072]
Next, defect determination is performed on the converted pixel data of each region (substep SS18). As the determination criterion, the above-described allowable threshold range is used. This defect determination is performed on the previously obtained defect light receiving element. The position information of the defective light receiving element is stored in the buffer 40. Only the pixel data of each corresponding area is read and determined. A defect in the area is detected by determining whether or not the area is a normal light receiving element that falls within the allowable threshold range. By this series of defect detection processing, the position information in the defect light receiving element, the information on the area where the defect has occurred, and the (conversion) coefficient for the normal area are stored in the defect data memory 48 as a set of defect detection data (sub data). Step SS20). When the defect detection mode is selected and set, the defect data memory 48 first saves the defect detection data currently stored under the control of the system control unit 46, and thereafter erases the contents. Defect detection data is written in the defect data memory 48 in accordance with the reading order under the control of the system control unit 46.
[0073]
When the writing and saving to the defect data memory 48 is completed, the process proceeds to return and the subroutine SUB1 is terminated.
[0074]
In this way, defect detection can be performed even if divided into regions. In this embodiment, the defect detection is performed by imaging twice, but the imaging may be performed only once. In this case, signal charges are read out for each region. The process for determining the position of the light receiving element may be performed later.
[0075]
With the configuration described above, the potential barrier 12c is formed in the light receiving element 10 and divided into regions, so that leakage of dark current from a defective region to a normal region can be prevented. By converting the pixel data obtained from the normal area into the entire photosensitive area 12, it is possible to suppress an error that occurs more than the defect correction obtained by the averaging process. Since it is not necessary to perform complicated discrimination processing, the processing can be simplified.
[0076]
Defect detection applies the division of the photosensitive area in the light receiving element, detects the defect position based on whether or not the allowable threshold range is satisfied, and converts the pixel data of each area of the defective light receiving element into the entire photosensitive area. The defective area is detected based on whether the allowable threshold range is satisfied. Thereby, defect position information and a defect area | region can be obtained, and such information can be advantageously used for defect correction.
[0077]
In the defect correction, the signal charge is read separately for each region, and pixel data for each region is added to a normal light receiving element to generate pixel data in all photosensitive regions. By multiplying the pixel data obtained from the normal area by a coefficient and converting it to pixel data in the entire photosensitive area, it is possible to perform correction with less error than conventional averaging processing using peripheral pixel data. it can. This correction is advantageous in that the pixel data at the defect position can be substantially reproduced without performing complicated boundary determination.
[0078]
【The invention's effect】
As described above, according to the solid-state imaging device of the present invention, the potential well of each light receiving element is partitioned by a barrier, thereby preventing supply of dark current caused by a defect and disposed adjacent to the partitioned potential well. By separately reading the signal charges accumulated through the charge reading means, the signal charges can be obtained from the region while corresponding to one position information as one light receiving element. Thereby, one position information is assigned to each area. For example, even if a defect occurs in one of the areas, signal charges obtained from other normal areas can have the same position information. Furthermore, if the imaging data obtained with the area of the normal region is converted into the imaging data obtained with the area of the entire photosensitive region in the light receiving element, it is possible to receive the light without performing complicated discrimination processing. Pixel data without defects can be easily obtained from the element, and the processing can be simplified. The pixel data is superior in that it can be made data with less error compared to the conventional averaging process using peripheral pixel data.
[0079]
Further, according to the defect detection method in the solid-state imaging device of the present invention, when an image of uniform brightness is captured, it is determined whether there is imaging data within the threshold range, and not only the position information of the light receiving element, Information on the defective area can also be stored, and a normal area and an abnormal (defect) area can be distinguished from each other to promote use from the normal area.
[0080]
Further, according to the defect correction method in the solid-state imaging device of the present invention, the imaging data of each area is stored, and the imaging data of the normal area is obtained from the light receiving element including the defect based on the position information of the defect and the information of the defect area. By taking out and correcting this imaging data by converting it to the entire photosensitive area of the light receiving element, the imaging data originally obtained by the light receiving element even though a defect has occurred without performing complicated determination processing It is possible to obtain almost accurately by suppressing the error.
[Brief description of the drawings]
FIG. 1 is a plan view seen from above of a light receiving element to which a solid-state imaging element of the present invention is applied, and a cross-sectional view of the light receiving element cut along a broken line Ib-Ib.
FIG. 2 is a diagram showing an example of a photosensitive region division and a transfer gate formation position of the light receiving element of FIG. 1;
3 is a block diagram showing a schematic configuration of a digital camera to which the light receiving element of FIG. 1 is applied.
4 is a flowchart for explaining an operation procedure in a defect detection mode and an imaging mode in the digital camera of FIG. 3;
FIG. 5 is a flowchart for explaining an operation procedure in the defect detection mode of FIG. 4;
[Explanation of symbols]
10 Light receiving element
12, 12a, 12b Photosensitive area
12c Potential barrier
12d, 12e transfer gate
26 color filters
28 On-chip micro lens
30 Digital camera
36 Image sensor
40 buffers
42 Signal processor
46 System controller
48 Defective data memory
460 Defect detection function
462 Defect correction function

Claims (4)

In a solid-state imaging device in which light receiving elements that convert incident light into a signal charge in a photosensitive region that receives incident light and store the signal charge in a potential well formed are arranged in a two-dimensional array, the device includes:
A potential well formed in a photosensitive region in which a barrier for partitioning a potential well formed immediately below the photosensitive region provided in one light receiving element is formed, and an area of the photosensitive region is divided into an integer ratio by the formed barrier. The charge reading means for reading the signal charges accumulated in each of the two is independently provided,
An image having a uniform brightness is picked up, the presence / absence of abnormality of each of the light receiving elements is determined based on imaging data obtained by the imaging, and the defect of the light receiving element is detected by the determination, and the light receiving element in which the defect is detected Control means for converting to normal imaging data based on an integer ratio representing the area of the normal photosensitive area and the area of the abnormal photosensitive area in the imaging data obtained from the normal photosensitive area;
Storage means for storing position information of the light-receiving element in which the defect is detected and information on a region where a defect occurs in the light-receiving element;
The control means includes defect detection means for determining whether or not imaging data obtained by imaging exists within an allowable threshold range;
Using the imaging data obtained from the normal photosensitive area with the defective light receiving element based on the positional information of the light receiving element in which the defect is detected in advance and the information on the defective area, the area of the normal photosensitive area And a defect correction means for correcting the imaging data of the abnormal photosensitive region to normal imaging data by converting the data into an integer ratio of the area of the abnormal photosensitive region.   2. The solid-state imaging device according to claim 1, wherein the charge reading unit selectively reads signal charges accumulated in a normal photosensitive region. A defect correction method for a solid-state imaging device that detects and corrects a defect in a solid-state imaging device in which light-receiving elements that convert the incident light into signal charges in a photosensitive region that receives incident light are arranged in a two-dimensional array. Is
Using a solid-state imaging device that partitions the potential well region for storing the signal charge, and independently reads out the signal charge stored in the partitioned region,
A light-receiving element that picks up an image of uniform brightness, determines whether there is an abnormality based on whether there is imaging data obtained by the imaging within a pre-allowed threshold range, and detects a defect by the determination A first step of storing the position information of and the information of the region where the defect occurs in the light receiving element;
After the actual imaging, the stored imaging data is normal for the defective light receiving element based on the positional information of the light receiving element in which the defect indicating the abnormality of the solid-state imaging element is detected and the information indicating the defective area of the light receiving element. Using the imaging data obtained from the photosensitive region, the imaging data of the abnormal photosensitive region is corrected to normal imaging data by converting the area of the normal photosensitive region and the area of the abnormal photosensitive region by an integer ratio. A defect correction method for a solid-state imaging device, comprising: a second step. 4. The method according to claim 3, wherein the first step is to take an image with uniform brightness and determine whether there is imaging data obtained by the imaging within a threshold range allowed in advance. 3 steps,
A fourth step of reading each of the signal charges from a photosensitive region partitioned in the light receiving element corresponding to abnormal imaging data that deviates from an allowable threshold range;
A fifth step of converting data obtained from each of the read signal charges into the photosensitive region area of the light receiving element;
A sixth step in which the converted data is determined to be defective if the converted data is out of the threshold range;
And a seventh step of storing a set of conversion factors represented by position information of the defective light receiving element, information on the defective photosensitive region, and an area ratio of the abnormal photosensitive region to the normal photosensitive region. A defect correction method for a solid-state imaging device.

Images