JP2006121164A - Imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively reproduce deterioration of image quality due to crosstalk of an image sensor by simple control.
A crosstalk correction coefficient corresponding to an output level of an image sensor is read to control a luminance signal correction circuit.
[Selection] Figure 1

Description

  The present invention relates to an imaging apparatus using an image sensor, and more particularly to an imaging apparatus that corrects crosstalk (charge leakage) to adjacent pixels of an image sensor.

  In the image sensor, crosstalk occurs in which a signal leaks to adjacent pixels. There are two main causes: incident light optically leaks into adjacent pixels and signal charge generated inside the image sensor substrate leaks into adjacent pixels due to diffusion.

  It is generally known that this crosstalk increases as the incident light increases. Due to crosstalk, in a single-plate camera system in which multiple types of color filters are mounted on a single image sensor, color reproducibility is deteriorated by causing color mixing, and the three image sensors are respectively red (R) and green (G ), In a three-plate camera system that transmits blue (B), crosstalk becomes equivalent to a low-pass filter, which degrades the high-frequency component of the luminance signal.

  This crosstalk is more prominent in the CMOS image sensor than in the CCD image sensor.

For this situation, for example, Patent Document 1 includes a data storage unit that stores a correction coefficient for correcting a color mixture component, and a color mixture correction unit that corrects the color mixture component based on the correction coefficient. A correction coefficient is read in accordance with the F value to correct the color mixture component (crosstalk).
Japanese Patent Laid-Open No. 10-271519 JP-A-10-155100

  However, in the above prior art, since the same correction coefficient is read unless the F value (aperture) of the lens is changed, if the change in the aperture cannot follow the change in the amount of incident light, the correction output from the data storage means Since the coefficient is originally different from a necessary value, there is a problem that appropriate crosstalk correction cannot be performed.

  The present invention has been made paying attention to the above points. According to the output level of the image pickup means, the crosstalk correction value is output from the data storage means, and the crosstalk correction is performed, so that the amount of incident light is adjusted. Another object of the present invention is to provide an imaging apparatus that performs appropriate crosstalk correction and can easily reproduce high-frequency components of an image lost due to crosstalk.

  In the present invention, in view of the above-described conventional problems, the crosstalk correction value is output from the data storage unit according to the output level of the imaging unit, and the crosstalk correction is performed, so that the appropriate crosstalk correction according to the incident light amount is performed. Is to do.

  Further, the first invention of the present invention will be described as follows.

  (1) Imaging means, data storage means for storing the crosstalk correction value of the imaging means, luminance signal component means for generating a luminance signal component from the output of the imaging means, and a luminance signal for correcting the luminance signal component Correction means, and outputs the crosstalk correction value from the data storage means according to the output level of the imaging means, and controls the luminance signal correction means with the crosstalk correction value. Imaging device.

  According to the present invention, the crosstalk correction value is output from the data storage unit according to the output level of the image pickup unit, and the crosstalk correction is performed to perform an appropriate crosstalk correction according to the amount of incident light. Therefore, the high frequency component of the lost image can be easily reproduced.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to examples.

An image pickup apparatus to which the first embodiment of the present invention is applied is shown in FIG. FIG. 1 shows a three-plate image pickup apparatus, and a CMOS sensor is used as an image pickup element. Input signals obtained by photoelectrically converting red (R), green (G), and blue (B) by the CMOS sensors 1 to 3 are subjected to noise removal, gain adjustment, and analog-digital by analog front end (AFE) circuits 4 to 6, respectively. (AD) conversion or the like is performed. Thereafter, for example, the luminance signal generation circuit 7
Y = 0.3 * R + 0.59 * G + 0.11 * B Formula (1)
A luminance signal is generated by the calculation of Expression (1). The output of the luminance signal generation circuit 7 is processed by the luminance signal correction circuit 8 such as aperture correction and gamma correction.

  Outputs of the AFE circuits 4 to 6 are also sent to the color signal processing circuit 10 to perform processing such as gamma correction and white balance, and generate color difference signals RY and BY. The output of the color signal processing circuit 10 is multiplexed with the output of the luminance signal correction circuit 8 by the YC multiplexing circuit 11.

  The outputs of the AFE circuits 4 to 6 are also input to the crosstalk amount storage means 9. The crosstalk amount storage means 9 stores the crosstalk amount by an operation described later in accordance with an instruction from the microcomputer 12 and outputs a crosstalk correction value k to the luminance signal correction circuit 8 at the time of crosstalk correction.

  FIG. 2 shows an internal circuit of the crosstalk amount storage means 9. Output signals of the AFE circuits 4 to 6 are added by the level synthesis circuit 91. Although the addition ratio is not particularly limited, the level synthesis circuit 91 is not provided and the output of the luminance signal generation circuit 7 when the addition ratio common to the equation (1) used in the luminance signal generation circuit 7 is used. May be input to the crosstalk amount storage means 9.

  The output of the level synthesis circuit 91 is stored in the memory 92 and simultaneously input to the selector 94.

  The counter 93 receives a signal CEN for allowing the counter to be incremented from the microcomputer 12, performs an increment every time a clock (not shown) is input, and sends the value to the selector 94 while the CEN is valid. The selector 94 inputs from the microcomputer 12 a TEST signal that becomes “1” when the crosstalk amount is measured. When the TEST = “0” is normally used, the output of the level synthesis circuit 91 is selected and the TEST = “1”. When measuring the amount of crosstalk, the output of the counter 93 is selected. The output of the selector 94 becomes an address signal of the memory 92. The microcomputer 12 also sends to the memory 92 WEN that permits writing to the memory 92.

  FIG. 3 shows an internal circuit of the luminance signal correction circuit 8. In the aperture correction circuit 81, a high-pass filter (HPF) 811 extracts the high frequency of the signal. The gain generation circuit 812 generates an aperture gain APCG by means of a hardware configuration or a ROM table or the like based on the output k of the crosstalk amount storage unit 9 of FIG. 1, and a multiplier 813 multiplies the output of the HPF 811 and the APCG. . An adder 814 adds the output of the luminance signal generation circuit 7 and the output of the multiplier 842 to perform aperture correction. The output of the aperture correction circuit 81 is sent to the gamma correction circuit 82 for gamma correction.

  FIG. 4 shows an example of the relationship between the amount of incident light and the amount of crosstalk at that time. In order to obtain the characteristics as shown in FIG. 4, it is necessary to measure the crosstalk with respect to the incident light during factory adjustment or the like and store it in the memory 92. The memory 92 is preferably a non-volatile memory, but is not limited to this.

  FIG. 8 is an explanatory diagram of the principle of crosstalk measurement. 8A, 8B, and 8C are extracted from three pixels of the CMOS sensors 1 to 3, and each pixel includes a microlens 801, a color filter 802, a wiring layer 803, a photodiode 804, and a silicon substrate 805. At the time of crosstalk measurement, a part of the CMOS sensors 1 to 3 is covered with a light shielding plate 810. FIG. 8 shows an example in which the pixel 8A is covered with a light shielding plate 810, and the pixels 8B and 8C receive the incident light 811 as usual. Since crosstalk is leakage of electric charge to adjacent pixels through the silicon substrate 805, the amount of crosstalk with respect to the incident light 811 can be measured by measuring the output value of the light-shielded pixel 8A.

  FIG. 11 is a flowchart showing the procedure of crosstalk measurement. When the measurement is started in step (S) 1101, the microcomputer 12 sets the TEST signal to “1”. In S1102, a partial area of the CMOS sensor is shielded by the light shielding plate 810. In step S1103, the microcomputer 12 outputs a WEN signal to enable writing to the memory 92. Next, in S1104, a certain level of incident light 711 is irradiated. In S1105, the output level of the pixel at the end pixel (8A in FIG. 8) of the shielded region is measured. As shown in FIG. 1, when a CMOS sensor is used as an image sensor, a pixel at a desired position can be read freely. The signal read here is written in the memory 92 in S1106. The address value at this time is the output value of the counter 93 controlled by the microcomputer 12. The address value at this time is made to coincide with the level value of the incident light 711. If it is determined in S1107 that the measurement is continued, the address is incremented (S1108). In S1109, the light source (not shown) is controlled so as to change the irradiation level corresponding to the increment of the address, and irradiation is performed again in S1104.

  If it is determined in S1107 that the measurement is to be terminated, the WEN signal from the microcomputer 12 is changed in S1110 to disable writing to the memory 92. In step S1111, the TEST signal is set to “0” in response to an instruction from the microcomputer 12, and the crosstalk measurement ends.

  During normal use, crosstalk is corrected using crosstalk characteristics measured during factory adjustment. The operation will be described below.

  In the three-plate camera system as shown in FIG. 1, when crosstalk occurs, a luminance signal component leaks into adjacent pixels. This is equivalent to passing the image through a low-pass circuit (LPF). Therefore, the high frequency component of the image is deteriorated, resulting in a blurred image. Therefore, in the present invention, the aperture correction circuit 81 and the gamma correction circuit 82 are controlled by the measured crosstalk amount to reproduce the deteriorated high frequency component, thereby performing the crosstalk correction.

  By the above-described crosstalk measurement operation, the crosstalk amount storage unit 9 outputs the crosstalk amount k with respect to the input signal level, so that the optimum crosstalk correction can be performed for the level of the input signal.

  In the present invention, the input / output characteristics of the gain generation circuit 812 are as shown in FIG. In FIG. 6A, the aperture gain APC is increased as the value of the crosstalk amount k increases. As a result, when the amount of crosstalk is relatively large and the high frequency of the image has dropped significantly, the aperture correction is strengthened to reproduce the high frequency component lost by the crosstalk.

  Further, the input / output characteristics of the gain generation circuit 812 may be determined by a plurality of straight lines as shown in FIG. 6B.

  When the aperture correction is increased, the added edge component is likely to saturate to the high level side, and the edge portion may become white or glaring. Therefore, it is necessary to control the gamma correction to prevent edge component saturation. The input / output characteristics of the gamma correction circuit 82 at this time are shown in FIG. When the value of k is low, that is, when the amount of crosstalk is small, the aperture gain APCG is low and edge saturation is not likely to occur, so there is no need to suppress the output in the high luminance region, but when the value of k is high, When the amount of crosstalk is large, the aperture gain APCG is high and edge saturation is likely to occur. Therefore, it is necessary to change the gamma input / output characteristics to suppress the output in the high luminance region. Note that the implementation method of the gamma correction circuit 82 may be configured by hardware or a ROM table.

  FIG. 12 shows the internal configuration of the crosstalk amount storage means 9 applied in the second embodiment of the present invention. In the first embodiment, as a method of generating the address of the memory 92 at the time of crosstalk measurement, a value corresponding to the incident light 811 is generated by the counter 93. However, in this embodiment, the crosstalk amount is shielded. In the clock period immediately before reading out from the pixel 8A, the signal of the pixel 8B that is not shielded, that is, the incident light amount is read out and accumulated in the D flip-flop (DFF) 95, and in the next clock period, crosstalk is performed from the shielded pixel 8A. The amount is read and written to the memory 92 using the output of the DFF 95 as an address. With this operation, a look-up table that outputs a crosstalk amount k corresponding to the amount of incident light can be realized as in the first embodiment.

  FIG. 9 shows an internal circuit of the luminance signal correction circuit 8 applied in the third embodiment of the present invention.

  In general, when crosstalk occurs inside an image sensor, the way in which high frequency components of an image are deteriorated varies depending on the crosstalk amount k. That is, since the LPF characteristic equivalent to crosstalk changes, in this embodiment, the filter characteristic of the aperture correction circuit 81 is switched according to the crosstalk amount k.

  When the crosstalk amount k is large, that is, when k ≧ a, the HPF 811 having the frequency characteristics shown in FIG.

  In this case, since the level difference from the adjacent pixels is likely to disappear due to the leakage of electric charges, the highest range of the luminance signal is deteriorated. Therefore, a filter having frequency characteristics as shown in FIG. 7A is used to reproduce the lost highest band.

  When the crosstalk amount k is small, that is, when k <a, the selector 816 selects the BPF 815 having the frequency characteristics shown in FIG. At this time, the level difference from the adjacent pixels is kept to some extent even though there is a charge leakage, so that a filter having frequency characteristics as shown in FIG. 7B is used to correct the middle range.

  FIG. 10 shows an internal configuration of the camera signal processing circuit 8 applied in the fourth embodiment of the present invention. In the third embodiment, an example in which two filters and a selector are used as an example of switching the filter characteristics of the aperture correction circuit 81 in accordance with the crosstalk amount k has been described. However, in this embodiment, the filter 818 uses the parameter k. Switch the tap coefficient or number of taps.

  Other operations are the same as those of the third embodiment.

The figure which shows the 1st Example of this invention Configuration diagram of crosstalk amount storage means Configuration diagram of luminance signal correction circuit Input / output characteristics diagram showing an example of the relationship between the amount of incident light and the amount of crosstalk Input / output characteristics of gamma correction circuit Aperture gain characteristics Filter frequency characteristics Diagram showing the principle of crosstalk measurement Configuration diagram of luminance signal correction circuit used in the third embodiment of the present invention Configuration diagram of luminance signal correction circuit used in the fourth embodiment of the present invention Flow chart showing the procedure for crosstalk measurement Configuration diagram of crosstalk amount storage means used in the second embodiment of the present invention

Explanation of symbols

1, 2, 3 MOS sensor 4, 5, 6 AFE
7 luminance signal generation circuit 8 luminance signal correction processing circuit 9 crosstalk amount storage means 10 color signal processing circuit 11 YC multiplexing circuit 12 microcomputer

Claims (5)

Imaging means, data storage means for storing the crosstalk correction value of the imaging means, luminance signal component means for generating a luminance signal component from the output of the imaging means, and luminance signal correction means for correcting the luminance signal component Have
An imaging apparatus, wherein the crosstalk correction value is output from the data storage unit in accordance with the output level of the imaging unit, and the luminance signal correction unit is controlled by the crosstalk correction value.   2. The imaging apparatus according to claim 1, wherein an aperture correction unit and a gamma correction unit in the luminance signal correction unit are controlled by the crosstalk correction value.   The imaging apparatus according to claim 2, wherein an aperture gain in the aperture correction unit is controlled by the crosstalk correction value.   The imaging apparatus according to claim 2, wherein a frequency characteristic of a filter in the aperture correction unit is switched by the crosstalk correction value.   The imaging means includes a first imaging element that transmits first color light, a second imaging element that transmits second color light, and a third imaging element that transmits third color light. The imaging apparatus according to claim 1, wherein:

Images