Processing method and device of dispersion band diagram and angle-resolved system using the same

Abstract

This application discloses a method, apparatus, and angle-resolved system for processing dispersive band diagrams. The method includes: performing nonlinear calibration on the image to be processed; detecting the number of image stitching segments; aligning the image stitching segments; and stitching and reconstructing the image stitching segments to obtain a complete image. The method, apparatus, and angle-resolved system disclosed in this application are compatible with dispersive band diagram testing of both photonic crystal structures without active regions and photonic crystal structures containing active regions. Through nonlinear calibration processing, measurement errors in large-angle regions are significantly reduced. Through the detection and alignment of the number of image stitching segments, image stitching errors are reduced. The method enables the stitching and reconstruction of multi-segment band structure test data, improving test accuracy.

Description

Technical Field

[0001] This application relates to the field of semiconductor laser technology, and more particularly to a method, apparatus, and angle-resolved system for processing dispersive band diagrams. Background Technology

[0002] Surface-emitting photonic crystal semiconductor lasers have broad application prospects in new energy, laser information processing, laser biology and medicine, laser printing, communications, laser chemistry, and laser detection and metrology. Surface-emitting photonic crystal semiconductor lasers have significant advantages in terms of size, beam quality, and integration, and have great application potential in industry and communications.

[0003] Currently, the technology and development of surface-emitting photonic crystal semiconductor lasers are still immature, and the transverse and longitudinal modes of these lasers are limited by the band structure of the photonic crystal. Existing methods for testing dispersive band structures are limited, generally used for testing non-active photonic crystal structures, and only include dispersive band structure testing structures with a single light source. Furthermore, existing angle-resolved systems suffer from low accuracy, slow speed, poor signal-to-noise ratio, and significant stitching errors in the generated images. All of these factors significantly impact the accuracy of test results for surface-emitting photonic crystal semiconductor lasers. Summary of the Invention

[0004] To address one or more of the above problems, this application proposes a method for verifying semiconductor laser materials.

[0005] According to one aspect of this application, a method for processing dispersive band diagrams is provided, comprising the following steps: performing nonlinear calibration on the image to be processed; detecting the number of image stitching segments; aligning the image stitching segments; and stitching and reconstructing the image stitching segments to obtain a complete image.

[0006] In some implementations, nonlinear calibration of the image to be processed includes: Calculate the angle of the diffracted light in real space based on the pixel coordinates; Calculate the angular space coordinates based on the angle of the diffracted light in real space.

[0007] In some implementations, the number of image stitching segments detected includes: The wavelength step sequence is obtained by performing adjacent difference calculations on the wavelength sequence in the spectral data. The change between adjacent sampling points is obtained based on the wavelength step sequence; Based on the preset criteria for judging abnormal stitching points, determine whether the sampling point is an abnormal stitching point in the image. The number of image stitching points and the number of image stitching segments are calculated based on the number of abnormal points in the image stitching, and the wavelength of the stitching boundary is also calculated.

[0008] In some implementations, aligning image stitching segments includes: Extract the curve of each splicing boundary, and extract one curve to the left and one curve to the right of each splicing boundary; The two curves extracted from each splicing boundary are processed to remove DC noise; Define the cross-correlation function for the two curves corresponding to each splicing boundary; Each image stitching segment is defined as a sub-image. Based on the cross-correlation function of the two curves corresponding to each stitching boundary, the optimal translation amount between adjacent image stitching segments and the cumulative translation amount of the current image stitching segment relative to the first image stitching segment are calculated.

[0009] In some implementations, stitching and reconstructing image segments includes: Calculate the number of pixels that the sub-image needs to be translated based on the optimal translation amount between adjacent image stitching segments. Based on the number of pixels that need to be translated from the sub-images, the image stitching segments are stitched together and reconstructed to obtain the overall image.

[0010] According to a second aspect of this application, an apparatus for processing dispersive band diagrams is provided, comprising the following steps: The nonlinear calibration module is used to perform nonlinear calibration on the image to be processed. The sub-image quantity detection module is used to detect the number of image stitching segments, and defines each image stitching segment as a sub-image; The sub-image alignment module is used to align image stitching segments; The sub-image stitching module is used to stitch and reconstruct image segments to obtain the overall image.

[0011] According to a third aspect of this application, an angle-resolved system is provided, comprising: a dispersive band diagram processing apparatus, wherein the dispersive band diagram processing apparatus applies the processing method of the dispersive band diagram provided in any of the above claims.

[0012] In some embodiments, the angular resolution system further includes a multi-channel light source device, an angular resolution imaging device, and a momentum space imaging device. The multi-channel light source device includes at least a first light source, a second light source, and a light source switching module. The light source switching module is used to control the light incident angle resolution imaging device of the first light source, the light incident angle resolution imaging device of the second light source, or the light incident angle resolution imaging device without light. An angle-resolved imaging device includes at least a five-axis adjustment module, a power supply module, a switchable objective lens module, a switchable beam splitter module, and an imaging tube. The sample to be tested can be placed on the five-axis adjustment module, which is used to level or switch the angle of the sample to be tested. The power supply module is used to supply power to the sample to be tested. The light emitted from the multi-channel light source device can pass through the switchable beam splitter module and the switchable objective lens module in sequence before entering the sample to be tested. The light emitted from the sample to be tested can pass through the switchable objective lens module, the switchable beam splitter module, and the imaging tube in sequence before entering the momentum space imaging device. A momentum space imaging device includes at least an imaging lens and a spectrometer. The light emitted from the imaging tube is collected by the spectrometer after passing through the imaging lens, and the spectrometer can output a dispersive band structure. The dispersive band diagram processing device processes the dispersive band diagram.

[0013] In some embodiments, the angle-resolved imaging device further includes a real-space imaging module disposed on the rear focal plane of the imaging tube, which is used to image the sample to be tested.

[0014] In some embodiments, the angle-resolved imaging device further includes an imaging switching module, which is disposed between the imaging tube and the real space imaging module. The imaging switching module is used to control the light emitted from the imaging tube to enter the real space imaging module or the momentum space imaging device.

[0015] The dispersive band structure processing method, apparatus, and angle-resolved system disclosed in this application are compatible with dispersive band structure testing of both photonic crystal structures without active regions and photonic crystal structures containing active regions. Through nonlinear calibration processing, measurement errors in large-angle regions are significantly reduced. Image stitching errors are reduced through image segment number detection and alignment processing. The system can stitch and reconstruct test data from multiple band structures, improving test accuracy. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for processing dispersive band diagrams according to an embodiment of this application.

[0017] Figure 2 This indicates the image obtained after linear calibration in the dispersive band diagram processing method provided in an embodiment of this application when NA equals 0.22.

[0018] Figure 3 This indicates the image obtained after nonlinear calibration in the dispersive band diagram processing method provided in an embodiment of this application when NA equals 0.22.

[0019] Figure 4 This means that when NA equals 0.22, Figure 2 and Figure 3 The image shown is the result of interpolation followed by difference.

[0020] Figure 5 This indicates the image obtained after linear calibration in the dispersive band diagram processing method provided in an embodiment of this application when NA equals 0.42.

[0021] Figure 6 This indicates the image obtained after nonlinear calibration in the dispersive band diagram processing method provided in an embodiment of this application when NA equals 0.42.

[0022] Figure 7 This means that when NA equals 0.42, Figure 5 and Figure 6 The image shown is the result of interpolation followed by difference.

[0023] Figure 8 This refers to a dispersive band diagram that has not been processed by the dispersive band diagram processing method provided in the embodiments of this application.

[0024] Figure 9 This indicates the dispersive band diagram after processing by the dispersive band diagram processing method provided in the embodiments of this application.

[0025] Figure 10 This is a schematic diagram of the structure of a dispersive band pattern processing device provided in an embodiment of this application.

[0026] Figure 11 This is a schematic diagram of the structure of an angle resolution system provided in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only some, not all, of the embodiments of this invention, and are used merely to explain the invention and are not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0028] Example 1: Reference manual attached Figure 1 This invention provides a method for processing dispersive band diagrams, which may specifically include the following steps: Step 1.0: Perform nonlinear calibration on the image to be processed; Step 2.0: Detect the number of image stitching segments; Step 3.0: Align the image stitching segments; Step 4.0: Combine and reconstruct the image segments to obtain the overall image.

[0029] In this application, the image to be processed can be obtained by a spectrometer in an angle-resolved system. The stitched image can be the raw data collected by the spectrometer. The angle-resolved system should also include at least an objective lens, an imaging tube lens, and an imaging lens. The light emitted by the sample to be tested passes through the objective lens, the imaging tube lens, and the imaging lens in sequence and is then collected by the spectrometer to obtain the stitched image.

[0030] In step 1.0, performing nonlinear calibration on the image to be processed may include the following steps: Calculate the angle of the diffracted light in real space based on the pixel coordinates; Calculate the angular space coordinates based on the angle of the diffracted light in real space.

[0031] Specifically, angular space coordinates The expression is as follows:

[0032] In the formula, This indicates the focal length of the objective lens in an angle-resolved system. n Represents the refractive index in the object space. This represents the angle of diffracted light in real space.

[0033] The imaging tube and imaging lens form a relay lens, which transmits the objective lens's back focal plane coordinates, i.e., angular space coordinates. Image to slit plane angular coordinates The lateral magnification from the back focal plane of the objective lens to the slit plane of the spectrometer is defined as... ,but:

[0034] in This represents the physical coordinates on the slit plane along the angle axis (the direction of the slit length).

[0035] Since the spectrometer primarily achieves imaging from the slit image to its CCD along the angular axis, and the CCD is the image sensor of the spectrometer, the lateral magnification from the slit plane to the CCD (along the angular axis) is defined as... Then the physical coordinate Y on the CCD of the spectrometer satisfies the following formula:

[0036] Assume the pixel size of the spectrometer CCD is p ,when When the angle is 0°, the corresponding pixel coordinates are Then the following expression exists:

[0037] Simplifying the above formulas yields the following expression:

[0038] Therefore, the expression for the change of pixel coordinate y with respect to the angle θ of the diffracted light in real space is obtained as follows:

[0039] Right now:

[0040] In the formula, This represents the angle of diffracted light in real space. Indicates the focal length of the objective lens. This indicates the lateral magnification from the back focal plane of the objective lens to the slit plane of the spectrometer. , Indicates the focal length of the imaging tube. Indicates the focal length of the imaging lens. This represents the lateral magnification from the slit plane of the spectrometer to the CCD of the spectrometer. The value can be 1.

[0041] Therefore, through angular space coordinates The expression for the pixel coordinate y and the expression for the change of the angle θ of the diffracted light in real space can convert the pixel coordinates of the image to be processed into angular space coordinates through nonlinear calibration, eliminating the error generated in the large angle region and improving the accuracy of angular resolution testing.

[0042] Calculate the effective maximum angle of the image to be processed Assuming the corresponding equivalent effective numerical aperture is NA eff Then we have the following formula:

[0043]

[0044] In the formula, n Represents the refractive index in the object space. NA eff It is the effective numerical aperture in an angle-resolved system; The value is limited by both the imaging area of ​​the spectrometer's CCD (including manual selection) and the numerical aperture of the objective lens; assuming the maximum effective angle corresponding to the CCD's imaging area is... The maximum effective angle corresponding to the numerical aperture of the objective lens is The effective angle of the image to be processed. The value is and The smaller value in the formula is as follows:

[0045] Among them, the maximum angle corresponding to the CCD of the spectrometer. And the maximum effective angle corresponding to the numerical aperture of the objective lens is The expressions are as follows:

[0046]

[0047] In the formula, NA represents the numerical aperture of the objective lens.

[0048] In nonlinear calibration, the relationship between the angle θ(y) of the image to be processed and the pixel coordinate y can be written as:

[0049]

[0050] Therefore, the angle change corresponding to each pixel in the nonlinear calibration is obtained as follows: .

[0051] In an optional embodiment, the real space coordinates of the pixels in the image to be processed can also be converted into angular space coordinates through linear calibration. The specific method of linear calibration is as follows: Since the distance between each pixel is very small, the angle between adjacent pixels can be used in linear calibration. Set as a constant, the angle between adjacent pixels. The expression is:

[0052] Therefore, the angle change corresponding to each pixel in the linear calibration is:

[0053] Assuming the maximum angle is obtained by nonlinear calibration and linear calibration If they are the same, then the difference between the angle change of each pixel after non-linear calibration and the angle change of each pixel after linear calibration is as follows:

[0054]

[0055]

[0056] At the maximum angle and Under the same conditions, linear calibration assumes that the angular interval between pixels is constant, while nonlinear calibration obtains... As the angle changes, and with It is proportional to the angle accuracy of the dispersive band diagram, so even nonlinear calibration in the large-angle region can ensure the accuracy of the test results.

[0057] Reference manual attached Figure 2 To the instruction manual Figure 4 As shown. The instruction manual is attached. Figure 2 This represents the image obtained after linear calibration of the image acquired by the spectrometer when NA is 0.22. (See attached instruction manual.) Figure 3 This represents the image obtained after nonlinear calibration of the same image acquired by the spectrometer when NA is 0.22. (See attached instruction manual.) Figure 4 Instruction manual attached Figure 2 Included with instruction manual Figure 3 The image shown is the result of interpolation followed by difference.

[0058] Reference manual attached Figure 5 To the instruction manual Figure 7 As shown. The instruction manual is attached. Figure 5 This represents the image obtained after linear calibration of the image acquired by the spectrometer when NA is 0.42. (See attached instruction manual.) Figure 6 This represents the image obtained after nonlinear calibration of the same image acquired by the spectrometer when NA is 0.42. (See attached instruction manual.) Figure 7 Instruction manual attached Figure 5 Included with instruction manual Figure 6 The image shown is the result of interpolation followed by difference.

[0059] Based on the comparison of the two sets of images above, it can be concluded that as the angular span of the dispersive band diagram increases, the accuracy of linear calibration decreases, while the accuracy of nonlinear calibration is not affected by the angular span.

[0060] In step 2.0, detecting the number of image stitching segments may include the following steps: Step 2.1: Perform adjacent difference calculations on the wavelength sequence in the spectral data to obtain the wavelength step sequence; Step 2.2: Obtain the change between adjacent sampling points based on the wavelength step sequence; Step 2.3: Determine whether the sampling point is an image stitching anomaly point based on the preset stitching anomaly judgment conditions; Step 2.4: Calculate the number of image stitching points and the number of image stitching segments based on the number of image stitching anomalies, and calculate the stitching boundary wavelength.

[0061] Specifically, in step 2.1, performing adjacent difference calculations on the wavelength sequence in the spectral data to obtain the wavelength step sequence may include: Obtain the spectral data matrix M obtained by the spectrometer, where the spectral data matrix M contains N r Line N c Column, read wavelength, assuming the spectral data is located in the j-th row, where 1≤j≤N r If we place the wavelength group in the first row, then the wavelength array λ is the element of all columns in the first row of the spectral data matrix M. Then, we perform an adjacent difference operation on the wavelength array λ to obtain the wavelength step sequence. (i): .

[0062] In the formula, 1≤ i ≤N c -1.

[0063] Specifically, in step 2.2, obtaining the change between adjacent sampling points based on the wavelength step sequence includes: Since the range of values ​​for i is 1≤ i ≤ N c -1, then wavelength step sequence Δλ The length is N c -1, wavelength step sequence Δλ Sort in ascending order to obtain the sorted sequence. ,satisfy:

[0064] when N c When -1 is an odd number, the change between adjacent sampling points The element that is in the middle position after sorting:

[0065] when N c When -1 is an even number, the change between adjacent sampling points The average of the two middle elements:

[0066] In step 2.3, determining whether a sampling point is an image stitching anomaly point based on preset stitching anomaly point judgment conditions includes: The expression for setting the anomaly detection threshold T is:

[0067] In the formula, 'a' represents the proportionality coefficient. The anomaly detection threshold T must satisfy the following conditions:

[0068] in This represents the theoretical wavelength interval at wavelength λ for a single pixel point in the spectrometer. A sampling point is determined to be an image stitching anomaly point when it meets the following conditions;

[0069] In step 2.4, the number of image stitching points and the number of image stitching segments are calculated based on the number of image stitching anomalies, and the stitching boundary wavelength is calculated, including: The expression for the set of indexes of outliers in image stitching is as follows:

[0070] Using image stitching anomalies as segmentation points, the original spectral data sequence is divided into multiple continuous data segments. Each data segment is an image stitching segment, and each image stitching segment is defined as a sub-image. The expression for the number of image stitching points is as follows:

[0071] That is, the number of image stitching points is equal to the number of image stitching anomalies.

[0072] The expression for the number of image stitching segments, i.e., sub-images, is as follows:

[0073] That is, the number of image stitching segments is equal to the number of image stitching points plus one.

[0074] The expression for the splicing boundary wavelength is as follows:

[0075] In the formula, idx represents the set of indices of image stitching anomalies. The set of index values ​​representing the wavelength of the abrupt change is used to obtain the column data of the wavelength at the image stitching point, which is the intensity data.

[0076] In step 3.0, aligning the image stitching segments includes: Step 3.1: Extract the curve of each splicing boundary, and extract one curve on the left and one on the right of each splicing boundary; Step 3.2: Perform DC noise removal processing on the two curves extracted from each splicing boundary; Step 3.3: Define the cross-correlation function for the two curves corresponding to each splicing boundary; Step 3.4: Define each image stitching segment as a sub-image. Based on the cross-correlation function of the two curves corresponding to each stitching boundary, calculate the optimal translation amount between adjacent image stitching segments and the cumulative translation amount of the current image stitching segment relative to the first image stitching segment.

[0077] In step 3.1, extracting the curve for each splicing boundary may include: Sort the data in the idx set in ascending order to get a new set B, and the expression for B is as follows:

[0078] In the formula, b m ∈idx, where m satisfies b m This indicates the m-th image stitching anomaly.

[0079] Consider each element in set B as a stitching boundary. The number of stitching boundaries plus one equals the number of image stitching segments. Select one column on each side of the stitching boundary, forming two curves. The curves on the left and right sides of the stitching boundary can then be represented by the following expression:

[0080]

[0081] In the formula, 2≤r≤N r , N r This represents the total number of rows in the spectral data matrix M.

[0082] In step 3.2, the DC noise removal process for the two curves extracted at each splicing boundary includes:

[0083] In step 3.3, the cross-correlation function for the two curves corresponding to each splicing boundary is defined as follows:

[0084] In the formula, z represents the translation of the curve in pixels. It represents both the number of rows in the spectral data matrix and the number of pixels in the angular space.

[0085] In step 3.4, calculating the optimal translation between adjacent image stitching segments and the cumulative translation of the current image stitching segment relative to the first image stitching segment includes: Since the two image segments are approximately considered continuous at the splicing point, when When the maximum value is reached, the two curves are most correlated. Therefore, the optimal translation amount between adjacent image stitching segments can be determined by searching for the maximum value of the cross-correlation function of the two curves corresponding to each stitching boundary. .

[0086] Optimal translation amount The expression is:

[0087] To achieve alignment of all image stitching segments in a unified reference coordinate system, it is necessary to calculate the cumulative translation of the m-th image stitching segment relative to the first image stitching segment, corresponding to the m-th image stitching anomaly point. The m-th image stitching segment refers to the data between the m-th and (m+1)-th image stitching anomaly points. The expression for the cumulative translation is as follows: .

[0088] In step 4.0, the image stitching segments are stitched and reconstructed, including: Step 4.1: Calculate the number of pixels the sub-image needs to be translated based on the optimal translation amount between adjacent image stitching segments. Step 4.2: Based on the number of pixels that the sub-images need to be translated, stitch and reconstruct the image stitching segments to obtain the overall image.

[0089] In step 4.1, the formula for calculating the number of pixels the sub-image needs to be translated is as follows:

[0090] Where 2≤ s ≤ N segment ,and =0.

[0091] In step 4.2, the image stitching segments are stitched together according to the number of pixels that the sub-images need to be translated, and the resulting formula is as follows:

[0092] In the formula, Indicates the background noise intensity. This represents the original s-th sub-image. Let represent the s-th sub-image after alignment. The pixel coordinates and angular space coordinates w obtained in step 1.0 satisfy the mapping relationship w = w(r), where x represents the wavelength coordinate. Therefore, Can be converted The above formula shows that during the stitching process, the size of each sub-image remains unchanged, and background noise can be used to fill in areas that exceed the boundaries or are blank.

[0093] The final reconstructed overall image expression is as follows: .

[0094] Reference manual attached Figure 8 Included with instruction manual Figure 9 Instruction manual attached Figure 8The diagram shows a dispersive band structure that has not been processed by the dispersive band structure processing method provided in the embodiments of this application. (See attached specification.) Figure 8 It is evident that there are obvious image stitching marks at wavelengths of 920nm and 930nm. (See attached instruction manual.) Figure 9 This shows a dispersive band diagram processed by the dispersive band diagram processing method provided in the embodiments of this application, as shown in the appendix to the specification. Figure 9 It can be seen that there are no image stitching marks at wavelengths of 920nm and 930nm, which indicates that the dispersive band diagram processed by the dispersive band diagram processing method provided in this application embodiment effectively reduces the image stitching error, improves the authenticity and accuracy of the dispersive band diagram, and improves the test accuracy.

[0095] The dispersive bandgap image processing method provided in this application eliminates errors generated in large-angle regions by establishing a nonlinear mapping relationship between pixel coordinates and true angles, ensuring that the image is not distorted in large-angle regions, thereby improving image realism and angular resolution test accuracy. By detecting and aligning the number of image stitching segments, it reduces image stitching errors and improves image accuracy. It can realize the stitching and reconstruction of multi-segment bandgap structure test data, thereby improving test accuracy.

[0096] Example 2: Reference manual attached Figure 10 This invention provides a processing apparatus for dispersive band diagrams. Applying any one of the dispersive band diagram processing methods provided in the above embodiments, the processing apparatus for dispersive band diagrams may include: Nonlinear calibration module 41 is used to perform nonlinear calibration on the image to be processed; The sub-image quantity detection module 42 is used to detect the number of image splicing segments and defines each image splicing segment as a sub-image; Sub-image alignment module 43 is used to align image splicing segments; The sub-image stitching module 44 is used to stitch and reconstruct the image stitching segments to obtain the overall image.

[0097] Other parts that are the same as in Example 1 will not be described again.

[0098] Example 3: Reference manual attached Figure 11 The present invention provides an angle resolution system, which includes at least the dispersion band diagram processing device 4 proposed in any one of the above embodiments. The dispersion band diagram processing device 4 is capable of applying the dispersion band diagram processing method provided in any one of the above embodiments.

[0099] The angle-resolved system may also include a multi-channel light source device 1, an angle-resolved imaging device 2, and a momentum space imaging device 3.

[0100] The multi-channel light source device 1 includes at least a first light source 11, a second light source 12, and a light source switching module 13. The light source switching module 13 is used to control the light incident angle resolution imaging device of the first light source 11, or the light incident angle resolution imaging device of the second light source 12, or the light-incident angle resolution imaging device.

[0101] In an optional embodiment, the first light source 11 can be a narrowband pump laser, which can be used to pump a photonic crystal structure with a gain medium. A narrowband filter can be disposed after the first light source 11 to filter out excess lasing spectrum and narrow the spectrum of the laser output from the first light source 11.

[0102] In an optional embodiment, the second light source 12 can be a broadband light source, which can be used to detect the dispersive band of the photonic crystal structure itself. A bandpass filter can be provided after the second light source 12, which can filter out excess emission spectrum of the second light source 12, so that the spectrum of the light output by the second light source 12 falls within a preset range.

[0103] The light source switching module 13 may include a first total reflection mirror. Specifically, the output light directions of the first light source 11 and the second light source 12 can be set perpendicularly, and the first total reflection mirror in the light source switching module 13 can be set in the light emission direction of the first light source 11 and the second light source 12. The light source switching module 13 may include three working modes: first light source output mode, second light source output mode, and no light source output mode. When the light source switching module 13 is in the first light source output mode, the light emitted by the multi-channel light source device 1 is the light emitted by the first light source 11. When the light source switching module 13 is in the second light source output mode, the light emitted by the multi-channel light source device 1 is the light emitted by the second light source 12. When the light source switching module 13 is in the no light source output mode, the multi-channel light source device 1 has no light output.

[0104] The angle-resolved imaging device 2 includes at least a five-axis adjustment module 21, a power supply module, a switchable objective lens module 23, a switchable beam splitter module 24, and an imaging tube 25. The sample to be tested can be placed on the five-axis adjustment module 21, which is used to level or switch the angle of the sample to be tested. The power supply module is used to supply power to the sample to be tested. The light emitted from the multi-channel light source device 1 can pass through the switchable beam splitter module 24 and the switchable objective lens module 23 in sequence before being incident on the sample to be tested. The light emitted from the sample to be tested can pass through the switchable objective lens module 23, the switchable beam splitter module 24, and the imaging tube 25 in sequence before being incident on the momentum space imaging device 3.

[0105] When the multi-channel light source device 1 outputs light, the light output from the multi-channel light source device 1 can directly enter the switchable beam splitter module 24, which enables the light output from the multi-channel light source device 1 to be incident on the sample to be tested. The sample to be tested is placed on the five-axis adjustment module 21, which can level the sample. When there are multiple samples to be tested, the five-axis adjustment module 21 can perform scanning tests on multiple samples, enabling the angle-resolved system to generate an angle-resolved array diagram. The five-axis adjustment module 21 can also quickly switch 360° angles. When the sample to be tested is a sample with a photonic crystal, it is used to measure the dispersive band diagram of multiple high-symmetry points in reciprocal space to analyze the performance of the sample to be tested.

[0106] When the sample to be tested is an electric pump chip, the power supply module is used to supply power to the sample to be tested.

[0107] The sample to be tested can be a freshly etched wafer, a wafer that has undergone secondary epitaxy after etching, or a packaged chip arranged according to a specific rule.

[0108] In the angle-resolved imaging device 2, the switchable objective lens module 23 may include at least a first objective lens, a second objective lens, and a third objective lens. The first objective lens may be a low-magnification, low-numerical-aperture objective lens with a magnification of no more than 30x and a numerical aperture (NA) of less than 0.4. The second objective lens may be a low-magnification, high-numerical-aperture objective lens with a magnification of no more than 30x and a numerical aperture (NA) of no less than 0.4 and no more than 0.8. The third objective lens may be a high-magnification, high-numerical-aperture objective lens with a magnification of more than 30x and a numerical aperture (NA) of more than 0.8.

[0109] The diameter D of the image on the back focal plane of the objective lens, i.e. the image in momentum space, is related to the focal length and numerical aperture of the objective lens.

[0110] When the emission angle of the sample to be tested is less than 23.57°, using a low-magnification, low-numerical-aperture objective lens can improve the angular resolution of the angular resolution system because low magnification usually corresponds to a longer focal length of the objective lens. At the same time, the lower numerical aperture can match the target size of the camera on the spectrometer 32 in the momentum space imaging device 3, thereby improving the system utilization rate and presenting a magnified image of momentum space on the camera of the spectrometer 32.

[0111] When the emission angle of the sample to be tested is not less than 23.57° and not greater than 53°, using a small magnification, large numerical aperture objective lens can improve the angular resolution of the angular resolution system because a small magnification usually corresponds to a longer focal length of the objective lens. At the same time, a large numerical aperture can match the emission angle of the sample and also match the target size of the camera on the spectrometer 32 in the momentum space imaging device 3, thereby improving the utilization rate of the system and presenting a moderate momentum space image on the camera of the spectrometer 32.

[0112] When the emission angle of the sample to be tested is greater than 53°, when using a high-magnification, large numerical aperture objective lens, since high magnification usually corresponds to a shorter focal length of the objective lens, by sacrificing a certain angular resolution of the angular resolution system, it is possible to match the particularly large emission angle of the sample. At the same time, since the target surface of the camera on the spectrometer 32 in the momentum space imaging device 3 does not change, but the angle of the displayed momentum space image has increased, thus presenting a compressed momentum space image.

[0113] Therefore, depending on the emission angle of the sample to be tested, an objective lens with appropriate parameters can be selected in the switchable objective lens module 23 to match the emission angle of the sample to be tested, thereby improving the accuracy of the test.

[0114] In the angle-resolved imaging device 2, the switchable beam splitter module 24 may include an edge-pass dichroic mirror and a broadband beam splitter. The light output from the multi-channel light source device 1 enters the switchable beam splitter module 24, and the switchable beam splitter module 24 switches its operating mode according to the light emitted from the multi-channel light source device 1. The switchable beam splitter module 24 may include at least three operating modes: an edge-pass dichroic mirror operating mode, a broadband beam splitter operating mode, and a beam splitter-free operating mode, which correspond to the light emitted from the first light source 11 in the multi-channel light source device 1, the light emitted from the second light source 12 in the multi-channel light source device 1, and no light emitted from the multi-channel light source device 1, respectively.

[0115] When the multi-channel light source device 1 emits light, the light emitted from the multi-channel light source device 1 is split by the side-pass dichroic mirror or the broadband beam splitter, and a portion of the light enters the test sample placed on the five-axis adjustment module 21. When the test sample placed on the five-axis adjustment module 21 is an electric pump chip, and the multi-channel light source device 1 emits no light, the switchable beam splitter module 24 switches to the lensless mode, the power supply module supplies power to the electric pump chip, and the light emitted by the electric pump chip is imaged after passing through the switchable objective lens module 23 and the imaging tube lens 25.

[0116] The imaging tube 25 works in conjunction with the first, second, or third objective lens in the switchable objective lens module 23 to form an image, while simultaneously collecting the detected signal.

[0117] In an optional embodiment, the imaging tube 25 can be a single convex lens, an achromatic convex lens, or a lens group, and the focal length of the imaging tube 25 should match that of the objective lens.

[0118] The angle-resolved imaging device 2 may also include an imaging switching module 26 and a real-space imaging module 27. Light emitted from the imaging tube 25 can be directed by the imaging switching module 26 into the real-space imaging module 27 or the momentum-space imaging device 3. The real-space imaging module 27 is located on the back focal plane of the imaging tube 25 and is used to image the sample to be tested.

[0119] In an optional embodiment, the imaging switching module 26 may include a second total internal reflection mirror. The imaging switching module 26 has two modes: a total internal reflection mirror mode and a mirrorless mode. When the real space imaging module 27 is positioned in the light emission direction of the imaging tube 25, the light incident direction of the momentum space imaging device 3 can be set at 90° to the light emission direction of the imaging tube 25. When the imaging switching module 26 switches to the total internal reflection mirror mode, the light emitted from the imaging tube 25 is reflected by the second total internal reflection mirror and then enters the momentum space imaging device 3. When the imaging switching module 26 switches to the mirrorless mode, the light emitted from the imaging tube 25 enters the real space imaging module 27.

[0120] The real-space imaging module 27 may include a first filter, a first attenuator, and a real-space imaging camera. Light emitted from the imaging tube 25 passes sequentially through the first filter and the first attenuator before entering the real-space imaging camera. The first filter filters out excess pump light, ensuring that only broadband excitation light can enter the imaging camera; the first attenuator prevents overexposure or damage to the real-space imaging camera; the real-space imaging camera can image the structure of the photonic crystal or markings on the wafer. The real-space imaging module 27 is a small field-of-view imaging module, which can approximate the position of the sample under test and, by adjusting the five-axis adjustment module 21, calibrate the objective lens, bringing the objective lens in the switchable objective lens module 23 to its working focal length.

[0121] The momentum space imaging device 3 may include a second attenuator, an aperture, a second filter, an imaging lens 31, a third filter, and a spectrometer 32. The light emitted from the imaging tube 25 passes through the second attenuator, the aperture, the second filter, the imaging lens 31, and the third filter in sequence before entering the spectrometer 32.

[0122] The second attenuator can be a variable optical attenuator. The second attenuator can adjust the strength of the signal in the angle resolution system. On the one hand, when the sample to be tested is an electric pump chip, it can attenuate the laser signal generated by the sample to be tested. On the other hand, when the multi-channel light source device 1 outputs a broadband light source, it can slightly attenuate or even not attenuate the signal generated by the broadband light source.

[0123] The aperture can filter out extra stray light in the signal, while preventing the extra spectral signal reflected by the second filter from returning to the angle-resolved system and causing noise amplification.

[0124] The second filter can be a bandpass filter, which can filter out redundant spectral signals in the signal, thereby improving the signal-to-noise ratio.

[0125] The imaging lens 31 is used to transform the image in real space into momentum space. The imaging lens 31 and the imaging tube 25 should satisfy the 4f system distribution.

[0126] The third filter can be a broadband filter, thus enabling the angle-resolved system to perform full-angle dispersive bandgap testing.

[0127] Therefore, by setting a second attenuator, an aperture, and a second filter, stray light can be effectively suppressed and the quality of the detection signal can be improved.

[0128] The detector of spectrometer 32 can be an area array camera, and it also has a slit and shutter at the spectrometer entrance for imaging dispersive band diagrams.

[0129] Reference manual attached Figure 5 The angle resolution system provided in this embodiment can realize at least three different test modes: The first method is for laser-pumped dispersive bandgap imaging. The first light source 11 uses a narrow-band pump laser. The light emitted from the first light source 11 passes through a narrow-band filter to narrow the lasing spectrum. At this time, the light source switching module 13 switches to the first light source output mode. In this embodiment, the light source switching module 13 is in a lens-free state in the first light source output mode, that is, the first total reflection mirror is removed from the output light paths of the first light source 11 and the second light source 12. While ensuring the laser incident on the first light source 11 in the angle-resolved imaging device 2, no additional beam combining device is needed. This avoids the slight change in the polarization characteristics of the laser caused by the introduction of multiple dichroic mirrors, which would lead to a decrease in the signal-to-noise ratio of the angle-resolved system. The switchable beam splitter module 24 switches to the side-pass dichroic mirror working mode. The laser emitted from the multi-channel light source device 1 is reflected by the side-pass dichroic mirror into the switchable objective lens module 23, and then passes through the sample under test to generate a broad-spectrum excitation light different from the incident light. The switchable objective module 23 selects the appropriate objective according to the sample to be tested. After the objective collects the signal light, it converts the real space information into momentum space information. Then, after passing through the switchable beam splitter module 24 and the imaging tube 25 in sequence, the momentum space image is converted into a real space image. Under the action of the imaging switching module 26, the image is projected into the real space imaging module 27 or the momentum space imaging device 3.

[0130] The second method involves dispersive bandgap imaging using a broadband light source. The second light source 12 is a broadband light source. The light emitted from the second light source 12 passes through a bandpass filter to purify the light spectrum. At this time, the light source switching module 13 switches to the second light source output mode. In this embodiment, the light source switching module 13 is in mirror mode when the second light source is in output mode, meaning the first total internal reflection mirror is positioned in the light path emitted from the second light source 12 to ensure the laser incident on the laser of the second light source 12 is within the angle-resolved imaging device 2. The switchable beam splitter module 24 switches to broadband beam splitter mode. The laser emitted from the multi-channel light source device 1 is reflected by the broadband beam splitter into the switchable objective lens module 23, and then passes through the sample under test to generate a broadband excitation light different from the incident light. Since broadband imaging typically produces weak signals, the switchable objective lens module 23 generally selects a low-magnification, high-numerical-aperture objective lens to ensure angular resolution while collecting more signal light. In the angle-resolved system provided in this embodiment, the broadband light source emitted from the multi-channel light source device 1 is focused onto the sample under test by the objective lens in the switchable objective lens module 23, which can enhance the signal intensity on the surface of the sample under test. The broadband light source generates an excitation light of a different broadband spectrum than the incident light after passing through the sample under test. After the objective lens collects the signal light, it converts the real space information into momentum space information. Then, after passing through the switchable beam splitter module 24 and the imaging tube 25 in sequence, the momentum space image is converted into a real space image. Under the action of the imaging switching module 26, it is injected into the real space imaging module 27 or the momentum space imaging device 3.

[0131] The third method is for dispersive band imaging of an electric pump chip. The light source switching module 13 switches to a no-light-source output mode, causing no light to be emitted from the multi-channel light source device 1. At this time, the switchable beam splitter module 24 switches to a beam-splitter-free operating mode. The power supply module supplies power to the sample under test. The sample under test is an active device and can be in either pulse or continuous operation mode. When the current is below a threshold, the sample under test generates broadband excitation light; when the current is not below the threshold, the sample under test generates laser light. The switchable objective module 23 selects the appropriate objective lens according to the sample under test. The light emitted from the sample under test passes sequentially through the objective lens of the switchable objective module 23 and the imaging tube 25, converting the momentum space image into a real space image. Under the action of the imaging switching module 26, the image is then projected into the real space imaging module 27 or the momentum space imaging device 3.

[0132] The dispersive band diagram acquired by the spectrometer 32 can be processed by the dispersive band diagram processing device 4.

[0133] Other content that is the same as in Examples 1 and 2 will not be repeated here.

[0134] The angle-resolved laser system disclosed in this application integrates narrowband pump lasers and broadband light sources, enabling switching between different testing modes. It can be used for photonic crystal band structure testing, optical pumping testing, and electrical pumping emission testing. Utilizing an objective lens, imaging tube lens, and imaging lens to form a 4f system, it converts the real-space information of the sample under test into momentum-space information, achieving angle-resolved dispersion testing. The dispersion band structure processing device establishes a nonlinear mapping relationship between pixel coordinates and the true angle, eliminating errors generated in large-angle regions and improving the accuracy of angle-resolved testing. The dispersion band structure processing device reduces image stitching errors through image segment number detection and alignment processing, enabling the stitching and reconstruction of multi-segment band structure test data, thus improving testing accuracy. It achieves wafer-level non-destructive testing and can rapidly acquire photonic crystal band structures, providing important experimental basis for the design and optimization of photonic crystal surface-emitting lasers and improving device development efficiency.

[0135] The above description is only an optional implementation of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for processing dispersive band diagrams, characterized in that, Includes the following steps: Perform nonlinear calibration on the image to be processed; Detect the number of image stitching segments; Align the image stitching segments; The image segments are stitched together and reconstructed to obtain the overall image.

2. The method for processing dispersive band diagrams according to claim 1, characterized in that, The nonlinear calibration of the image to be processed includes: Calculate the angle of the diffracted light in real space based on the pixel coordinates; Calculate the angular space coordinates based on the angle of the diffracted light in real space.

3. The method for processing dispersive band diagrams according to claim 1, characterized in that, The number of image stitching segments includes: The wavelength step sequence is obtained by performing adjacent difference calculations on the wavelength sequence in the spectral data. The change between adjacent sampling points is obtained based on the wavelength step sequence; Based on the preset criteria for judging abnormal stitching points, determine whether the sampling point is an abnormal stitching point in the image. The number of image stitching points and the number of image stitching segments are calculated based on the number of abnormal points in the image stitching, and the wavelength of the stitching boundary is also calculated.

4. The method for processing dispersive band diagrams according to claim 3, characterized in that, The alignment of the image stitching segments includes: Extract the curve of each splicing boundary, and extract one curve to the left and one curve to the right of each splicing boundary; The two curves extracted from each splicing boundary are processed to remove DC noise; Define the cross-correlation function for the two curves corresponding to each splicing boundary; Each image stitching segment is defined as a sub-image. Based on the cross-correlation function of the two curves corresponding to each stitching boundary, the optimal translation amount between adjacent image stitching segments and the cumulative translation amount of the current image stitching segment relative to the first image stitching segment are calculated.

5. The method for processing dispersive band diagrams according to claim 4, characterized in that, The process of stitching and reconstructing the image segments includes: Calculate the number of pixels that the sub-image needs to be translated based on the optimal translation amount between adjacent image stitching segments. Based on the number of pixels that need to be translated from the sub-images, the image stitching segments are stitched together and reconstructed to obtain the overall image.

6. An apparatus for processing dispersive band diagrams, characterized in that, The method for processing the dispersive band diagram according to any one of claims 1-5 includes: The nonlinear calibration module is used to perform nonlinear calibration on the image to be processed. The sub-image quantity detection module is used to detect the number of image stitching segments, and defines each image stitching segment as a sub-image; The sub-image alignment module is used to align image stitching segments; The sub-image stitching module is used to stitch and reconstruct image segments to obtain the overall image.

7. An angular resolution system, characterized in that, include: Dispersion band diagram processing apparatus (4), wherein the dispersion band diagram processing apparatus (4) applies the dispersion band diagram processing method according to any one of claims 1-5.

8. The angle resolution system according to claim 7, characterized in that, Also includes: Multi-channel light source device (1), angle resolution imaging device (2) and momentum space imaging device (3). The multi-channel light source device (1) includes at least a first light source (11), a second light source (12) and a light source switching module (13). The light source switching module (13) is used to control the light incident angle resolution imaging device of the first light source (11) or the light incident angle resolution imaging device of the second light source (12) or the light-incident angle resolution imaging device. The angle-resolved imaging device (2) includes at least a five-axis adjustment module (21), a power supply module, a switchable objective lens module (23), a switchable beam splitter module (24), and an imaging tube (25). The sample to be tested can be placed on the five-axis adjustment module (21). The five-axis adjustment module (21) is used to level or switch the angle of the sample to be tested. The power supply module is used to supply power to the sample to be tested. The light emitted from the multi-channel light source device (1) can pass through the switchable beam splitter module (24) and the switchable objective lens module (23) in sequence and then be incident on the sample to be tested. The light emitted from the sample to be tested can pass through the switchable objective lens module (23), the switchable beam splitter module (24), and the imaging tube (25) in sequence and be incident on the momentum space imaging device (3). The momentum space imaging device (3) includes at least an imaging lens (31) and a spectrometer (32). The light emitted from the imaging tube (25) can be collected by the spectrometer (32) after passing through the imaging lens (31). The spectrometer (32) can output a dispersive band structure. The dispersive band diagram processing device (4) processes the dispersive band diagram.

9. The angle resolution system according to claim 8, characterized in that, The angle-resolved imaging device (2) further includes a real space imaging module (27), which is disposed on the back focal plane of the imaging tube (25) and is used to image the sample to be tested.

10. The angle-resolved system according to claim 9, characterized in that, The angle-resolved imaging device (2) further includes an imaging switching module (26), which is located between the imaging tube (25) and the real space imaging module (27). The imaging switching module (26) is used to control the light emitted from the imaging tube (25) to enter the real space imaging module (27) or the momentum space imaging device (3).

Images