A s-snom sub-surface defect fine imaging method and system based on polarization laser scattering guidance

Abstract

The application discloses a s-SNOM sub-surface defect fine imaging method and system based on polarization laser scattering guidance, belongs to the technical field of optical nondestructive testing, and is realized based on a laser, a polarizer, a polarization modulator, a to-be-detected sample, a light splitter array, a polarimeter array and a detector which are sequentially arranged along a laser propagation direction, and the to-be-detected sample contains a sub-surface defect. Firstly, polarization scattering fast detection is carried out on the to-be-detected sample to obtain polarization scattering data; secondly, quantitative analysis and risk assessment are carried out on sub-surface structure non-uniformity to obtain a region of interest; thirdly, region of interest extraction and detection task scheduling are carried out on a high-risk region; and finally, high-resolution near-field scanning imaging is carried out on the region of interest. On the basis of maintaining the high-resolution imaging advantage of the s-SNOM, the prior guidance mechanism is introduced, cross-scale cooperation between polarization scattering information and near-field imaging can be realized, and efficiency and precision are taken into account in large-area sample detection.

Description

Technical Field

[0001] This invention belongs to the technical field of optical nondestructive testing, and relates to a fine imaging method and system for s-SNOM subsurface defects based on polarized laser scattering guidance. Background Technology

[0002] With the rapid development of high-power laser systems, precision optical components, and semiconductor devices, the requirements for the integrity of the internal structure and the stability of optical performance of materials are becoming increasingly stringent. Especially in hard and brittle materials such as fused silica, optical glass, crystalline materials, and semiconductor substrates, subsurface regions often introduce defects such as microcracks, densification layers, pores / gases, and inclusions during processing, polishing, or service. Although these subsurface defects are small in scale, they significantly alter the local refractive index, scattering characteristics, and energy deposition behavior of the material, thereby reducing the laser damage threshold (LIDT), optical uniformity, and long-term reliability of the device. This has become a crucial factor restricting the manufacturing and application of high-performance optoelectronic devices.

[0003] Near-field scanning optical microscopy (s-SNOM) has attracted widespread attention for its ability to overcome the diffraction limit and achieve nanometer-level spatial resolution, addressing the need for detection and characterization of subsurface defects. This technique enables fine imaging of the optical response of material surfaces and near-surface regions, providing a powerful tool for high-resolution characterization of subsurface microstructures. However, s-SNOM typically employs a point-by-point scanning method, resulting in a limited effective imaging area per scan and low scanning efficiency. Furthermore, in the absence of prior information, it often requires blind scanning of large-area samples, making it difficult to balance detection efficiency with engineering practicality.

[0004] On the other hand, polarized laser scattering (PLS) technology, with its sensitive response to anisotropy, stress concentration, and microstructural disturbances within materials, can rapidly acquire scattering characteristic information of subsurface defects over a large scale, offering advantages such as high detection speed and large coverage area. However, this type of method mainly relies on far-field scattering signals, which limits spatial resolution and makes it difficult to accurately distinguish the fine morphology, scale, and type of defects. It is usually only suitable for qualitative or semi-quantitative assessment and cannot meet the needs for fine imaging and quantitative analysis of subsurface defects.

[0005] In existing technologies, near-field scanning optical microscopy and polarized laser scattering methods are mostly used independently, and an effective cross-scale collaborative detection mechanism has not yet been formed. On the one hand, there is a lack of technical solutions to unify and correlate large-scale scattering information with nanoscale near-field imaging results; on the other hand, the coordinate systems of different detection modules are independent, making it difficult to achieve accurate mapping and automated directional scanning of defect areas. In addition, existing detection processes generally rely on manual experience to select scanning areas, lacking adaptive scanning strategies guided by physical features, making it difficult to significantly improve detection efficiency while ensuring imaging accuracy.

[0006] Chinese invention patent CN119619553A discloses "A Method for Detecting Densification of Hard and Brittle Materials Based on Near-Field Optical Amplitude and Phase Analysis." The inventors utilize a scanning near-field optical microscopy system to acquire near-field amplitude or phase information of the sample surface and establish a correspondence between the near-field optical response and the material densification rate, thereby achieving high-precision, non-destructive testing of densification levels at different locations. However, this method still primarily relies on the point-by-point scanning imaging mechanism of the near-field scanning optical microscopy system, resulting in a limited single detection area. The overall detection efficiency is significantly constrained by the scanning speed and range. When dealing with large-size optical components or large-area hard and brittle material samples, long-duration, large-area area-by-area scanning is still required, which is insufficient to meet the demands of rapid screening and high-throughput detection in engineering applications. Furthermore, this method, based on the direct mapping relationship between near-field optical signals and densification rate, lacks methods for predicting the macroscopic distribution characteristics of defects. In actual testing, it often requires manual experience to select the scanning area, leading to certain blind scanning problems.

[0007] Therefore, there is an urgent need to develop a fine imaging method and system for subsurface defects based on polarized laser scattering-guided s-SNOM. While maintaining the advantages of high-resolution s-SNOM imaging, a fast and effective prior guidance mechanism should be introduced to achieve cross-scale synergy between polarized scattering information and near-field imaging, and a unified coordinate mapping and automated directional scanning process should be constructed to balance efficiency and accuracy in large-area sample detection. Summary of the Invention

[0008] To address key issues such as low detection efficiency, insufficient scale range, and lack of prior guidance mechanisms in existing methods, this invention aims to propose a fine imaging method and system for subsurface defects based on polarized laser scattering guidance (s-SNOM). While maintaining the high-resolution imaging advantages of s-SNOM, this invention introduces a fast and effective prior guidance mechanism to achieve cross-scale synergy between polarized scattering information and near-field imaging, constructing a unified coordinate mapping and automated directional scanning process, thereby balancing efficiency and accuracy in large-area sample detection.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0010] A fine imaging method for subsurface defects based on polarized laser scattering (s-SNOM) is disclosed. This method comprises a laser, a polarizer, a polarization modulator, a sample under test, a beam splitter array, an analyzer array, and a detector, arranged sequentially along the laser propagation direction. The sample under test contains subsurface defects, which cause changes in the polarization characteristics of the scattered light under polarized laser illumination. The s-SNOM fine imaging method includes the following steps:

[0011] The first step involves constructing a physical detection layer using the s-SNOM subsurface defect fine imaging system to rapidly detect polarization scattering of the sample under test and obtain polarization scattering data.

[0012] The second step involves constructing a polarization feature and risk assessment layer based on the polarization scattering dataset obtained in the first step. This layer performs quantitative analysis and risk assessment of the subsurface structural inhomogeneity of the sample under test, thereby obtaining the region of interest. The polarization feature and risk assessment layer includes a polarization parameter calculation unit, a spatial reconstruction unit, a risk map generation unit, and a parameter configuration and threshold management unit.

[0013] The third step involves constructing a decision-making and scheduling layer based on the two-dimensional subsurface defect risk map generated in the second step. This layer extracts regions of interest (ROIs) and schedules detection tasks for high-risk areas. The decision-making and scheduling layer includes an ROI extraction unit and an ROI screening and priority evaluation unit.

[0014] The fourth step is to construct a cross-system coordinate mapping and control layer to achieve coordinate unification and scan control between the polarized laser scattering detection system and the s-SNOM imaging system; the cross-system coordinate mapping and control layer includes a reference marker recognition unit, a coordinate transformation calculation unit, and a scan command generation unit;

[0015] The fifth step is to construct an s-SNOM fine imaging layer to perform high-resolution near-field scanning imaging on the region of interest determined in the third step. The s-SNOM fine imaging layer includes a probe and near-field excitation unit, an adaptive scanning control unit, and a near-field signal acquisition unit.

[0016] Furthermore, in the first step: the laser, polarizer, and polarization modulator constitute a laser and polarization modulation unit, used to generate a laser with controllable polarization state and modulate the polarization state of the incident light. The controllable polarization laser irradiates the surface of the sample under test. The sample under test is placed in the sample and scanning unit, and under the drive of the scanning mechanism, a scanning detection of the sample surface is achieved. During the scanning process, the sample under test generates a scattered light signal to the incident laser, which is received by the scattered signal acquisition unit. The scattered light after beam splitting is analyzed and synchronously acquired in different polarization directions, and the scattering response information under each polarization component is extracted and correlated with the scanning position information to obtain polarization scattering data that simultaneously contains spatial position information and polarization information, forming a polarization scattering dataset for quickly identifying subsurface anomalies of the sample under test. The polarization scattering data includes scattering response information under different polarization states, used to reflect the perturbation characteristics of the subsurface structure of the sample under test to the polarization state of the incident light.

[0017] Furthermore, in the second step:

[0018] The polarization scattering data collected at different spatial locations is analyzed by the polarization parameter calculation unit. By normalizing, differentiating, ratioing, or statistically combining the scattered light intensities of different polarization components, polarization characteristic parameters used to characterize the inhomogeneity of the subsurface structure are obtained. The polarization characteristic parameters include, but are not limited to, depolarization rate, degree of polarization, anisotropy parameters, or combinations thereof.

[0019] The spatial reconstruction unit maps the polarization feature parameters to the spatial position information recorded during the scanning process. Based on the scanning path and sampling order, the polarization feature parameters are reconstructed in the two-dimensional plane to form a polarization feature distribution matrix that is consistent with the coordinate system of the sample surface.

[0020] Based on the polarization feature distribution matrix, the risk map generation unit combines the preset or adaptively updated risk criteria in the parameter configuration and threshold management unit to perform hierarchical or mapping processing on the polarization feature parameters at each spatial location, thereby generating a two-dimensional subsurface defect risk map that reflects the degree of potential subsurface defect risk in different regions of the sample surface.

[0021] Ultimately, this enables rapid coarse-screening risk assessment of large areas of the sample to be tested, which is used to identify high-risk areas that may have subsurface defects and to provide prior guidance for subsequent high-resolution fine detection.

[0022] Furthermore, the third step specifically includes:

[0023] The two-dimensional subsurface defect risk map is analyzed by the ROI extraction unit. Based on the risk value distribution, continuous or discrete regions with risk values ​​higher than a preset threshold are identified and extracted as one or more regions of interest. The risk value is determined by a comprehensive evaluation index calculated based on polarization feature parameters. The determination method includes: normalizing different polarization feature parameters and performing weighted fusion according to preset weights or adaptive weights to obtain a risk score characterizing the degree of subsurface structural anomaly. The risk level information is divided according to the risk score and a preset or dynamically adjusted risk level interval to form a graded risk criterion for guiding detection decisions.

[0024] The ROI screening and priority evaluation unit combines the spatial location, size, and risk level information of the region of interest to schedule and sort subsequent high-resolution detection tasks, generating a detection priority list. This enables centralized allocation of detection resources and avoids ineffective fine scanning of low-risk areas.

[0025] Furthermore, the fourth step specifically includes:

[0026] One or more reference marks are set on the surface of the sample to be tested or on the sample stage to establish the spatial mapping relationship between the polarized laser scattering detection coordinate system, the sample stage coordinate system and the s-SNOM imaging coordinate system; in the fourth step, the reference marks are set according to the geometric features of the sample to be tested and the detection requirements; preferably, they are arranged in a position that does not affect the effective detection area of ​​the sample to be tested, and can be stably identified in both polarized laser scattering detection and s-SNOM imaging.

[0027] The reference marker identification unit acquires the position information of the reference markers under different detection systems; the coordinate transformation calculation unit calculates the transformation parameters between different coordinate systems based on the correspondence of the reference markers in each coordinate system; the scanning command generation unit converts the spatial position information of the region of interest obtained in the third step into scanning coordinates that can be recognized by the s-SNOM imaging system according to the coordinate transformation relationship, and generates the corresponding scanning command; ensuring accurate connection between rapid coarse screening and fine imaging.

[0028] Furthermore, the fifth step specifically includes:

[0029] Based on the spatial location information of the region of interest (ROI) and its corresponding risk level obtained in the third step, s-SNOM scanning imaging is performed only on the ROI. The probe and near-field excitation unit are used to generate local near-field excitation on the surface of the sample and interact with the sample in the near field. The near-field signal acquisition unit is used to acquire the near-field amplitude signal and near-field phase signal formed by the probe scattering. The scanning parameters are dynamically adjusted by the adaptive scanning control unit according to the risk level of the ROI. Finally, the near-field amplitude image and near-field phase image of the corresponding ROI are obtained for fine characterization and analysis of subsurface defects.

[0030] The scanning parameters include scanning step size, scanning density, or scanning path. The dynamic adjustment involves using a small scanning step size or high scanning density when the risk level of the region of interest is high, achieving fine scanning of that region; and using a relatively large scanning step size or low scanning density when the risk level of the region of interest is low, achieving rapid scanning. Through this adaptive scanning method, while ensuring imaging resolution in high-risk areas, the scanning time consumption in low-risk areas is reduced, further improving overall detection efficiency. Finally, near-field amplitude and near-field phase images of the corresponding region of interest are obtained, used for fine characterization and analysis of subsurface defect 5.

[0031] A fine imaging system for subsurface defects based on polarized laser scattering (s-SNOM) is disclosed. This system enables fine imaging of s-SNOM subsurface defects by combining large-area, high-speed polarized laser scattering detection with high spatial resolution scanning near-field optical microscopy (s-SNOM) imaging. This constructs a cross-scale collaborative detection system of "coarse screening and positioning—directional fine scanning," comprising a laser, a polarizer, a polarization modulator, a sample to be tested, a beam splitter array, an analyzer array, and a detector. The laser, polarizer, polarization modulator, sample to be tested, beam splitter array, analyzer array, and detector are sequentially arranged along the laser propagation direction, forming the optical path structure for polarized laser scattering detection. Specifically:

[0032] The polarizer is positioned between the laser and the polarization modulator to define the polarization state of the laser output from the laser. The polarization modulator is positioned between the polarizer and the sample to be tested, and its angle is adjustable. During use, it needs to be adjusted to the corresponding size to modulate the polarization state of the laser incident on the sample to be tested.

[0033] The sample to be tested is placed downstream of the polarization modulator, and its interior or near-surface region contains subsurface defects. These subsurface defects cause changes in the polarization characteristics of the scattered light under polarization laser irradiation.

[0034] The polarization modulator is angle-adjustable and is used to modulate the polarization state of the laser incident on the sample under test, wherein the sample under test contains subsurface defects in its interior or near-surface region.

[0035] The beam splitter array is positioned on the propagation path of the scattered light from the sample under test, and is used to split the scattered light from the sample under test to form multiple scattered light signals.

[0036] The polarizer array is located downstream of the beam splitter array and includes multiple polarizers, each of which corresponds to a different polarization analysis direction. It is used to extract the scattered light component in the corresponding polarization direction and obtain multiple sets of polarization-related scattering data through the detector, so as to finally realize the location and characterization of subsurface defects or densification features.

[0037] The detector is positioned downstream of the analyzer array to receive the scattered light signals analyzed by each analyzer, obtaining multiple sets of polarization-correlated scattering data. Based on the relative changes in the intensity of different polarization components, the detector performs polarization response analysis on the scattered light components, constructing polarization response parameters that reflect the local optical anisotropy and scattering characteristics of the material. By analyzing the spatial distribution characteristics of the polarization response parameters, the detector can locate and characterize subsurface defects or densification features, providing prior guidance for subsequent high-resolution near-field scanning.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] (1) This invention introduces rapid polarized laser scattering detection as a priori guiding method, using polarized scattering signals to quickly coarsely screen a large area of ​​the sample under test, constructing a two-dimensional subsurface defect risk map, and automatically extracting the region of interest based on the risk assessment results, performing s-SNOM fine imaging only on areas with risk values ​​higher than the threshold. Through the detection process of "coarse screening and positioning - directional fine scanning", the blind near-field scanning of low-risk areas is effectively avoided, fundamentally reducing the number of scanning points and scanning time, and achieving a significant improvement in detection efficiency, which is especially suitable for large-size samples or large-area detection scenarios.

[0040] (2) This invention organically integrates millimeter- to centimeter-scale polarized laser scattering detection with nanoscale s-SNOM near-field imaging to construct a unified cross-scale detection framework. On the one hand, polarized laser scattering technology can sensitively reflect the macroscopic distribution characteristics of material subsurface structure inhomogeneity, anisotropy, and stress disturbance, which can be used to quickly locate potential defect areas. On the other hand, s-SNOM technology can perform high-resolution fine imaging of the local optical response of subsurface defects by acquiring near-field amplitude and near-field phase information. By establishing the mapping relationship between the polarized scattering coordinate system, the sample stage coordinate system, and the s-SNOM imaging coordinate system, this invention achieves a precise spatial correspondence between detection results at different scales, enabling macroscopic risk assessment results to directly guide nanoscale near-field imaging. This allows for the simultaneous acquisition of spatial distribution information and fine structural features of defects in the same detection process, significantly improving the completeness and accuracy of subsurface defect characterization.

[0041] (3) This invention achieves full automation of the entire process from risk assessment, region of interest extraction, detection task scheduling to near-field scanning parameter adjustment by constructing a polarization feature and risk assessment layer, a decision and scheduling layer, and an adaptive scanning control mechanism. During the detection process, there is no need to rely on manual experience to select the scanning area. The extraction of the region of interest, priority ranking, and scanning strategy are all automatically completed by the system based on the polarization scattering risk assessment results. At the same time, the polarization scattering parameters and s-SNOM near-field response used in this invention are based on the universal physical response mechanism of materials to optical excitation. Therefore, it is not only applicable to hard and brittle optical materials such as fused silica and optical glass, but also to semiconductor materials and other material systems with subsurface structural inhomogeneities. It can effectively detect and finely image various subsurface defect types such as microcracks, densification layers, pores, and inclusions, and has good versatility and engineering application prospects. Attached Figure Description

[0042] Figure 1 This is a flowchart of the subsurface defect fine imaging method of the present invention;

[0043] Figure 2 This is a schematic diagram of the polarization laser scattering detection and data acquisition method of the present invention;

[0044] In the figure: 1 Laser; 2 Polarizer; 3 Polarization modulator; 4 Sample under test; 5 Subsurface defect; 6 Beam splitter array; 7 Analyzer array; 8 Detector. Detailed Implementation

[0045] The invention will now be further described with reference to the accompanying drawings. Figure 1 As shown, in order to make the technical means, creative features, objectives and effects of the present invention easier to understand, the present invention will be further described below in conjunction with specific embodiments.

[0046] This embodiment provides a fine imaging system for subsurface defects based on polarized laser scattering-guided s-SNOM, including a laser 1, a polarizer 2, a polarization modulator 3, a sample under test 4, a subsurface defect 5, a beam splitter array 6, an analyzer array 7, and a detector 8. The laser 1, polarizer 2, polarization modulator 3, sample under test 4, beam splitter array 6, analyzer array 7, and detector 8 are arranged sequentially along the laser propagation direction, forming the optical path structure for polarized laser scattering detection. Specifically:

[0047] In this embodiment, the polarizer 2 is disposed between the laser 1 and the polarization modulator 3. The polarizer 2 is preferably a Glan-Taylor prism type polarizer with an extinction ratio of not less than 10. 5 :1, used to define the polarization state of the laser output by laser 1; the polarization modulator 3 is disposed between polarizer 2 and sample 4 under test, and is an electrically rotating waveplate structure with a rotation angle range of 0°–180°. The angle adjustment accuracy is preferably 0.1°. It needs to be adjusted to the corresponding size during use to modulate the polarization state of the laser incident on sample 4 under test.

[0048] In this embodiment, the sample 4 to be tested is a single-crystal silicon wafer or an optical glass substrate with a thickness of 0.5–1 mm and a size of 100 mm × 100 mm. It is located downstream of the polarization modulator 3, and its internal or near-surface region contains subsurface defects 5. The depth range of the subsurface defects 5 is preferably 10 nm–5 μm, which cause changes in the polarization characteristics of the scattered light under polarized laser irradiation.

[0049] In this embodiment, the beam splitter array 6 is set on the propagation path of the scattered light from the sample 4 under test. Specifically, it is set into 8 groups, arranged in a ring array with a ring diameter of 80–120 mm. The array is symmetrically distributed around the normal axis of the sample, and the angular interval between adjacent beam splitters is 45°. It is used to split the scattered light from the sample 4 under test to form multiple scattered light signals.

[0050] In this embodiment, the polarizer array 7 is located downstream of the beam splitter array 6, specifically in 8 groups. The polarization analysis directions of the polarizers are set to 0°, 22.5°, 45°, 67.5°, 90°, 112.5°, 135°, and 157.5°, respectively. Each polarizer corresponds to a different polarization analysis direction and is used to extract the scattered light component in the corresponding polarization direction.

[0051] In this embodiment, the detector 8 is positioned downstream of the analyzer array 7, with a response band covering 400–800 nm. It is used to receive the scattered light signals analyzed by each analyzer, obtaining multiple sets of polarization-correlated scattering data. Based on the relative changes in the intensity of different polarization components, the scattered light is analyzed for polarization response, constructing polarization response parameters that reflect the local optical anisotropy and scattering characteristics of the material. Since subsurface defects 5 or densification regions cause significant changes in the polarization characteristics of the scattered light, by analyzing the spatial distribution characteristics of the polarization response parameters, the location and characterization of subsurface defects 5 or densification features can be achieved, providing prior guidance for subsequent high-resolution near-field scanning.

[0052] A fine imaging method for subsurface defects based on polarized laser scattering-guided s-SNOM, implemented using the aforementioned system, includes the following steps:

[0053] The first step involves constructing a physical detection layer using the s-SNOM subsurface defect fine imaging system to rapidly detect polarization scattering of the sample under test (4), obtaining polarization scattering data. Specifically, laser 1, polarizer 2, and polarization modulator 3 constitute a laser and polarization modulation unit, used to generate a laser with controllable polarization and modulate the polarization state of the incident light. This controllable polarization laser illuminates the surface of the sample under test (4). The sample under test (4) is placed in the sample and scanning unit, and rapid scanning detection of its surface is achieved under the drive of the scanning mechanism. During the scanning process, the sample under test (4) generates scattered light signals to the incident laser. The scattered light is received by the scattering signal acquisition unit, which performs polarization analysis and synchronous acquisition on the split scattered light in different polarization directions, extracting scattering response information under each polarization component and correlating it with the scanning position information to obtain polarization scattering data that simultaneously contains spatial position and polarization information. This forms a polarization scattering dataset for rapidly identifying subsurface anomalies in the sample under test (4). The polarization scattering data includes scattering response information under different polarization states, reflecting the perturbation characteristics of the subsurface structure of the sample under test (4) to the polarization state of the incident light.

[0054] The second step, based on the polarization scattering dataset obtained in the first step, constructs a polarization feature and risk assessment layer to quantitatively analyze and assess the inhomogeneity of the four subsurface structures of the sample under test, thereby obtaining the region of interest. The polarization feature and risk assessment layer includes a polarization parameter calculation unit, a spatial reconstruction unit, a risk map generation unit, and a parameter configuration and threshold management unit. Specifically:

[0055] The polarization parameter calculation unit analyzes the multi-polarization scattering response data collected at different spatial locations. By normalizing, differentiating, ratioing, or statistically combining the scattered light intensities of different polarization components, polarization characteristic parameters used to characterize the inhomogeneity of the subsurface structure are obtained. These polarization characteristic parameters include, but are not limited to, depolarization rate, degree of polarization, anisotropy parameters, or combinations thereof.

[0056] The spatial reconstruction unit maps the polarization feature parameters to the spatial position information recorded during the scanning process. Based on the scanning path and sampling order, the polarization feature parameters are reconstructed in a two-dimensional plane to form a polarization feature distribution matrix that is consistent with the coordinate system of the surface of the sample 4 under test.

[0057] The risk map generation unit, based on the polarization feature distribution matrix and combined with the preset or adaptively updated risk criteria in the parameter configuration and threshold management unit, performs hierarchical or mapping processing on the polarization feature parameters at each spatial location to generate a two-dimensional subsurface defect risk map reflecting the risk level of potential subsurface defects 5 in different regions of the surface of the sample 4 under test. In this way, a rapid, coarse-screening risk assessment of a large area of ​​the sample 4 under test is achieved, identifying high-risk areas where subsurface defects 5 may exist, and providing prior guidance for subsequent high-resolution fine detection.

[0058] The third step involves constructing a decision-making and scheduling layer based on the two-dimensional subsurface defect risk map generated in the second step. This layer extracts regions of interest (ROIs) and schedules detection tasks for high-risk areas. The decision-making and scheduling layer includes an ROI extraction unit and an ROI screening and priority evaluation unit. Specifically:

[0059] The ROI extraction unit analyzes the two-dimensional subsurface defect risk map, identifies continuous or discrete regions with risk values ​​exceeding a preset threshold based on the risk value distribution, and extracts them as one or more regions of interest (ROIs). The ROI screening and priority evaluation unit combines the spatial location, region size, and risk level information of the ROIs to schedule and prioritize subsequent high-resolution detection tasks, generating a detection priority list. This achieves centralized allocation of detection resources and avoids ineffective fine scanning of low-risk areas.

[0060] The risk value is a comprehensive evaluation index calculated based on polarization characteristic parameters. Its determination method includes: normalizing different polarization characteristic parameters and weighting and fusing them according to preset weights or adaptive weights to obtain a risk score characterizing the degree of subsurface structure anomaly; the risk level information is divided according to the risk score and the preset or dynamically adjusted risk level range to form a graded risk criterion for guiding detection decisions.

[0061] The fourth step involves constructing a cross-system coordinate mapping and control layer to achieve coordinate unification and scan control between the polarized laser scattering detection system and the s-SNOM imaging system. This cross-system coordinate mapping and control layer includes a reference marker recognition unit, a coordinate transformation calculation unit, and a scan command generation unit. Specifically:

[0062] One or more reference markers are set on the surface of the sample 4 or on its stage to establish a spatial mapping relationship between the polarized laser scattering detection coordinate system, the sample 4 stage coordinate system, and the s-SNOM imaging coordinate system. The reference markers are positioned so as not to affect the effective detection area of ​​the sample 4 and can be stably identified during both polarized laser scattering detection and s-SNOM imaging. A reference marker identification unit acquires the position information of the reference markers under different detection systems; a coordinate transformation calculation unit calculates the transformation parameters between different coordinate systems based on the correspondence of the reference markers in each coordinate system; and a scanning command generation unit converts the spatial position information of the region of interest obtained in the third step into scanning coordinates recognizable by the s-SNOM imaging system according to the coordinate transformation relationship and generates corresponding scanning commands. This step ensures precise connection between rapid coarse screening and fine imaging.

[0063] The fifth step involves constructing an s-SNOM fine imaging layer to perform high-resolution near-field scanning imaging of the region of interest identified in the third step. The s-SNOM fine imaging layer includes a probe and near-field excitation unit, an adaptive scanning control unit, and a near-field signal acquisition unit. Specifically:

[0064] Based on the spatial location information of the region of interest (ROI) and its corresponding risk level obtained in the third step, s-SNOM scanning imaging is performed only on the ROI. The probe and near-field excitation unit generate local near-field excitation on the surface of the sample 4 and interact with it in the near field; the near-field signal acquisition unit acquires the near-field amplitude and phase signals scattered by the probe. The adaptive scanning control unit dynamically adjusts the scanning parameters, including the scanning step size, scanning density, or scanning path, according to the risk level of the ROI. When the risk level of the ROI is high, a smaller scanning step size or a higher scanning density is used to achieve fine scanning of the region; when the risk level of the ROI is low, a relatively larger scanning step size or a lower scanning density is used to achieve rapid scanning. Through this adaptive scanning method, while ensuring the imaging resolution of high-risk areas, the scanning time consumption of low-risk areas is reduced, improving the overall detection efficiency. Finally, the near-field amplitude and near-field phase images of the corresponding ROI are obtained for fine characterization and analysis of the subsurface defect 5.

[0065] As described above, those skilled in the art can make various other corresponding changes and modifications based on the technical solutions and concepts of this invention, and all such changes and modifications should fall within the protection scope of the claims of this invention.

Claims

1. A fine imaging method for subsurface defects based on polarized laser scattering-guided s-SNOM, characterized in that, The s-SNOM subsurface defect fine imaging method is based on an s-SNOM subsurface defect fine imaging system, which includes a laser (1), a polarizer (2), a polarization modulator (3), a sample to be tested (4), a beam splitter array (6), an analyzer array (7), and a detector (8) arranged sequentially along the laser propagation direction. The sample to be tested (4) contains subsurface defects (5), and the subsurface defects (5) cause changes in the polarization characteristics of the scattered light under polarized laser illumination. The s-SNOM subsurface defect fine imaging method includes the following steps: The first step is to construct a physical detection layer using the s-SNOM subsurface defect fine imaging system, and to perform rapid polarization scattering detection on the sample to be tested (4) to obtain polarization scattering data; The second step involves constructing a polarization feature and risk assessment layer based on the polarization scattering dataset obtained in the first step. This layer is used to perform quantitative analysis and risk assessment of the subsurface structural inhomogeneity of the sample (4) to obtain the region of interest. The polarization feature and risk assessment layer includes a polarization parameter calculation unit, a spatial reconstruction unit, a risk map generation unit, and a parameter configuration and threshold management unit. The third step involves constructing a decision-making and scheduling layer based on the two-dimensional subsurface defect risk map generated in the second step. This layer extracts regions of interest (ROIs) and schedules detection tasks for high-risk areas. The decision-making and scheduling layer includes an ROI extraction unit and an ROI screening and priority evaluation unit. The fourth step is to construct a cross-system coordinate mapping and control layer to achieve coordinate unification and scan control between the polarized laser scattering detection system and the s-SNOM imaging system; the cross-system coordinate mapping and control layer includes a reference marker recognition unit, a coordinate transformation calculation unit, and a scan command generation unit; The fifth step is to construct an s-SNOM fine imaging layer to perform high-resolution near-field scanning imaging on the region of interest determined in the third step. The s-SNOM fine imaging layer includes a probe and near-field excitation unit, an adaptive scanning control unit, and a near-field signal acquisition unit.

2. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 1, characterized in that, The first step is specifically: The laser (1), polarizer (2) and polarization modulator (3) constitute a laser and polarization modulation unit, which is used to generate a laser with controllable polarization state and to modulate the polarization state of the incident light. The controllable polarization laser irradiates the surface of the sample to be tested (4). Under the drive of the scanning mechanism, the surface of the sample to be tested (4) is scanned. During the scanning process, the sample to be tested (4) generates a scattered light signal to the incident laser. The scattered light is received by the scattering signal acquisition unit. The scattered light after beam splitting is analyzed and synchronously acquired in different polarization directions. The scattering response information under each polarization component is extracted and matched with the scanning position information to obtain polarization scattering data that simultaneously contains spatial position information and polarization information. A polarization scattering dataset is formed for quickly identifying the subsurface abnormal features of the sample to be tested (4). The polarization scattering data includes scattering response information under different polarization states, which is used to reflect the perturbation characteristics of the subsurface structure of the sample under test (4) on the polarization state of the incident light.

3. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 2, characterized in that, The second step is specific to: The polarization scattering data collected at different spatial locations is analyzed by the polarization parameter calculation unit. By normalizing, differentiating, ratioing or statistically combining the scattered light intensities of different polarization components, polarization characteristic parameters used to characterize the inhomogeneity of subsurface structures are obtained. The polarization feature parameters are mapped one-to-one with the spatial position information recorded during the scanning process by the spatial reconstruction unit. The polarization feature parameters are reconstructed in the two-dimensional plane according to the scanning path and sampling order to form a polarization feature distribution matrix consistent with the coordinate system of the surface of the sample (4) to be tested. The risk map generation unit is based on the polarization feature distribution matrix and combines the preset or adaptively updated risk criteria in the parameter configuration and threshold management unit to classify or map the polarization feature parameters at each spatial location, and generate a two-dimensional subsurface defect risk map that reflects the risk level of potential subsurface defects (5) in different regions of the surface of the sample to be tested (4). Ultimately, a rapid coarse-screening risk assessment of a large area of ​​the sample to be tested (4) is achieved, which is used to identify high-risk areas where subsurface defects (5) may exist, and to provide prior guidance for subsequent high-resolution fine detection.

4. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 3, characterized in that, The polarization characteristic parameters in the second step include depolarization rate, degree of polarization, anisotropy parameters, or combinations thereof.

5. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 3, characterized in that, The third step specifically includes: The two-dimensional subsurface defect risk map is analyzed by the ROI extraction unit. Based on the risk value distribution, continuous or discrete regions with risk values ​​higher than the preset threshold are identified and extracted as one or more regions of interest. The ROI screening and priority evaluation unit is used to combine the spatial location, size and risk level information of the region of interest to schedule and sort subsequent high-resolution detection tasks, generate a detection priority list, and realize the centralized allocation of detection resources.

6. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 5, characterized in that, The risk value in the third step is determined by a comprehensive evaluation index calculated based on polarization feature parameters. The determination method includes: normalizing different polarization feature parameters and weighting and fusing them according to preset weights or adaptive weights to obtain a risk score characterizing the degree of subsurface structure anomaly; the risk level information is divided according to the risk score and the preset or dynamically adjusted risk level interval to form a graded risk criterion for guiding detection decisions.

7. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 5, characterized in that, The fourth step specifically includes: One or more reference marks are set on the surface of the sample to be tested (4) or on the stage of the sample to be tested (4) to establish the spatial mapping relationship between the polarization laser scattering detection coordinate system, the stage coordinate system of the sample to be tested (4) and the s-SNOM imaging coordinate system; the reference marks in the fourth step are set according to the geometric features of the sample to be tested (4) and the detection requirements; The reference marker identification unit acquires the position information of the reference marker under different detection systems; The coordinate transformation calculation unit calculates the transformation parameters between different coordinate systems based on the correspondence of the reference markers in each coordinate system; the scanning command generation unit converts the spatial location information of the region of interest obtained in the third step into scanning coordinates that can be recognized by the s-SNOM imaging system according to the coordinate transformation relationship, and generates the corresponding scanning command; ensuring accurate connection between rapid coarse screening and fine imaging.

8. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 6, characterized in that, The fifth step specifically includes: Based on the spatial location information of the region of interest and its corresponding risk level obtained in the third step, s-SNOM scanning imaging is performed only on the region of interest; the probe and the near-field excitation unit are used to generate local near-field excitation on the surface of the sample to be tested (4) and interact with the sample to be tested (4) in the near field; The near-field signal acquisition unit is used to acquire the near-field amplitude signal and near-field phase signal formed by probe scattering; the adaptive scanning control unit dynamically adjusts the scanning parameters according to the risk level of the region of interest; and finally obtains the near-field amplitude image and near-field phase image of the corresponding region of interest, which are used for fine characterization and analysis of subsurface defects (5).

9. The fine imaging method for subsurface defects based on polarized laser scattering guided by s-SNOM as described in claim 6, characterized in that, In the fifth step, the scanning parameters include scanning step size, scanning density, or scanning path; the dynamic adjustment is as follows: when the risk level of the region of interest is high, a small scanning step size or a high scanning density is used to achieve fine scanning of the region; when the risk level of the region of interest is low, a relatively large scanning step size or a low scanning density is used to achieve fast scanning.

10. A fine imaging system for subsurface defects based on polarized laser scattering-guided s-SNOM, characterized in that, The s-SNOM subsurface defect fine imaging system described in any one of claims 1-9 implements the s-SNOM subsurface defect fine imaging method based on polarized laser scattering guidance, comprising a laser (1), a polarizer (2), a polarization modulator (3), a sample to be tested (4), a beam splitter array (6), a polarizer array (7), and a detector (8). The polarizer (2) defines the polarization state of the laser output from the laser (1); The polarization modulator (3) is angle-adjustable and is used to modulate the polarization state of the laser incident on the sample (4) to be tested, wherein the sample (4) to be tested contains subsurface defects (5) inside or near the surface. The beam splitter array (6) is positioned on the propagation path of the scattered light from the sample (4) to split the scattered light from the sample (4) into multiple scattered light signals. Each analyzer in the analyzer array (7) corresponds to a different polarization analysis direction, used to extract the scattered light component in the corresponding polarization direction, and obtain multiple sets of polarization-related scattering data through the detector (8), ultimately realizing the location and characterization of subsurface defects (5) or densification features. The detector (8) analyzes the polarization response of the scattered light component based on the relative change relationship between the light intensities of different polarization components, and constructs polarization response parameters that reflect the local optical anisotropy and scattering characteristics of the material. By analyzing the spatial distribution characteristics of the polarization response parameters, the detector can locate and characterize the subsurface defects (5) or densification features, and provide prior guidance for subsequent high-resolution near-field scanning.

Images