Image sensor based on spad and method of operation of the image sensor

Abstract

The present disclosure provides an image sensor and an operating method thereof, wherein the image sensor includes a pixel array including a plurality of pixels. Each of the plurality of pixels includes a SPAD that senses a photon and generates a hit signal, a counter that counts a number of received pulse signals, a multiplexer that selects one of a plurality of counting bit cells of the counter according to a mode selection signal and outputs an overflow signal according to a bit value of the selected counting bit cell, and a front-end module that converts the hit signal into a pulse signal and transmits a global clock signal converted into a pulse signal to the counter when the overflow signal is received. Thus, it is possible to sense light with high accuracy at low power consumption, and an interval using an extrapolation method can be variable, so it is possible to sense light while adjusting accuracy.

Description

Technical Field

[0001] This disclosure relates to an image sensor and a method for operating the image sensor, and more specifically, to a SPAD-based image sensor and a method for operating the image sensor. Background Technology

[0002] An image sensor is a device that realizes an image by sensing incident light and converting it into an electrical signal. It can include various sensing elements capable of sensing light. Existing optical sensing elements mainly use CMOS or CCD elements. However, with the increasing demand for ultra-high sensitivity optical sensors, single-photon avalanche diodes (SPADs) that can directly detect the smallest unit of light, a single photon, have recently been developed. Research on image sensors based on SPAD elements is booming.

[0003] SPAD elements, by applying a reverse bias voltage exceeding their breakdown voltage, can be triggered by a single photon to induce avalanche breakdown, thus enabling ultra-high sensitivity in light sensing. When a SPAD element senses a photon and triggers avalanche breakdown, it generates a large instantaneous current, outputting an impulse waveform signal. Subsequently, the SPAD element undergoes quenching after outputting the impulse signal, returning to its previous state. Therefore, SPAD-based image sensors can detect the intensity of incident light and acquire an image by counting the number of impulse signals output by the SPAD elements in each pixel. That is, when a large number of photons are incident, increasing the count of impulse signals, a bright image is obtained; conversely, when the count of impulse signals decreases, a dim image is obtained.

[0004] In a SPAD-based image sensor, each pixel is equipped with a SPAD element and a counter for counting the number of impact signals generated by the SPAD element. The counting range and number of bits of the counter need to be predetermined according to the application scenario of the image sensor.

[0005] In this SPAD-based image sensor, the SPAD element, as previously described, operates through an avalanche breakdown phenomenon induced by applying a high reverse bias voltage. This results in significant power consumption each time an impact signal is generated. Specifically, there is a problem of extremely high power consumption when detecting bright images generated by frequent repetition of impact signals.

[0006] Furthermore, image sensors need to detect light intensity in different ranges depending on the application, but the number of bits in the counters of existing SPAD-based image sensors is fixed and predetermined. This necessitates manufacturing image sensors with counters of different bit lengths for different application scenarios, thereby increasing manufacturing costs. Summary of the Invention

[0007] The technical problem to be solved

[0008] The purpose of this disclosure is to provide an image sensor capable of sensing light with ultra-high precision while reducing power consumption, as well as a method for operating the image sensor.

[0009] The purpose of this disclosure is to provide an image sensor capable of adjusting the accuracy of light sensing according to its application, and a method for operating the image sensor.

[0010] Invention Effects

[0011] The image sensor and its operation method disclosed herein are based on an extrapolation method. This method allows for estimation of the number of impact signals generated by the SPAD element without continuously counting them during the exposure time, thereby shortening the drive time of the SPAD element and enabling ultra-high precision light sensing with low power consumption. Furthermore, the range using the extrapolation method can be varied according to the application scenario, allowing for light sensing while adjusting accuracy. Attached Figure Description

[0012] Figure 1 This shows a schematic structure of a SPAD-based image sensor.

[0013] Figure 2 express Figure 1 A schematic diagram of the pixel structure.

[0014] Figure 3 A schematic structure of pixels is shown in one embodiment.

[0015] Figure 4 This is to illustrate the application to having Figure 3 The figure shows the extrapolation method of the SPAD-based image sensor for the pixels shown.

[0016] Figure 5 A schematic structure of pixels representing another embodiment.

[0017] Figure 6 This describes an embodiment of an operation method for a SPAD-based image sensor.

[0018] Technical means to solve the problem

[0019] According to one embodiment of this disclosure, an image sensor includes a pixel array comprising a plurality of pixels. Each of the plurality of pixels includes: a SPAD that senses photons and generates an impulse signal; a counter that counts the number of received pulse signals; a multiplexer that selects one of a plurality of counting bit units of the counter according to a mode selection signal and outputs an overflow signal according to the bit value of the selected counting bit unit; and a front-end module that converts the impulse signal into the pulse signal, and when the overflow signal is received, converts a global clock signal into the pulse signal and transmits it to the counter.

[0020] When performing light sensing operations, the front-end module can drive the SPAD; when the front-end module receives the overflow signal, it can stop driving the SPAD.

[0021] When performing light sensing operation, the front-end module can convert the impact signal or the global clock signal received within the set exposure time into the pulse signal and transmit it to the counter.

[0022] The front-end module can respond to the overflow signal once within a set exposure time and receive the global clock signal.

[0023] The front-end module can initialize the counter after the set exposure time.

[0024] The multiplexer can be implemented as an integral part of either the front-end module or the counter.

[0025] The image sensor also includes a control module that generates the mode selection signal and, after a set exposure time, can receive a count value from the counter to obtain pixel values.

[0026] The control module can obtain the overflow quantity and overflow time from the count value based on the counter bit unit selected according to the mode selection signal, and calculate and obtain the pixel value by extrapolation based on the obtained overflow quantity and overflow time and the exposure time.

[0027] The control module can select one from multiple timing gain tables based on the selected counting bit unit, and then search for the pixel value corresponding to the count value in the selected timing gain table and obtain the pixel value. The multiple timing gain tables are tables that are pre-calculated and store the pixel values ​​corresponding to each count value.

[0028] The control module independently sets the selected counting bit unit in each of the plurality of pixels and generates the mode selection signal.

[0029] An image sensor operation method according to another embodiment of the present disclosure, wherein the image sensor includes a pixel array and a control module, the pixel array comprising a plurality of pixels each including a SPAD and a counter. The operation method includes: generating a mode selection signal for selecting a counting bit unit from a plurality of counting bit units of the counter for outputting an overflow signal; counting the number of impulse signals generated by photons sensed by driving the SPAD; generating an overflow signal in response to the bit value of a specific counting bit unit, the specific counting bit unit being selected from the counter according to the mode selection signal; and counting a global clock signal when the overflow signal is generated. Detailed Implementation

[0030] Hereinafter, specific embodiments of the present disclosure are described with reference to the accompanying drawings. The following detailed description is intended to help to fully understand the methods, apparatus, and / or systems described herein. However, these are merely examples, and the invention is not limited thereto.

[0031] In describing embodiments of this disclosure, detailed descriptions of prior art related to the present invention may be omitted if they unnecessarily obscure the key points of the embodiments. Furthermore, the terminology used below is based on the functional definition of the present invention and may vary depending on the intent or convention of the user or operator. Therefore, its definition should be based on the overall content of this specification. The terminology used in the specific embodiments is only for describing some embodiments and should not be considered limiting. Unless otherwise expressly stated, singular expressions include the meaning of plural forms. In this description, expressions such as "comprising" or "having" are intended to refer to certain features, quantities, steps, operations, elements, a portion thereof, or combinations thereof, and should not be construed as excluding the existence or possibility of one or more other features, quantities, steps, operations, elements, a portion thereof, or combinations thereof besides those described. Furthermore, terms such as "...part," "...device," "module," and "block" used in the specification refer to a unit that processes at least one function or operation, which can be implemented by hardware, software, or a combination of hardware and software.

[0032] Figure 1 A schematic diagram showing the structure of a SPAD-based image sensor. Figure 2 express Figure 1 A schematic diagram of the pixel structure.

[0033] Reference Figure 1 The SPAD-based image sensor includes a pixel array 10 and a control module 20. The pixel array 10 includes a plurality of pixels px arranged in an array, for example, it may include an M × N arrangement of pixels px. Figure 2 express Figure 1 An example of the structure of a pixel 30 in a pixel array 10, which consists of multiple pixels px.

[0034] Figure 2 The pixel 30 shown may be composed of a SPAD 31, a front-end module 32, and a counter 33.

[0035] SPAD 31 can employ a reverse bias voltage V above the applied breakdown voltage. SPAD (For example, 23V) and implemented using an avalanche diode operating in Geiger mode. The SPAD 31 can respond to photons with ultra-sensitivity, triggering an avalanche breakdown phenomenon to generate an impulse signal.

[0036] The front-end module 32 controls the drive of SPAD 31 according to the control module 20D, and converts the impact signal output from the driven SPAD 31 into a pulse signal and transmits it to the counter 33.

[0037] As an example, the front-end module 32 applies a specified level of drive control voltage to the anode of SPAD 31 according to the control of the control module 20, so that the voltage difference between the two ends of SPAD 31 is below the breakdown voltage, thereby preventing SPAD 31 from being driven. However, there are various variations in the method of controlling the drive of SPAD 31.

[0038] Furthermore, the front-end module 32 can perform level shifting to ensure that the pulse signal has a lower voltage level than the impulse signal before transmitting the pulse signal to the counter 33. As described above, when a high voltage (e.g., 23V) reverse bias voltage V is applied to SPAD 31... SPAD When photons are incident and avalanche breakdown is triggered, a high-voltage impulse signal can be generated. To prevent such impulse signals from being directly transmitted to the counter 33, the front-end module 32 can shift the applied high-voltage impulse signal level to a low-voltage level (e.g., 3V) and convert it into a pulse signal to be transmitted to the counter 33.

[0039] In addition, the front-end module 32 can quickly quench the SPAD 31 that has experienced avalanche breakdown and restore it to its previous state, and reset the counter 33, thereby initializing the pixel value counted by the counter 33.

[0040] The front-end module 32 may include multiple switches, multiple logic elements and level shifting circuits implemented using transistors, but the constituent elements and circuit structure of the front-end module 32 can have various designs, and detailed descriptions are omitted here.

[0041] Counter 33 counts the number of pulse signals received from front-end module 32 and obtains the pixel value obtained by measuring the amount of light (or intensity of light) incident on pixel 30. In this case, counter 33 can be implemented using an N-bit counter, and the number of bits N of counter 33 is preset according to the application scenario of image sensor.

[0042] The control module 20 controls the front-end module 32 of each pixel in the pixel array 10 to control the driving of the SPAD 31 included in each pixel px. For example, the control module 20 can enable the multiple pixels px of the pixel array 10 to sense light simultaneously or to sense light in rows or columns. Depending on the situation, it can also enable only a portion of the multiple pixels px of the pixel array 10 to selectively sense light according to different combinations. Furthermore, the control module 20 can adjust the exposure time by controlling the front-end module 32, which represents the time for each pixel in the multiple pixels px to sense photons.

[0043] Furthermore, the control module 20 can receive pixel values ​​transmitted from the counter 33 corresponding to each pixel to form an image. That is, the control module 20 not only operates as a control circuit for controlling the pixel array 10, but also as an image processing circuit.

[0044] SPAD 31, front-end module 32 and counter 33 can all be implemented on the same layer, but depending on the situation, SPAD 31, front-end module 32 and counter 33 can be implemented on different layers.

[0045] exist Figure 2 In the case of the pixels shown, the number of bits N in counter 33 is preset and fixed, therefore the range of pixel values ​​output by each pixel 30 is 0 to 2. N -1.

[0046] Therefore, although the number of impact signals generated by the SPAD 31 of each of the multiple pixels 30 during the exposure time is unlimited, the counter 33 can count a maximum of 2. N -1 impulse signal. Therefore, when the value counted by counter 33 reaches the maximum value of 2... N When the value is -1, the front-end module 32 stops driving SPAD 31 to reduce power consumption.

[0047] Figure 3 A schematic structure of pixels is shown in one embodiment.

[0048] Figure 3 It also shows Figure 1 An example of the composition of each pixel in the multiple pixels px of the pixel array 10 in the image sensor shown. Figure 3 Pixel 40 and Figure 2Similarly, it may also include SPAD 41, front-end module 42, and counter 43.

[0049] Here, SPAD 41 and Figure 2 SPAD 31 has the same structure, so it will not be described in detail. However, the front-end module 42 and the counter 43 can adopt the same... Figure 2 The front-end module 32 and the counter 33 are constructed in different ways. Specifically, the counter 43 is basically the same as... Figure 2 Counter 33 is implemented using a fixed N-bit counter, just like other counters. However, in... Figure 3 In counter 43, when the value counted based on the received pulse signal exceeds a specified reference value, an overflow signal OF can be transmitted to front-end module 42. As an example, it is assumed that counter 43 transmits the overflow signal OF to front-end module 42 based on the value of the most significant bit (MSB). That is, when the value counted in N-bit counter 43 reaches 2... N-1 When MSB is set to 1, the overflow signal OF can be output to the front-end module 42. However, the bit position of the output overflow signal OF can be preset to different positions according to the application scenario.

[0050] Furthermore, when the front-end module 42 receives the overflow signal OF from the counter 43, it stops driving the SPAD 41. Therefore, the SPAD 41 no longer generates impulse signals. Instead, the front-end module 42 receives the global clock signal φ. GCLK Instead of an impulse signal, it can be based on the received global clock signal φ GCLK A pulse signal is transmitted to counter 33. That is, when the front-end module 42 receives the overflow signal OF, it can transmit a pulse signal according to the global clock signal φ. GCLK Instead of using an impact signal to generate a pulse signal, the pulse signal is transmitted to the counter 43.

[0051] Therefore, counter 43 first counts the number of pulse signals generated based on the impulse signal generated by SPAD 41, and then counts the number of pulse signals generated based on the global clock signal φ. GCLK The number of pulse signals generated is counted. Since it can be counted up to 2... N The N-bit counter 43, with a value of -1, can only count up to 2 for pulses based on the impulse signal. N-1 The value of , therefore, counter 43 has 2 remaining. N-1 +1 ~ 2 N Within the range of -1, based on the global clock signal φ GCLK The number of pulses is counted.

[0052] Furthermore, the front-end module 42 can, under the control of the control module 20, adjust the exposure time T.EXP The internal clock signal φ will be used as the basis for the operation. GCLK The pulse signal is transmitted to counter 43.

[0053] Figure 4 This is to illustrate the application to having Figure 3 The figure shows the extrapolation method of the SPAD-based image sensor for the pixels shown.

[0054] In pixel array 10, multiple pixels (px) are generated by Figure 2 In the case of a configuration of pixels 30, each pixel 30 can, under the control of the control module 20, adjust the exposure time T according to the preset setting. EXP The number of incident photons during the period is counted and the pixel value N is obtained. PH .

[0055] Here, the exposure time T EXP It is pre-set; let's assume that in each pixel (px), the photon concentration during exposure time T... EXP The radiation is uniformly incident during this period. Therefore, as... Figure 4 As shown, even without adjusting the exposure time T EXP During this period, all incident photons are counted. Once it is known that the number of incident photons reaches a certain threshold (in this example, the overflow number N), the count is determined. OF The time of overflow is T. OF This allows for easy estimation of the exposure time T using a simple proportional formula based on extrapolation. EXP The number of photons incident during the period, i.e., the pixel value N. PH .

[0056] Reference Figure 4 Pixel value N based on extrapolation method PH It can be calculated according to mathematical formula 1.

[0057]

Mathematical Formula 1

[0058]

[0059] The overflow quantity N in mathematical formula 1 OF and exposure time T EXP It is preset, so as long as the overflow time T can be confirmed... OF This allows for easy calculation of pixel value N. PH .

[0060] Furthermore, in Figure 3 In the 40-pixel structure, the overflow time T OF This refers to the time when the overflow signal OF is generated. However, in order to use the global clock signal φ... GCLK To directly measure the overflow time T of the overflow signal OF. OFIn addition to the counter 43 that needs to count the number of pulse signals based on the impact signal, it also needs to count the global clock signal φ. GCLK A separate counter is used to count the number of items. However, due to the exposure time T... EXP It is preset, therefore utilizing the global clock signal φ GCLK To measure the overflow time T from the generation of the overflow signal OF OF Then until the exposure time T EXP up to the time (T) MEA ), and then from the exposure time T EXP Subtract the measured time T MEA (T) EXP - T MEA The overflow time T can be easily calculated. OF .

[0061] As mentioned above, due to in Figure 3 In the pixel 40 shown, counter 43 is assumed to only work for a maximum of N overflows. OF That is, 2 N-1 The value is based on the impact signal and is counted accordingly. Therefore, in the subsequent 2 N-1 +1 ~ 2 N The range of -1 refers to the global clock signal φ. GCLK The counter 43, which is set up to count the number of impact signals, can count the number of pulses during the overflow time T. OF It was then reused for the global clock signal φ GCLK The count of the number of clocks, global clock signal φ GCLK Used to determine measurement time T MEA Therefore, the pixel value N can be estimated by extrapolation using only a counter 43. PH .

[0062] Pixel value N PH The count value can be estimated by control module 20, which receives the count value obtained by counter 43. Here, since counter 43 is assumed to output an overflow signal OF based on the bit value of MSB, control module 20 can confirm the counted global clock signal φ from the bit values ​​other than MSB output by counter 43. GCLK The quantity. And, it can be determined from the global clock signal φ. GCLK The quantity is used to estimate the overflow time T OF And the estimated overflow time T OF Substituting into mathematical formula 1, we can obtain the pixel value N. PH However, for each pixel in the multiple pixels px of the pixel array 10, the pixel value N is... PHEstimation requires a lot of computation, making it inefficient. Therefore, the control module 20 is configured to directly obtain the pixel value N using a timing-gain table. PH This significantly reduces the computational load. The timing gain table is pre-calculated and stores the pixel value N corresponding to the count value. PH The table.

[0063] exist Figure 4 In pixel 40, at overflow time T OF Then until the exposure time T EXP Until then, the front-end module 42 stops driving SPAD 41, thus greatly reducing the power consumption of the pixel array 10.

[0064] Figure 5 A schematic structure of pixels representing another embodiment.

[0065] Figure 5 The 50 pixels are basically the same as Figure 4 Like pixel 40, it may also include SPAD 51, front-end module 52, and counter 53. SPAD 51 and front-end module 52 can be adopted with... Figure 4 The pixel 40 operates in the same way as the SPAD 41 and the front-end module 42.

[0066] That is, when SPAD 51 is driven by the front-end module 52 and senses photons, it outputs an impact signal generated by the avalanche breakdown phenomenon. Furthermore, the front-end module 52 controls the driving of SPAD 51 and converts the impact signal received from SPAD 51 into a pulse signal, which is then output to counter 53. When an overflow signal OF is received from counter 53, the driving of SPAD 51 is stopped, and the global clock signal φ is received. GCLK Until exposure time T EXP And based on the global clock signal φ GCLK The pulse signal is output to counter 53.

[0067] Counter 53 can be used with Figure 2 and Figure 4 Counters 33 and 43 are implemented using an N-bit counter capable of counting N values, and count the number of pulse signals transmitted from the front-end module 52. Furthermore, counter 53 outputs an overflow signal OF to the front-end module 52 based on the count value. However, Figure 5 The counter 53 does not output the overflow signal OF based solely on the bit value of a pre-set specific bit (e.g., MSB), but rather outputs the overflow signal OF based on the bit value specified by the mode selection signal sel.

[0068] Therefore, such as Figure 5As shown, counter 53 also includes a multiplexer (MUX) 54 for receiving a mode selection signal sel. The multiplexer 54 is connected to a designated count bit cell among the multiple count bit cells in counter 53 that store bit values ​​respectively, receives the bit values, selects the bit value of a specific count bit cell according to the mode selection signal sel, and outputs an overflow signal OF.

[0069] exist Figure 5 The following example illustrates how multiplexer 54 selects one of the bit values ​​of the (k+m)th, kth, and kmth bits according to the mode selection signal sel, and outputs an overflow signal OF based on the selected bit value. Here, the (k+m)th bit can be the MSB of an N (=k+m) bit counter 53. When multiplexer 54 selects the (k+m)th bit according to the mode selection signal sel, Figure 5 The pixels can be used with Figure 4 Operate in the same way for pixels.

[0070] However, multiplexer 54 can select the k-th or km-th bit based on the mode selection signal sel and output an overflow signal OF. That is, Figure 5 The counter 53 in the middle can be operated in a similar way to the variable bit counter, which adjusts the count value of the output overflow signal OF according to the mode selection signal sel.

[0071] Therefore, counter 53 can adjust the overflow time T. OF The number of countable impact signals. Furthermore, when the number of countable impact signals is adjusted, during the measurement time T using the extrapolation method... MEA Measurable global clock signal φ during the period GCLK The quantity can also be adjusted. As mentioned above, Figure 4 and Figure 5 The pixel structure is configured such that using a single counter 43, 53 not only allows for monitoring of overflow time T OF It can count the number of impact signals within the system and also count the global clock signal φ. GCLK The quantity is also counted together.

[0072] Therefore, when the number of countable impulse signals is adjusted, the countable global clock signal φ GCLK The quantity can also be adjusted. For example, when the multiplexer 54 selects the km-th bit from the previously set k-th bit according to the mode selection signal sel, causing the number of countable impulse signals to decrease, the overflow time T used for directly counting the number of input photons is adjusted. OF It will shorten, and according to the global clock signal φ GCLK The quantity is used to extrapolate the measurement time T. MEA This will increase. In this case, compared to using only extrapolation... Figure 4 Compared to the previous version, while it can further reduce the power consumption of SPAD 51, the accuracy may be reduced accordingly.

[0073] When multiplexer 54 selects the (k+m)th bit according to the mode selection signal sel, causing the number of countable impulse signals to increase, the overflow time T... OF It will increase, while the measurement time T MEA This will shorten the time. In this case, although the power consumption of SPAD 51 increases, the accuracy may improve accordingly.

[0074] Therefore, by adjusting the overflow time T according to the application scenario OF A trade-off can be made between power consumption and accuracy. When using a variable-bit counter, the countable range of the counter itself changes, therefore the overflow time T... OF and measurement time T MEA These will increase or decrease together, causing power consumption and accuracy to increase or decrease together, and vice versa. Figure 5 As shown, when configured to select the bit used for the output overflow signal OF, the overall countable range of the counter remains unchanged, thereby enabling a balance between power consumption and accuracy.

[0075] Although it is stated here that the multiplexer 54 is located within the counter 53, the multiplexer 54 can be located in the front-end module 52, or it can be configured independently of the counter 53 or the front-end module 52.

[0076] Furthermore, when the bit used to output the overflow signal OF is not the MSB, the bit value may change again as the counter 53 increases, causing a change in the state of the overflow signal OF. However, this can be mitigated by setting the front-end module 52 to operate at the exposure time T. EXP This problem can be easily solved by responding only once to the overflow signal OF during the exposure time T, and then not responding again. The front-end module 52 responds only once to the overflow signal OF during the exposure time T. EXP When the counter 53 is initialized later, it can be reset to respond to the overflow signal OF again.

[0077] The control module 20 applies a mode selection signal sel to the multiplexer 54. Furthermore, it calculates the pixel value N according to mathematical formula 1. PH However, the overflow time T after adjusting the mode selection signal sel can be considered. OF and measurement time T MEA Calculate pixel value N in different ways PH However, it can also have multiple timing gain tables, and after selecting one of the multiple timing gain tables according to the mode selection signal sel, the pixel value N corresponding to the count value of counter 53 is found and obtained. PHThe timing gain table is pre-calculated and stores the pixel value N corresponding to the count value. PH The table.

[0078] Depending on the situation, the control module 20 can cause a portion of the pixels (e.g., pixels in the region of interest) in the pixel array 10 to select the high bit (e.g., the k+m bit) of the mode selection signal sel to improve accuracy, and cause the remaining pixels to select the low bit (e.g., the km bit) to reduce power consumption, thereby achieving a balance between overall power consumption and accuracy in the pixel array 10.

[0079] Therefore, since the SPAD-based image sensor of this disclosure is configured to output an overflow signal OF by using the bit value of one of the multiple counting bit units of the counter 53 of each pixel px, which is variably selected by the mode selection signal sel, adjustments can be made to reduce power consumption or improve accuracy depending on the application scenario.

[0080] In the embodiments described, each structure may have different functions and capabilities in addition to those described below, and may include additional structures not mentioned below. Furthermore, in the embodiments, each structure may be implemented by one or more physically separate devices, or by one or more processors or a combination of one or more processors and software. Unlike the illustrated examples, they may not be clearly distinguishable in actual operation.

[0081] Figure 6 This describes an embodiment of an operation method for a SPAD-based image sensor.

[0082] Figure 6 Also based on Figure 5 To explain using pixel structure Figure 1 The operation of the image sensor. (Refer to...) Figure 6 The control module 20 first sets a counting bit unit for outputting an overflow signal OF in the multiple counting bit units of the counter 53 of each of the multiple pixels px in the pixel array 10, and generates a mode selection signal sel according to the set counting bit unit and transmits it to the multiplexer 54 of each pixel px (S110). Here, the control module 20 can either make the same setting for all pixels px in the pixel array 10, or make different settings for different pixels.

[0083] Each pixel px can be configured to include SPAD 51, front-end module 52, counter 53 and multiplexer 54. The multiplexer 54 of the pixel px that receives the mode selection signal sel selects the counting bit unit corresponding to the mode selection signal sel from the multiple counting bit units of the counter 53.

[0084] Furthermore, the control module 20 controls the front-end module 52 to drive the SPAD 51 of each of the multiple pixels px in the pixel array 10 (S120). At this time, the control module 20 can control all the pixels px in the pixel array 10 to be driven simultaneously, or it can control them to be driven at different times.

[0085] The driven SPAD 51 repeatedly generates impact signals in response to incident photons. The front-end module 52 converts the impact signals generated by the SPAD 51 into pulse signals and transmits them to the counter 53. The counter 53 counts the number of received pulses (S130). As the counter 53 counts the number of pulses, the bit value changes starting from the lowest bit of the multiple counting bit units. When the bit value of the set counter bit unit changes, the multiplexer 54 transmits the overflow signal OF to the front-end module.

[0086] Based on this, the front-end module 52 determines whether an overflow signal OF is generated. If no overflow signal OF is generated, the front-end module 52 continues to drive SPAD 51 and converts the impact signal generated by SPAD 51 into a pulse signal. The counter 53 continuously counts the number of pulses (S120, S130).

[0087] However, if an overflow signal OF is generated, the front-end module 52 stops driving SPAD 51 and receives the global clock signal φ. GCLK This generates a pulse signal and transmits it to counter 53, which then switches to responding to the global clock signal φ. GCLK The number of signals is counted, rather than the number of impact signals (S150).

[0088] Next, determine whether the exposure time T has been reached. EXP (S160). If the exposure time T has not been reached... EXP Then it will still receive the global clock signal φ GCLK Continue counting (S150). If it is determined that the exposure time T has been reached... EXP Then, obtain the count value of the current counter 53 in each pixel px (S170).

[0089] Furthermore, the pixel value N is obtained from the acquired count value. PH (S180). At this time, the pixel value N PH Exposure time T can be used EXP Overflow quantity N OF And from the measurement time T MEA Estimated overflow time T OF The measurement time T is calculated according to mathematical formula 1. MEAIt is obtained from the count value based on the set count bit unit. However, it can also be obtained from the pixel value N, which is pre-calculated and stored according to the set count bit unit and corresponds to each count value. PH After selecting one from multiple timing gain tables, find and obtain the pixel value N corresponding to the count value of counter 53. PH .

[0090] If we obtain the pixel values ​​N from multiple pixels px... PH Then, these can be combined to obtain an image corresponding to the light incident on the image sensor (S190).

[0091] Although Figure 6 The description indicates that the steps are performed sequentially, but this is merely an illustrative example. Those skilled in the art can make further modifications without departing from the essential characteristics of the embodiments of the present invention. Figure 6 The recorded order can be changed, or more than one step can be executed in parallel, or other steps can be added, thus allowing for various modifications and variations.

[0092] Although the present invention has been described in detail through the above representative embodiments, those skilled in the art will understand that various modifications and equivalent embodiments can be derived therefrom. Therefore, the true scope of protection of the present invention should be defined by the technical concept of the appended claims.

Claims

1. An image sensor, comprising a pixel array, the pixel array including a plurality of pixels, Each of the plurality of pixels includes: SPAD, which senses photons and generates impact signals; A counter that counts the number of received pulse signals; The multiplexer selects one of the multiple counting bit units of the counter according to the mode selection signal, and outputs an overflow signal according to the bit value of the selected counting bit unit. as well as The front-end module converts the impact signal into the pulse signal, and when it receives the overflow signal, it converts the global clock signal into the pulse signal and transmits it to the counter.

2. The image sensor according to claim 1, wherein, When performing light sensing operation, the front-end module drives the SPAD; when the front-end module receives the overflow signal, it stops driving the SPAD.

3. The image sensor according to claim 1, wherein, When performing light sensing operation, the front-end module converts the impact signal or the global clock signal received within the set exposure time into the pulse signal and transmits it to the counter.

4. The image sensor according to claim 1, wherein, The front-end module responds to the overflow signal once within the set exposure time and receives the global clock signal.

5. The image sensor according to claim 1, wherein, The front-end module initializes the counter after the set exposure time.

6. The image sensor according to claim 1, wherein, The multiplexer is implemented as an integral part of either the front-end module or the counter.

7. The image sensor according to claim 1, wherein, The image sensor also includes a control module. The control module generates the mode selection signal and, after the set exposure time, receives a count value from the counter to obtain the pixel value.

8. The image sensor according to claim 7, wherein, The control module obtains the overflow quantity and overflow time from the count value based on the counter bit unit selected according to the mode selection signal, and calculates and obtains the pixel value by extrapolation based on the obtained overflow quantity and overflow time and the exposure time.

9. The image sensor according to claim 7, wherein, The control module selects one of the multiple timing gain tables based on the selected counting bit unit, then searches for the pixel value corresponding to the count value in the selected timing gain table and obtains the pixel value. The multiple timing gain tables are tables that are pre-calculated and store the pixel values ​​corresponding to each count value.

10. The image sensor according to claim 7, wherein, The control module independently sets the selected counting bit unit in each of the plurality of pixels and generates the mode selection signal.

11. A method for operating an image sensor, wherein, The image sensor includes a pixel array and a control module. The pixel array consists of multiple pixels, each including a SPAD and a counter. The operation method includes: The step of generating a mode selection signal, wherein the mode selection signal is used to select one of the multiple counting bit units of the counter for outputting an overflow signal; The step of counting the number of impact signals, which are generated by photons sensed by driving the SPAD; The step of generating an overflow signal in response to the bit value of a specific counting bit unit, wherein the specific counting bit unit is a counting bit unit selected from the counter according to the mode selection signal; and The step of counting the global clock signal when the overflow signal is generated.

12. The method of operating the image sensor according to claim 11, wherein, In the step of counting the global clock signal When the overflow signal is generated, stop driving the SPAD.

13. The method of operating the image sensor according to claim 11, wherein, In the step of counting the global clock signal The global clock signal is counted within the set exposure time.

14. The method of operating the image sensor according to claim 11, wherein, In the step of counting the global clock signal Within the set exposure time, respond to the overflow signal once, receive the global clock signal, and count the received global clock signal.

15. The method of operating an image sensor according to claim 11, wherein, In the step of counting the global clock signal After the set exposure time, the counter's count value is initialized.

16. The method of operating the image sensor according to claim 11, wherein, The operation method of the image sensor further includes the step of receiving the count value of the counter and obtaining the pixel value after a set exposure time.

17. The method of operating an image sensor according to claim 16, wherein, In the step of obtaining the pixel value Based on the counter bit unit selected according to the mode selection signal, the overflow quantity and overflow time are obtained from the count value. Based on the obtained overflow quantity and overflow time, as well as the exposure time, the pixel value is calculated and obtained by extrapolation.

18. The method of operating an image sensor according to claim 16, wherein, In the step of obtaining the pixel value One of the selected counting bit units is chosen from a plurality of timing gain tables, which are pre-calculated and store pixel values ​​corresponding to each counting value. Find the pixel value corresponding to the count value from the selected timing gain table and obtain the pixel value.

19. The method of operating an image sensor according to claim 11, wherein, In the step of generating the mode selection signal The counting bit unit selected in each of the plurality of pixels is independently set, and the mode selection signal is generated.

Images