JP2004343623A - Solid-state imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a solid-state imaging device capable of reducing capacity of a frame memory without substantially lowering image quality, promptly processing calibration correction and/or performing calibration correction without redundancy. <P>SOLUTION: The solid-state imaging device is provided with a LOG image sensor 11, an operational amplifier 12 which adjusts a dynamic range of an output signal of the sensor 11, an AD converter 13 for performing AD conversion of an output signal from the amplifier 12, a frame memory 15 which holds calibration data and an image processing unit 14 which generates image data and outputs it to a monitor. The number M of bits of the converter 13 and the number N of bits of the memory 15 are in relation of M>N and the image processing unit 14 performs correction of the output signal from the converter 13 based on the calibration data and generates image data. The number N of bits to be used for correction to the number M of bits of the AD conversion of the converter 13 may be M>N. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a solid-state imaging device, and more particularly to a solid-state imaging device that outputs image data to a monitor or the like based on an output of an image sensor.
[0002]
[Prior art and problems]
Patent Document 1: JP-A-11-298798 Patent Document 2: JP-A-2001-197369
In general, in a solid-state imaging device, since an image sensor has a variation in output characteristics of each pixel, uniform light is made incident on the image sensor in advance, and output data of each pixel is corrected with calibration data (correction data such as shading and fixed pattern noise). ) Is stored in a memory, correction is performed based on the calibration data at the time of actual imaging, and output as image data.
[0004]
Particularly, in a recently developed LOG image sensor (for example, see Patent Document 1), since an output signal that changes logarithmically according to the amount of incident light is generated, the output characteristics of each pixel greatly vary, and calibration is performed. It is essential.
[0005]
For this purpose, Patent Document 2 proposes a solid-state imaging device in which shading data when a LOG image sensor is irradiated with uniform light is compressed and stored, and when correction is performed, the stored data is expanded. .
[0006]
That is, in the conventional solid-state imaging device, if the number of bits of the AD converter that performs AD conversion of the sensor output is M, the number of bits of the frame memory that holds the calibration data is also M, and the frame memory is (total pixels). It required a capacity of (several × M) bits.
[0007]
Further, when the number of bits of the AD converter is increased in order to precisely perform the A / D conversion of the imaging data, the number of bits of the frame memory is increased in proportion to this. That is, when the number of bits of the AD converter is increased by 2 bits, the capacity of the frame memory needs to be increased by (total number of pixels × 2) bits. Further, since an increase in the number of bits means an increase in the number of gradation steps after AD conversion, more precise calibration is possible. However, even if the calibration is performed with a certain accuracy or higher, the image quality of the obtained image is not significantly improved, so that a redundant calibration is performed.
[0008]
Therefore, it is an object of the present invention to (1) reduce the capacity of the frame memory, (2) quickly process the calibration correction, and / or (3) redundancy without substantially reducing the image quality. It is an object of the present invention to provide a solid-state imaging device capable of performing calibration correction without error.
[0009]
Configuration, operation and effect of the present invention
In order to achieve the above object, a solid-state imaging device according to a first aspect of the present invention includes an image sensor that generates an output signal according to an incident light amount, an operational amplifier that amplifies an output signal of the image sensor, and an output of the operational amplifier. An A / D converter for performing A / D conversion, a frame memory for holding calibration data, and an image processor for correcting an output from the A / D converter based on the calibration data held in the frame memory to generate image data Wherein the number of bits M of the AD converter and the number of bits N of the frame memory are in a relationship of M> N.
[0010]
The calibration data obtained by irradiating the image sensor with uniform light is data according to the number of bits of the AD converter. However, the data of the lower number of bits is not so important as the calibration data, and is ignored. Even if the correction is made, the image quality does not decrease in practical use. In the solid-state imaging device according to the first aspect, by setting the number of bits M of the AD converter and the number of bits N of the frame memory in a relationship of M> N, the capacity of the frame memory can be reduced without deteriorating the image quality. {Total number of pixels × (M−N)} bits can be reduced, and the cost of the solid-state imaging device can be reduced.
[0011]
A solid-state imaging device according to a second aspect of the present invention provides an image sensor that generates an output signal according to the amount of incident light, an operational amplifier that amplifies the output signal of the image sensor, an AD converter that performs AD conversion on the output of the operational amplifier, A frame memory for holding image data, and an image processor for correcting the output from the AD converter based on the calibration data held in the frame memory to generate image data. Wherein the number of bits N used for the calibration correction is M> N for the number of bits M of the AD conversion.
[0012]
In the solid-state imaging device according to the second aspect, since the number of bits N used for calibration correction is M> N with respect to the number of bits M of the AD conversion of the AD converter, the image quality is not reduced. Calibration correction can be processed quickly.
[0013]
A solid-state imaging device according to a third aspect of the present invention provides an image sensor that generates an output signal according to the amount of incident light, an operational amplifier that amplifies an output signal of the image sensor, an AD converter that performs AD conversion on the output of the operational amplifier, A frame memory for holding image data, and an image processor for correcting the output from the AD converter based on the calibration data held in the frame memory to generate image data. Wherein the number of bits N of the calibration data is smaller than the number of bits M of the AD conversion by 2 or more.
[0014]
In the solid-state imaging device according to the third aspect of the invention, the number of bits N of the calibration data is smaller than the number of bits M of the AD conversion by the AD converter by 2 or more. The total number of pixels × 2) bits or more can be reduced, and the calibration correction can be performed quickly.
[0015]
A solid-state imaging device according to a fourth aspect of the present invention provides an image sensor that generates an output signal corresponding to an amount of incident light, an operational amplifier that amplifies an output signal of the image sensor, an AD converter that performs AD conversion on the output of the operational amplifier, A frame memory for holding the calibration data, and an image processor for correcting the output from the AD converter based on the calibration data held in the frame memory to generate image data. The data is created by reducing the number of bits of the original data output from the AD converter for all pixels by the same processing method.
[0016]
In the solid-state imaging device according to the fourth aspect of the invention, the calibration data is created by reducing the number of bits of the original data output from the AD converter by the same processing method for all pixels, so that the image quality can be reduced without reducing the image quality. The memory capacity can be reduced by (the total number of pixels × the number of reduced bits) bits, and the calibration correction can be performed quickly.
[0017]
In the solid-state imaging device according to the first, second, third, and fourth inventions, the image sensor may be a LOG image sensor that generates an output signal that changes logarithmically according to the amount of incident light. Since the LOG image sensor has a greater variation in output characteristics of each pixel than other sensors (such as a CCD image sensor), calibration correction is indispensable, and the present invention can be applied effectively.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of a solid-state imaging device according to the present invention will be described with reference to the accompanying drawings.
[0019]
(1st Embodiment, see FIG. 1)
FIG. 1 shows a block diagram of a solid-state imaging device 1 according to a first embodiment of the present invention. The solid-state imaging device 1 includes a LOG image sensor 11, an operational amplifier 12, an AD converter 13, an image processor 14, and a frame memory 15, and a monitor such as a CRT is connected to the image processor 14.
[0020]
The LOG image sensor 11 has a LOG characteristic as a photoelectric conversion characteristic, generates an output signal that changes logarithmically according to the amount of incident light, and can image a range of six digits in luminance at a time. For example, the solid-state imaging device described in Patent Document 1 can be used.
[0021]
The operational amplifier 12 is for amplifying the output signal of the LOG image sensor 11, and the gain is adjusted so that the range of 2 to 4 digits of the dynamic range becomes a displayable range of a monitor such as a CRT. This is because the LOG image sensor 11 has a dynamic range of six digits with respect to the luminance, but the illuminance distribution of the subject is about four digits at the maximum, and in most cases can be accommodated in two to three digits.
[0022]
The AD converter 13 is for AD-converting the output of the operational amplifier 12, and its bit number is M.
[0023]
The frame memory 15 holds calibration data. The calibration data is data relating to variations in output characteristics of each pixel of the LOG image sensor 11 and is obtained by irradiating the LOG image sensor 11 with uniform light. It is assumed that the number of bits of the frame memory 15 is N.
[0024]
The image processor 14 performs a necessary correction on the imaging data after the AD conversion, in particular, a correction of the imaging data based on the calibration data read from the frame memory 15 for each pixel in real time. Generate and output to monitor.
[0025]
In the solid-state imaging device 1 having the above configuration, the number M of bits of the AD converter 13 and the number N of bits of the frame memory 15 are set to have a relationship of M> N. When the calibration data obtained by irradiating the LOG image sensor 11 with the uniform light is stored in the frame memory 15, the upper N bits of the output data (M bits) of the AD converter 13 are stored as the calibration data.
[0026]
On the other hand, when correcting the imaging data with the calibration data in order to remove fixed pattern noise or the like from the imaging data of the LOG image sensor 11, the calibration data stored in the frame memory 15 is changed to (M−N). The data is shifted to the bit MSB side, that is, multiplied by 2 ^(M−N) , converted into M-bit data, and subtracted from the imaging data.
[0027]
By the way, the substantial number of bits necessary for removing the noise component from the M-bit image data is a part of the higher-order bits of the calibration data stored in the frame memory 15. Assuming that the number of bits required to remove this noise component is N, the number of bits required for the frame memory 15 is N, and the frame memory 15 stores the upper N bits of the M bits output from the AD converter. It is only necessary to store bit data, and even if data corresponding to the lower (MN) bits is truncated, there is substantially no adverse effect on image quality.
[0028]
It is clear from experiments that even if such correction is performed uniformly for all pixels, the image quality of the LOG image sensor 11 does not deteriorate.
[0029]
Specifically, as an example, an example in which the 10-bit AD converter 13 and the 6-bit frame memory 15 are used to perform calibration correction using the upper 6 bits is shown.
[0030]
When illuminating the LOG image sensor 11 with uniform light and storing the calibration data in the frame memory 15, the upper 6 bits of the 10 bits output from the AD converter 13 are stored in the frame memory 15. That is, the upper 6 bits are uniformly selected and stored for the AD output data of all the pixels.
[0031]
When removing fixed pattern noise or the like from image data captured by the LOG image sensor 11 and outputting the image data as image data, the image processor 14 converts the 6-bit calibration data read from the frame memory 15 into 4 bits MSB side. shifted (multiplication 2 ^4), into a 10-bit data, is subtracted from the image data.
[0032]
By performing such processing, the capacity of the frame memory 15 can be reduced without substantially lowering the image quality of the image data.
[0033]
(Second embodiment, see FIG. 2)
FIG. 2 shows a block diagram of a solid-state imaging device 2 according to a second embodiment of the present invention. As in the first embodiment, the solid-state imaging device 2 includes a LOG image sensor 11, an operational amplifier 12, an AD converter 13, an image processor 14, and a frame memory 15 '.
[0034]
The LOG image sensor 11, the operational amplifier 12, the AD converter 13, and the image processor 14 are the same as those in the first embodiment, and a description thereof will be omitted.
[0035]
The frame memory 15 'holds calibration data similarly to the frame memory 15. In the second embodiment, a general-purpose device having M bits or more is used. The number is N.
[0036]
In the solid-state imaging device 2 having the above configuration, the number of bits M of the AD converter 13 and the number of bits N used for calibration correction are set to have a relationship of M> N. When the calibration data obtained by irradiating the LOG image sensor 11 with the uniform light is stored in the frame memory 15 ′, the output data (M bits) of the AD converter 13 is stored as the calibration data.
[0037]
On the other hand, when correcting the imaging data with the calibration data in order to remove fixed pattern noise or the like from the imaging data of the LOG image sensor 11, the image processor 14 converts the calibration data stored in the frame memory 15 ′ into noise. The 6-bit data required for the removal of the component is read, and the data is shifted to the (M−N) bits MSB side, that is, multiplied by 2 ^(M−N) , converted to M-bit data, and imaged. Subtract from data.
[0038]
As described above, the substantial number of bits necessary for removing the noise component from the M-bit image data is a part of the higher-order bits of the calibration data stored in the frame memory 15 '. In the second embodiment, even if a general-purpose device having M bits or more is used for the frame memory 15 ', the number of bits used for calibration correction is suppressed to N, and the calibration correction is not substantially adversely affected. Efficiency can be improved.
[0039]
In the second embodiment, the general-purpose device used as the frame memory 15 ′ does not necessarily need to have M bits or more, but need only have N bits or more, which is the minimum necessary for calibration correction.
[0040]
In addition, M-bit calibration data is stored in the frame memory 15 ′, and calibration correction is performed by selecting N bits in a monochrome mode, for example, and in a color mode according to an imaging mode. It is possible to select to perform calibration correction based on M-bit data.
[0041]
Specifically, as an example, an example is shown in which correction is performed using 6 bits of the AD-converted calibration data using a 10-bit AD converter 13 and a 16-bit frame memory 15 ′.
[0042]
When illuminating the LOG image sensor 11 with uniform light and storing the calibration data in the frame memory 15 ', the output 10 bits of the AD converter 13 are changed to 16 bits, that is, 0 is stored in the upper 6 bits. Then, it is stored in the frame memory 15 '. This process is performed uniformly for the AD output data of all pixels.
[0043]
When removing fixed pattern noise or the like from the image data captured by the LOG image sensor 11 and outputting the image data as image data, the image processor 14 sets the LSB side of the 16-bit calibration data stored in the frame memory 15 ′. read a more 5 of 6 bits by selecting from bit to 10 bit data, and bit-shifting four bits MSB side (multiplication 2 ^4), into a 10-bit data, it is subtracted from the image data.
[0044]
By performing such processing, the calibration correction can be performed at high speed without substantially lowering the image quality of the image data, and the calibration correction according to the imaging mode can be selected.
[0045]
(Other embodiments)
The solid-state imaging device according to the present invention is not limited to the above-described embodiment, and can be variously modified within the scope of the gist.
[0046]
For example, although the solid-state imaging devices according to the first and second embodiments are configured by an external operational amplifier, an AD converter, an image processor, and a frame memory, these are built into a LOG image sensor to be used. It may be a chip. In addition, some elements in the solid-state imaging devices of the first and second embodiments may be configured as an inseparable single element integrated in terms of configuration or function.
[0047]
In the solid-state imaging devices according to the first and second embodiments, the case where the number of bits M of the AD conversion of the AD converter is 10 and the number of bits N used for the calibration correction is 6 has been described. N may be set to be smaller than M, and is preferably set to be smaller by 2 or more.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a first embodiment of a solid-state imaging device according to the present invention.
FIG. 2 is a block diagram illustrating a second embodiment of the solid-state imaging device according to the present invention.
[Explanation of symbols]
1, 2, solid-state imaging device 11, LOG image sensor 12, operational amplifier 13, AD converter 14, image processors 15, 15 ', frame memory

Claims (5)

An image sensor that generates an output signal according to the amount of incident light;
An operational amplifier for amplifying an output signal of the image sensor;
An AD converter that performs AD conversion on the output of the operational amplifier;
A frame memory for holding calibration data,
An image processor that corrects the output from the AD converter based on the calibration data held in the frame memory and generates image data,
A solid-state imaging device, wherein the number M of bits of the AD converter and the number N of bits of the frame memory have a relationship of M> N. An image sensor that generates an output signal according to the amount of incident light;
An operational amplifier for amplifying an output signal of the image sensor;
An AD converter that performs AD conversion on the output of the operational amplifier;
A frame memory for holding calibration data,
An image processor that corrects the output from the AD converter based on the calibration data held in the frame memory and generates image data,
A solid-state imaging device, wherein the number of bits N used for the calibration correction is M> N with respect to the number of bits M of the AD conversion of the AD converter. An image sensor that generates an output signal according to the amount of incident light;
An operational amplifier for amplifying an output signal of the image sensor;
An AD converter that performs AD conversion on the output of the operational amplifier;
A frame memory for holding calibration data,
An image processor that corrects the output from the AD converter based on the calibration data held in the frame memory and generates image data,
A solid-state imaging device, wherein the number of bits N of the calibration data is smaller by 2 or more than the number of bits M of AD conversion of the AD converter. An image sensor that generates an output signal according to the amount of incident light;
An operational amplifier for amplifying an output signal of the image sensor;
An AD converter that performs AD conversion on the output of the operational amplifier;
A frame memory for holding calibration data,
An image processor that corrects the output from the AD converter based on the calibration data held in the frame memory and generates image data,
The solid-state imaging device according to claim 1, wherein the calibration data is generated by reducing the number of bits of the original data output from the AD converter for all pixels by the same processing method. The solid-state imaging device according to claim 1, wherein the image sensor is a LOG image sensor that generates an output signal that changes logarithmically according to the amount of incident light. .

Images