JP2006033036A - Imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus the image quality of an image imaged by which is not deteriorated due to uneven luminance, colors of the image and a partial increase or the like in defective pixels while maintaining a frame rate and consecutive shot performance even at a temperature rise of part of its imaging element. <P>SOLUTION: Even when a temperature difference exists in the imaging element, a plurality of temperature sensors are arranged around the imaging element in order to excellently correct the uneven luminance and colors and the defective pixels, and a correction value applied to an output value from the imaging element is changed in response to a temperature measurement result by the plurality of temperature sensors to excellently correct the output of the imaging element. Or the correction value of the defective pixels of the imaging element is used for the correction value applied to the output value to excellently correct the defective pixels caused by a dark current of the imaging element. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image pickup apparatus using an image pickup element such as a CCD or a CMOS image sensor, and more particularly to improvement of image quality of a picked up image.

  In an imaging apparatus such as a digital camera or a video camera, a CCD or a CMOS image sensor is generally used as an imaging element.

  It is well known that the characteristics of the image pickup device include fixed pattern noise due to the dark current variation of the photodiode for each pixel and defective pixels caused by the dark current, which deteriorate the image quality of the imaged image. ing.

  The dark current is generally said to increase by a factor of 2 with a temperature increase of 8 ° C. The fixed pixel noise due to the dark current variation of the photodiode and the defective pixel due to the dark current have temperature characteristics. It is also well known that fixed pattern noise and defective pixels increase at higher temperatures.

  As described above, as a method for reducing fixed pattern noise due to dark current of a photodiode having temperature characteristics, Patent Document 1 previously stores fixed pattern noise for each predetermined temperature range and detects the temperature of the image sensor. A technique for correcting fixed pattern noise by reading out fixed pattern noise corresponding to a detected temperature and subtracting it from an actual captured image signal is disclosed.

  As for a defective pixel caused by dark current, a technique for storing the address of the defective pixel and interpolating the defective pixel information based on the surrounding pixel information is known. It is also conceivable to correct the defective pixel by storing the address of the defective pixel.

  In addition, by performing the imaging operation in a state where the entire image sensor is shielded immediately before or after the actual imaging operation, a captured image of only the dark current component is obtained and subtracted from the actual captured image signal. There is known a technique for reducing a fixed pattern noise caused by dark current including a temperature change and a reduction in image quality due to defective pixels caused by dark current (correction by actual dark subtraction).

  In addition, Patent Document 2 also discloses a method of detecting fixed temperature by detecting the ambient temperature of the image sensor based on the average value of the optical black region of the image sensor.

Japanese Laid-Open Patent Publication No. 1-147973 JP 2001-174329 A JP-A-10-155100

  However, in Patent Document 2 (Japanese Patent Laid-Open No. 2001-174329), it is very effective when the entire image sensor undergoes a uniform temperature change, but a partial circuit inside the image sensor or a peripheral circuit outside the image sensor. If the temperature of a part of the image sensor rises due to excessive heat generation, the dark current will be different between the part where the temperature of the image sensor is high and the part where the temperature is low. I can't let you. In addition, the captured image at this time causes deterioration in image quality such as luminance unevenness and color unevenness due to a dark current difference in the screen.

  Further, when there is a temperature difference in the image sensor as described above, if the entire screen is corrected with a defective pixel address for each predetermined temperature range as in Patent Document 1 (Japanese Patent Application Laid-Open No. 1-147973), excessive correction or A part with insufficient correction will occur. Since defective pixel correction is often performed by interpolation from surrounding pixel information, it is well known that when the correction is excessive, the reproducibility of the pattern deteriorates when a fine pattern is imaged. If the correction is insufficient, the noise on the screen increases and the image quality deteriorates.

  In recent years, as the number of pixels of an image sensor has increased, the chip size of the image sensor itself has increased. In particular, some imaging devices such as a digital still camera that require high image quality have a chip size equivalent to the screen size of a 35 mm film of a silver halide camera. The problem of unevenness and defective pixels becomes particularly significant.

  In addition, since the correction method based on actual dark subtraction performs two imaging operations in one scene, the shooting time of one scene is doubled, and the frame rate of the video camera and the continuous shooting performance of the digital still camera are greatly reduced. become.

  The object of the present invention is to solve the above-mentioned problems, and maintain the frame rate and continuous shooting performance even when the temperature of the image sensor partially rises. An object of the present invention is to provide an imaging device that does not deteriorate image quality.

  Even when there is a temperature difference in the image sensor as described above, in order to satisfactorily correct luminance unevenness, color unevenness, and defective pixels, a plurality of temperature sensors are arranged around the image sensor, and the plurality of temperature sensors By changing the correction value for the output value from the image sensor in accordance with the temperature measurement result of, the image sensor output can be corrected satisfactorily.

  Further, the correction value for the output value is set as the correction value for the defective pixel of the image sensor, so that the defective pixel correction due to the dark current of the image sensor can be performed satisfactorily.

  Further, by dividing the pixel range of the image sensor into a plurality of areas and changing the correction value of the defective pixel for each area, it is possible to perform defective pixel correction with higher accuracy.

  Further, by making the correction value for the output value a correction value for the dark current value of the image sensor, it is possible to satisfactorily correct luminance unevenness and color unevenness.

  Further, by constructing the temperature sensor on the same substrate as the image pickup device, it is possible to satisfactorily correct luminance unevenness, color unevenness, and defective pixels without increasing the size and cost of the apparatus.

  Further, the light shielding pixels arranged around the effective pixel portion of the image sensor are divided into a plurality of areas, the dark current values of the light shielding pixels are detected, and the dark current values from the plurality of light shielding pixel areas are detected. By measuring the temperature and changing the correction value for the output value from the image sensor in accordance with the temperature measurement results of the plurality of regions, the image sensor output can be favorably corrected.

  According to the present invention, even when there is a temperature difference in the image sensor, a plurality of temperature sensors are arranged around the image sensor in order to satisfactorily correct luminance unevenness, color unevenness, and defective pixels, and measure the temperature. Since the correction value of the defective pixel of the image sensor and the correction value of the dark current value of the image sensor are changed according to the result, even when there is a temperature difference in the image sensor, luminance unevenness, color unevenness and defect Pixel correction can be performed satisfactorily, and it can be said that the effect is very large particularly in a large image sensor.

  Further, by dividing the pixel range of the image sensor into a plurality of areas and changing the correction value of the defective pixel for each area, it is possible to perform defective pixel correction with higher accuracy.

  Further, by constructing the temperature sensor on the same substrate as the image pickup device, it is possible to satisfactorily correct luminance unevenness, color unevenness, and defective pixels without increasing the size and cost of the apparatus.

  Further, the light shielding pixels arranged around the effective pixel portion of the image sensor are divided into a plurality of areas, the dark current values of the light shielding pixels are detected, and the dark current values from the plurality of light shielding pixel areas are detected. Since the temperature is measured and the correction value for the output value from the image sensor is changed according to the temperature measurement results of the plurality of regions, even when there is a temperature difference in the image sensor, luminance unevenness and color unevenness and It is possible to correct defective pixels satisfactorily, and it can be said that the effect is very large particularly in a large-sized image sensor.

  Hereinafter, a method for correcting an imaging signal and an imaging apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a pixel section circuit of a CMOS area sensor according to an embodiment of the present invention.

  In FIG. 1, a photodiode 1, a transfer switch 2, a reset switch 3, a pixel amplifier 10, and a row selection switch 6 are provided in the pixel, and the gate of the transfer switch 2 is ΦTX (n, n + 1) from the vertical scanning circuit 14. ), The gate of the reset switch 3 is connected to ΦRES (n, n + 1) from the vertical scanning circuit 14, and the gate of the row selection switch 6 is connected to ΦSEL (n, n + 1) from the vertical scanning circuit 14. Yes.

  Photoelectric conversion is performed by the photodiode 1, and the transfer switch 2 is in an off state during the accumulation period of the light amount charge, and the charge photoelectrically converted by the photodiode 1 is applied to the gate of the source follower 10 constituting the pixel amplifier. Not transferred. The gate 11 of the source follower 10 constituting the pixel amplifier is initialized to an appropriate voltage by turning on the reset switch 3 before starting the accumulation. That is, this is a dark level. Next or simultaneously, when the row selection switch 6 is turned on, the source follower circuit composed of the load current source 7 and the pixel amplifier 10 is in an operating state. Here, the transfer switch 2 is turned on so that the photodiode is turned on. The charge accumulated in 1 is transferred to the gate of the source follower 10 constituting the pixel amplifier. Here, 4 is a reset power source, and 5 is a power source for driving the source follower 10.

  Here, the output of the selected row is generated on the vertical output line 13. This output is stored in the signal storage unit 15 through the transfer gates 15a and 15b. The output temporarily stored in the signal storage unit 15 is sequentially read out to the output amplifier unit by the horizontal scanning circuit 16.

  FIG. 2 is an operation timing chart of the CMOS area sensor of FIG.

  At the timing of the all-pixel reset period T1, ΦTX (n) and ΦTX (n + 1) become active, and the charges of the photodiodes 1 of all the pixels are transferred to the gate of the source follower 10 via the transfer switch 2. The photodiode 1 is reset. This state is a state where the cathode charge of the photodiode 1 is shifted to the gate of the source follower 10 and is averaged. By increasing the capacitance component of the capacitor 9 at the gate of the source follower 10, the cathode of the photodiode 1 is It becomes the same as the reset level.

  At this time, a mechanical shutter (not shown) that guides the amount of light of the target image is open, and at the end of time T1, accumulation starts simultaneously for all pixels. The mechanical shutter remains open during the period T3, and this period is the accumulation period of the photodiode 1.

  After the elapse of T3 time, the mechanical shutter is closed at the timing of T4, and the photocharge accumulation of the photodiode 1 is completed. In this state, charges are accumulated in the photodiode 1. Next, reading starts for each line. That is, the Nth row is read after the N-1th row is read.

  The period ΦSEL (n) at time T5 becomes active, the row selection switch 6 is turned on, and the source follower circuit composed of the pixel amplifiers 10 of all the pixels connected to the nth row is in an operating state. . Here, the gate of the source follower constituted by the pixel amplifier 10 becomes ΦRES (n) active in the period T2, the reset switch 3 is turned on, and the gate 11 of the source follower 10 is initialized. That is, the dark level signal is output to the vertical output line 13.

  Next, ΦTN (n) becomes active, the transfer gate 15 b is turned on, and is held in the signal storage unit 15. This operation is executed simultaneously in parallel for all the pixels connected to the N rows. When the transfer of the dark level signal storage unit 15 is completed, the signal charge stored in the photodiode 1 is made active by making ΦTX (n) active, so that the transfer switch 2 is turned on. Transfer to the gate 11 of the configured source follower. At this time, the potential of the source follower gate 11 constituted by the pixel amplifier 10 fluctuates from the reset level by an amount corresponding to the transferred signal charge, and the signal level is determined.

  Here, ΦTS becomes active, the transfer gate 15 a is turned on, and the signal level is held in the signal storage unit 15. This operation is executed simultaneously in parallel for all the pixels connected to the N rows. Here, the signal storage unit 15 holds the dark level and the signal level of all the pixels connected to the N rows. By taking the difference between the dark level and the signal level between the pixels, The fixed pattern noise (FPN) due to the threshold voltage Vth variation of the follower and the KTC noise generated when the reset switch 3 is reset can be canceled to obtain a signal from which a noise component having a high S / N is removed. Here, the output from the pixel in which the photodiode 1 is optically shielded is only a dark current component caused by a surface defect or the like of the photodiode 1.

  The horizontal scanning circuit 16 horizontally scans the difference signal between the dark level and the signal level stored in the signal storage unit 15 by the horizontal scanning circuit 16, and outputs the signal in time series at the timing of T7. This completes the output of N rows. Similarly, by driving ΦSEL (n + 1), ΦRES (n + 1), ΦTX (n + 1), ΦTN, and ΦTS in the same way as the Nth row as shown in FIG.

  FIG. 3 schematically shows the chip 20 of the CMOS area sensor of FIG. Here, 20a indicates an effective pixel portion of the area sensor, and 20b is an OB pixel portion in which the pixel portion is shielded from light. Reference numeral 20c denotes an amplifier for amplifying and outputting the signal of the signal storage unit 15 shown in FIG.

  FIG. 3 also shows an image output from the CMOS area sensor of FIG. 1. In this case, the image is in a state where the uniform luminance surface is exposed slightly under, and a dark current is applied to a part of the effective portion 20a. The resulting defective pixel is shown as a white spot.

  FIG. 4 explains FIG. 3 in more detail, and shows the state of a defective pixel due to luminance unevenness, color unevenness, and dark current due to heat generated by the operation of the amplifier 20c. Since the temperature at the upper left of the screen is high, the screen brightness level is apparently high due to the dark current and the number of defective pixels is large.

In order to measure temperatures at a plurality of locations in the chip, temperature sensors A to J are built in the outer peripheral portion of the pixel portion in the chip 20 of the CMOS type area sensor. Further, the entire pixel range is divided into 6 (af), and the average temperature of each part as shown in FIG.
Part a temperature = (A + J) / 2
Part b temperature = (B + J + D) / 3
Part c temperature = (C + D) / 2
d part temperature = (H + I) / 2
e part temperature = (G + I + E) / 3
f part temperature = (E + F) / 2
It is said.

  FIG. 6 shows a table of defective pixel addresses for each temperature, and is prepared for each of the six divisions (af) shown in FIG. That is, if different defective pixel address tables are used depending on the temperature of each part shown in FIG. 5, even if there is a temperature difference in the image sensor, the problem of excessive correction or insufficient correction of defective pixels is solved, and optimal correction is performed. An effect can be obtained.

  On the other hand, regarding the luminance unevenness in the screen, a table of dark current values with respect to temperature is prepared. Next, since the temperatures of the a part to the f part are measured in the description of FIGS. 4 to 5, the measured temperature is the temperature of the pixel located at the central coordinate of each part (af) as shown in FIG. a1 to f1). Then, each pixel temperature within the range of all pixels (p × m pixels) is approximately determined by the least square method using the values of a1 to f1.

  If each pixel temperature is known, the dark current of each pixel is obtained from a table of dark current values with respect to the temperature. Therefore, subtracting the dark current from each pixel output is optimal even if there is a temperature difference in the image sensor. A correction effect can be obtained and image quality deterioration such as luminance unevenness and color unevenness can be prevented.

  Here, as the temperature sensors A to J shown in FIG. 4, a diode or the like may be arranged outside the imaging device and the temperature characteristics thereof may be used. However, the manufacturing process of the imaging device on the same substrate as the imaging device. NPN transistors and diodes that can be manufactured in the above are arranged, and the temperature characteristics thereof may be used.

  In FIG. 4, the temperature sensor has 10 temperature sensors A to J and the number of divisions of the entire pixel range is 6 to a to f. However, by increasing the number within the possible range, more accurate correction is possible. It goes without saying that.

  Further, in place of the temperature sensor in FIG. 4, as shown in FIG. 8, a 30b shading pixel (OB) is provided on the outer peripheral portion of the effective pixel portion 30a in the chip 20 of the CMOS type area sensor, and at the portions A to J. It is possible to divide and measure the temperatures of the portions A to J from the table of dark current values of the OB portion and dark current values with respect to temperature.

  In this embodiment, a CMOS area sensor is described as an example, but any area sensor such as a CCD area sensor may be used.

It is a pixel part circuit diagram of the CMOS type area sensor in the embodiment of the present invention. FIG. 2 is an operation timing chart of the CMOS area sensor of FIG. 1. It is a chip | tip schematic diagram of the CMOS type area sensor of FIG. FIG. 2 is a detailed view of a chip of the CMOS area sensor of FIG. 1. It is the figure which showed the temperature conversion example of each part in the area sensor in embodiment of this invention. It is a table of defective pixel addresses for each temperature in the embodiment of the present invention. It is a figure explaining the brightness nonuniformity correction in a screen in an embodiment of the invention. It is a figure which shows other embodiment of the temperature measurement method in this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Photodiode 2 Transfer switch 3 Reset switch 4,5 Reference power supply 6 Row selection switch 7 Load current source 9 Source follower gate capacitor 10 Source follower 10 Pixel amplifier 11 Source follower gate 13 Vertical output line 14 Vertical scanning circuit 15 Signal storage unit 15a, 15b Transfer gate 16 Horizontal scanning circuit 20, 30 CMOS type area sensor chip 20a, 30a CMOS type area sensor effective unit 20b, 30b Shading pixel (OB) part 20c of CMOS type area sensor 20a Amplifier

Claims (9)

  An imaging apparatus comprising: a plurality of temperature sensors arranged around an imaging element; and a correction value for an output value from the imaging element is changed according to a temperature measurement result by the plurality of temperature sensors.   The imaging apparatus according to claim 1, wherein the correction value for the output value is a correction value for a defective pixel of the imaging device.   The imaging apparatus according to claim 2, wherein the pixel range of the imaging element is divided into a plurality of areas, and the correction value is changed for each area.   The imaging apparatus according to claim 1, wherein the correction value for the output value is a correction value for a dark current value of the imaging element.   2. The imaging apparatus according to 1, wherein the temperature sensor is configured on the same substrate as the imaging element.   The light-shielding pixel arranged around the effective pixel portion of the image sensor is divided into a plurality of areas, and includes a means for detecting a dark current value of the light-shielding pixel, and a table means for dark current values with respect to temperature. The means for measuring the temperature of a plurality of regions from the dark current value from the light-shielding pixel area, and the correction value for the output value from the image sensor is changed according to the temperature measurement result of the plurality of regions. An imaging device that is characterized.   The image pickup apparatus according to claim 6, wherein the correction value for the output value is a correction value for a defective pixel of the image pickup device.   8. The imaging apparatus according to claim 7, wherein the pixel range of the imaging element is divided into a plurality of areas, and the correction value is changed for each area.   The imaging apparatus according to claim 6, wherein the correction value for the output value is a correction value for a dark current value of the imaging element.

Images