JP2002125155A - Solid-state image pickup device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly correct noise depending on a state of noise at photographing. SOLUTION: The solid-state image pickup device is provided with a noise generating means that generates noise component data to correct the noise of an image pickup element. The noise generating means generates the noise component data by each photographing of one image pattern to apply noise correction to an obtained image signal, thereby obtaining the noise component data corresponding to the photographing state and correcting the noise depending on the state of noise at photographing.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a solid-state imaging device used for a video camera or the like.

[0002]

2. Description of the Related Art In a solid-state imaging device used in a video camera or the like, it is known that noise is generated due to various factors in a signal processing process. inside that,
Conventionally, two types of fixed pattern noise in a solid-state image sensor have been known: random uneven noise that produces an image like frosted glass, and streak noise arranged vertically or horizontally.

[0003] As a method of correcting these noises,
Conventionally, a method using a circuit as shown in FIG. 9 has been used. That is, the noise component data of the image sensor for one screen is stored in the frame memory 52 in advance.
Then, when outputting the imaging signal generated by the imaging element 51 to the process circuit 7, the subtractor 6 subtracts the noise component stored in the frame memory 52 from the imaging signal and outputs the result.

[0004]

However, in the above-mentioned conventional example, a frame memory for one screen having a large memory amount is required to store the noise component data of the image pickup device. And the problem of cost increase. By the way, in order to reduce the size of the device, noise correction may be performed using a line buffer that can save a memory amount. But,
In this case, there are the following problems. That is, since streak-like noise is generally most noticeable among fixed pattern noises, it is necessary to remove this noise effectively. However, when a line buffer is used, there is a problem that the streak-like noise cannot be accurately corrected because it is affected by uneven noise, which is another fixed pattern noise of the image sensor. .

Also, since the noise component data of the image pickup device stored in the frame memory is fixed, it is not possible to effectively remove, for example, dark current uneven noise that changes according to a change in temperature. There was a problem. Further, in the above-described conventional example, even when it is not necessary to correct the imaging signal, the correction is uniformly performed. In this case, instead of improving the image quality,
On the contrary, there is a problem that it deteriorates.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and has as its object to enable appropriate noise correction according to the state of noise at the time of photographing.

[0007]

According to the present invention, there is provided a solid-state imaging device comprising: an image sensor for generating an image signal by photoelectric conversion; an image memory for storing an image signal generated by the image sensor; A noise generator configured to generate noise component data for performing the correction, and a subtractor configured to subtract the noise component data generated by the noise generator from the image signal stored in the image memory; Is stored in the image memory, noise component data is generated by the noise generation means, and the noise component data is subtracted from the image signal by the subtractor and output.

[0008] The noise generating means includes:
Light shielding means for shielding light from a subject to the image sensor, wherein the noise component data is generated by the image sensor in a dark state in which light from the subject is shielded by the light shielding means. Is also good.

Further, a decision means for deciding whether or not to perform the subtraction processing according to the state of the image signal stored in the image memory may be provided. Further, when increasing the gain of the image signal stored in the image memory, the above-described subtraction process may be performed without fail.

Another feature of the present invention is that an imaging area for generating an image signal by photoelectric conversion, light shielding means for shielding light from a subject to the imaging area, and A driving unit for performing a first operation of reading an image signal generated in the imaging region, a second operation of reading a noise signal generated in the imaging region in a state where the light is shielded by the light shielding unit, and a noise read by the driving unit. It is characterized by comprising a correcting means for correcting a read image signal using a signal, and a control means for controlling whether or not to read the noise signal according to an imaging condition.

When reading out the noise signal, the control means performs the first operation by the driving means, performs the second operation after the first operation, and performs the first operation. And after performing the second operation, controlling the image signal to be corrected by the correction unit, and when not reading out the noise signal, performing the first operation by the driving unit and performing the first operation. After the first operation, control may be performed such that the second operation by the driving unit and the correction of the image signal by the correction unit are not performed.

As described above, according to the present invention, the noise generating means for generating the noise component data of the image sensor is provided.
By generating noise component data for each screen shot, it is possible to obtain noise component data according to the shooting situation, and it is possible to perform noise correction according to the state of noise during shooting. Furthermore, it is determined whether or not to perform the noise correction according to the state of the image signal of the subject. When the correction is not necessary, the noise correction is not performed. By performing the correction, optimal noise correction according to the state of the photographed subject can be performed. In addition, by not performing the noise correction according to the shooting conditions, the shooting time can be reduced when the noise correction is not performed.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the solid-state imaging device according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a solid-state imaging device according to a first embodiment of the present invention. In FIG. 1, the portions denoted by the same reference numerals as those in the conventional solid-state imaging device shown in FIG. 9 indicate the same functional portions as those in the conventional solid-state imaging device. In the first embodiment, a signal line to which a photosensitive pixel is not connected is provided in a part of an image sensor, and a signal obtained from this signal line is used as noise component data of the image sensor, so that an influence of the photosensitive pixel is obtained. This makes it possible to perform noise correction without any noise, and to reliably remove noise components in the image sensor without increasing the size of the entire apparatus.

The image pickup device 1 is driven by a driver 8. The driver 8 is driven by a drive pulse PDR generated from a synchronization signal generation circuit 9. The output signal from the image sensor 1 is switched by the switch 2, and when the movable terminal 2 a is switched to the fixed terminal A side, the terminal A
Is applied to one end of the subtractor 6 as an imaging output Y _S.
When the movable terminal 2a is switched to the fixed terminal B side, the output signal from the image sensor 1 is transmitted from the terminal B to the A / D
The data is input to the converter 3 and stored in the line buffer 4 as digital noise component data.

This noise component data is D / A converted by the D / A converter 5 and is converted into analog noise component data Y.
_N is input to the other end of the subtractor 6. Then, the noise component data Y _N is subtracted from the above-described imaging output Y _S to obtain a correction signal Y. The correction signal Y generated in this manner is input to the process circuit 7 and subjected to processing such as white clipping and aperture correction to produce the video signal S.
output as v.

FIG. 2 is a diagram showing a detailed configuration of the image pickup device 1, in which a MOS type image pickup device is taken as an example. That is, one pixel includes a MOS-FET (MOS type field effect transistor) 11 and a photodiode 12. A predetermined number of pixels are regularly arranged in the horizontal direction and the vertical direction, respectively, to form a photosensitive portion. The charge generated in the photosensitive section is read out via a horizontal signal line 13 and a vertical signal line 14 connected to each pixel.

The charge read operation will be described in more detail. First, in the vertical direction, an address signal ADR specifying a read line is input to the decoder 15, and the specified read line is transmitted to the vertical register 16. Then, the designated horizontal lines 13 are sequentially scanned in accordance with the address signal ADR. Next, in the horizontal direction, the signal of each pixel connected to one horizontal line specified by the address signal ADR is applied to the horizontal register 1 via the vertical signal line 14.
7 are sequentially output.

The present embodiment is characterized in that a dummy line 18 to which no photosensitive pixel is connected is provided at the top of this horizontal line. The dummy line 18 is designated when a fixed pattern noise in the vertical direction described later is captured.

Next, the overall operation of the solid-state imaging device of this embodiment shown in FIG. 1 will be described in detail with reference to the timing chart of FIG.

FIG. 3A shows a vertical synchronization signal VD.
And (b) shows the vertical blanking signal VB. First, when the timing enters a vertical blanking period as shown in FIG. 3B, the address signal ADR becomes
As shown in (c), the dummy line 18 shown in FIG. 2 is specified. As described above, since no photosensitive pixel is connected to the dummy line 18, the signal output from the dummy line 18 is only a noise component, and is used as fixed pattern noise of the image sensor 1.

In the vertical blanking period,
As shown in FIG. 3D, the switch signal PSW is supplied to the switch 2 so that the movable terminal 2a selects the fixed terminal B side. Thereby, the noise component data output from the image sensor 1 is input to the A / D converter 3. Furthermore,
The A / D converter pulse PCNT1 is turned on as shown in FIG. 3E, and the A / D converter 3 starts operating. Thus, the noise component data for one horizontal line is A / D converted, and this is stored in the line buffer 4 as a digital signal.

Next, when the vertical blanking period ends and the valid period starts, as shown in FIG. 3D, the switch signal PSW is applied to the switch 2 so that the movable terminal 2a selects the fixed terminal A side. Given. As a result, the imaging output Y _S in the effective line designated by the address signal ADR during the effective period is input to one end of the subtractor 6.

During this valid period, the D / A converter pulse PCNT2 for driving the D / A converter 5 is:
As shown in FIG. 3 (f), the noise component data stored in the line buffer 4 during the above-described vertical blanking period is converted into analog noise component data Y _N and input to the other end of the subtractor 6 as shown in FIG. Is done. In the subtractor 6, Y =
Y _S -Y _N becomes operation is performed, thereby the fixed pattern noise correction signal Y component is removed is obtained. Then, the correction signal Y is input to the process circuit 7, where a synchronizing signal and the like are added thereto and output as a video signal Sv.

In the first embodiment described above, the case where the subtraction processing in the subtractor 6 is performed using an analog signal has been described. However, the subtraction processing may be performed using a digital signal. FIG. 4 is a diagram illustrating a solid-state imaging device according to a second embodiment in which the subtraction process is performed digitally in this manner. In FIG. 4, the same functional portions as those of the solid-state imaging device shown in FIG. 1 are denoted by the same reference numerals.

In FIG. 4, an output signal from the image sensor 1 is A / D converted by an A / D converter 3 to be a digital signal. The output signal from the image pickup device 1 is an image pickup output or noise component data of the image pickup device 1 which is switched and output at the operation timings as shown in FIGS. .

When the digital signal output from the A / D converter 3 is obtained by A / D converting the image output of the image sensor 1, this digital signal is imaged at one end of a subtractor 20 by switching a switch 19. Directly input as output Y _SD . The digital signal output from the A / D converter 3 converts the noise component data of the image sensor 1 into an A / D signal.
When the digital signal is converted, the digital signal is input to the line buffer 4 by switching of the switch 19, temporarily stored, and then converted as noise component data Y _{ND into the} subtracter 20.
Is input to the other end.

Then, the above-described image pickup output Y _{SD is} output by the subtractor 20.
The noise component data Y _ND is subtracted from the
_D is obtained. The correction signal Y _D is input to the D / A converter 5 after being added with a synchronization signal and the like by the process circuit 7, is converted into an analog signal, and is output as the video signal Sv.

In the first and second embodiments described above, the number of times of taking in the noise component data from the dummy line 18 is not described in detail, but the number is not limited to one. For example, if the number of captures is increased and the average value is taken, more accurate noise component data can be obtained.

As described above, according to the first and second embodiments, the signal obtained in units of one line from the dummy line 18 to which the photosensitive pixels are not connected is converted into the line buffer 4 having a smaller memory amount than the frame memory. Since the noise correction is stored as noise component data and noise correction is performed based on the noise component data, noise correction can be performed without being affected by photosensitive pixels without unnecessarily increasing the size of the device. Good image quality from which components have been reliably removed can be obtained.

Next, a third embodiment of the present invention will be described. In the third embodiment, optimal noise correction can be performed according to the state of noise at the time of shooting. FIG. 5 is a block diagram illustrating the solid-state imaging device according to the present embodiment. First, the overall operation of this embodiment will be briefly described with reference to FIG.

In FIG. 5, when a shutter button (not shown) is turned on, the light-shielding plate 31 is opened, and an object image is formed on the image pickup surface of the CCD 32 by an optical system (not shown) provided in front of the light-shielding plate 31. The image is photoelectrically converted by the CCD 32. The image signal thus obtained is subjected to A / D conversion by the A / D converter 33 and then stored in the frame memory 34 as an image signal.

Next, photoelectric conversion is performed by the CCD 32 in the dark state with the light shielding plate 31 closed. Image signal obtained thereby, since is obtained by photoelectric conversion performed by blocking all light from the outside of the CCD 32, it is used as the noise component data S _N of the fixed pattern noise of the CCD 32.

The noise component data _SN obtained in this way is directly input to one end of the subtractor 36 without being subjected to A / D conversion or the like by the A / D converter 33. The reading of the noise component data S _N is not performed when it is determined that the noise component data S _N is unnecessary by a determination operation described later.

Then, in synchronization with the input of the noise component data S _N to the subtractor 36, the frame memory 34
The image signal of the subject stored in the D / A converter 35
And is input to the other end of the subtractor 36 as an analog image signal _SD . Then, S =
A subtraction process of S _D -S _N is performed to obtain a correction signal S from which noise components of the CCD 32 have been removed.

The correction signal S is supplied to the terminal B of the switch 37.
Is input to The uncorrected image signal _SD is input to the terminal A of the switch 37, and the system controller 42
, The correction signal S or the uncorrected signal _SD is selected. The method for this determination will be described later. Then, the selected signal is given to the recording device 39 after predetermined processing is performed in the process circuit 38, and is recorded.

Next, a method of determining the terminal selection of the switch 37 by the system controller 42 will be described with reference to FIG.
This will be described with reference to the flowchart of FIG. First, step S1
When the shutter is turned on at step S2, the light shielding plate 3 is set at step S2.
1 is opened, the photoelectric conversion of the subject is performed in step S3, and the imaging signal (subject data) is stored in the frame memory 34.
It is taken in.

In step S4, a histogram process of the luminance distribution of the fetched subject data is performed, and a highlight point HP and a dark point DP of the image are determined from the processing result. FIG. 7 is a diagram illustrating an example of a result of the histogram processing. In this processing result, pixels are classified into 256 steps based on the captured subject data and arranged in the order of brightness. The value of 99% of the whole is defined as the highlight point HP, and the value of 1% is defined as the dark point DP.

Further, from the combination of the highlight point HP and the dark point DP obtained in this way, an index (INDEX) is determined based on a chart as shown in FIG. 8, for example. Here, the index is O
If N, the noise component is read out, and if the index is OFF, the noise component is not read out.

Next, the index is turned on in step S5.
Or OFF. If it is determined that the index is ON, the process proceeds to close the light-shielding plate 31 in step S6, reads the noise component data S _N in step S7 CCD 32 is inputted to one end of the subtractor 36. Next, in step S8, the movable terminal 37a of the switch 37 is switched to the terminal B side, whereby the subtracter 3
The correction signal S generated in step 6 is selected, and the correction signal S is recorded in the recording device 39 in step S10.

In step S5, the index is OF.
If it is determined to be F, the process proceeds to step S9 without reading out the noise component data _SN in steps S6 and S7, and the uncorrected signal _SD is selected by switching the movable terminal 37a to the terminal A side. In S10, the uncorrected signal _SD is recorded in the recording device 39.

In the above description of the present embodiment, the image pickup signal itself in the dark state captured at one time is
Although the case of using as the noise component data S _N of No. 2 has been described, the present invention is not limited to this. For example, by averaging image signals in a dark state captured a plurality of times using a line buffer or the like and performing noise correction using the average, the effect of random noise can be reduced.

Further, the selection of whether to read the noise component data S _N has been described for the case of performing on the basis of the result of histogram processing, when increasing the gain of the image signal, the noise component data S _N It is also possible to always read and perform noise correction. Further, the fact that the subtraction operation in the subtractor 36 may be performed digitally is the same as in the case of the above-described second embodiment.

As described above, according to the third embodiment, the image signal in the dark state generated by closing the light shielding plate is used as noise component data of the image sensor, and correction of fixed pattern noise of the image sensor is performed based on the data. So that
It is possible to obtain noise component data according to the shooting situation for each shooting of one screen, and it is possible to perform optimal noise correction according to the noise state at the time of shooting. Further, it is determined whether or not to perform the noise correction according to the state of the image signal stored in the frame memory, and when the correction is not required, the noise correction is not performed. Optimum noise correction can be performed, and the photographing time of one image can be reduced.

[0044]

As described above, according to the present invention, 1
Since noise component data is generated every time a picture is taken and noise correction is performed based on the generated noise component data, optimal noise correction can be performed, and good image quality can always be obtained. In a solid-state imaging device capable of not performing noise correction depending on conditions, when noise correction is not performed, the photographing time can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a solid-state imaging device according to a first embodiment.

FIG. 2 is a diagram illustrating a detailed configuration of an image sensor.

FIG. 3 is a timing chart showing the overall operation timing of the solid-state imaging device.

FIG. 4 is a block diagram illustrating a solid-state imaging device according to a second embodiment.

FIG. 5 is a block diagram illustrating a solid-state imaging device according to a third embodiment.

FIG. 6 is a flowchart for explaining the operation of the third embodiment.

FIG. 7 is a diagram illustrating an example of a result of the histogram processing.

FIG. 8 is a diagram showing an index determined from a result of the histogram processing.

FIG. 9 is a block diagram illustrating a configuration of a conventional solid-state imaging device.

[Description of Signs] 1 Image sensor 2 Switch 2a Movable terminal 3 A / D converter 4 Line buffer 5 D / A converter 6 Subtractor 7 Process circuit 9 Synchronous signal generation circuit 13 Horizontal signal line 14 Vertical signal line 18 Dummy line 19 Switch Reference Signs List 20 subtractor 31 light shielding plate 32 CCD 33 A / D converter 34 frame memory 35 D / A converter 36 subtractor 37 switch 37a movable terminal 38 process circuit 39 recording device 42 system controller

Claims (6)

[Claims] An image sensor that generates an image signal by photoelectric conversion, an image memory that stores an image signal generated by the image sensor, and a noise that generates noise component data for performing noise correction of the image sensor Generating means, and a subtractor for subtracting the noise component data generated by the noise generating means from the image signal stored in the image memory. After storing the image signal in the image memory, generating the noise Means for generating noise component data by means, subtracting the noise component data from the image signal by the subtractor, and outputting the result. 2. The noise generating means includes: the image sensor; and light blocking means for blocking light from the subject to the image sensor. The noise component data includes light from the subject by the light blocking means. The solid-state imaging device according to claim 1, wherein the solid-state imaging device is generated by the imaging element in a light-shielded dark state. 3. The solid-state imaging device according to claim 1, further comprising a determination unit that determines whether to perform the subtraction processing according to a state of an image signal stored in the image memory. apparatus. 4. The solid-state imaging device according to claim 1, wherein the subtraction process is always performed when increasing the gain of the image signal stored in the image memory. 5. An imaging area for generating an image signal by photoelectric conversion, a light shielding unit for shielding light from a subject to the imaging area, and reading an image signal generated in the imaging area in a state where light is not shielded. A driving unit for performing a first operation and a second operation of reading a noise signal generated in the imaging region in a state where the light is shielded by the light shielding unit; and an image signal read using the noise signal read by the driving unit. A solid-state imaging device comprising: a correction unit configured to correct the noise signal; and a control unit configured to control whether to read the noise signal according to an imaging condition. 6. The control means, when reading out the noise signal, performs the first operation by the driving means, performs the second operation after the first operation,
After performing the first operation and the second operation, control is performed so that the image signal is corrected by the correction unit. When the reading of the noise signal is not performed, the first operation is performed by the driving unit. 6. The solid-state imaging device according to claim 5, wherein after the first operation, control is performed such that the second operation by the driving unit and the correction of the image signal by the correction unit are not performed.