A method for verifying pattern reconstruction and recognition capabilities of infrared photodetectors

Abstract

This invention relates to the field of photoelectric detection technology, and particularly to a method for verifying the pattern reconstruction and recognition capabilities of infrared photodetectors. This invention constructs a two-dimensional distribution map by scanning a pattern mask with an infrared laser and collecting photocurrent under zero bias voltage. Combined with edge contour feature extraction and standard pattern comparison analysis, it achieves a systematic verification of pattern reconstruction and spatial recognition capabilities. This solves the problem that traditional single-parameter tests are insufficient to reflect the overall performance of actual imaging, improves the accuracy of spatial resolution and image fidelity assessment under low-light conditions, simplifies the testing process, and enhances repeatability, providing reliable quantitative support for the research, development, and industrialization of high-performance infrared detectors.

Description

Technical Field

[0001] This invention relates to the field of photoelectric detection technology, and in particular to a method for verifying the pattern reconstruction and recognition capabilities of infrared photoelectric detectors. Background Technology

[0002] With the increasing penetration of infrared detection technology into fields such as night vision security, medical diagnosis, and industrial monitoring, the verification requirements for the imaging performance of photodetectors are becoming increasingly stringent. Current performance evaluations of infrared photodetectors largely focus on single-parameter testing, lacking systematic methods for verifying pattern reconstruction and spatial recognition capabilities, making it difficult to accurately reflect the comprehensive performance of devices in actual imaging scenarios. In particular, novel self-powered heterostructure detectors exhibit significantly different photoelectric response characteristics in zero-bias operation compared to traditional devices, and existing testing methods often cannot effectively assess their spatial resolution and image fidelity under low-light conditions. Furthermore, traditional scanning imaging testing systems are complex in structure and cumbersome in operation, making it difficult to achieve high-precision, repeatable quantitative analysis, thus hindering the research, development, and industrialization of high-performance infrared detectors.

[0003] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for pattern reconstruction and recognition capability verification of infrared photodetectors. The technical solution of this method is as follows:

[0005] An infrared laser source is controlled to emit an infrared laser, and the infrared laser is irradiated onto a pattern mask, the pattern mask having a predefined pattern.

[0006] The scanning device is controlled to move at least one of the pattern mask and the infrared photodetector under test, so that the laser spot transmitted or reflected from the pattern mask scans the photosensitive surface of the infrared photodetector under test along a preset path.

[0007] When the infrared photodetector under test is in zero bias voltage working mode, the photocurrent signal corresponding to each scanning position is collected and recorded, and each photocurrent signal is integrated into a two-dimensional current distribution map according to the corresponding scanning position information.

[0008] The two-dimensional current distribution map is processed to extract the edge contour features of the reconstructed pattern, and the edge contour features are compared with the standard contour features of the predefined pattern. The spatial resolution parameters are calculated based on the intensity change gradient of the edge transition region.

[0009] The pattern reconstruction capability of the infrared photodetector under test is evaluated based on the overlap between the edge contour features and the standard contour features. The spatial recognition capability of the infrared photodetector under test is evaluated based on the value of the spatial resolution parameter. The performance evaluation result of the infrared photodetector under test is generated based on the evaluation results of the pattern reconstruction capability and the spatial recognition capability.

[0010] Furthermore, the step of controlling the infrared laser source to emit infrared laser light and irradiating the pattern mask with the infrared laser light further includes:

[0011] The infrared laser source is controlled to generate infrared laser of a specific wavelength, and the infrared laser is collimated and then incident vertically onto the surface of the pattern mask;

[0012] The pattern mask is a metal mask etched with the predefined pattern, which includes feature geometry for verifying spatial resolution.

[0013] Furthermore, the step of controlling the scanning device to move at least one of the pattern mask and the infrared photodetector under test, such that the laser spot transmitted or reflected from the pattern mask scans the photosensitive surface of the infrared photodetector under test along a preset path, further includes:

[0014] The infrared photodetector to be tested is fixed on the support platform of the scanning device;

[0015] The scanning device is controlled to drive the carrier platform to move in two dimensions along a preset path, the preset path covering the complete projection area of ​​the predefined pattern of the pattern mask on the photosensitive surface;

[0016] The scanning device is controlled to pause at each step position of the preset path for a preset sampling time, and the scanning device is controlled to coordinate the movement of the carrier platform with the irradiation of the laser spot, so as to complete the point-by-point scanning of the predefined pattern on the photosensitive surface.

[0017] Furthermore, the step of acquiring and recording the photocurrent signal corresponding to each scanning position when the infrared photodetector under test is in zero-bias operating mode, and integrating each photocurrent signal into a two-dimensional current distribution map according to the corresponding scanning position information, further includes:

[0018] During each step position of the scanning device, the control data acquisition unit acquires the photocurrent signal generated by the infrared photodetector under test in the zero bias voltage operating mode;

[0019] The control unit associates and stores each acquired photocurrent signal with its corresponding step position coordinates. After scanning all step positions, the associated photocurrent signals are arranged into a two-dimensional matrix according to the stored step position coordinates.

[0020] The two-dimensional matrix is ​​normalized to generate the two-dimensional current distribution map; wherein, the pixels in the two-dimensional current distribution map have a one-to-one correspondence with the step position coordinates.

[0021] Furthermore, the step of performing image processing on the two-dimensional current distribution map, extracting the edge contour features of the reconstructed pattern, comparing the edge contour features with the standard contour features of the predefined pattern, and calculating the spatial resolution parameters based on the intensity change gradient of the edge transition region, further includes:

[0022] The two-dimensional current distribution map is subjected to filtering and noise reduction processing;

[0023] An edge detection algorithm is applied to the processed two-dimensional current distribution map to extract the edge contour features of the reconstructed pattern;

[0024] Spatial registration and comparison are performed between the edge contour features and the standard contour features. The edge transition region is identified in the registered edge contour features, and the intensity change gradient of the edge transition region is calculated.

[0025] The spatial resolution parameter is calculated based on the intensity change gradient, wherein the spatial resolution parameter R is calculated using the following formula:

[0026] ,

[0027] In the formula, This represents the difference in photocurrent intensity between two adjacent sampling points along the scanning direction within the edge transition region. This represents the actual spatial distance between two adjacent sampling points on the preset scanning path. This represents the average intensity gradient along the scanning direction in the edge transition region. This represents the maximum value of the intensity gradient along the scanning direction in the edge transition region. This indicates the maximum intensity value in the edge transition region. This represents the minimum intensity value of the edge transition region.

[0028] Furthermore, the step of evaluating the pattern reconstruction capability of the infrared photodetector under test based on the degree of overlap between the edge contour features and the standard contour features further includes:

[0029] Calculate the area of ​​the overlapping region between the edge contour feature and the standard contour feature, and the total contour area;

[0030] Extract the normal gradient magnitude of each edge point in the edge contour features;

[0031] A comprehensive contour overlap evaluation value is calculated based on the area of ​​the overlapping region, the area of ​​the total contour region, and the magnitude of the normal gradient.

[0032] The first level corresponding to the pattern reconstruction capability of the infrared photodetector under test is determined based on the comprehensive contour overlap evaluation value.

[0033] The comprehensive contour overlap evaluation value S is calculated using the following formula:

[0034] ,

[0035] In the formula, This represents the area of ​​the overlapping region between the edge contour feature and the standard contour feature. This represents the total area of ​​the contour region of the standard contour feature. This represents the total number of edge points. Indicates the first The magnitude of the normal gradient at each edge point and These are the preset weighting coefficients.

[0036] Furthermore, the step of evaluating the spatial recognition capability of the infrared photodetector under test based on the value of the spatial resolution parameter further includes:

[0037] Obtain the spatial resolution parameters of all edge transition regions in the edge contour features;

[0038] Calculate the statistical distribution characteristics of the spatial resolution parameters of all edge transition regions;

[0039] The spatial recognition capability evaluation value is calculated based on the statistical distribution characteristics of the spatial resolution parameters.

[0040] The second level corresponding to the spatial recognition capability of the infrared photodetector under test is determined based on the spatial recognition capability evaluation value.

[0041] The spatial recognition capability evaluation value Q is calculated using the following formula:

[0042] ,

[0043] In the formula, This represents the average spatial resolution parameter of all edge transition regions. This represents the standard deviation of the spatial resolution parameters for all edge transition regions. The third central moment represents the spatial resolution parameter of all edge transition regions. The fourth central moment represents the spatial resolution parameter of all edge transition regions. , , and These are the preset weighting coefficients.

[0044] Furthermore, the step of generating the performance evaluation result of the infrared photodetector under test based on the evaluation results of the pattern reconstruction capability and the spatial recognition capability further includes:

[0045] The first level corresponding to the pattern reconstruction capability and the second level corresponding to the spatial recognition capability are used as input parameters, and the first level and the second level are weighted and fused based on a preset comprehensive evaluation rule.

[0046] The comprehensive performance score of the infrared photodetector under test is determined based on the weighted fusion calculation results;

[0047] The comprehensive performance score is compared with a preset performance level threshold to generate the final performance evaluation result of the infrared photodetector under test.

[0048] Furthermore, the preset path is a matrix of scanning points, and the spacing between adjacent scanning points is smaller than the pixel size of the infrared photodetector under test.

[0049] Furthermore, the final performance evaluation result includes three levels: qualified, basically qualified, and unqualified; wherein, the qualified level corresponds to a comprehensive performance score greater than or equal to the first threshold, the basically qualified level corresponds to a comprehensive performance score less than the first threshold but greater than or equal to the second threshold, and the unqualified level corresponds to a comprehensive performance score less than the second threshold.

[0050] The technical solution of this invention constructs a two-dimensional distribution map by scanning a pattern mask with infrared laser and collecting photocurrent under zero bias voltage. Combined with edge contour feature extraction and standard pattern comparison analysis, it realizes the systematic verification of pattern reconstruction and spatial recognition capabilities, solves the problem that traditional single-parameter testing is difficult to reflect the actual imaging performance, improves the accuracy of spatial resolution and image fidelity evaluation under low illumination conditions, simplifies the testing process and enhances repeatability, and provides reliable quantitative support for the research and development and industrialization of high-performance infrared detectors.

[0051] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0053] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0054] Figure 1 This is a schematic flowchart of an embodiment of a pattern reconstruction and recognition capability verification method for an infrared photodetector according to the present invention. Detailed Implementation

[0055] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0056] Figure 1 This diagram illustrates a flowchart of an embodiment of a pattern reconstruction and recognition capability verification method for an infrared photodetector provided by the present invention, executed by a control terminal. Figure 1 As shown, it includes the following steps:

[0057] S1. Control the infrared laser source to emit infrared laser and irradiate the pattern mask with a predefined pattern.

[0058] Infrared laser source refers to a device capable of emitting infrared laser light; for example, a semiconductor laser with an output wavelength of 1550 nanometers can be used as an infrared laser source. Infrared laser refers to a coherent beam of light with a wavelength in the infrared band; for example, an infrared laser with a wavelength of 1550 nanometers can be generated by a semiconductor laser and used to illuminate a pattern mask. Pattern mask refers to a transparent or opaque material with a specific pattern etched on its surface; for example, a mask made of metal with a predefined pattern composed of the letter "H" and triangles etched on it. Predefined pattern refers to a standard pattern pre-designed to verify the imaging capability of a detector; for example, a test pattern composed of the letter "H" and triangles etched on a mask.

[0059] S2. Control the scanning device to move at least one of the pattern mask and the infrared photodetector under test, so that the laser spot transmitted or reflected from the pattern mask scans the photosensitive surface of the infrared photodetector under test along a preset path.

[0060] The scanning device refers to a mechanical platform system capable of precise positioning and movement; for example, a precision translation stage driven by an XY dual-axis stepper motor. The infrared photodetector under test refers to an infrared photoelectric device whose imaging performance needs to be verified; for example, a self-powered infrared photodetector fabricated using a lead selenide / graphene / silicon heterostructure. The laser spot refers to the illumination area formed after a laser beam passes through a mask; for example, a 1550 nm infrared laser beam passing through an "H" pattern on a mask forms a corresponding spot projection on the detector surface. The preset path refers to a pre-set scanning trajectory; for example, a matrix scanning path covering 14 × 10 pixels is set as the preset path. The photosensitive surface refers to the effective area of ​​the detector for receiving light signals; for example, the sensitive layer surface on the detector used for photoelectric conversion.

[0061] S3. When the infrared photodetector under test is in zero bias working mode, the photocurrent signal corresponding to each scanning position is collected and recorded, and each photocurrent signal is integrated into a two-dimensional current distribution map according to the corresponding scanning position information.

[0062] Zero-bias operating mode refers to the detector's operating state without an external bias voltage; for example, operating the detector at 0 volts bias voltage and utilizing its self-powered characteristics for signal acquisition. Scan position refers to the specific coordinates of the laser spot on the photosensitive surface during scanning; for example, the (7,5)th scan position in the preset path corresponds to a specific coordinate point of the spot on the detector's photosensitive surface. Photocurrent signal refers to the electrical signal generated by the detector after being illuminated; for example, the nanoampere-level current signal generated by the detector after being irradiated by a 1550 nm laser under zero-bias conditions. Two-dimensional current distribution map refers to an image formed by arranging the photocurrent signals according to their spatial positions; for example, a grayscale image generated by arranging the photocurrent signals from 140 scan positions in a 14×10 matrix.

[0063] S4. Perform image processing on the two-dimensional current distribution map, extract the edge contour features of the reconstructed pattern, compare the edge contour features with the standard contour features of the predefined pattern, and calculate the spatial resolution parameters based on the intensity change gradient of the edge transition region.

[0064] The reconstructed pattern refers to the mask pattern reconstructed from a two-dimensional current distribution map; for example, the outline of the letter "H" extracted from the two-dimensional current distribution map is called the reconstructed pattern. Edge contour features refer to the geometric shape information of the pattern edges; for example, the continuous curve features of the stroke edges of the letter "H" in the reconstructed pattern. Standard contour features refer to the ideal edge contour of a predefined pattern; for example, the standard geometric contour of the "H" pattern design on the mask. Edge transition region refers to the region where light intensity changes at the edge of the pattern; for example, the band-shaped region in the reconstructed pattern where the light intensity changes from high to low at the edge of the strokes of the letter "H". Intensity gradient refers to the rate of change of light intensity in space; for example, the change in photocurrent intensity per unit distance in the edge transition region. Spatial resolution parameter refers to a quantitative indicator characterizing the detector's resolving power; for example, the value 0.85 obtained by calculating the intensity gradient of the edge transition region is used as the spatial resolution parameter.

[0065] S5. Based on the overlap between the edge contour features and the standard contour features, evaluate the pattern reconstruction capability of the infrared photodetector under test; based on the value of the spatial resolution parameter, evaluate the spatial recognition capability of the infrared photodetector under test; and generate the performance evaluation result of the infrared photodetector under test based on the evaluation results of the pattern reconstruction capability and the spatial recognition capability.

[0066] The following parameters are used to evaluate the detector's capabilities: **Overlap Rate:** This refers to the degree of matching between the reconstructed contour and the standard contour. For example, a 92% overlap between the reconstructed "H" contour and the standard contour is considered a 92% overlap. **Pattern Reconstruction Capability:** This refers to the detector's ability to accurately reconstruct pattern shapes. For example, a 92% overlap rate indicates excellent pattern reconstruction capability. **Spatial Recognition Capability:** This refers to the detector's ability to distinguish detailed features. For example, a spatial resolution parameter of 0.85 indicates good spatial recognition capability. **Pattern Reconstruction Capability Evaluation Result:** This is a quantitative evaluation of the pattern reconstruction capability. For example, a "Grade A" rating in the first level. **Spatial Recognition Capability Evaluation Result:** This is a quantitative evaluation of the spatial recognition capability. For example, a "Grade B" rating in the second level. **Performance Evaluation Result:** This is a comprehensive evaluation of the detector's overall performance. For example, combining Grade A pattern reconstruction capability and Grade B spatial recognition capability results in an "Excellent" performance evaluation.

[0067] The technical solution of this embodiment constructs a two-dimensional distribution map by scanning a pattern mask with infrared laser and collecting photocurrent under zero bias voltage. Combined with edge contour feature extraction and standard pattern comparison analysis, it realizes the systematic verification of pattern reconstruction and spatial recognition capabilities, solves the problem that traditional single-parameter testing is difficult to reflect the actual imaging performance, improves the accuracy of spatial resolution and image fidelity evaluation under low illumination conditions, simplifies the testing process and enhances repeatability, and provides reliable quantitative support for the research and development and industrialization of high-performance infrared detectors.

[0068] In one alternative approach, S1 specifically includes:

[0069] The infrared laser source is controlled to generate an infrared laser of a specific wavelength, and the infrared laser is collimated and then incident vertically onto the surface of the pattern mask.

[0070] The pattern mask is a metal mask etched with the predefined pattern, which includes feature geometry for verifying spatial resolution.

[0071] Here, "specific wavelength" refers to the defined wavelength value emitted by the laser; for example, 1550 nanometers is a specific wavelength output by an infrared laser source. "Metal mask" refers to a patterned mask made of metallic material; for example, a metal mask made by etching test patterns onto a 0.5 mm thick stainless steel plate. "Feature geometry" refers to the typical geometry used for testing; for example, feature geometry such as lines of varying widths and acute-angled triangles contained within the mask pattern.

[0072] Specifically, the infrared laser source is controlled to generate an infrared laser with a wavelength of 1550 nanometers, the collimating optical component is controlled to collimate the infrared laser to form a parallel beam, and the optical adjustment mechanism is controlled to make the collimated infrared laser incident at an angle perpendicular to the surface of the pattern mask, ensuring that the central axis of the laser beam coincides with the normal of the surface of the pattern mask.

[0073] In the above-mentioned optional methods, by further controlling the collimation and vertical incidence of infrared lasers of specific wavelengths, and in conjunction with etching a metal mask with a predefined pattern, stable illumination conditions and sharp pattern edges are ensured, providing a standardized optical input basis for subsequent high-precision pattern reconstruction and spatial recognition capability verification.

[0074] In one alternative approach, S2 specifically includes:

[0075] The infrared photodetector to be tested is fixed on the support platform of the scanning device.

[0076] The term "supporting platform" refers to the component in a scanning device that holds the detector in place; for example, a metal supporting platform on an XY translation stage used to hold the detector.

[0077] The scanning device is controlled to drive the carrier platform to move in two dimensions along a preset path, the preset path covering the complete projection area of ​​the predefined pattern of the pattern mask on the photosensitive surface.

[0078] Two-dimensional stepping movement refers to moving at fixed intervals within a plane; for example, the platform moves sequentially in the X and Y directions with a step size of 10 micrometers to complete two-dimensional stepping movement. Complete projection area refers to the entire imaging area of ​​the mask pattern on the photosensitive surface; for example, the entire spot area formed by the "H" letter pattern on the photosensitive surface of the detector.

[0079] The scanning device is controlled to pause at each step position of the preset path for a preset sampling time, and the scanning device is controlled to coordinate the movement of the carrier platform with the irradiation of the laser spot, so as to complete the point-by-point scanning of the predefined pattern on the photosensitive surface.

[0080] The preset sampling time refers to the signal acquisition duration at each scanning position; for example, setting the sampling time to 100 milliseconds at each scanning position. Point-by-point scanning means scanning each position sequentially; for example, scanning 140 positions sequentially from the top left corner to the bottom right corner to complete the point-by-point scanning.

[0081] Specifically, the scanning device is controlled to pause for a preset sampling time at each step position of the preset path, and the scanning device is controlled to drive the carrier platform to move along the preset path. At the same time, it is ensured that the laser spot stably illuminates the photosensitive surface of the infrared photodetector under test after being transmitted or reflected from the pattern mask, so that the movement step of the carrier platform is synchronized with the irradiation position of the laser spot, and the predefined pattern is sequentially covered in all areas on the photosensitive surface, thereby completing the point-by-point scanning of the predefined pattern.

[0082] Among the above-mentioned optional methods, a method is further adopted in which the detector under test is fixed on the support platform and driven to move in two dimensions. The preset path covers the entire projection area and stops to sample at each step position, so as to realize the precise point-by-point scanning of the photosensitive surface of the detector under test and ensure the spatial integrity and positional accuracy of the data acquisition.

[0083] In one alternative approach, S3 specifically includes:

[0084] During each step position of the scanning device, the control data acquisition unit acquires the photocurrent signal generated by the infrared photodetector under test in the zero bias voltage operating mode.

[0085] The data acquisition unit refers to a device that collects and records electrical signals; for example, a high-precision current acquisition device can be used as the data acquisition unit.

[0086] Specifically, during the period when the scanning device reaches each step position and remains stationary, the control data acquisition unit establishes an electrical connection with the infrared photodetector under test. Under the condition that the infrared photodetector under test maintains a zero bias voltage working mode, the photocurrent signal generated by the laser spot irradiation on the photosensitive surface of the infrared photodetector under test is collected in real time, and the collected photocurrent signal value is transmitted to the data acquisition unit for buffering.

[0087] The control unit associates and stores each acquired photocurrent signal with its corresponding step position coordinates. After scanning all step positions, the associated photocurrent signals are arranged into a two-dimensional matrix according to the stored step position coordinates.

[0088] A two-dimensional matrix refers to a set of data arranged in rows and columns; for example, 140 photocurrent signals are arranged in 14 rows and 10 columns to form a two-dimensional matrix.

[0089] Specifically, when acquiring each photocurrent signal, the current step position coordinates of the scanning device are simultaneously obtained. Each photocurrent signal value and its corresponding step position coordinates are paired to form a data pair and an associated storage relationship is established. The associated data pairs are temporarily stored in the memory according to the scanning order. After scanning all step positions is completed, the corresponding photocurrent signal values ​​are arranged in the order of rows and columns according to the spatial distribution relationship of the step position coordinates to form a two-dimensional matrix corresponding to the spatial layout of the scanning dot matrix.

[0090] The two-dimensional matrix is ​​normalized to generate the two-dimensional current distribution map; wherein, the pixels in the two-dimensional current distribution map have a one-to-one correspondence with the step position coordinates.

[0091] Specifically, normalization is performed on the value of each element in the two-dimensional matrix, the maximum and minimum values ​​of all elements in the two-dimensional matrix are calculated, the minimum value is subtracted from the value of each element and then divided by the difference between the maximum and minimum values, the normalized value is mapped to a preset grayscale value range, and the processed value is converted into a grayscale image according to the row and column structure of the two-dimensional matrix to generate a two-dimensional current distribution map, in which the row and column position of each pixel maintains a one-to-one spatial mapping relationship with the step position coordinates.

[0092] In the above-mentioned optional methods, the zero-bias photocurrent signal is further acquired at each step position and stored in association with the coordinates. After being arranged into a two-dimensional matrix, it is normalized to ensure that the generated two-dimensional current distribution map maintains a one-to-one correspondence with the scanning position, thus establishing a standardized data representation form.

[0093] In one alternative approach, S4 specifically includes:

[0094] The two-dimensional current distribution map is subjected to filtering and noise reduction processing.

[0095] Among them, filtering and noise reduction processing refers to algorithms for eliminating noise in signals; for example, using a Gaussian filtering algorithm to perform filtering and noise reduction processing on a two-dimensional current distribution map.

[0096] Specifically, a Gaussian filtering algorithm is applied to the two-dimensional current distribution map. A Gaussian kernel with a preset size and standard deviation is used to perform convolution operation on the two-dimensional current distribution map. The weighted average value of each pixel and its neighboring pixels is calculated, and the original pixel values ​​are replaced to smooth the image and suppress noise, thus generating a filtered and denoised two-dimensional current distribution map.

[0097] An edge detection algorithm is applied to the processed two-dimensional current distribution map to extract the edge contour features of the reconstructed pattern.

[0098] Edge detection algorithms refer to computer algorithms that extract the edges of an image; for example, the Canny operator can be used to perform edge detection on a filtered image.

[0099] Specifically, the Canny edge detection algorithm is applied to the two-dimensional current distribution map after filtering and noise reduction to calculate the gradient magnitude and direction of each pixel in the image. The local gradient maximum point is retained by non-maximum suppression. Then, the double threshold method is used to distinguish strong edge and weak edge pixels. Strong edge pixels are connected to form continuous edge segments, thereby extracting the edge contour features of the reconstructed pattern as the edge pixel set.

[0100] The edge contour features are spatially registered and compared with the standard contour features. The edge transition region is identified in the registered edge contour features, and the intensity change gradient of the edge transition region is calculated.

[0101] The intensity gradient refers to the rate of change of light intensity in space. For example, in the edge transition region, the change in photocurrent intensity from 10 nanoamps to 2 nanoamps per micrometer along the scanning direction is the intensity gradient.

[0102] Specifically, the edge contour features are spatially registered with the standard contour features. By matching feature points and transforming coordinates, the edge contour features and the standard contour features are aligned in space. Then, the geometric differences between the aligned edge contour features and the standard contour features are compared. The regions in the registered edge contour features where the light intensity changes significantly are located as edge transition regions. The derivative of the photocurrent intensity in the edge transition region with position is calculated along the scanning direction to obtain the intensity change gradient.

[0103] The spatial resolution parameter is calculated based on the intensity change gradient, wherein the spatial resolution parameter R is calculated using the following formula:

[0104] ,

[0105] In the formula, This represents the difference in photocurrent intensity between two adjacent sampling points along the scanning direction within the edge transition region. This represents the actual spatial distance between two adjacent sampling points on the preset scanning path. This represents the average intensity gradient along the scanning direction in the edge transition region. This represents the maximum value of the intensity gradient along the scanning direction in the edge transition region. This indicates the maximum intensity value in the edge transition region. This represents the minimum intensity value of the edge transition region.

[0106] It should be noted that the formula for the spatial resolution parameter quantifies the detector's detail resolution capability by using the intensity change gradient in the edge transition region. The numerator of the formula is the product of the average and maximum absolute values ​​of the intensity gradients to comprehensively reflect the overall level and local extrema of the gradient change. The denominator is the square of the difference between the maximum and minimum intensity to eliminate the influence of the intensity dynamic range, thus achieving standardization. This formula transforms the intensity change characteristics of the edge region into a quantifiable parameter for objectively evaluating the detector's ability to distinguish subtle features in the spatial dimension. A higher parameter value indicates better spatial resolution performance of the detector.

[0107] In the above-mentioned optional methods, filtering and noise reduction and edge detection algorithms are further applied to the two-dimensional current distribution map to extract contour features. Combined with spatial registration and standard contour comparison, edge transition regions are identified and intensity change gradients are calculated. The resolution evaluation is numericalized through the quantization formula of the spatial resolution parameter R.

[0108] In an alternative approach, the step of evaluating the pattern reconstruction capability of the infrared photodetector under test based on the degree of overlap between the edge contour features and the standard contour features further includes:

[0109] Calculate the area of ​​the overlapping region between the edge contour feature and the standard contour feature, and the total contour area.

[0110] The overlapping area refers to the area of ​​the part where the two contours overlap; for example, the number of pixels in the overlapping part of the reconstructed "H" contour and the standard contour is converted into the overlapping area. The total contour area refers to the total area of ​​the standard contour; for example, the total number of pixels contained in the standard "H" contour is converted into the total contour area.

[0111] Specifically, on the spatially registered edge contour features and standard contour features, the overlapping pixel regions are identified, the number of overlapping pixels is counted and multiplied by the actual area of ​​a single pixel to obtain the area of ​​the overlapping region; at the same time, the total number of pixels contained in the standard contour features is counted and multiplied by the actual area of ​​a single pixel to obtain the total contour region area.

[0112] Extract the normal gradient magnitude of each edge point in the edge contour feature.

[0113] The normal gradient magnitude refers to the rate of change of intensity along the normal direction of the edge; for example, the normal gradient magnitude can be obtained by calculating the intensity difference in the normal direction of each edge point of the "H" letter outline.

[0114] Specifically, for each edge point in the edge contour feature, the normal direction is determined according to the gradient direction of the point. The photocurrent intensity difference between adjacent pixels is calculated along the normal direction in the two-dimensional current distribution map. The intensity change rate is obtained by dividing by the pixel spacing. The absolute value of the intensity change rate is used as the normal gradient magnitude of the edge point.

[0115] The comprehensive contour overlap evaluation value is calculated based on the area of ​​the overlapping region, the area of ​​the total contour region, and the magnitude of the normal gradient.

[0116] The comprehensive contour overlap evaluation value refers to the contour matching score that takes into account multiple factors; for example, a comprehensive contour overlap evaluation value of 0.92 is obtained by combining the overlapping area and the edge gradient.

[0117] The first level corresponding to the pattern reconstruction capability of the infrared photodetector under test is determined based on the comprehensive contour overlap evaluation value.

[0118] The first level refers to the rating classification of pattern reconstruction capability; for example, a comprehensive contour overlap evaluation value of 0.92 corresponds to "Level A" in the first level.

[0119] Specifically, the calculated comprehensive contour overlap evaluation value is compared with multiple preset first-level threshold ranges. Based on the threshold range in which the comprehensive contour overlap evaluation value is located, the first level corresponding to the pattern reconstruction capability of the infrared photodetector under test is determined. The first level includes multiple performance levels arranged from high to low.

[0120] The comprehensive contour overlap evaluation value S is calculated using the following formula:

[0121] ,

[0122] In the formula, This represents the area of ​​the overlapping region between the edge contour feature and the standard contour feature. This represents the total area of ​​the contour region of the standard contour feature. This represents the total number of edge points. Indicates the first The magnitude of the normal gradient at each edge point and These are the preset weighting coefficients.

[0123] It should be noted that the formula for the comprehensive contour overlap evaluation value integrates both pattern shape matching and edge quality. The base part uses the ratio of the overlapping area to the total contour area to measure the similarity of the contour geometry. The adjustment part introduces the arithmetic mean and squared mean of the normal gradient magnitude and weights them with coefficients to enhance the sensitivity to edge sharpness and consistency. This formula calculates a comprehensive score through multi-feature coupling, comprehensively characterizing the degree of matching between the detector-reconstructed pattern and the standard pattern in terms of contour overlap and edge sharpness. The higher the value, the better the accuracy and fidelity of the pattern reconstruction.

[0124] In the above-mentioned optional methods, the pattern reconstruction capability is further transformed into a comparable numerical index by calculating the area of ​​the overlapping region between the edge contour and the standard contour, the total contour area, and extracting the normal gradient magnitude of each edge point, and by using a quantitative formula for the comprehensive contour overlap evaluation value, so as to achieve an objective evaluation of the reconstruction accuracy.

[0125] In one alternative approach, the step of evaluating the spatial recognition capability of the infrared photodetector under test based on the value of the spatial resolution parameter further includes:

[0126] Obtain the spatial resolution parameters of all edge transition regions in the edge contour features.

[0127] Specifically, all edge transition regions are identified in the edge contour features. For each edge transition region, the intensity gradient data along the scanning direction is extracted, the average and maximum values ​​of the absolute values ​​of the intensity gradient are calculated, and the maximum and minimum intensity values ​​of the region are obtained. Then, the spatial resolution parameters of each edge transition region are calculated according to the spatial resolution parameter calculation formula.

[0128] Calculate the statistical distribution characteristics of the spatial resolution parameters of all edge transition regions.

[0129] Among them, statistical distribution characteristics refer to the statistical properties of multiple parameters; for example, the average and standard deviation of the spatial resolution parameters of all edge transition regions are calculated as statistical distribution characteristics.

[0130] Specifically, spatial resolution parameters of all edge transition regions are collected, and the arithmetic mean of these spatial resolution parameters is calculated as the average. The square root of the arithmetic mean of the squares of the deviations of each spatial resolution parameter from the average is calculated as the standard deviation. The arithmetic mean of the cube of each spatial resolution parameter from the average is calculated as the third central moment. The arithmetic mean of the fourth power of each spatial resolution parameter from the average is calculated as the fourth central moment, thereby obtaining the statistical distribution characteristics of the spatial resolution parameters.

[0131] The spatial recognition capability evaluation value is calculated based on the statistical distribution characteristics of the spatial resolution parameters.

[0132] Among them, the spatial recognition ability evaluation value refers to the quantitative score of spatial recognition ability; for example, a spatial recognition ability evaluation value of 85 points is obtained by calculating through statistical distribution characteristics.

[0133] The second level corresponding to the spatial recognition capability of the infrared photodetector under test is determined based on the spatial recognition capability evaluation value.

[0134] The second level refers to the rating and classification of spatial recognition ability; for example, a spatial recognition ability rating of 85 points corresponds to "Level B" in the second level.

[0135] Specifically, the spatial recognition capability evaluation value is compared with multiple preset second-level threshold ranges. Based on the specific threshold range in which the spatial recognition capability evaluation value is located, the second level corresponding to the spatial recognition capability of the infrared photodetector under test is determined. The second level includes multiple performance levels arranged from high to low.

[0136] The spatial recognition capability evaluation value Q is calculated using the following formula:

[0137] ,

[0138] In the formula, This represents the average spatial resolution parameter of all edge transition regions. This represents the standard deviation of the spatial resolution parameters for all edge transition regions. The third central moment represents the spatial resolution parameter of all edge transition regions. The fourth central moment represents the spatial resolution parameter of all edge transition regions. , , and These are the preset weighting coefficients.

[0139] It should be noted that the formula for evaluating spatial recognition capability is based on the multi-dimensional characteristics of the statistical distribution of spatial resolution parameters. The formula includes the mean to reflect the overall resolution level, the standard deviation to measure resolution consistency, the third central moment to capture distribution skewness, and the fourth central moment to describe distribution kurtosis. These statistics are integrated through preset weighting coefficients. This formula comprehensively evaluates the stability and overall quality of the detector's recognition performance at different spatial locations from four aspects: central tendency, dispersion, asymmetry, and sharpness. Higher values ​​indicate stronger and more uniform spatial recognition capability of the detector.

[0140] In the above-mentioned optional methods, the spatial resolution parameters of all edge transition regions are further obtained and their statistical distribution characteristics are calculated. The average value, standard deviation and higher-order moment information are comprehensively reflected by the quantitative formula of the spatial recognition capability evaluation value to fully reflect the spatial recognition consistency of the detector under test at different positions.

[0141] In an optional embodiment, the step of generating a performance evaluation result for the infrared photodetector under test based on the evaluation results of the pattern reconstruction capability and the spatial recognition capability further includes:

[0142] The first level corresponding to the pattern reconstruction capability and the second level corresponding to the spatial recognition capability are used as input parameters, and the first level and the second level are weighted and fused based on a preset comprehensive evaluation rule.

[0143] Among them, the preset comprehensive evaluation rules refer to the pre-set comprehensive evaluation methods; for example, setting the pattern reconstruction ability weight at 60% and the spatial recognition ability weight at 40% as the preset comprehensive evaluation rules.

[0144] Specifically, the first level corresponding to pattern reconstruction capability and the second level corresponding to spatial recognition capability are mapped to preset numerical scores respectively. Based on the weight coefficients defined in the preset comprehensive evaluation rules, the numerical scores of pattern reconstruction capability and spatial recognition capability are weighted and summed to obtain the weighted fusion calculation result.

[0145] The overall performance score of the infrared photodetector under test is determined based on the weighted fusion calculation results.

[0146] The overall performance score refers to the quantitative score of the overall performance; for example, a score of 90.5 obtained through weighted fusion calculation is used as the overall performance score.

[0147] Specifically, the weighted fusion calculation result is used as the basic value. The basic value is standardized according to the preset comprehensive evaluation rules. The processed value is then mapped to the preset comprehensive performance scoring range to determine the comprehensive performance score of the infrared photodetector under test.

[0148] The comprehensive performance score is compared with a preset performance level threshold to generate the final performance evaluation result of the infrared photodetector under test.

[0149] The preset performance level thresholds refer to the pre-defined grading standards; for example, setting 90 points or above as "Excellent" and 80-90 points as "Good" as preset performance level thresholds. The final performance evaluation result refers to the conclusion after comprehensive evaluation; for example, a comprehensive performance score of 90.5 points corresponds to a final performance evaluation result of "Excellent".

[0150] Specifically, the comprehensive performance score is compared with the preset performance level threshold. Based on whether the comprehensive performance score is greater than or equal to the first threshold, between the first and second thresholds, or less than the second threshold, the final performance evaluation result of the infrared photodetector under test is determined to be qualified, basically qualified, or unqualified.

[0151] In the above-mentioned optional methods, the pattern reconstruction capability level and spatial recognition capability level are further used as input parameters. Based on the preset comprehensive evaluation rules, a weighted fusion calculation is performed. By comparing with the performance level threshold, a final evaluation result including qualified, basically qualified or unqualified is generated, forming a complete evaluation conclusion.

[0152] In one alternative approach, the preset path is a matrix of scanning points, where the spacing between adjacent scanning points is smaller than the pixel size of the infrared photodetector under test.

[0153] In this context, a matrix-arranged scanning dot matrix refers to scanning positions arranged in a matrix manner; for example, 140 scanning points evenly distributed in 14 rows and 10 columns constitute a matrix-arranged scanning dot matrix. Pixel size refers to the physical size of a single pixel in the detector; for example, the size of a single pixel in the detector is 15 micrometers × 15 micrometers.

[0154] In the above-mentioned optional methods, the preset path is further set as a matrix of scanning points and the distance between adjacent scanning points is smaller than the pixel size of the detector under test, thereby increasing the spatial sampling density, ensuring that the tiny pattern features are completely captured, and avoiding the loss of spatial information.

[0155] In one optional approach, the final performance evaluation result includes three levels: qualified, basically qualified, and unqualified; wherein, the qualified level corresponds to the comprehensive performance score being greater than or equal to a first threshold, the basically qualified level corresponds to the comprehensive performance score being less than the first threshold but greater than or equal to a second threshold, and the unqualified level corresponds to the comprehensive performance score being less than the second threshold.

[0156] The first threshold refers to the upper limit of the performance level classification; for example, 90 points is set as the first threshold to distinguish between "excellent" and "good". The second threshold refers to the lower limit of the performance level classification; for example, 80 points is set as the second threshold to distinguish between "good" and "pass".

[0157] In the above-mentioned optional methods, the final performance evaluation results are further clearly divided into three levels: qualified, basically qualified, and unqualified, which correspond to different comprehensive performance scoring threshold ranges, making the evaluation output clear and easy to understand, and facilitating a quick judgment of the performance level of the detector under test.

[0158] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0159] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0160] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for pattern reconstruction and identification capability verification of an infrared photodetector, characterized in that, The method comprises: controlling an infrared laser source to emit infrared laser and irradiate the infrared laser to a pattern mask having a predefined pattern; controlling a scanning device to move at least one of the pattern mask and an infrared photodetector to be tested, so that a laser spot transmitted or reflected from the pattern mask scans a photosensitive surface of the infrared photodetector to be tested according to a preset path; under a zero bias voltage operation mode of the infrared photodetector to be tested, collecting and recording a photocurrent signal corresponding to each scanning position, and integrating each photocurrent signal into a two-dimensional current distribution map according to corresponding scanning position information; performing image processing on the two-dimensional current distribution map, extracting edge profile features of a reconstructed pattern, comparing the edge profile features with standard profile features of the predefined pattern, and calculating a spatial resolution parameter based on an intensity variation gradient of an edge transition region; based on a coincidence degree between the edge profile features and the standard profile features, evaluating a pattern reconstruction capability of the infrared photodetector to be tested, based on a numerical value of the spatial resolution parameter, evaluating a spatial recognition capability of the infrared photodetector to be tested, and generating a performance evaluation result of the infrared photodetector to be tested according to an evaluation result of the pattern reconstruction capability and an evaluation result of the spatial recognition capability; the step of performing image processing on the two-dimensional current distribution map, extracting edge profile features of a reconstructed pattern, comparing the edge profile features with standard profile features of the predefined pattern, and calculating a spatial resolution parameter based on an intensity variation gradient of an edge transition region further comprises: performing filtering and noise reduction processing on the two-dimensional current distribution map; performing an edge detection algorithm on the processed two-dimensional current distribution map to extract the edge profile features of the reconstructed pattern; spatially registering and comparing the edge profile features with the standard profile features, identifying the edge transition region in the registered edge profile features, and calculating the intensity variation gradient of the edge transition region; calculating the spatial resolution parameter based on the intensity variation gradient, wherein the spatial resolution parameter R is calculated by the following formula: ， In the formula, denotes the difference of the photocurrent intensity between two adjacent sampling points along the scanning direction in the edge transition region, denotes the actual spatial distance between two adjacent sampling points along the scanning direction in the preset scanning path, denotes the average value of the intensity gradient along the scanning direction in the edge transition region, denotes the maximum value of the intensity gradient along the scanning direction in the edge transition region, denotes the maximum intensity value of the edge transition region, denotes the minimum intensity value of the edge transition region.

2. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 1, characterized in that, the step of controlling an infrared laser source to emit infrared laser and irradiate the infrared laser to a pattern mask further comprises: controlling the infrared laser source to generate the infrared laser of a specific wavelength, and collimating the infrared laser and vertically irradiating the surface of the pattern mask; wherein the pattern mask is a metal mask plate etched with the predefined pattern, and the predefined pattern comprises a feature geometric pattern for verifying spatial resolution.

3. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 2, characterized in that, the step of controlling a scanning device to move at least one of the pattern mask and an infrared photodetector to be tested, so that a laser spot transmitted or reflected from the pattern mask scans a photosensitive surface of the infrared photodetector to be tested according to a preset path further comprises: fixing the infrared photodetector to be tested on a bearing platform of the scanning device; controlling the scanning device to drive the carrying platform to move in a two-dimensional step-by-step manner according to a preset path, the preset path covering a complete projection area of the predefined pattern of the pattern mask on the photosensitive surface; controlling the scanning device to stay at each step position of the preset path for a preset sampling time, and controlling the scanning device to coordinate the movement of the carrying platform with the irradiation of the laser spot to complete the point-by-point scanning of the predefined pattern on the photosensitive surface.

4. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 3, characterized in that, The step of collecting and recording the photocurrent signals corresponding to each scanning position and integrating each photocurrent signal into a two-dimensional current distribution map according to the corresponding scanning position information when the to-be-tested infrared photodetector is in the zero-bias working mode further comprises: controlling a data acquisition unit to collect the photocurrent signals generated by the to-be-tested infrared photodetector in the zero-bias working mode during the stay of the scanning device at each step position; controlling the data acquisition unit to store each collected photocurrent signal in association with the corresponding step position coordinates, and arranging the associated photocurrent signals into a two-dimensional matrix according to the stored step position coordinates after the scanning at all step positions is completed; normalizing the two-dimensional matrix to generate the two-dimensional current distribution map; wherein the pixels in the two-dimensional current distribution map have a one-to-one correspondence with the step position coordinates.

5. The method for pattern reconstruction and identification capability verification of infrared photodetectors according to claim 1, characterized in that, The step of evaluating the pattern reconstruction capability of the to-be-tested infrared photodetector based on the degree of coincidence between the edge profile feature and the standard profile feature further comprises: calculating the overlapping area and the total profile area between the edge profile feature and the standard profile feature; extracting the normal gradient amplitude of each edge point in the edge profile feature; calculating a comprehensive profile coincidence evaluation value based on the overlapping area, the total profile area, and the normal gradient amplitude; determining the first grade corresponding to the pattern reconstruction capability of the to-be-tested infrared photodetector according to the comprehensive profile coincidence evaluation value; wherein the comprehensive profile coincidence evaluation value S is calculated using the following formula: ， wherein, represents an area of an overlapping region between the edge profile feature and the standard profile feature, represents a total profile area of the standard profile feature, represents a total number of edge points, represents a normal gradient amplitude of the th edge point, and is a preset weight coefficient.

6. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 5, wherein, The step of evaluating the spatial recognition capability of the to-be-tested infrared photodetector based on the numerical value of the spatial resolution parameter further comprises: obtaining the spatial resolution parameters of all edge transition regions in the edge profile feature; calculating the statistical distribution characteristics of the spatial resolution parameters of all edge transition regions; calculating a spatial recognition capability evaluation value based on the statistical distribution characteristics of the spatial resolution parameters; determining the second grade corresponding to the spatial recognition capability of the to-be-tested infrared photodetector according to the spatial recognition capability evaluation value; wherein the spatial recognition capability evaluation value Q is calculated using the following formula: ， wherein an average of the spatial resolution parameters of all edge transition regions, a standard deviation of the spatial resolution parameters of all edge transition regions, a third central moment of the spatial resolution parameters of all edge transition regions, a fourth central moment of the spatial resolution parameters of all edge transition regions, , , and are preset weight coefficients.

7. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 6, characterized in that, The step of generating the performance evaluation result of the to-be-tested infrared photodetector according to the evaluation results of the pattern reconstruction capability and the spatial recognition capability further comprises: taking the first grade corresponding to the pattern reconstruction capability and the second grade corresponding to the spatial recognition capability as input parameters, and performing weighted fusion calculation on the first grade and the second grade based on a preset comprehensive evaluation rule; determine a comprehensive performance score of the to-be-tested infrared photodetector according to the weighted fusion calculation result; compare the comprehensive performance score with a preset performance level threshold to generate a final performance evaluation result of the to-be-tested infrared photodetector.

8. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 3, wherein, The preset path is a matrix-arranged scanning point array, and a spacing between adjacent scanning points is less than a pixel size of the to-be-tested infrared photodetector.

9. The method for pattern reconstruction and identification capability verification of an infrared photodetector according to claim 7, wherein, The final performance evaluation result includes three levels of qualified, basically qualified and unqualified; wherein the qualified level corresponds to the comprehensive performance score being greater than or equal to a first threshold, the basically qualified level corresponds to the comprehensive performance score being less than the first threshold and greater than or equal to a second threshold, and the unqualified level corresponds to the comprehensive performance score being less than the second threshold.

Images