JP2010183481A - Solid-state imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reliably detect a smear oversaturation even if there is noise. <P>SOLUTION: A solid-state imaging device has a lens unit with an electronic eye, a CCD imaging unit that photoelectrically converts light incident through the lens unit and outputs an electric signal, an A/D converter that converts the electric signal into a digital signal, and a smear correction means that corrects a smear signal contained in the digital signal. The solid-state imaging device includes a smear position detection unit 512 that detects a horizontal position of the smear signal from a dark line contained in the digital signal, a decision unit 515 that detects an oversaturation state from the data about the bottom line of an image period contained in the digital signal within the detected horizontal position, and a control unit that controls to narrow down the electronic eye when the oversaturation state is detected. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a solid-state imaging device, and more particularly to a solid-state imaging device with improved detection accuracy of a supersaturated state.

  In a solid-state imaging device using an EM-CCD (Electron Multiplying-Charge Coupled Device) imaging device, it is known that when intense light is incident on the light receiving device, a smear supersaturation state occurs and the video signal read from the light receiving device deteriorates. It is done. Hereinafter, a smear generation mechanism will be described.

FIG. 1 is a diagram showing a basic driving principle of an IT (Interline Transfer) CCD. A frame indicated by a dotted line indicates one cell (a structure corresponding to one pixel), and a circle in the frame is a light receiving portion (photodiode).
(A) is during a video period, and the light received by the light receiving unit is stored as a charge e.
(B) is a vertical blanking period, in which charges are read to a vertical CCD (vertical transfer path) all at once.
(C) is the initial stage of the next video period, in which vertical transfer is performed for one stage, and the leading charge enters a horizontal CCD (horizontal transfer path).
In (D), horizontal transfer is repeated until the horizontal CCD becomes empty.
When the horizontal CCD becomes empty, (C) vertical transfer and (D) horizontal transfer are repeated. A signal obtained by amplifying the horizontally transferred charge corresponds to the signal level of each pixel.

  FIG. 2 is a diagram showing the principle of smear generation, and shows only one vertical CCD. In FIG. 2, when strong light enters the light receiving portion as shown in (A), a non-negligible amount of charge ΔP overflows from the light receiving portion to the vertical CCD. As shown in (B), the charge from the light receiving section continues to overflow during vertical transfer. Accordingly, as shown in (C), ΔP remaining in the vertical CCD at the time of the next charge reading is added to the original charge e. Alternatively, ΔP is added later to the charge e transferred from the vertical CCD. When this is viewed in the playback image, it appears as white vertical streaks. This is smear.

  When the electric charge overflows due to the occurrence of smear, if more intense light enters, the smear becomes supersaturated. When smear supersaturation occurs, a high-intensity deterioration signal from the lowest line in the video period (the horizontal line that is finally transferred to the horizontal transfer path in one frame and read from the CCD, which is the highest line in FIG. 1). Begin to see.

FIG. 3 is a diagram for explaining a smear supersaturated state in a FIT (Frame Interline Transfer) type CCD. In the FIT type CCD, a memory unit for accumulating charges is provided between the vertical CCD and the horizontal CCD. The vertical CCD can transfer the number of vertical CCDs within a time much shorter than one frame time.
Usually, the vertical CCD is swept out immediately before the charge e of the light receiving unit is read out to the vertical CCD, and the vertical CCD and the memory unit are transferred immediately after the readout. These sweep-out and transfer are performed at the same speed and in a time much shorter than one frame time. Therefore, the main cause of smear is only the electric charge leaking from the light receiving unit to the vertical CCD during sweeping and vertical transfer, and the smear is a bright line at a certain level from the upper end to the lower end of the screen regardless of the incident position of the high brightness light. appear.

  When very strong light enters and smear is oversaturated, a deterioration signal begins to appear on the smear from the lowest line in the video period, and the range of the deterioration signal expands upward as the light becomes stronger. This phenomenon is that when a large amount of charge overflowing the vertical CCD reaches the memory part, the overflowing charge is accumulated in the vicinity due to the phenomenon that all of the charge is not transferred to the memory part (overflow or transfer leakage), and there is an amount of accumulated charge When the value is reached, it is considered that this accumulated charge is added to the subsequent transfer charge.

FIG. 4 is an example of a signal reproduction screen in a smear supersaturated state. Although known smear correction is performed and no noticeable smear is seen, the level of the lowest few lines in the video period is even higher than normal smear due to smear oversaturation, so that part cannot be corrected completely It appears as a bright line.
It is impossible in principle to completely prevent the occurrence of smear supersaturation, and it is extremely difficult to accurately predict and compensate for deterioration due to smear supersaturation. Therefore, when the occurrence of supersaturation is detected, the lens (iris) is stopped to reduce the amount of light until no supersaturation occurs.

  FIG. 5 is a diagram showing a flowchart of conventional smear oversaturation avoidance control, and FIG. 6 is an example of a time chart thereof. This control is performed for each frame, and when a signal exceeding a predetermined detection level is detected in the video signal (RAW signal) in the lowest line, the lens is controlled to be further narrowed down from the current value. The lens is stopped until there is no signal exceeding the detection level.

JP 2007-150770 A JP 2007-110413 A JP 2007-116335 A

  However, the above-described smear supersaturation state detection method has a problem that a smear supersaturation state is erroneously determined when an EM-CCD having an electron multiplying effect is operated at a high gain. This is because when the camera sensitivity is increased, noise at the time of high sensitivity increases, and a high-level noise component is erroneously detected as a deteriorated signal.

  FIG. 7 is a schematic diagram for explaining erroneous detection of a deteriorated signal. The left side of FIG. 7 shows a playback screen, the right side is a graph of actual sample data of the lowest horizontal line, and the vertical axis is the signal level. In the graph, noise exceeding the detection level can be confirmed. If it is determined that the smear is supersaturated due to such noise, the lens is erroneously controlled so as to squeeze the lens more than necessary or lower the EM gain.

The following two points can be considered as countermeasures against erroneous determination.
(A) Noise reduction by filter
(B) Gate processing based on smear correction information If the filter is strongly applied, the above-mentioned erroneous determination is eliminated only by (A). However, if the filter is applied too strongly, the original oversaturation may not be detected. If the filter is applied weakly, an erroneous determination occurs. Therefore, it is difficult to prevent erroneous determination using only the filter.

In the present invention, the method (B) is adopted.
That is, smear detection is performed from lines other than the lowest line used for detection of supersaturation. Thereafter, the level of the deteriorated signal in the lowest (final) line is determined. At that time, determination is made only at the same position as the position where smear is detected in the dark line, and when a deterioration signal is detected, a smear supersaturated state detection signal is output.
The line used for detection of supersaturation is preferably the lowest line, but the same effect can be obtained with several lines from the lowest.
As a line used for smear detection, a dark line (horizontal optical black) is preferably used, but is not limited thereto.
In order to further reduce false detection of a smear supersaturated state, a signal to be detected may be filtered to suppress noise due to random components.

  According to the present invention, since the deterioration signal is detected only at the same horizontal position as the smear, it is possible to prevent erroneous determination of the smear supersaturated state even in a situation where there is a lot of noise as in the case of high sensitivity.

The figure which shows the basic drive principle of IT type CCD Diagram showing the smear generation principle The figure explaining the smear supersaturation state in FIT type CCD Example of signal playback screen in smear supersaturated state (conventional) Flow chart of smear supersaturation avoidance control (conventional example 1) Example of time chart of smear oversaturation avoidance control (conventional example 1) Schematic diagram explaining false detection of degraded signals Schematic diagram explaining the detection range of the degraded signal Configuration of EM-CCD camera (Example 1) Configuration diagram of a smear supersaturation detection circuit 51 provided in the DSP 6 (Example 1)

FIG. 8 is a schematic diagram for explaining a detection range of a deteriorated signal in the embodiment of the present invention.
As shown on the right side of FIG. 8, the present invention utilizes the property that a deterioration signal of smear supersaturation to be detected appears superimposed on the smear at the horizontal position where smear is generated. Then, the smear generation range is specified, and the deterioration signal is detected only at the same position as the smear.
For detection of smear, a dark line (horizontal optical black) existing in the vertical blanking period is used. Since there is no video signal of light incident from the lens in the dark line, only the smear signal can be detected. In the example of FIG. 8, the black portion at the top of the video is a dark line, and consists of several horizontal lines that are first transferred to the horizontal transfer path.
A signal of a certain level or more is detected as a smear signal existing in the dark line, and a signal of a certain level or more is detected as a deterioration signal at the same horizontal position of the final line of the video period using the detected horizontal position of the smear signal. When a signal is detected, a supersaturated state detection signal is output. When it is expected that the smear position is slightly shifted between the dark line and the final line, the detection range of the degradation signal may be expanded from the smear generation range.

FIG. 9 is a configuration diagram of an EM-CCD camera that is an embodiment of the present invention.
The lens unit 1 focuses light from the subject on the imaging surface of the EM-CCD. The lens unit 1 has an automatic diaphragm, and the diaphragm is variable according to control from the DSP 6.
The EM-CCD 2 is a FIT type two-dimensional image sensor, which photoelectrically converts incident light from the lens unit 1 and sequentially outputs signals obtained by electron multiplication.
The timing generator (TG) 3 generates a frequency-controlled clock signal, gives a driving pulse to the EM-CCD, and gives an operation clock to a DSP (Digital Signal Processor) 5 or the like.
The A / D converter (A / D) 4 converts the electrical signal from the EM-CCD 2 from analog to digital after performing correlated double sampling and adjusting the gain.

An FPGA (Field Programmable Gate Array) 5 performs arithmetic processing such as smear correction, shading correction, noise reduction (cyclic filter) on the digital signal input from the A / D 4, and temporarily outputs it to the DSP 6. Further, the video signal processed signal input from the DSP 6 is subjected to superimposition processing for OSD (On Screen Display) display of various states and settings (for example, focus gate range), and is output.
The DSP 6 controls the entire EM-CCD camera including the smear oversaturation avoidance control of the present invention, and performs video signal processing such as color balance.
The TV encoder (ENC) 7 converts the signal input from the FPGA 5 into, for example, an NTSC (National Television System Committee) analog video signal and outputs it to the outside.

FIG. 10 is a configuration diagram of the smear supersaturation detection circuit 51 provided in the DSP 6. The smear supersaturation detection circuit 51 receives the raw signal obtained by A / D conversion from the image sensor and operates in a one-frame cycle.
The noise reduction filter 511 performs filtering on the input RAW signal, for example, within one line to reduce noise.
The smear position information detection unit 512 has a memory for one line, compares the signal level with a predetermined threshold value for each pixel in the dark line (horizontal optical black) signal period, and is true when the threshold value is exceeded. When it does not exceed, store a false logical value. Then, it is read and output during the last line period. In this example, the dark line existing in the vertical blanking period is used, and the timing of the dark line and the final line is given to the smear oversaturation detection circuit 51 as a pulse generated by counting the clock from TG2. In addition to the above, various known methods can be used for smear position detection, and the smear correction and processing performed by the DSP 6 may be shared or performed separately.
The final line extraction unit 513 extracts and outputs a signal as shown in FIG. 7 during the final line period.
The gate unit 514 passes the signal from the final line extraction unit 513 only when the logical value read from the smear position information detection unit is true.
The degradation signal determination unit 515 compares the signal of the final line that has passed through the gate unit 514 with a predetermined detection level, and outputs a supersaturation detection signal to the DSP 6 when a signal (degradation signal) that is equal to or higher than the detection level is detected.

  Next, smear oversaturation avoidance control in the DSP 6 will be described with reference to FIG. The DSP 6 controls the aperture and exposure time of the lens unit 1 as well as AGC (Auto Gain Control) as normal control for maintaining the brightness of the video appropriately. In this example, in addition to the normal aperture control value, an oversaturation aperture control value is provided. When the supersaturated state is reached, the supersaturation aperture control value is given to the lens 1 instead of the normal aperture control value. The aperture control value may be either the absolute value of the aperture or the drive amount (change amount) of the aperture.

As an example of the former case, when the supersaturation detection signal is continuously received from the FPGA 5 in the most recent predetermined number of frames, it is determined that the state is supersaturated, and the aperture is controlled by a predetermined amount to the aperture control value given to the lens 1 immediately before. Only the value to be closed is added to obtain a supersaturation aperture control value, which is output to the lens 1 instead of the normal aperture control value. Even after the supersaturated state is released, a value for opening the aperture by a predetermined amount is added to the supersaturated aperture control value for each frame and updated until there is no difference between the supersaturated aperture control value and the normal aperture control value. Output to the lens 1.
As an example of the latter case, when the supersaturation detection signal is continuously received in the most recent predetermined number of frames, the supersaturation aperture control value for closing the aperture by a predetermined amount is output to the lens 1 instead of the normal aperture control value. That's fine.

  When the aperture control value is forcibly set in this way, control loops such as EM gain and digital gain operate so as to compensate for insufficient illuminance. However, if the EM gain is increased too much, the image quality (S / N) is deteriorated or the element of the EM-CCD itself is deteriorated (in a situation where a smear in a substantially saturated state is generated). Therefore, when it is determined that the state is supersaturated, the upper limit of the EM gain control range may be lowered.

As described above, in the example of the present invention, a dark line is obtained from the image sensor, and smear level determination is performed on the dark line portion. Thereafter, the level of the deteriorated signal in the final line is determined. At that time, determination is made only at the same position as the position where smear is detected in the dark line, a deterioration signal is detected, and a smear supersaturation state detection signal is output.
After detecting the smear supersaturated state, the smear supersaturated state is eliminated by reducing the aperture of the lens, lowering the gain, and decreasing the amount of light incident on the lens.

  The present invention is suitable for FIT type and IT type CCDs, but can also be applied to other CCDs.

1 lens part, 2 EM-CCD,
3 Timing generator (TG), 4 A / D converter (A / D),
5 FPGA, 6 DSP,
7 TV encoder (ENC),
51 smear supersaturation detection circuit, 511 noise reduction filter,
512 smear position information detection unit, 513 final line extraction unit,
514 Gate part, 515 Deterioration signal determination part.

Claims (2)

A lens unit having an automatic aperture; a CCD image pickup unit that photoelectrically converts light incident through the lens unit to output an electrical signal; an A / D converter that converts the electrical signal into a digital signal; In a solid-state imaging device having smear correction means for correcting the included smear signal,
A smear position detector that detects a horizontal position of the smear signal from data of the first horizontal line included in the digital signal;
A solid-state imaging device comprising: a determination unit that detects a supersaturated state from data of a second horizontal line included in the digital signal based on the detected horizontal position. The CCD imaging unit is an IT-type or FIT-type EM-CCD,
The first horizontal line is a dark line that is separate from the second horizontal line;
The second horizontal line is a line near the bottom of the video period,
The determination unit detects the supersaturated state when the data of the second horizontal line exceeds a predetermined detection level within the range of the detected horizontal position,
The solid-state imaging device according to the first embodiment, further comprising a control unit that controls to reduce the automatic aperture when the supersaturated state is detected.

Images