Restoration method of imaging distortion of superconducting nanowire single-photon imager

Abstract

The present invention discloses a method for restoring imaging distortion of a superconducting nanowire single-photon imager, comprising: constructing a superconducting nanowire single-photon detector with separate detection units and delay units, and connecting delay lines in series between adjacent detection units; extracting N real pixel data from the original photon detection data to obtain a real pixel image, and extracting N-1 virtual pixel data to obtain a virtual pixel image; interpolating multi-photon detection data of N-2 real pixels by taking the average of adjacent pixels of the N-1 pixel data of the virtual pixel image, inserting the first pixel data of the virtual pixel data before the first data of the N-2 multi-photon data, and inserting the last pixel data of the virtual pixel data after the last data, to form an N-pixel multi-photon interpolation image; and subtracting the multi-photon interpolation image from the real pixel image to obtain a restored image. This method can solve the problem of single-photon imaging distortion of superconducting nanowire single-photon imagers.

Description

Technical Field

[0001] The present invention belongs to the technical field of superconducting nanowire single-photon imager imaging, and particularly relates to a method for recovering single-photon imaging distortion caused by multi-photon detection readout ambiguity of a superconducting nanowire single-photon imager. Background Art

[0002] Superconducting nanowire single-photon imagers utilize the high dynamic inductance of superconducting nanowires to reduce microwave propagation speed to 1-2% of the speed of light. This allows spatial resolution based on the time difference between the arrival of the response signal at two ports. This simplifies the readout of a superconducting single-photon detector array to two readout ports, reducing the complexity and power consumption of the readout circuit as the array expands. Therefore, they have enormous application potential in single-photon detection. Superconducting nanowire single-photon imagers, which separate the detection unit from the delay unit, build on these advantages by spatially separating pixels and connecting them with delay lines. By tailoring the delay line length, they can achieve crosstalk-free responses between pixels. However, superconducting nanowire single-photon imagers lack the ability to resolve multi-photon detection events. When two or more photons simultaneously strike different pixels in the imager, readout errors occur. This inability to resolve multi-photon detection events leads to detection errors that cause image distortion during single-photon imaging.

[0003] In the prior art document (“A scalable multi-photon coincidence detector based on superconducting nanowires”), a superconducting nanowire single-photon imager with separate detection units and delay units is used as a multi-channel photon coincidence counter, and a scheme is proposed to use the arrival time difference of the signals at the two ports to distinguish the two-photon events of the superconducting single-photon imager, but the premise is that the time from the emission of the photon to the incidence on the imager is known. The document also proposes that multi-photon events can be distinguished by extracting the signal waveform fingerprint, and demonstrates the resolution of multi-photon events by waveform fingerprint on a 4-pixel imager. The above scheme has low practicality when the imager is used for single-photon imaging, because the premise of its use is that the photon arrival time is known, which cannot be met in imaging applications, especially 3D imaging, and the method of extracting the signal waveform fingerprint is too complicated and difficult to implement for large array imagers.

[0004] To date, single-photon imaging image distortion caused by multiphoton detection readout errors in superconducting nanowire single-photon imagers has received insufficient attention, nor has a relevant image restoration solution been developed. By modeling the response of this type of imager and analyzing its image distortion characteristics, we can perform multiphoton calibration and restoration of image distortion. Summary of the Invention

[0005] In view of the above problems existing in the prior art, the present invention provides a method for restoring single-photon imaging image distortion caused by multi-photon detection readout errors of a superconducting nanowire single-photon imager.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for recovering imaging distortion of a superconducting nanowire single-photon imager, wherein the imaging distortion is caused by multi-photon detection readout errors of the superconducting nanowire single-photon imager, comprises the following steps:

[0008] Step 1: Construct a superconducting nanowire single-photon detector with separate detection units and delay units, and connect delay lines in series between adjacent detection units;

[0009] Step 2: Extract N real pixel data from the original photon detection data to obtain a real pixel image, and extract N-1 virtual pixel data to obtain a virtual pixel image;

[0010] Step 3: Multi-photon detection data of N-2 real pixels are interpolated by taking the average of the adjacent pixels of the N-1 pixels of the virtual pixel image. The first pixel data of the virtual pixel data is inserted before the first data of the interpolated N-2 multi-photon data, and the last pixel data of the virtual pixel data is inserted after the last data to form a multi-photon interpolation image of N pixels.

[0011] Step 4: Subtract the multi-photon interpolation image from the real pixel image to obtain the restored image.

[0012] Furthermore, the multi-photon detection refers to different pixels of the superconducting nanowire single-photon imager responding to two or more photons within the same time window, and the time window is determined by the recovery time of the imager.

[0013] Furthermore, the adjacency in step 3 refers to the physical adjacency of device pixels.

[0014] Furthermore, single-photon detection events and multi-photon detection events of the superconducting nanowire single-photon imager can be distinguished through the original photon detection data.

[0015] Furthermore, multi-photon detection events of the superconducting nanowire imager may lead to readout errors; multi-photon detection errors may not only be misidentified as other real pixels, but also as half-pixel points between real pixels that do not exist. These non-existent half-pixel points are called virtual pixels.

[0016] Furthermore, the responses at the real pixels of the superconducting nanowire single-photon imager include both correct single-photon detection events and multi-photon detection events that are mistakenly identified as such, and the responses at the virtual pixels are all multi-photon detection events that are mistakenly identified as such.

[0017] Furthermore, for an N-pixel superconducting nanowire single-photon imager, the probability of a multiphoton detection event being misclassified as occurring at the i-th real pixel is the sum of the probabilities of a series of response combinations, where both a pair of real pixels symmetrically centered around pixel i respond to photons, while none of the pixels outside the physical connection to the pixel pair responds. The probability of multiphoton detection at the i-th pixel can be expressed as:

[0018]

[0019] Among them, λ i Refers to the average number of photons incident on the i-th pixel.

[0020] Furthermore, for an N-pixel superconducting nanowire single-photon imager, the probability that a multiphoton detection event is misjudged as occurring at the virtual pixels i-0.5 and i+0.5 on either side of the i-th real pixel is the sum of the probabilities of a series of response combinations. These response combinations refer to events in which a pair of real pixels symmetrically centered around the virtual pixel i-0.5 (or i+0.5) both respond to photons, while the pixels do not respond to pixels outside of their physical connection. The multiphoton detection probability at the virtual pixel i-0.5 and the virtual pixel i+0.5 can be expressed as:

[0021]

[0022] Among them, λ i Refers to the average number of photons incident on the i-th pixel.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention provides a method for restoring image distortion from superconducting nanowire single-photon imagers. This method models the response of the superconducting nanowire single-photon imager, analyzes its image distortion characteristics, and performs multiphoton calibration of the image distortion. This method effectively addresses the problem of single-photon imaging distortion caused by multiphoton detection readout errors in superconducting nanowire single-photon imagers. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a diagram of the device structure and working principle of the superconducting nanowire single-photon imager;

[0026] Figure 2 This is a schematic diagram of the principle of multi-photon event generation in a superconducting nanowire single-photon imager;

[0027] Figure 3 This is a diagram distinguishing the real and virtual pixel responses of a superconducting nanowire single-photon imager.

[0028] Figure 4 Image restoration flow chart of the method for restoring image distortion from single-photon imaging according to the present invention;

[0029] Figure 5 is an image restoration result diagram of the embodiment. DETAILED DESCRIPTION

[0030] The present invention will be described in further detail below with reference to the accompanying drawings.

[0031] This embodiment is a method for recovering single-photon imaging distortion caused by multi-photon detection readout ambiguity of a superconducting nanowire single-photon imager.

[0032] A method for recovering single-photon imaging distortion caused by multi-photon detection readout ambiguity in a superconducting nanowire single-photon imager (SNPI) is disclosed. The method includes constructing a superconducting nanowire single-photon imager with separate detection and delay units, extracting data from real and virtual pixels of the SNPI, and performing a multi-photon interpolation method to recover distorted images from the SNPI. (Hereinafter, SNSPI refers to superconducting nanowire single-photon imager.)

[0033] In this embodiment, the device structure of SNSPI is as follows: Figure 1 As shown, each pixel is constructed from two 100nm wide and 5μm long superconducting nanowires connected in parallel, with a sufficient length of 400nm wide meandering delay line connected in series between adjacent pixels. Only the pixel location responds to photons, while the wider delay line is insensitive to incident photons. When a photon is detected by a pixel, two pulses, one positive and one negative, are generated at the response location and transmitted to the device's two ports. A readout device connected to each port determines the photon's response location and timing based on the arrival times of the two pulses.

[0034] When the SNSPI device responds to more than two photons simultaneously, readout errors may occur. Figure 2 Schematic diagrams of three different response conditions of the N-pixel SNSPI are given: single-photon response, two-photon response, and three or more photon response. Since the N pixels of the SNSPI device are connected in series, the device is directly expanded into a straight line with N nodes and a length of L in the figure.

[0035] For the case of a photon response, assume that the photon p The moment when the pixel is responded by the i-th pixel, the time when the pair of positive and negative pulses arrive at both ends is expressed as: τ is the transmission time of the pulse on the delay line between pixels. The arrival position and arrival time of photons can be obtained by differential time and sum time mapping respectively. For the case of two-photon response, assume that two photons are respectively detected by the i-th pixel and the j-th pixel at time t p (i < j). At each response point, a pair of positive and negative pulses are generated and transmitted to both ends. Two negative pulses are transmitted to the left end, and two positive pulses are transmitted to the right end. Since the response pulse tail time is much longer than the signal transmission time on the nanowire, the two readout ports can only record the time t1 when the negative pulse generated by the i-th pixel arrives at the left end and the time t2 when the positive pulse of the j-th pixel arrives at the right end. The arrival times of the other two pulses are later, and their waveforms are mixed with the previous waveforms, making it difficult to extract the rising edge and falling edge time information. The differential time t diff and sum time t sum representing the arrival position and arrival time of photons are respectively expressed as: The situation of three or more photon responses is similar to that of two-photon responses. Only the arrival times of the negative pulse of the response pixel closest to the left side and the positive pulse of the response pixel closest to the right side are recorded. From the expression of t diff for two-photon responses, it can be seen that the photon response position is misjudged to the ((j - i) / 2)-th pixel. When j - i is even, this multi-photon detection event is wrongly superimposed on the response of the ((j - i) / 2)-th pixel of the device; when j - i is odd, this multi-photon detection error event is wrongly judged to the half-pixel point between pixels i and j. Since this half-pixel point is not a real pixel, the half-pixel points that appear due to multi-photon detection error readout are called virtual pixels hereinafter. By comparing the expressions of photon position and arrival time for one-photon response and multi-photon responses, it can be found that not only is the photon response position misjudged in multi-photon detection, but there are also differences in the sum time t sum representing the photon arrival time.

[0036] As Figure 3 shown is the statistical histogram of the differential time t diff of the response events of pixels 33 - 40 of the SNSPI device. From the statistical histogram of all detection events in the figure, it can be seen that response events are distributed on both real pixels (pixel indices are integers) and virtual pixels (half-pixels covered by shadows in the figure). The single-photon detection events and multi-photon detection events in all detection events are separated, and their statistical histogram distributions are shown in the figure. Among them, the separation of single-photon detection events and multi-photon detection events is based on the differential time t sumThe detection events on real pixels include single-photon detection events and multi-photon detection events. The detection events on virtual pixels are all multi-photon detection events.

[0037] Imaging distortion is mainly due to the incorrect superposition of multi-photon detection events on the response time of real pixels. In order to restore the distorted image, the multi-photon detection events at the real pixels need to be removed. For N-pixel SNSPI, the probability of a multi-photon detection event being misjudged to the i-th real pixel is the sum of the probabilities of a series of response combinations. These response combinations refer to events in which a pair of real pixels symmetrical about pixel i respond to photons, while pixels outside the physical connection do not respond. The probability of multi-photon detection at the i-th pixel can be expressed as:

[0038]

[0039] Among them, λ i is the average number of photons incident on the i-th pixel. The probability of multi-photon detection events at the two virtual pixels adjacent to the i-th real pixel can be expressed as:

[0040]

[0041]

[0042] The pixel positions involved in the cumulative terms in the expression for the probability of a multiphoton detection event at the real pixel and the expression for the probability of a multiphoton detection event at the two virtual pixels to the left and right differ by one pixel, respectively. Therefore, the average of the multiphoton event probabilities of the two virtual pixels can be used as the multiphoton event probability of the middle real pixel. The multiphoton detection event is then subtracted from all the detection events at the real pixel to obtain the restored image.

[0043] like Figure 4 As shown, the specific steps of the method for restoring image distortion of single-photon imaging of the present invention include:

[0044] 1) Extract N real pixel data from the original photon detection data to obtain a real pixel image, and extract N-1 virtual pixel data to obtain a virtual pixel image. Since the number of virtual pixels is one less than the number of real pixels, the virtual pixel image has one less pixel in the lower left corner;

[0045] 2) Using the mean of the adjacent pixels of the N-1 pixels of the virtual pixel image, interpolate the multi-photon detection data of N-2 real pixels. Insert the first pixel data of the virtual pixel data before the first data of the interpolated N-2 multi-photon data, and insert the last pixel data of the virtual pixel data after the last data, to form a multi-photon interpolation image of N pixels. The adjacent here refers to the physical connection of the serial pixels.

[0046] 3) The restored image is obtained by subtracting the multi-photon interpolation image from the real pixel image.

[0047] Figure 5 The proposed image distortion restoration method demonstrates its effectiveness. Figures (a)-(c) show the distorted image, multiphoton interpolated image, and restored image for an incident light intensity of 2 photons / pulse. Figure (d) shows the distorted image and restored image quality when the light intensity varies from 0.5 to 10 photons / pulse. The mean structural similarity (MSSIM) between the measured image and the original image is used as the image quality evaluation metric. It can be seen that when the light intensity is 2 photons / pulse, the proposed restoration method can effectively remove image blur and ghosting. When the light intensity is 5 photons / pulse, the MSSIM quality evaluation factor of the restored images increases by 1.3-9.3 times. This demonstrates that this method can effectively address the single-photon imaging image distortion problem caused by multiphoton detection readout errors in superconducting nanowire single-photon imagers.

[0048] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for restoring imaging distortion of a superconducting nanowire single-photon imager, characterized in that: The imaging distortion is caused by a multi-photon detection readout error of a superconducting nanowire single-photon imager. The method comprises the following steps: Step 1: Construct a superconducting nanowire single-photon detector with separate detection units and delay units, and connect delay lines in series between adjacent detection units; Step 2: Extract N real pixel data from the original photon detection data to obtain a real pixel image, and extract N-1 virtual pixel data to obtain a virtual pixel image; Step 3: Multi-photon detection data of N-2 real pixels are interpolated by taking the average of the adjacent pixels of the N-1 pixels of the virtual pixel image. The first pixel data of the virtual pixel data is inserted before the first data of the interpolated N-2 multi-photon data, and the last pixel data of the virtual pixel data is inserted after the last data to form a multi-photon interpolation image of N pixels. Step 4: Subtract the multi-photon interpolation image from the real pixel image to obtain the restored image.

2. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 1, characterized in that: The multi-photon detection refers to that different pixels of the superconducting nanowire single-photon imager respond to two or more photons within the same time window, and the time window is determined by the recovery time of the imager.

3. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 1, characterized in that: The adjacency in step 3 refers to the physical connection adjacency of device pixels.

4. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 1, characterized in that: Single-photon detection events and multi-photon detection events of superconducting nanowire single-photon imagers can be distinguished through the original photon detection data.

5. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 1, characterized in that: Multi-photon detection events of the superconducting nanowire single-photon imager may lead to readout errors; Multi-photon detection errors are not only misidentified as other real pixels, but also misidentified as half-pixel points between real pixels that do not exist. These non-existent half-pixel points are called virtual pixels.

6. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 5, characterized in that: The responses at the real pixels of the superconducting nanowire single-photon imager include both correct single-photon detection events and multi-photon detection events that are erroneously identified as such, and the responses at the virtual pixels are all multi-photon detection events that are erroneously identified as such.

7. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 6, characterized in that: For an N-pixel superconducting nanowire single-photon imager, the probability that a multiphoton detection event is misjudged to the i-th real pixel is the sum of the probabilities of a series of response combinations, where both a pair of real pixels symmetrical about pixel i respond to photons, while the pixels do not respond to pixels outside the physical connection. The multiphoton detection probability at the i-th pixel is p i_mpd_real Expressed as: Among them, λ i Refers to the average number of photons incident on the i-th pixel.

8. The method for restoring imaging distortion of a superconducting nanowire single-photon imager according to claim 6, characterized in that: For an N-pixel superconducting nanowire single-photon imager, the probability that a multiphoton detection event is misjudged to the virtual pixels i-0.5 and i+0.5 on both sides of the i-th real pixel is the sum of the probabilities of a series of response combinations. These response combinations refer to events in which a pair of real pixels symmetrical about the virtual pixel i-0.5 or i+0.5 respond to photons, while the pixels do not respond to pixels outside the physical connection. The multiphoton detection probability p at the virtual pixel i-0.5 is i-0.5_mpd_idle and the multiphoton detection probability p at virtual pixel i+0.5 i+0.5_mpd_idle Respectively expressed as: Among them, λ i Refers to the average number of photons incident on the i-th pixel.

Images