Image sensor

Abstract

The image sensor comprises a photon counting detector element (10) and a first set of pulse shapers (20) for converting a signal (12) of the photon counting detector element (10) into a pulse shaper output stream (22), where the pulse (23) has at least a first pulse length (t1). A connection unit (30) combines the plurality of pulse shaper output streams (22) into at least one macro-pixel stream (32). The at least one macro-pixel stream (32) is processed by at least one monostable / non-stable circuit (40) which outputs a processed stream (42) in which the pulses (43) have a second pulse length (t2). The second pulse length (t2) is shorter than the first pulse length (t1) by a time span tPF such that a pulse-free interval (PF) at least within the time span tPF is generated after each pulse (43) in the processed stream (42).

Description

Technical Field The present disclosure relates to an image sensor including a photon counting detector element and a method of operating an image sensor. Background Art For example, WO2020 / 151838A1 and WO2020 / 207571A1 as well as the following prior art documents all introduce image sensors with photon counting detector elements such as single photon avalanche diodes (hereinafter referred to as SPADs):

[0001] : A. Carimatto et al., “Versatile, Fully Integrated 128x128 Event-Driven MD-SiPM with 512 16-Bit TDCs, 45ps LSB, and 20ns Gating in 40nm CMOS Technology,” IEEE SOLID-STATECIRCUITS LETTERS, vol. 1, no. 12, December 2018, doi: 10.1109 / LSSC.2019.2911043.

[0002] : LHC Braga et al., “All-digital 8x16 SiPM array with per-pixel TDC and real-time energy output for PET applications,” IEEE J. Solid-State Circuits, vol. 49, no. 1, pp. 301-314, January 2014, doi: 10.1109 / JSSC.2013.2284351.

[0003] : M.-J. Lee et al., “High-Performance Back-Illuminated 3D Stacked Single-Photon Avalanche Diodes Implemented in 45nm CMOS Technology,” IEEE J. Sel. Top. Quantum Electronics, vol. 24, no. 6, pp. 1–9, November 2018, doi: 10.1109 / JSTQE.2018.2827669.

[0004] : S. Lindner, C. Zhang, I. Mantolovic, M. Wolf, and E. Charbon: “252 × 144 SPAD Pixel Flash LiDAR with 1728 Dual-Clocked 48.8P STDC, Integrated Histogramming, and 14.9-to-1 Compression in 180NM CMOS Technology,” in 2018 IEEE VISE Symposium on Circuits, June 2018, pp. 69–70, doi: 10.1109 / VLSIC.2018.8502386. Such image sensors consist of an array of detectors connected to electronics that measure and process the timing and intensity information of the detected photons. The processed information is transmitted to an external system. Raw data is rarely sent because this would quickly increase the requirements of the communication system to impractical limits. As technology rapidly advances, the number of SPADs in a SPAD array increases, and the interconnection between the SPADs and the electronics becomes increasingly difficult to achieve. In this case, it becomes impractical to provide an electronic module that serves a single SPAD. Architectures that combine the pulse signals of multiple SPADs into a single signal as an input signal to the electronic device are very common. SPADs whose outputs are combined together are said to form a group or a macropixel, which has the aggregated information of its individual SPADs. In this architecture, the timing and intensity are preserved, but the address or position of the SPAD in its group is usually lost. Many applications do not need to distinguish which of the several SPADs in the macropixel generated the pulse. The electronics that serve the macropixel may include time-to-digital converters (TDCs), counters, level comparators, triggers, and scalers. Whichever module is required for the application, a failure in the macropixel output line may cause module instability. The photons that trigger the SPADs arrive at the sensor randomly and asynchronously. The combination of pulses generated (usually but not limited to using an OR connection) may produce spurious pulses that are smaller in voltage or faster / shorter in time than the expected pulse from the electronics. In the system, the memory (bank) of the electronics is usually double buffered and swapped over time, and if this switching coincides in time with the arrival of the photons, spurious pulses may also appear. These undesirable pulses or failures do not meet the input ratings of the subsequent electronics and may cause them to fail at varying severity levels that may be unacceptable to the target field. For easier understanding, see Figure 1 , the figure shows a related image sensor 1', which includes an array of photon counting detector elements 10 (i.e., SPADs). The signals 12 of the photon counting detector elements 10 are processed by a pulse shaper (i.e., a monoflop 20'). For consistency, the output of the pulse shaper / monoflop will be referred to as a pulse shaper output stream 22. A pulse 13 in one of the signals 12 represents a registered photon. The monoflop 20' can shorten the pulse length so that a pulse 23 in the pulse shaper output stream 22 has a first pulse length t1 that is shorter than the original pulse 13. A connection unit 30 (eg, an OR element) combines the pulse shaper output streams 22 from the monostable flip-flops 20' into a macro-pixel stream 32. Figure 2 This is described in more detail in Figure 2 As shown in more detail, Figure 2Two pulse shaper output streams 22 (originating from two monoflops 20' and hence from two photon counting detector elements 10) are depicted, wherein each pulse shaper output stream 22 comprises several pulses 23 representing detected photons. The macropixel stream 32 comprises all pulses 23 of the two pulse shaper output streams 22. Figure 3 shows a similar situation, however, the pulses 23 from different SPADs are close to each other. This has the result that the gap between two pulses 23 has a short time period t_short in the macropixel stream 32. The time period t_short can be any value, since photons can hit the SPAD at any time. However, if t_short is too small, errors may occur in the subsequent electronic modules: Figure 4 shows the rise and fall time requirements for electronic modules such as counters, TDCs or level detectors. The pulses of the electronic module input signal 52 must have a minimum length tp and the gaps between pulses must have a minimum length tn. The minimum length tn of the pulse-free interval is required as a recovery time to restore the internal state settings of the electronic module to the initial state so that the next pulse can be processed correctly. Otherwise the electronic module may malfunction or output wrong results. The exact values ​​of tp and tn depend on the specific electronic module. Reference Figure 3 , it can be ensured that the pulse 23 always meets the tp requirement, because the pulse length t1 can be selected by the appropriate design of the monostable trigger 20', so as to always meet the requirement of t1 ≥ tp. However, the time span t_short of the gap between pulses 23 can be any value, so t_short may be shorter than the required minimum time span tn, which may cause errors in the electronic module. This is an example of a conventional image sensor that generates faults or pulses that do not meet the requirements of the electronic module, which may cause errors in subsequent processing steps, such as metastable states of the electronic module, incorrect photon count values, or reduced signal-to-noise ratio. The present invention can solve the problem caused by irregular pulses and avoid malfunctions in downstream electronic devices. Summary of the invention The object of the present invention is to provide an image sensor and a method to reduce or eliminate irregular pulses or failures in the image sensor, especially in its macro-pixel output lines, or in electronic modules connected to the macro-pixel output lines. The above objects are achieved by an image sensor and a method comprising the features of the respective independent claims. An image sensor according to an embodiment of the present invention comprises photon counting detector elements and a first group of pulse shapers (e.g., monostable triggers), the first group of pulse shapers being arranged to convert signals of the photon counting detector elements into a pulse shaper output stream, wherein pulses in the pulse shaper output stream have at least a first pulse length. The pulse shapers may be arranged in parallel with each other, i.e., each pulse shaper of the first group receives a signal from another photon counting detector element. The number of pulse shapers of the first group may match the number of photon counting detector elements. Each output of any pulse shaper of the first group is referred to as a pulse shaper output stream. For each photon counting detector element, a corresponding pulse shaper output stream may be generated. The image sensor further comprises a connection unit configured to combine a plurality of pulse shaper output streams into at least one macropixel stream. The image sensor further comprises at least one monostable / astable circuit (i.e., at least one monostable circuit and / or at least one astable circuit), which is arranged to receive at least one macropixel stream and output a processed stream, wherein pulses in the processed stream have a second pulse length. The at least one monostable / astable circuit is configured such that the second pulse length is shorter than the first pulse length by a time span t_PF so as to generate a pulse-free interval at least within the time span t_PF after each pulse in the processed stream. In one embodiment of the image sensor operation method of the present invention, photons are received by photon counting detector elements, and the photon counting detector elements output signals indicating the received photons. A first group of pulse shapers converts the signals from the photon counting detector elements into a pulse shaper output stream, wherein the pulses have a first pulse length. A connecting unit combines a plurality of pulse shaper output streams into at least one macro-pixel stream. At least one monostable / astable circuit processes / receives the at least one macro-pixel stream and outputs a processed stream, wherein the pulses have a second pulse length, and the second pulse length is shorter than the first pulse length by a time span t_PF. In this way, after each pulse in the processed stream, a pulse-free interval is formed at least within the time span t_PF. The monostable / astable circuit ensures a minimum distance between each pulse and the next one. This makes it possible to avoid that two pulses in the processed stream are too close to each other. The pulse-free interval is created by using monostable flip-flops arranged in series, where the latter monostable flip-flop (i.e. the monostable / astable circuit) returns to a stable state in a shorter time than the first monostable flip-flop (i.e. the pulse shaper). The pulse-free interval of t_PF reduces the risk of errors in the subsequent electronics, which require a certain minimum pause between pulses. If t_PF is selected according to the required minimum pause time, i.e. t_PF is at least as large as the required minimum pause time, the risk of two consecutive pulses being too close to each other for correct processing can be completely excluded. For the sake of clarity, it should be noted that the pulse-free interval is created only if the monostable / astable circuit is arranged downstream of the connection unit so that it receives the macropixel stream. In fact, the arrangement of the monostable / astable circuit leads to a dead time during which no photon registration can be transmitted, but this dead time avoids or reduces the risk of failure of other electronics that may process the processed stream. Alternative Implementation Advantageous variants of the image sensor according to the invention and of the method according to the invention are the subject matter of the dependent claims and are explained in more detail in the following description. Electronic module and pulse-free interval length The image sensor may include an electronics module for processing a signal stream including information from the photon counting detector element. The electronics module may be mounted on the same chip as the photon counting detector element. In general, the electronic module can be any electronic device or circuit that performs strength or time measurement on a received signal. The electronic module may include or contain at least one of the following: a time-to-digital converter (TDC), a digital counter, a level comparator, a trigger, a scaler, an inverter, a buffer, a pull-up, a latch, a level monostable trigger, a delay line, a clock divider, a phase detector, a prescaler for reducing the frequency of an input signal to a downstream component, or any other digital circuit. The TDC outputs digital data indicating, for example, the arrival time of a detected photon. The digital counter outputs a count value representing the number of photons detected. The level comparator can be configured to output data indicating whether or when the level in the received signal is above a specific threshold. The scaler can be configured to aggregate pulses in the received signal into a single output pulse; the single output pulse can specifically represent a predetermined number of input pulses. Other types of scalers can be configured to convert the voltage level of a received signal to a different voltage level. Whereas a conventional image sensor may include an electronics module that may be arranged to directly receive the combined output of the pulse shaper (referred to herein as the macropixel stream), in various embodiments of the present invention, the electronics module may be arranged to receive a processed stream, i.e., a signal output by a monostable / astable circuit downstream of the pulse shaper. The recovery time of the electronic module may be tn, i.e., the electronic module requires a pause (pulse-free interval) of at least tn between two pulses in the processed stream in order to distinguish or correctly process the pulses in the processed stream. The "required" pause may be understood as meaning that if tn is not met, the pulse will be processed incorrectly, or if the pause is shorter than tn, the risk of incorrect processing will exceed a predetermined threshold. The at least one monostable / astable circuit can be configured such that t_PF is at least as long as tn, i.e. t_PF ≥ tn. In other words, the second pulse length t2 created by the at least one monostable / astable circuit is at least tn shorter than the first pulse length t1 of the pulse shaper. In this case, it is always ensured that the two pulses are sufficiently far apart so as to meet the recovery time of the electronic module. To avoid unnecessarily long dead time, an upper limit of the minimum pulse-free interval t_PF may be defined. In particular, the first set of pulse shapers and at least one monostable / astable circuit may be configured such that t_PF is less than 2 times tn (or 3 times tn). The electronic module may also require that the pulses in the processed stream have a minimum pulse length of tp. This requirement may be met if at least one monostable / astable circuit is configured so that the second pulse length t2 is at least as long as tp. Optionally, the first set of pulse shapers may be configured so that the first pulse length t1 is at least as long as tp+tn to allow for sufficiently long pulse-free intervals. Optionally, t1 is less than 2*(tp+tn) to avoid unnecessarily long dead times. If the processed streams provided to the electronic modules have different recovery times tn, the longest tn should be selected in the described embodiment. The image sensor may generate multiple processed streams for input to different electronic modules (such as TDC and counter), which have different requirements for tp and tn. In this case, the monostable / astable circuits can be different from each other to generate different second pulse lengths t2, depending on the respective tp and tn of the electronic modules described below. All pulse shapers of the first group of a macropixel (i.e., pulse shapers connected to the same monostable / astable circuit) can generate pulses with the same pulse length t1; however, the t1 of each macropixel may be different, i.e., the t1 of the first group of pulse shapers connected to one monostable / astable circuit may be different relative to the t1 of the first group of pulse shapers connected to another monostable / astable circuit. In this way, t1 and t2 can be selected according to the respective requirements for tp and tn. Optionally, the first group of pulse shapers and / or monostable / astable circuits can be configured to flexibly adjust their t1 and t2 during the operation of the image sensor. In this case, the t1 and t2 values ​​can be flexibly adjusted according to the current connection set by the connection unit, and the connection unit can adjustably generate and transmit the processed stream to different electronic modules. Considering the flexible instant formation of the macro pixel, it is possible to set a suitable t1 and t2 according to the respective tn and tp of each electronic module. Pulse shapers such as monostable flip-flops and monostable / astable circuits A pulse shaper may be understood as an electronic component that outputs a pulse for each received pulse, wherein the length of the output pulse is at least t1. All output pulses may have the same length or may have different lengths equal to or greater than t1. A pulse shaper may specifically be or include a monostable trigger, which may be understood as an electronic circuit or unit that is configured to receive an input signal and process the input signal to generate an output signal that changes between a stable state (no signal / no pulse) and an unstable state (pulses are generated when photons are recorded), wherein the monostable trigger has a specific time until its output changes from an unstable state to a stable state. The specific time results in a specific pulse length. Therefore, the photons recorded by the detector element will generate pulses of a common length defined by the monostable trigger / pulse shaper. The monostable trigger may also be called a monostable circuit or a monostable multivibrator. The pulse shapers of the first group of pulse shapers may be, for example, level monostables, which are triggered when the level of the signal from the photon counting detector element exceeds a predetermined threshold. In principle, the pulse shapers may also be edge-delay monostables (edge ​​detector monostables), which are triggered by an edge of the received signal, i.e. the start of a pulse representing the detection of a photon. Some or all of the pulse shapers may also be integrated with the corresponding photon counting detector element, i.e. the photon counting detector element and the pulse shapers may form an integrated unit. Generally speaking, a monostable / astable circuit can be understood as a monostable circuit or an astable circuit. An astable circuit is a circuit or electronic component that generates an output, which switches between at least two states, and none of the states is permanently stable, however, it switches when receiving the first side of a received pulse, and switches back after a time shorter than t1 (or switches back after a time t2, where t2 can represent a specific value or a span value, all of which are shorter than t1). Multiple monostable / astable circuits (each circuit is used to process one of the macro pixel streams) may include different types of monostable or astable states, or some monostable states and some stable states. The monostable / astable circuit can specifically be a level monostable trigger or an edge delayed monostable trigger (EDM). However, the level monostable trigger usually has a recovery time and needs to reset its internal state after receiving a pulse. Therefore, the level monostable trigger requires that the first pulse and the second pulse received are separated by a sufficiently long time, otherwise only the first pulse will be correctly processed as the required pulse length (here t2), while the second pulse will be incorrectly processed as a pulse length shorter than t2. Such short stray pulses may cause errors in the subsequent electronic modules. This problem can be avoided if at least one monostable / astable circuit is an EDM, which is triggered by the edge of each new pulse and can output pulses with constant length even if there is no pause between the first pulse and the second pulse. In addition, the second pulse may also start when the first pulse has not completely terminated, that is, the rising edge of the second pulse may overlap with the falling edge of the first pulse to a certain extent. If the EDM can still recognize the overlapping edges, two pulses with a length of t2 will be output. If the overlap of the first pulse and the second pulse is so strong that the EDM cannot recognize the rising edge of the second pulse as a new pulse, then the EDM will only output one pulse instead of two pulses, and will not generate any stray pulses that may cause errors in the subsequent electronic devices. This effect of suppressing the power supply voltage by about half will produce the effect of additional dead time, which can be as short as 200ps. Connection unit The connection unit is configured to combine a plurality of asynchronous pulse shaper output streams (of the first group of pulse shapers) into at least one macropixel stream. Thus, a macropixel stream is a combination of two or more pulse shaper output streams and comprises all pulses of these pulse shaper output streams. If one of these pulse shaper output streams contains a pulse at a specific time, the macropixel stream will also contain a pulse at this specific time. If none of these pulse shaper output streams contains a pulse at a specific time, the macropixel stream will also not contain a pulse at this specific time. If pulses of different pulse shaper output streams overlap each other in time, the macropixel stream will include a pulse having a length determined by the overlapping pulses. The connection unit may be configured to output a plurality of macropixel streams, which differ in the combination of the pulse shaper output streams. Since the connection unit combines the outputs of a plurality of pixels (= photon counting detector elements), it can be considered that the connection unit defines a macropixel consisting of photon counting detector elements, the photon counting detector element outputs being combined together, and the output of the connection unit is therefore referred to as a macropixel stream. The connection unit may be a connection matrix, which may be configured to flexibly combine any pulse shaper output streams to create reconfigurable macropixels. The connection matrix may include an OR matrix in which a signal is output if any of a plurality of inputs of a logical OR unit receives a signal. The connection matrix may also include other logic units in addition to OR connections to combine a plurality of inputs. The connection matrix can be configured so that which outputs are combined can be changed flexibly at run time, for example using internal memory. The outputs can be combined arbitrarily, for which the connection matrix can be equipped with SRAM memory and / or LUT (LUT: Look Up Table). The connection matrix can be changed flexibly to change the number of macropixels. Furthermore, the number of photon counting detector elements combined into one macropixel can be set flexibly. The connection matrix can also be set to forward only some pulse shaper output streams without combining with others. Assisted by artificial pulse (simulated photon) memory switching (Bank-switch) If a memory switch splits a pulse in two, creating one or two spurious pulses, this can cause errors in electronic components. This is explained below. When an electronic component (such as a counter) is enabled or disabled and receives a pulse at the same time, a stray pulse may be generated, which may cause the counter to enter a metastable state. To avoid metastable states in the counter, pulses during memory switching should be suppressed / rejected. This is achieved by introducing a signal that simulates photons ("simulated photons") before the switch, and lasting long enough until the switch is completed. This technique can be used in conjunction with the described fail-safe circuit to avoid failures if the simulated photon signal overlaps with the real photon pulse in time. In more detail: The electronic module may include at least two memories, such as two counters, wherein one counter counts the input pulses and the other counter can be read out. The image sensor may further include a memory switching unit configured to switch to which of the at least two memories the processed stream is forwarded. The memory switching unit is configured to switch at a specific memory switching time. Without further measures, the memory switching time may coincide with a pulse in the processed stream. This means that the first half of the pulse is transferred to one of the memories and the second half is transferred to the other memory. As a result, both memories receive only partial pulses. If both partial pulses are long enough to be processed as correct pulses, the number of pulses is incorrect, resulting in an incorrect count of the counter. If the partial pulses are short, they will not be processed correctly, but may cause a malfunction of the electronic module. Since photons can strike the photon counting detector element at any time, it is inevitable that the time when a photon strikes the photon counting detector element to generate a pulse coincides with the memory switching time. To solve this problem, an artificial pulse, also called a simulated photon, can be generated, which can block the time of memory switching, as will be explained in detail below. The image sensor includes a pulse generator, which is configured and arranged to introduce an artificial pulse (simulated photon) into at least one macropixel stream or any pulse shaper output stream. The length of the artificial pulse is represented by t_artificial. The monostable / astable circuit processes the artificial pulse so that its pulse length is shortened to a second pulse length t2, followed by a pulse-free interval, called the "artificial pulse-free interval". The artificial pulse-free interval has a time span of (at least) t_artificial minus the second pulse length t2. In this way, in the processed stream, each artificial pulse is followed by a pulse-free interval. The pulse generator is configured to introduce the artificial pulse at a timing so that the time of memory switching occurs during the artificial pulse-free interval. Therefore, the artificial pulse causes a pulse-free period during which memory switching occurs. In general, not all artificial pulses need to have the same length, but different values ​​of t_artificial may be used. The pulse generator can be configured to introduce an artificial pulse for each memory switch. The frequency of generating the artificial pulse can be matched to the frequency of the memory switch. In particular, the pulse generator and the memory switch can operate at a common frequency, such that the frequency of the memory switch is equal to the frequency of the artificial pulse. Each artificial pulse is timed such that its start time is less than t_artificial before any memory switch time and greater than the second pulse length t2 before any memory switch time; this ensures that the memory switch occurs during a pause period without any pulse present. The time span of t_artificial is greater than t2, for example, it can be in the range of 1.3*t2 < t_artificial < 3.0*t2. This creates a pause with a length between 0.3*t2 and 2.0*t2. In principle, t_artificial can also be longer, but this would inappropriately increase the stagnation time. In contrast, the aforementioned range ensures sufficient distance from any possible pulse when the memory switch occurs while maintaining a relatively short stagnation time. General features The photon counting detector elements can form at least one array (sensor array). A pixel should be understood as an element of a sensor or a sensor array, i.e., a photon counting detector element. For example, the photon counting detector element can be an APD (avalanche photodiode) or a SPAD element, and can output a pulse for each detected photon. The photon counting detector elements can be manufactured using CMOS technology, and / or can form one or more arrays, where each array can form a silicon photomultiplier (SiPM). Thus, the aforementioned "signal" from the photon counting detector element includes the (electrical) pulse generated when a photon is detected. The image sensor is particularly useful for applications that are crucial for the precise counting of single photons and / or timing information, such as for microscopes, LiDAR systems (LiDAR: Light Detection and Ranging), PET systems (PET: Positron Emission Tomography), or particle detectors. Thus, the exemplary embodiments also include an optical microscope, LiDAR, PET system, or particle detector with the image sensor described herein. The time periods t1 and t2 of the pulse can be defined as the time periods during which the pulse height is above a specific threshold. For example, the threshold can be 50% of each pulse height ( / the reference voltage defining the pulse height), or more generally a value between 10% and 100%. Similarly, the time span of the pulse-free time period can be defined as the time period during which the signal height is below a specific threshold; this threshold can be the same as or different from the threshold used to define t1 and t2. The intended use of the image sensor leads to variants of the method of the present invention. Similarly, the features described for the method of the present invention also form variants of the image sensor of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS A better understanding of the present invention and various other features and advantages of the present invention will become apparent from the following description taken in conjunction with the schematic diagrams, which are shown by way of example only and not by way of limitation, in which like reference numerals may refer to similar or substantially the same components: Figure 1 A schematic diagram showing the components of an image sensor; Figure 2 A schematic diagram showing the merging of two signals from different sensor elements into one signal is shown; Figure 3 A schematic diagram showing signals from different sensor elements that are too close to each other to be processed properly; Figure 4 Schematic diagram illustrating the required rise and fall times of electronic modules; Figure 5 A schematic diagram showing components of an image sensor of the present invention and example signals processed by the image sensor; Figure 6 A schematic diagram showing a monostable / astable circuit processing pulses is shown, the circuit being a level monostable circuit; Figure 7 A schematic diagram showing a monostable / astable circuit processing a pulse is shown, the circuit being an edge-delay monostable circuit; Figure 8 Schematic showing the interpretation of edge-delayed monostable as strongly overlapping pulses with only one pulse; Fig. 9 A schematic diagram showing memory switching for forwarding signal intervals to different memories; Fig.10 A schematic showing how memory switching splits the pulse; Fig.11 A schematic diagram showing components of another embodiment of an image sensor of the present invention; Fig.12 Shows Fig.11 Schematic diagram of image sensor signals; Fig.13 Shows Fig.11 Schematic diagram of other signals of the image sensor; Fig.14 a schematic diagram showing components of another embodiment of an image sensor of the present invention; and Fig.15 Examples of monostable / astable circuits used in various embodiments of the image sensor of the present invention are shown. DETAILED DESCRIPTION Figure 5 : Fail-safe circuit with monostable / astable circuit Figure 5 An image sensor 1 according to an embodiment of the invention is schematically shown. The image sensor 1 comprises a plurality of photon counting detector elements 10, such as SPADs, which are configured to output electrical pulses 13 upon registration of photons. Figure 5 A schematic diagram of two different photon counting detector element 10 output signals 12 is shown. Each signal 12 includes a pulse 13, indicating that a corresponding photon has been detected. For all pulses, the horizontal axis represents time and the vertical axis represents the electrical signal strength, such as voltage or current. Therefore, the photons that cause the two shown pulses 13 are received at different points in time, wherein the two pulses 13 overlap. The image sensor 1 comprises a first set of pulse shapers 20 (e.g. monoflops), wherein the output signal 12 of each photon counting detector element 10 is directed to a respective pulse shaper 20. In the present example, the pulse shapers 20 are monoflops and process the signals 12, wherein the output of the pulse shapers 20 will be referred to as a pulse shaper output stream 22. The pulse shapers 20 may change (here: shorten) the length of the pulses 13, thereby converting the pulses 13 into pulses 23 having a specific length t1, which is a common length for all pulses 23 in the pulse shaper output stream 22. In this way, the pulse shapers 20 preserve the number and timing (start time) of the pulses 13 while changing the pulse length. The image sensor 1 further comprises a combining gate or connection unit 30, which combines the pulse shaper output streams 22 into macropixel outputs / macropixel streams 32, which are output via macropixel output lines 31. For ease of understanding, the schematic diagram shows only a small number of photon counting detector elements 10, whose outputs are all combined into one macropixel stream 32. However, the entire image sensor 1 may include a considerable number of photon counting detector elements 10, for example, between 100 and 10,000, each element being provided with its own pulse shaper 20. The connection unit 30 may include a plurality of macropixel output lines 31 for outputting a plurality of macropixel streams 32, for example, between 10 and 1,000 macropixel streams. The macropixel streams 32 differ in which photon counting detector elements 10 outputs they combine. The connection unit 30 may be fixed or flexible. In a flexible design, the connection unit 30 may be adjusted to change which pulse shaper output streams 22 are combined into one macropixel stream 32. In a fixed design, it is predetermined which pulse shaper output streams 22 are combined into one macro-pixel stream 32. The connection unit 30 can combine the pulse shaper output streams 22 using a logical OR connection, i.e., a high level is output when any one of the received pulse shaper output streams 22 has a high level. Therefore, the macro-pixel stream 32 includes all pulses 23 of the received pulse shaper output streams 22. The connection unit 30 does not change the length of the pulse 23. As shown in this example, and in the Figure 3 and Figure 4 As mentioned in the introduction, the gap or time span t_short may be too short to be properly processed by the subsequent electronic module 50 (e.g. a digital counter or a time-to-digital converter TDC). This typically occurs if t_short is smaller than the recovery time tn required by the electronic module 50 to restore the original internal state after the end of a pulse. Since photons can hit the photon counting detector element 10 at any time, the gap t_short between two pulses 23 can have any value. The image sensor 1 further comprises a monostable or astable trigger, referred to as an astable circuit 40, for receiving the macropixel stream 32. In case there are multiple macropixel output lines 31, the same number of monostable / astable circuits 40 may be provided in parallel. The monostable / astable circuit 40 is configured to shorten the pulse length of each pulse 23 of the macropixel stream 32. Thus, the monostable / astable circuit 40 outputs a signal, referred to as a processed stream 42, in which each pulse 43 has a second pulse length t2 shorter than t1. The monostable / astable circuit 40 generally retains the number of pulses and their start times, except that the pulse length is shortened from t1 to t2. As a result, a pulse-free interval PF is created after each pulse 43. The time span t_PF of the created pulse-free interval PF may be equal to t1 minus t2. Thus, the overall gap between two consecutive pulses 43 is equal to t_PF plus the original gap t_short between the pulses 23. The additional pulse-free interval PF reduces the risk of a recovery time tn of the electronic module 50 being violated. In general, t_PF can be any value, since a small additional gap between pulses also helps to reduce the risk of errors in the electronic module. For the sake of clarity, the electronic module 50 can optionally be considered as part of the image sensor 1 of the present invention and is therefore drawn with a dotted line. If the electronic module 50 is part of the image sensor 1, the pulse shaper 20 and / or the monostable / astable circuit 40 can be configured according to the recovery time tn and the minimum pulse length tp required by the electronic module 50. If t1 and t2 are set to t1-t2≥tn, the recovery time tn of the electronic module always remains constant regardless of when a photon strikes the photon counting detector element 10, i.e., the risk of violating tn is not only reduced but completely eliminated. If t2 is set to t2≥tp, the required minimum pulse length is always met. In this case, t1 is set to t1≥tp++tn. It should be noted that the risk of tn being violated can only be reduced by using two consecutive monostable triggers (more generally: a pulse shaper 20 followed by a monostable / astable circuit 40), where the latter monostable trigger generates a shorter pulse. If the monostable / astable circuit 40 is not used and the pulse shaper 20 generates a pulse with a length of t2short, any gap t_short between two consecutive pulses is possible. The monostable / astable circuit 40 can in principle be any astable trigger or monostable trigger, such as a level monostable trigger or an edge-delay monostable trigger (EDM). However, EDM has certain advantages, as described in detail in the following figure. Figures 6 to 8 : Output of monostable / astable circuit Figure 6 A macropixel stream 32 is shown being provided to a monostable / astable circuit 40 (here a level monostable), which is here a level monostable flip-flop. In this example, the two pulses 23 in the macropixel stream 32 are not clearly separated from each other. If the recovery time of the level monostable flip-flop is not satisfactory, a shorter pulse is usually output. Therefore, the monostable / astable circuit 40 in this example only correctly processes the first pulse to output a pulse 43 having the required pulse length, while the second pulse results in a shorter pulse 43B. Figure 7 The macropixel stream 32 shown in the previous figure can be shown, but in this case, the macropixel stream 32 is provided to a monostable trigger, which is an EDM. In contrast to a level monostable trigger, the pulse width produced by the EDM is independent of the input pulse. This plays an important role in fault suppression. The EDM has no recovery time because it is not dependent on the input pulse, so the input pulses 23, if not separated, and may even overlap, will produce output pulses 43 of the same width as otherwise. As shown, the EDM (monostable / astable circuit 40) outputs a processed stream 42 in which the pulses 43 have the same length / width. Figure 8 A macropixel stream 32 is shown where two pulses 23 are too close together for the EDM to recognize them as two pulses. This can happen if the falling edge of the first pulse and the rising edge of the second pulse overlap so much that the signal does not fall sufficiently between the two pulses 23. In this case, the EDM sees the two pulses 23 as one pulse. The EDM refuses to accept the incomplete edge between the pulses 23 and will not trigger, thereby suppressing the second pulse. Figure 6 Compared to the level monostable trigger, the EDM has the advantage that it does not output any damaging pulses that may cause errors in subsequent electronic equipment. Fig. 9and 10 : Memory switching The electronics serving the photon counting detector element can be operated in double buffer mode. In this case, there are two memories of the electronics with the same function that switch after a predetermined time. The reason for this is to achieve a service time close to 100%, since one of the memories of the two electronics is in an active state, receiving pulses from the photon counting detector element, while the memory of the other electronics is in an inactive state and can be read by an external system. After each switch, the output signal alternately enters the first memory and the second memory. Fig. 9 Schematically, a macropixel stream 32 including pulses 23a and 23b is shown. Until the moment of memory switch s1, the macropixel stream 32 is provided to the first memory, so that the first memory receives a first part 33 of the macropixel stream 32, which only includes pulses 23a before memory switch s1, but does not include pulses 23b after memory switch s1. After the moment of memory switch s1, the macropixel stream 32 is provided to the second memory, which receives a second part 34 of the macropixel stream 32, which only includes pulses 23b, but does not include pulses 23a. However, the pulse can also occur at any random time, and therefore at the exact time when the memory switch occurs. Fig.10 This is the case shown, where a pulse 23 in a macropixel stream 32 occurs exactly at the time of a memory switch s1. The pulse 23 is divided into two pulse parts 24 and 25, the ratio of which can be arbitrary, depending on the start time of the pulse 23 relative to the memory switch s1. The first memory receives the first part 33 of the macropixel stream 32, including the pulse part 24, and the second memory receives the second part 34 of the macropixel stream 32, including the pulse part 25. The pulses formed by the pulse components 23 and / or 24 may not meet the tp requirement discussed above, i.e., the pulse components 23 and / or 24 are shorter than the shortest pulse time required by the subsequent electronic module. In particular, if the pulse 23 is generated by a monostable trigger, its length is not too long, for example, the length is tp or slightly higher than tp, then in most cases, the pulse components 23 and 24 will be shorter than tp. Therefore, the pulse components 23 and 24 may cause malfunction of the electronic module. This problem can be solved by the design shown in the figure below. Fig.11 and 12 : Memory switching assistance using simulated photons Fig.11Schematically, an image sensor 1 according to another embodiment of the present invention is shown. The image sensor 1 comprises a plurality of photon counting detector elements 10, a first set of pulse shapers 20 (e.g. monostable triggers), a connection unit 30, at least one monostable / astable circuit 40 and an electronic module 50, all of which can be arranged according to Figure 5 Configure as described above. The electronic module 50 includes two memories B1 and B2, for example, counter memories for counting pulses / photons. One counter can be read out while the other counter is working to count incoming pulses. The memory switch S of the electronic module 50 determines which of the memories B1 and B2 the processed stream 42 is input to. Examples of the electronic module 50 include: ripple counters, binary counters, LFSR counters, grayscale counters, prescalers, TDCs or any other electronic circuit susceptible to metastability. In order to overcome the problem of coincidence of the appearance of pulses in the processed stream 42 with the memory switching, it is necessary to generate an artificial pulse (simulated photon) AP which produces a pulse-free interval when the memory switches. In more detail: the pulse generator G provides an artificial pulse AP which is added to the macro-pixel stream of the connection unit 30. This can be achieved by an OR gate 35 (or other component that combines two inputs into a joint output) on the macro-pixel output line 31 between the connection unit 30 and the EDM / monostable / astable circuit 40. The monostable / astable circuit 40 can shorten the artificial pulse AP and produce a pause (pulse-free interval) after the shortened artificial pulse. The generation of the artificial pulse AP is timed according to the memory switching so that each memory switch occurs during the pulse-free interval produced by the monostable / astable circuit 40 directly after the shortened artificial pulse. The pulse-free interval of the memory switching is generated by the monostable / astable circuit 40 and is independent of the arrival time of any pulse caused by real photons, that is, independent of any pulse in the macro-pixel stream before the addition of the artificial pulse AP. If the macro-pixel stream contains a pulse that coincides with the memory switching, this pulse will overlap with the artificial pulse AP, so that only one pulse is generated, which will be shortened by the EDM, so that the pulse-free interval appears again at the memory switching. The design thus eliminates any stray pulses or glitches that may occur when the memory is swapped. Note that when the simulated photon pulse is present on the line, the EDM will not receive any real pulses; therefore, there is a loss in sensitivity equal to t_artificial / tb, where tb is the time a memory is in operation. In practice, this ratio is about 1%. Will refer to Fig.12 The effect of artificial pulse AP is further described. Fig.12 A graph of different signals is shown, where the horizontal direction represents time. The memory switching signal s indicates that a memory switch occurs at time s1. The pulse generator generates an output G_AP, which includes a pulse (artificial pulse AP) with a time length of t_artificial. The macropixel stream 32 output by the connection unit includes pulses 23, each of which has a length t1 (t1 is set by the first group of pulse shapers). The OR gate adds the pulse generator output G_AP to the macropixel stream 32, creating a "macropixel stream with added artificial pulses" 32', which includes all pulses 23 and artificial pulses AP of the macropixel stream 32. The "macropixel stream with added artificial pulses" 32' is provided to the monostable / astable circuit 40, which shortens all pulses to a length of t2, thereby generating a processed stream 42 schematically described. The processed stream 42 is provided to the electronic module and is segmented at the memory switching time s1. A first part 42a of the processed stream 42 comprises only all pulses until the group switching time s1 and is provided to the first memory, whereas a second part 42b of the processed stream 42 comprises only all pulses after the group switching time s1 and is provided to the second memory. The start time and length of the artificial pulse AP are selected to ensure that there is no pulse at the memory switching time s1. To this end, the start time of the artificial pulse AP is selected to be earlier than t2 before the memory switching time s1, and the pulse length t_artificial is selected to be greater than t2. The monostable / astable circuit 40 shortens the length of the artificial pulse AP to t2, and thus creates a pulse-free interval with a length of t_pause=t_artificial-t2. Since the length t2 of the artificial pulse AP is shortened, the artificial pulse AP in the processed stream 42 ends before the memory switching s1 occurs. The start time of the artificial pulse AP is selected to be less than t_artificial before the memory switching time, with the result that when the memory switching s1 occurs, the artificial pulse-free interval t_pause (after the shortened artificial pulse) has not yet ended. In addition, the selection of t2 and t1 can refer to Figure 5 In particular, t2 may be selected as t2≥tp, and t1 may be selected as t1≥tp+tn. When memory switching occurs, the artificial pulse AP will be reduced by the EDM so that the artificial pulse AP meets the tp requirement of the electronic device. Since the input of the EDM is set, any other real photons will be rejected during this switching time. In this way, a memory switching aid is provided that reliably excludes the possibility of any pulse being split by memory switching. Of course, the artificial pulse AP can be discarded or compensated by the electronic module. For example, in the case of a counter, each time a memory switch occurs (or each time an artificial pulse is generated), the count number can be reduced by one, thereby compensating for the count of the artificial pulse AP. Fig.13 : Pulse-free period during memory switching time Fig.13 The case where the memory switch s1 coincides with the pulse 23c in the macropixel stream 32 is illustrated. Since the artificial pulse AP also coincides with the memory switch s1, the pulse 23c and the artificial pulse AP will produce only one pulse in the combined output 32'. In the example shown, the artificial pulse AP completely covers the time of the pulse 23c, i.e., it starts before the pulse 23c and ends after the pulse 23c, so the combined output 32' contains only one pulse AP of length t_artificial, which is the same as the case described in the previous figure. This means that the pulse 23c that coincides with the silver memory switch s1 will be filtered out. Therefore, the artificial pulse will produce a dead time of t_artificial. For completeness, if pulse 23c only overlaps with artificial pulse AP but is not completely covered by it, these two pulses will still produce only one pulse in combined output 32', but its length is greater than t_artificial (at most t_artificial+t1). The monostable / astable circuit 40 will shorten this pulse again, thus creating a pulse-free interval during which memory switching occurs. Fig.14 : Introducing changes in simulated photons Fig.14 yes Fig.11 A variant of Fig.11 The difference is that the pulse generator G is not connected to the macro-pixel output line 31 (i.e., downstream of the connection unit 30); instead, the pulse generator G is connected to the input of the connection unit 30, thereby providing the artificial pulse AP to the connection unit 30, and the connection unit 30 combines the artificial pulse AP with any pulse output by the pulse shaper 20. Fig.11 and Fig.14 In both cases, the artificial pulse AP is finally included in the macro-pixel stream of the macro-pixel output line 31. Fig.11 The OR gate 35 in Fig.14 In the embodiment of Fig.14 In a further variation of the illustrated embodiment, artificial pulses AP may also be added to the input of one or more pulse shapers 20 . Fig.15 :Monostable / astable circuit with feedback loop Fig.15A monostable / astable circuit 40 similar to an EDM is schematically shown, which can be used in other embodiments described with reference to the drawings. The macropixel output line 31 is connected to the input of the monostable / astable circuit 40. In this example, the macropixel stream 32 transmitted by the macropixel output line 31 includes two strongly overlapping pulses. The monostable / astable circuit 40 has at least two output ports, one of which is connected to a feedback loop 45, which leads to an input of the monostable / astable circuit 40, in this example to a reset input R. The feedback loop 45 has a function f, namely an electronic component that modifies the signal in the feedback loop at least by changing / shifting the phase, generating a feedback signal 32_feedback. As shown in the figure, in the case of strong overlap of pulses, the latter pulse may not be recognized, so the feedback signal 32_feedback only includes one pulse for two strongly overlapping pulses in the macropixel stream 32. Otherwise, each pulse of the macropixel stream 32 would result in a pulse in the feedback signal 32_feedback. On output line 41, the monostable / astable circuit 40 outputs a processed stream 42 generated from the macropixel stream 32 and the feedback signal 32_feedback. Due to the operation of the monostable / astable circuit 40 and its reset after a specified delay, the pulses in the processed stream 42 have a shorter pulse length than the pulses in the macropixel stream 32. Features described in different figures may be combined. In addition, the complexity of the embodiments is reduced for ease of understanding. In particular, the image sensor may include multiple macro-pixel output lines. The features described refer to only one macro-pixel output line. Each macro-pixel output line may provide a similar structure. In particular, a monostable / astable circuit 40 and the introduction of an artificial pulse AP may be provided for each macro-pixel output line. Reference Symbols List 1 Image Sensor 1' Related Image Sensor 10 Photon counting detector elements 12 Signal from photon counting detector element 10 13 Pulses from photon counting detector element 10 20 The first pulse shaper 20' Monostable Trigger 22 Pulse shaper output stream 23, 23a-23c Pulses output by the first set of pulse shapers 20 / Pulses in the pulse shaper output stream 22 24, 25 Pulse portions generated by bank switching split pulse 23 30 connection unit for combining pulse shaper output stream 22 31 macro pixel output lines 32 Macropixel stream / macropixel output 32' Macro pixel stream with artificial pulses (simulated photons) 32_feedback Feedback signal of monostable / astable circuit 40 33 The first part of the macro pixel stream 32, before the bank switch 34 The second part of the macro pixel stream 32, after the memory bank conversion 35 OR gate combining artificial pulse and macro pixel stream 40 Monostable / Astable Circuits 41 Output line of monostable / astable circuit 40 42 Processed stream (output by monostable / astable circuit 40) 42a First part of the processed stream (before bank conversion) 42b Second part of the processed stream (after bank conversion) 43 Pulse (in processed stream 42) 43B Spurious pulse 45 Feedback loop of monostable / astable circuit 40 50 Electronic modules 52 Input signals of electronic modules AP Artificial Pulse B1, B2 storage, such as counter EDM Edge Delayed Monostable Flip-Flop f Function in the feedback loop 45 (electronic components) G Pulse generator that provides artificial pulse AP G_AP Output of the pulse generator PF Pulse-free interval (after pulse 43 in processed stream 42) R Reset entry for monostable / astable circuit S Bank Switch Unit s Bank switch signal s1 Bank switch time t1 First pulse length (i.e. the length of pulse 23) t2 Second pulse length (i.e. pulse length 43) TDC Time to Digital Converter tn Recovery time of the electronic module (i.e. the required pause length between pulses) tp Minimum pulse length required by the electronic module t_artificial Pulse length of artificial pulse AP t_pause artificial pulse-free interval t_PF is the length of the pulse-free interval; the difference between the lengths of the first and second pulses t1 and t2 t_short The gap between two pulses in the macro pixel stream

Claims

1. Image sensor, including a photon counting detector element (10); A first set of pulse shapers (20) arranged to convert signals (12) from photon counting detector elements (10) into a pulse shaper output stream (22) wherein pulses (23) have at least a first pulse length (t1) A connection unit (30) configured to combine a plurality of pulse shaper output streams (22) into at least one macropixel stream (32); characterized in that at least one monostable / astable circuit (40) arranged to receive at least one macropixel stream (32) and output a processed stream (42) wherein the pulses (43) have a second pulse length (t2), Wherein, at least one monostable / astable circuit (40) is configured such that the second pulse length (t2) is shorter than the first pulse length (t1) by a time span t_PF so as to create a pulse-free interval (PF) within at least the time span t_PF after each pulse (43) in the processed stream (42).

2. The image sensor according to claim 1, The first group of pulse shapers (20) is composed of a group of monostable triggers, and at least one monostable / astable circuit (40) is another monostable trigger.

3. The image sensor according to claim 1 or 2, further comprising An electronic module (50) for processing a processed stream (42), wherein the electronic module (50) requires a pause of at least tn in length between two pulses (43) in the processed stream (42), At least one monostable / astable circuit (40) is configured such that t_PF is at least as long as tn.

4. The image sensor according to claim 3, in, The electronic module (50) includes at least one of a time-to-digital converter (TDC), a digital counter, a level comparator, a flip-flop, a scaler, an inverter, a buffer, a pull-up, a latch, a level monostable flip-flop, a clock divider, and a phase detector.

5. The image sensor according to claim 3 or 4, in, The first group of pulse shapers (20) and at least one monostable / astable circuit (40) are configured such that t_PF<2*tn to limit the dead time of the image sensor.

6. The image sensor according to any one of claims 3 to 5, wherein the electronic module (50) requires that the pulses (43) in the processed stream (42) have a minimum pulse length of tp; wherein at least one monostable / astable circuit (40) is configured so that the second pulse length (t2) is at least as long as tp; in, The first set of pulse shapers (20) is configured such that the first pulse length (t1) is at least as long as tp+tn.

7. The image sensor according to claim 6, in, The first set of pulse shapers (20) is configured to make the first pulse length (t1) shorter than 2*[tp+tn] to limit the dead time of the image sensor.

8. The image sensor according to any one of claims 1 to 7, in, At least one monostable / astable circuit (40) is not a level monostable trigger, but an edge-delay monostable trigger (EDM).

9. The image sensor according to any one of claims 3 to 8, in, The electronic module (50) comprises at least two memories (B1, B2); wherein the image sensor further comprises a memory switching unit (S), the memory switching unit (S) being configured to transmit to which of the at least two memories the processed stream (42) is switched, wherein the memory switching unit (S) is configured to switch at a specific memory switching time (s1); The image sensor further comprises a pulse generator (G) configured and arranged to introduce artificial pulses (AP) into at least one macro-pixel stream (32) or any pulse shaper output stream (22), wherein the length of each artificial pulse (AP) is t_artificial; wherein at least one monostable / astable circuit (40) processes each artificial pulse (AP) so that its pulse length is shortened to a second pulse length (t2) followed by an artificial pulse-free interval (t_pause) having a time span of at least t_artificial minus the second pulse length (t2); The pulse generator (G) is configured to introduce the artificial pulse (AP) at a timed interval so that the memory switching time (s1) occurs during an artificial pulse-free interval (t_pause).

10. The image sensor according to claim 9, in, The pulse generator (G) is configured to introduce an artificial pulse (AP) for each memory switching, and each artificial pulse (AP) is timed so that its start time is less than t_artificial before any memory switching time (s1), and its start time is greater than a second pulse length (t2) before any memory switching time (s1).

11. The image sensor according to claim 9 or 10, in, The pulse generator (G) and the memory switch (S) operate at a common frequency, making the frequency of the memory switch time (s1) equal to the frequency of the artificial pulse (AP).

12. A measuring system, being any of: a microscope, a lidar system, a position emission tomography system or a particle detector, comprising an image sensor according to one of claims 1 to 11.

13. A method of operating an image sensor, comprising: receiving photons using a photon counting detector element (10) which outputs a signal (12) indicative of the received photons; using a first set of pulse shapers (20) to convert the signal (12) of the photon counting detector element (10) into a pulse shaper output stream (22) wherein the pulses (23) have a first pulse length (t1); A connection unit (30) is used to combine a plurality of pulse shaper output streams (22) into at least one macro-pixel stream (32); characterized in that Use of at least one monostable / astable circuit (40) which processes at least one macropixel stream (32) to output a processed stream (42) wherein pulses (43) have a second pulse length (t2) which is shorter than the first pulse length (t1) by a time span t_PF, thereby creating a pulse-free interval (PF) after each pulse (43) in the processed stream (42) for at least the time span t_PF. An image sensor according to claim 6.

Images