CN110677603B - Image sensor for single-frame segmented exposure and data acquisition method - Google Patents
Image sensor for single-frame segmented exposure and data acquisition method Download PDFInfo
- Publication number
- CN110677603B CN110677603B CN201910871176.7A CN201910871176A CN110677603B CN 110677603 B CN110677603 B CN 110677603B CN 201910871176 A CN201910871176 A CN 201910871176A CN 110677603 B CN110677603 B CN 110677603B
- Authority
- CN
- China
- Prior art keywords
- exposure
- signal
- image sensor
- inflection point
- pixel array
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/40—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled
- H04N25/44—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled by partially reading an SSIS array
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/50—Control of the SSIS exposure
- H04N25/57—Control of the dynamic range
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/70—SSIS architectures; Circuits associated therewith
- H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
Abstract
The invention discloses an image sensor for single-frame segmented exposure, which comprises an image sensor module, an inflection point reference input module, a memory and a digital signal processor, wherein in the acquisition process of first frame data, the inflection point reference input module writes an inflection point reference input signal into a pixel array, and the pixel array outputs a pseudo exposure signal and writes the pseudo exposure signal into the memory; starting from the acquisition of second frame data, carrying out segmented exposure on the pixel array, reading out an exposure signal of the pixel array and a pseudo exposure signal in a memory by the digital signal processor, and subtracting the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated; the image sensor for single-frame segmented exposure provided by the invention can shorten the processing time of each line of the image sensor, improve the frame rate of the image sensor and completely eliminate FPN.
Description
Technical Field
The invention relates to the field of image sensors, in particular to an image sensor for single-frame segmented exposure.
Background
The CMOS image sensor is widely used in daily life and processes due to its advantages of high integration, low power consumption, low cost, and the like. For a CMOS image sensor, both high dynamic range and high frame rate are important indicators for measuring the CMOS image sensor. For a CMOS image sensor with a small dynamic range, if a small signal needs to be captured, a long exposure time needs to be adopted, but the large signal is saturated, and large signal data cannot be obtained; on the contrary, if a large signal is captured, the exposure time needs to be shortened, but the small signal cannot be distinguished, and effective data of the small signal is obtained. High dynamic range is therefore an important indicator of CMOS image sensors.
At present, the common methods for improving the dynamic range mainly comprise three methods of multi-frame exposure fusion, different conversion gains and segmented exposure, but the multi-frame exposure fusion needs to obtain a large and small complete signal of one image through multi-frame exposure data, and the frame rate needs to be sacrificed; the gain conversion method is to convert the two frames with different gains to obtain an image, or to read data with different conversion gains for each pixel (also called pixel unit) in one frame, and the two methods can also reduce the frame rate by half. Both approaches of multi-frame exposure fusion and different conversion gains require additional ISP (digital signal processor) algorithmic processing. Redundant signals in the large signals overflow to a power supply in a single frame in a segmented exposure mode, so that the originally saturated large signals are not saturated, and the small signals are not influenced.
However, a light sensitive curve in a single-frame segmented exposure HDR (High-Dynamic Range) mode may form a knee point, and the knee point of each pixel is different due to different thresholds of the segmented exposure control tube of each pixel, that is, a knee point difference is formed; so that FPN (Fixed Pattern Noise) of the image increases. One of the prior art is to eliminate the inflection point difference of each line of output signals by pre-exposing each line and then collecting the reference output signals containing the inflection point difference, thereby achieving the effect of optimizing the FPN. However, this approach to eliminating FPN doubles the processing time per line, reducing the image sensor frame rate.
Referring to fig. 1, a frame diagram of a conventional image sensor is shown, in which an image sensor module is mainly composed of a pixel array, a row driving circuit, a column ADC, and an interface circuit, and the interface circuit is used as an output port of the image sensor and outputs signals of the image sensor to a digital signal processor. A single frame HDR FPN optimization structure based on full-well control is disclosed in patent US 10250832B 1, and as shown in fig. 2, compared to the conventional image sensor in fig. 1, a knee reference input module is added in fig. 2, and the knee reference input module is connected to the pixel array. Fig. 3 is a schematic diagram of a working process of the structure corresponding to fig. 2 for the knee elimination scheme. Each pixel includes three major processes: a low full well exposure EXP _ LFWC, a high full well exposure EXP _ HFWC, and a READ, the knee difference being generated during the low full well exposure. Whereas in the readout phase, each pixel comprises 5 processes: the pixel reset signal reading SHR _ VP, the pixel output signal reading SHS _ VP, the pseudo exposure VIRTUAL EXP, the inflection point reset signal reading SHR _ RP and the inflection point output signal reading SHS _ RP are respectively read, wherein the pseudo exposure is the inflection point reference signal input, and because the pixel needs to acquire two signals, signal writing needs to be carried out twice when the inflection point reference signal is written in the pseudo exposure process, and a high signal and a low signal are respectively written in. Finally, after pseudo exposure, the pixel is not exposed, but the pixel can obtain a high signal and a low signal similar to normal work, the high-signal analog pixel is reset, and the potential of the low-signal analog pixel is reduced after exposure. Because the reset signal read out by the pixel in normal exposure work does not contain inflection point difference, and only the pixel output signal contains inflection point difference, the inflection point difference can be introduced when a low signal is written in the pseudo exposure process, and the inflection point difference can also be introduced when the low signal is read out. However, the MOS transistor introduced by the inflection point and the MOS transistor introduced by the inflection point difference during the operation of the pixel HDR are the same MOS transistor, so that the inflection point difference can be eliminated. In the operation mode shown in fig. 3, two sets of data, i.e., pixel exposure data and dummy exposure data, need to be read out every row of pixel signal readout, so that the row period increases and the image sensor frame rate decreases.
Disclosure of Invention
The invention aims to provide an image sensor for single-frame segmented exposure, which can shorten the processing time of each line of the image sensor, improve the frame rate of the image sensor and completely eliminate FPN.
In order to achieve the purpose, the invention adopts the following technical scheme: an image sensor for single-frame segmented exposure comprises an image sensor module, an inflection point reference input module, a memory and a digital signal processor, wherein the image sensor module comprises a pixel array containing pixel units, an output port of the image sensor module is simultaneously connected with the memory and the digital signal processor, the memory is connected with the digital signal processor, and an output port of the inflection point reference input module is connected with MOS (metal oxide semiconductor) tubes which control segmented exposure in the pixel units;
in the process of collecting first frame data, the inflection point reference input module writes an inflection point reference input signal into the pixel array, the pixel array outputs a pseudo exposure signal, and the pseudo exposure signal is written into the memory; starting from the acquisition of second frame data, carrying out segmented exposure on the pixel array, reading out an exposure signal of the pixel array and a pseudo exposure signal in a memory by the digital signal processor, and subtracting the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated;
the inflection point difference times of the pseudo exposure signals in the memory in the writing and reading processes are the same as the inflection point difference times of the pseudo exposure signals in the pixel array exposure and reading processes.
Further, the exposure signal includes an exposure reset signal and an exposure output signal, and the dummy exposure signal includes a dummy exposure reset signal and a dummy exposure output signal.
Further, the inflection point reference input signal output by the inflection point reference input module includes a low inflection point reference input signal and a high inflection point reference input signal, which respectively correspond to the dummy exposure reset signal and the dummy exposure output signal.
Further, the segmented exposure comprises low full trap quantity exposure and high full trap quantity exposure, and the low full trap quantity exposure and the high full trap quantity exposure are realized by controlling the barrier height between a sensing diode and a power supply in the pixel unit.
Further, the image sensor module and the inflection point reference input module are integrated in an image sensor chip, and the memory and/or the digital signal processor are integrated outside the image sensor chip or integrated in the image sensor chip.
Further, the image sensor module further includes a row driving circuit, a column stage ADC, and an interface circuit.
Further, the digital signal processor reads out the exposure signals of the pixel units in the A row in the pixel array and the corresponding pseudo exposure signals in the memory at the same time, wherein A is an integer which is greater than 0 and less than or equal to the row number of the pixel array.
A method for data acquisition by using an image sensor with single-frame segmented exposure comprises the following steps:
s01: acquiring first frame data of the image sensor, wherein the pixel array is not exposed at the moment, the inflection point reference input module writes an inflection point reference input signal into the pixel array, the pixel array outputs a pseudo exposure signal, and the pseudo exposure signal is written into a memory;
s02: acquiring second frame data of the image sensor, wherein the pixel array is exposed at the moment, the digital signal processor reads out an exposure signal of the pixel array and a pseudo exposure signal in the memory, and subtracts the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated;
s03: and repeating the step S02 to collect the data of the third frame to the B frame of the image sensor until the data collection of all the frames is completed, wherein B is the total frame number of the data collection of the image sensor.
Further, in step S02, the digital signal processor simultaneously reads out the exposure signals of the pixel units in the row a in the pixel array and the corresponding dummy exposure signals in the memory, where a is an integer greater than 0 and less than or equal to the number of rows in the pixel array.
The invention has the beneficial effects that: acquiring a pseudo exposure signal during the acquisition of first frame data, acquiring an exposure signal from the acquisition of second needle data, wherein the pseudo exposure signal and the exposure signal both contain the same inflection point difference, and subtracting the exposure signal of a corresponding pixel from the pseudo exposure signal to obtain an exposure signal with the inflection point difference eliminated; the processing time of each line of the image sensor can be shortened, the frame rate of the image sensor can be improved, and FPN can be completely eliminated.
Drawings
Fig. 1 is a schematic diagram of a conventional image sensor frame.
Fig. 2 is a block diagram of a single frame HDR FPN optimization based on the full well control in the prior art.
Fig. 3 is a schematic diagram of a working process of single frame HDR FPN optimization based on full well control in the prior art.
FIG. 4 is a block diagram of an image sensor for single frame segmented exposure according to the present invention.
FIG. 5 is a schematic diagram of an operation process of an image sensor for single-frame segmented exposure according to the present invention.
Fig. 6 is a schematic structural diagram of a 5T pixel cell and corner reference input module.
Fig. 7 is a schematic diagram of the working process of performing pseudo exposure based on 5T pixel units.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in detail below with reference to the accompanying drawings.
As shown in fig. 4, the image sensor for single frame segmented exposure disclosed by the present invention comprises an image sensor module, an inflection point reference input module, a memory and a digital signal processor, wherein the image sensor module comprises a pixel array including a plurality of pixel units, an output port of the image sensor module is simultaneously connected with the memory and the digital signal processor, the memory is connected with the digital signal processor, and an output port of the inflection point reference input module is connected with a MOS transistor in the pixel unit for controlling segmented exposure.
The image sensor module also comprises a row driving circuit, a column-level ADC and an interface circuit, wherein the row driving circuit generates driving signals of each row of the pixel array and controls the pixel array to expose and read signals line by line; the column-level ADC circuit converts all column analog signals output by each row into digital signals and outputs the digital signals through the interface circuit; the output port of the interface circuit is the output port of the whole image sensor module. The image sensor module and the inflection point reference input module are integrated in an image sensor chip, the digital signal processor can be integrated outside the image sensor chip or integrated in the image sensor chip, and the memory can be integrated in the image sensor chip or integrated outside the image sensor.
The memory stores dummy exposure signals of all pixel arrays, including a knee point difference. Only the pixel array writes inflection point difference into the memory in the acquisition process of the first frame data, from the second frame, the pixel array is normally sensitized, and each frame only reads out the inflection point difference written in the first frame from the memory, and the memory is not written any more. One end of the memory is connected with the digital signal processor, and the value read from the memory from the second frame is input to the digital signal processor for processing. The digital signal processor does not perform the knee point difference elimination operation for the first frame. And after the second frame of image sensor is normally sensitive, the digital signal processor subtracts the exposure signal output by the image sensor interface from the pseudo exposure signal read out from the memory, and eliminates FPN caused by inflection point difference in the output data of the image sensor.
The image sensor is used for single-frame segmented exposure, and the principle of the single-frame segmented exposure is that the voltage value of a mos tube for controlling exposure in a low-full-trap-volume exposure stage is not the maximum voltage (relative to the ground voltage and the power voltage which are not the maximum voltage), the voltage in the low-full-trap-volume exposure stage is defined as the intermediate voltage, so that charges higher than the potential barrier height corresponding to the intermediate voltage can be discharged, and a broken line with an inflection point is reflected on a photosensitive sensitivity curve. At the intermediate voltage, due to the threshold difference of the MOS tube for controlling the exposure of each pixel unit, the potential barrier formed by each pixel unit at the intermediate voltage is different in height, so that the residual charge is different, and the inflection point of each pixel unit is different. That is, the inflection point is generated due to the intermediate voltage. Therefore, in the stage of reading the dummy exposure signal, in order to ensure that the read dummy exposure signal includes an inflection point difference, the corresponding exposure control MOS transistor should be in an intermediate state when the dummy exposure signal is read. Because the inflection point difference of the pixel units is determined by the threshold of the MOS tube for controlling the sectional exposure, and the threshold of the device on the chip is irrelevant to the time, the inflection point difference of each pixel unit of the first frame and the subsequent frame can be considered to be the same, and the inflection point difference can be eliminated by subtracting the inflection point difference of the first frame from the subsequent frame. The inflection point difference in the invention refers to the difference of deviation from a normal inflection point value in the output signal of the pixel array because the threshold values of MOS tubes for controlling the sectional exposure in different pixel units are different.
An inflection point reference input signal output by the inflection point reference input module must pass through an MOS tube generated by inflection point difference, namely the MOS tube for controlling the pixel unit to be exposed in a segmented mode; therefore, the inflection point difference can be eliminated when the exposure signal and the pseudo exposure signal are subtracted, and the inflection point difference times included in the writing and reading of the inflection point reference input signal in the memory are required to be consistent with the inflection point difference generation times in the exposure and reading processes of the pixel signal. If the pixel unit only generates inflection point difference once in the exposure stage, the inflection point difference can be generated only once by one operation in the whole process of writing and reading the inflection point reference input signal in the memory; if the pixel unit generates two inflection point differences in the exposure reading stage, the inflection point reference input signal must include two inflection point differences in the writing and reading processes in the memory, so that the image FPN caused by the inflection point differences can be eliminated to the maximum extent. The number of the MOS tubes for controlling the sectional exposure can be one or more, and when the number of the MOS tubes for controlling the sectional exposure is multiple, each MOS tube for controlling the sectional exposure is connected with an inflection point reference input module, so that all inflection points in a finally formed image are eliminated.
The output port of the inflection point reference input module is connected to an MOS tube for controlling the sectional exposure in the pixel unit, namely, the writing of an inflection point reference input signal must be ensured to pass through the MOS tube generated by inflection point difference. The specific inflection point reference input module is connected to each pixel unit in the pixel array, or a row of pixel units, or a pixel unit, and is required to depend on the exposure mode of the pixel array and the output mode of the pixel array. If the pixel array is subjected to global exposure, all pixel units of the pixel array are connected with a knee point reference input module, namely the whole image sensor only needs one knee point reference input module; if the pixel array is exposed line by line, each line of pixel units is connected with one inflection point reference input module, namely the number of the inflection point reference input modules of the whole image sensor is consistent with the number of lines of the image sensor; if the pixel array is independently exposed and output for each pixel, each pixel unit needs to be connected with a knee point reference input module, that is, the number of knee point reference input modules of the whole image sensor is consistent with the number of pixels of the image sensor.
The principle of eliminating inflection point difference of the image sensor in the invention is as follows: in the process of collecting first frame data, a knee point reference input module inputs a knee point reference input signal to a pixel array, the pixel array outputs a pseudo exposure signal, and the pseudo exposure signal is written into a memory; starting from the acquisition of second frame data, carrying out segmented exposure on the pixel array, reading out an exposure signal of the pixel array and a pseudo exposure signal in a memory by the digital signal processor, and subtracting the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated; the inflection point difference times included in the writing and reading processes of the pseudo exposure signal are the same as the inflection point difference times in the pixel array exposure and reading processes. In the whole chip working process, only the inflection point reference input signal needs to be written in the first frame data acquisition process, and then normal exposure is carried out on the frame without writing the inflection point reference input signal; similarly, the memory only needs to write the pseudo exposure signal in the first frame, and only needs to read the pseudo exposure signal in the following frame.
Referring to fig. 5, a correlated double sampling process is taken as an example for explanation, and the method for eliminating the inflection point in the present invention is also applicable to other sampling methods. As can be seen from fig. 5, the dummy exposure and corner difference reading is only performed in the first frame, and from the second frame onwards, the pixels have only 3 operating states: low full well exposure EXP _ LFWC, high full well exposure EXP _ HFWC, READ, and READ phase has only 2 operating states: the pixel reset signal readout SHR _ VP and the pixel output signal SHS _ VP are read out. The pseudo exposure process of the first frame is the same as that of fig. 3, and the inflection point reference input module respectively writes a high inflection point reference input signal and a low inflection point reference input signal twice to simulate the exposure of the pixel. There are two working states in the signal readout process corresponding to the pseudo exposure: the knee reset signal readout SHS _ KP and the knee output signal readout SHS _ KP.
The first frame pixel is not exposed, i.e. dummy exposure and pixel exposure are not performed in the same frame, but are done in the same frame in the prior art. From the second frame, the exposure process of the pixel comprises low full-trap quantity exposure and high full-trap quantity exposure, and the conversion of the low full-trap quantity and the high full-trap quantity is realized by controlling the barrier height between the photosensitive diode of the pixel and the power supply. Each pixel is exposed through the two exposure stages, but for large signals, the effective exposure time may be less than the sum of the exposure time for low full well and the exposure time for high full well, because the signals are more easily saturated during the low full well exposure stage, and the charges generated during the excess exposure time after saturation are absorbed to the power supply and are no longer accumulated by the pixel.
The process in the read-out phase of the first frame is also different from the read-out process of the subsequent frame. After the pseudo exposure reset signal and the pseudo exposure output signal are read out in the first frame, a pseudo exposure signal containing inflection point difference of each pixel can be obtained. Writing a pseudo exposure signal into each pixel unit, reading the pseudo exposure signal, and writing the read pseudo exposure signal into a memory. In a signal reading stage at the beginning of the second frame, after each pixel collects an exposure reset signal and an exposure output signal, an exposure signal of each pixel can be obtained, and the exposure signal contains inflection point difference. And outputting the obtained exposure signal to a digital signal processor, reading out a pseudo exposure signal of a corresponding pixel unit in a memory by the digital signal processor, and subtracting the exposure signal of the corresponding pixel unit from the pseudo exposure signal to obtain the exposure signal with the inflection point difference eliminated.
The invention provides a method for acquiring data by adopting a single-frame segmented exposure image sensor, which comprises the following steps:
s01: and acquiring first frame data of the image sensor, wherein the pixel array is not exposed at the moment, the inflection point reference input module writes an inflection point reference input signal into the pixel array, and the pixel array outputs a pseudo exposure signal and writes the pseudo exposure signal into a memory.
S02: and acquiring second frame data of the image sensor, wherein the pixel array is exposed, the digital signal processor reads out an exposure signal of the pixel array and a pseudo exposure signal in the memory, and the exposure signal of the corresponding pixel unit is subtracted from the pseudo exposure signal to obtain an exposure signal with the inflection point difference eliminated.
The digital signal processor reads out the exposure signals of the A-row pixel units in the pixel array and the corresponding pseudo-exposure signals in the memory at the same time, wherein A is an integer which is larger than 0 and less than or equal to the row number of the pixel array. That is, the digital signal processor reads the dummy exposure signal of a certain line and the subtraction of the exposure signal and the dummy exposure signal may be performed simultaneously with the exposure signals of other lines, which may improve the frame rate. Meanwhile, it is also within the scope of the present invention if the digital signal processor reads the dummy exposure signal of a certain line and the exposure signal and the dummy exposure signal are subtracted from each other independently, i.e. line-by-line processing.
S03: and repeating the step S02 to collect the data of the third frame to the B frame of the image sensor until the data collection of all the frames is completed, wherein B is the total frame number of the data collection of the image sensor.
The structure and the method for eliminating the inflection point difference only need to ensure that the output port of the inflection point reference input module is connected with the MOS tube for controlling the sectional exposure in the pixel unit, and the inflection point difference times contained in the input and output processes of the pseudo exposure signal are the same as the inflection point difference times in the pixel array exposure and output processes, so that the structure and the method can be suitable for image sensors with any structures on the basis. The following description is made of an image sensor having a 5T structure:
fig. 6 is a schematic diagram of the connection between the 5T knee reference input module and the pixel, and the 5T HDR is realized by the gate voltage adjustment of the AB transistor. When the pixel is exposed normally, the inflection point reference input module always outputs power supply voltage, and the pixel works normally. When pseudo-exposure occurs, the inflection point reference input module outputs a high inflection point reference input signal and a low inflection point reference input signal, respectively.
Fig. 6 is a diagram illustrating the dummy exposure and reading of the dummy exposure signal according to the present invention with reference to fig. 7. The output signal of the inflection point input module is VKP, the high inflection point reference input signal of the VKP is VH, the low inflection point reference input signal is VL, the pixel power supply voltage is VDDP, the AB grid high voltage is VH _ AB, the middle voltage is MID _ AB, the low voltage is VL _ AB, wherein MID _ AB is the AB grid voltage value in the low full trap exposure stage in normal exposure. When the inflection point input module outputs VH, the voltage of an AB gate is VH _ AB, the AB tube is completely conducted at the moment, the voltage at the PD is VPD1 which is VH, and the PD signal at the moment, namely a pseudo exposure reset signal Vrst _ KP, is collected; when the inflection point input module outputs VL, the gate voltage of the AB tube is MID _ AB, the AB tube is not completely conducted at the moment, the voltage at PD is VPD2 which is VL-Vth, and the Vth is the threshold value of the AB tube; collecting a PD signal at the moment, namely a pseudo exposure output signal Vsig _ KP; after the pseudo exposure data collection is finished, the inflection point input module outputs the power supply voltage VDDP, AB is pulled to be high by the value VH _ AB, and PD enters a reset state to prepare for the next exposure. The signal obtained by the dummy exposure is V _ KP ═ VH- (VL-Vth) ═ VH-VL + Vth, and thus it can be seen that: v _ KP is positively correlated with Vth.
In the normal exposure process, the acquired exposure reset signal is Vrst _ PD, and the acquired exposure output signal is Vsig _ PD. When the AB voltage is MID _ AB, the barrier height is related to (MID _ AB-Vth). The smaller Vth of the AB tube is, (MID _ AB-Vth) is, the lower the barrier height is, the less charge is accumulated, and the larger Vsig _ PD is obtained; conversely, the larger the Vth of the AB transistor, (MID _ AB-Vth), the higher the barrier height, the more charge accumulated, and the smaller the resulting Vsig _ PD. Vsig is thus seen to be negatively correlated with Vth. The obtained exposure signal V _ PD, Vrst _ PD-Vsig _ PD, is positively correlated with Vth, and thus (V _ PD-V _ KP) is independent of Vth, that is, the phenomenon of FPN degradation of the image finally formed due to the inflection point difference caused by the Vth difference is optimized.
The above description is only a preferred embodiment of the present invention, and the embodiment is not intended to limit the scope of the present invention, so that all equivalent structural changes made by using the contents of the specification and the drawings of the present invention should be included in the scope of the appended claims.
Claims (9)
1. An image sensor for single-frame segmented exposure is characterized by comprising an image sensor module, an inflection point reference input module, a memory and a digital signal processor, wherein the image sensor module comprises a pixel array containing pixel units, an output port of the image sensor module is simultaneously connected with the memory and the digital signal processor, the memory is connected with the digital signal processor, and an output port of the inflection point reference input module is connected with MOS (metal oxide semiconductor) tubes which control segmented exposure in the pixel units;
in the process of collecting first frame data, the inflection point reference input module writes an inflection point reference input signal into the pixel array, the pixel array outputs a pseudo exposure signal, and the pseudo exposure signal is written into the memory; starting from the acquisition of second frame data, carrying out segmented exposure on the pixel array, reading out an exposure signal of the pixel array and a pseudo exposure signal in a memory by the digital signal processor, and subtracting the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated;
the inflection point difference times contained in the writing and reading processes of the pseudo exposure signal in the memory are the same as the inflection point difference times in the pixel array exposure and reading processes.
2. The image sensor according to claim 1, wherein the exposure signal comprises an exposure reset signal and an exposure output signal, and the dummy exposure signal comprises a dummy exposure reset signal and a dummy exposure output signal.
3. The image sensor of claim 2, wherein the knee reference input signal output by the knee reference input module comprises a high knee reference input signal and a low knee reference input signal, respectively corresponding to the dummy exposure reset signal and the dummy exposure output signal.
4. The image sensor for single-frame segment exposure according to claim 1, wherein the segment exposure comprises low-full-well exposure and high-full-well exposure, and the low-full-well exposure and the high-full-well exposure are realized by controlling the barrier height between a photodiode and a power supply in a pixel unit.
5. The image sensor for single frame segmented exposure according to claim 1, wherein the image sensor module and the knee reference input module are integrated in an image sensor chip, and the memory and/or the digital signal processor are integrated outside the image sensor chip or integrated in the image sensor chip.
6. The image sensor for single-frame segmented exposure according to claim 1, wherein the image sensor module further comprises a row driving circuit, a column stage ADC and an interface circuit.
7. The image sensor for single-frame segmented exposure according to claim 1, wherein the digital signal processor simultaneously reads out the exposure signals of a rows of pixel units in the pixel array and the corresponding dummy exposure signals in the memory, a being an integer greater than 0 and less than or equal to the number of rows in the pixel array.
8. A method for data acquisition using the image sensor of claim 1, comprising the steps of:
s01: acquiring first frame data of the image sensor, wherein the pixel array is not exposed at the moment, the inflection point reference input module writes an inflection point reference input signal into the pixel array, the pixel array outputs a pseudo exposure signal, and the pseudo exposure signal is written into a memory;
s02: acquiring second frame data of the image sensor, wherein the pixel array is exposed at the moment, the digital signal processor reads out an exposure signal of the pixel array and a pseudo exposure signal in the memory, and subtracts the exposure signal of a corresponding pixel unit from the pseudo exposure signal to obtain an exposure signal with an inflection point difference eliminated;
s03: and repeating the step S02 to collect the data of the third frame to the B frame of the image sensor until the data collection of all the frames is completed, wherein B is the total frame number of the data collection of the image sensor.
9. A method for data acquisition as claimed in claim 8, wherein the digital signal processor simultaneously reads the exposure signals of a rows of pixel elements in the pixel array and the corresponding dummy exposure signals in the memory in step S02, wherein a is an integer greater than 0 and less than or equal to the number of rows in the pixel array.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910871176.7A CN110677603B (en) | 2019-09-16 | 2019-09-16 | Image sensor for single-frame segmented exposure and data acquisition method |
| PCT/CN2020/103762 WO2021051999A1 (en) | 2019-09-16 | 2020-07-23 | Method for removing fixed pattern noise |
| US17/760,767 US11611813B2 (en) | 2019-09-16 | 2020-07-23 | Method of removing fixed pattern noise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910871176.7A CN110677603B (en) | 2019-09-16 | 2019-09-16 | Image sensor for single-frame segmented exposure and data acquisition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110677603A CN110677603A (en) | 2020-01-10 |
| CN110677603B true CN110677603B (en) | 2021-09-14 |
Family
ID=69077951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910871176.7A Active CN110677603B (en) | 2019-09-16 | 2019-09-16 | Image sensor for single-frame segmented exposure and data acquisition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110677603B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021051999A1 (en) * | 2019-09-16 | 2021-03-25 | 上海集成电路研发中心有限公司 | Method for removing fixed pattern noise |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917590A (en) * | 2005-05-26 | 2007-02-21 | 红外线解决方案公司 | Method for fixed pattern noise reduction in infrared imaging cameras |
| WO2014174498A1 (en) * | 2013-04-26 | 2014-10-30 | Ecole Polytechnique Federale De Lausanne (Epfl) | A novel cmos image sensor with event/change detection and reduced data redundancy |
| CN106341579A (en) * | 2015-07-02 | 2017-01-18 | 豪威科技股份有限公司 | Method for reading image pixel array and High Dynamic Range (HDR) imaging system |
| WO2017201488A1 (en) * | 2016-05-20 | 2017-11-23 | Harpoon Therapeutics, Inc. | Single domain serum albumin binding protein |
| CN107592472A (en) * | 2017-10-20 | 2018-01-16 | 上海晔芯电子科技有限公司 | The image-forming correction method of multistage exposure imaging sensor |
| CN108139207A (en) * | 2015-10-16 | 2018-06-08 | 卡普索影像公司 | For capturing the single image sensor of the structure light image of mixing and regular image |
| US10171763B2 (en) * | 2017-01-19 | 2019-01-01 | Axis Ab | Method for fixed pattern noise reduction and use of such method |
| CN109389086A (en) * | 2018-10-09 | 2019-02-26 | 北京科技大学 | Detect the method and system of unmanned plane silhouette target |
| CN109922287A (en) * | 2018-05-02 | 2019-06-21 | 思特威电子科技（开曼）有限公司 | Reduce method, the imaging system of the fixed picture noise of image sensor pixel array |
-
2019
- 2019-09-16 CN CN201910871176.7A patent/CN110677603B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917590A (en) * | 2005-05-26 | 2007-02-21 | 红外线解决方案公司 | Method for fixed pattern noise reduction in infrared imaging cameras |
| WO2014174498A1 (en) * | 2013-04-26 | 2014-10-30 | Ecole Polytechnique Federale De Lausanne (Epfl) | A novel cmos image sensor with event/change detection and reduced data redundancy |
| CN106341579A (en) * | 2015-07-02 | 2017-01-18 | 豪威科技股份有限公司 | Method for reading image pixel array and High Dynamic Range (HDR) imaging system |
| CN108139207A (en) * | 2015-10-16 | 2018-06-08 | 卡普索影像公司 | For capturing the single image sensor of the structure light image of mixing and regular image |
| WO2017201488A1 (en) * | 2016-05-20 | 2017-11-23 | Harpoon Therapeutics, Inc. | Single domain serum albumin binding protein |
| US10171763B2 (en) * | 2017-01-19 | 2019-01-01 | Axis Ab | Method for fixed pattern noise reduction and use of such method |
| CN107592472A (en) * | 2017-10-20 | 2018-01-16 | 上海晔芯电子科技有限公司 | The image-forming correction method of multistage exposure imaging sensor |
| CN109922287A (en) * | 2018-05-02 | 2019-06-21 | 思特威电子科技（开曼）有限公司 | Reduce method, the imaging system of the fixed picture noise of image sensor pixel array |
| CN109389086A (en) * | 2018-10-09 | 2019-02-26 | 北京科技大学 | Detect the method and system of unmanned plane silhouette target |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110677603A (en) | 2020-01-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9899447B2 (en) | Solid-state imaging device, method for driving the same, and imaging device | |
| CN107786821B (en) | Pixel circuit and operation method thereof | |
| EP2288142B1 (en) | Solid-state image sensor | |
| US7050094B2 (en) | Wide dynamic range operation for CMOS sensor with freeze-frame shutter | |
| US20100271517A1 (en) | In-pixel correlated double sampling pixel | |
| US7746398B2 (en) | Wide dynamic range active pixel with knee response | |
| CN102685403B (en) | Method for expanding dynamic range of time-delay integration-complementary metal oxide semiconductor (TDI-CMOS) image sensor | |
| JP2008028678A (en) | Imaging element | |
| KR20010042768A (en) | Active pixel cmos sensor with multiple storage capacitors | |
| CA3047698A1 (en) | Global shutter scheme that reduces the effects of dark current | |
| TWI622301B (en) | Method and system for reducing analog-to-digital conversion time for dark signals | |
| US10051216B2 (en) | Imaging apparatus and imaging method thereof using correlated double sampling | |
| CN110677603B (en) | Image sensor for single-frame segmented exposure and data acquisition method | |
| US20030231252A1 (en) | Image sensor with improved noise cancellation | |
| CN110708481B (en) | Fixed pattern noise removing method based on difference reset | |
| JP4517638B2 (en) | Solid-state imaging device, image input device, and noise removal method for solid-state imaging device | |
| CN108924445B (en) | Image reading circuit based on ROI, image reading method and image detector | |
| KR20240042621A (en) | Self-correcting barrier modulation pixel with multiple barriers, double conversion gain and low area | |
| WO2021051999A1 (en) | Method for removing fixed pattern noise | |
| JP2023540051A (en) | Event sensors and methods for generating signal streams containing event data | |
| CN114363537B (en) | Pixel structure and array, circuit and method, image sensor, and storage medium | |
| CN117813835A (en) | Pixel device, image sensor, and method of operating pixel device | |
| Willems et al. | High-speed VGA resolution CMOS image sensor with global shutter | |
| CN117793540A (en) | Photosensitive processing apparatus and method | |
| JP2002094882A (en) | Mos solid-state image pickup device and method for driving the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |