CMOS camera damage sensing system and method based on damage plasma spectrum information

Abstract

The invention discloses a CMOS camera damage sensing system and method based on damage plasma spectrum information, and belongs to the technical field of laser and material interaction detection. The CMOS camera damage sensing system based on damage plasma spectrum information comprises a laser source module, an optical module, a CMOS detector module, a plasma spectrum acquisition module, a laser and spectrum signal synchronous control module, a data processing and model building module, a thermal imaging system and a positioning and adjusting device. The laser-induced CMOS target surface is used for generating plasma, and high-resolution spectrum acquisition and thermal imaging synchronous monitoring are combined, so that real-time and non-contact accurate sensing of the damage process is realized; according to the invention, high-efficiency signal coupling of the beam splitter prism and the optical fiber is utilized, rapid capture and transmission of a plasma spectrum are ensured, multi-modal fusion analysis of spectral characteristics and thermal imaging data is realized through construction of a high-speed data acquisition card and a computer model, and the accuracy and reliability of damage assessment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of laser-material interaction detection, and in particular to a CMOS camera damage perception system and method based on damaged plasma spectrum information. Background Art

[0002] Optical detectors, such as CMOS and CCD, are widely used in industrial production and scientific research. However, in high-energy laser environments, detectors are highly susceptible to damage caused by laser irradiation. This damage can include surface coating ablation, pixel failure, and permanent damage to sensitive components, directly impacting detector performance and even causing complete failure.

[0003] At present, detector damage detection methods mainly include the following categories:

[0004] Offline detection method: The damaged detector surface morphology is observed using an optical microscope, scanning electron microscope (SEM), or atomic force microscope (AFM). Although this method can obtain high-resolution surface morphology information, it requires removing the detector from the working environment and cannot achieve real-time monitoring.

[0005] Optical interferometry: This method uses the principle of laser interference to detect detector damage. By analyzing the fringe variations in the interference pattern, damage depth and morphology can be inferred. However, interferometry is susceptible to fringe jitter, particularly in dynamic measurement environments, where mechanical vibration or thermal disturbances can cause fringe blurring, reducing detection accuracy.

[0006] Electrical performance testing: The damage level is inferred by measuring changes in detector parameters such as dark current and responsivity. While this method is suitable for performance evaluation, it cannot directly obtain spatially resolved damage information.

[0007] Although the above methods have solved the problem of detector damage assessment to a certain extent, many methods require offline detection after laser irradiation, which is difficult to meet the needs of real-time monitoring. Some methods also require direct contact with the detector, which may cause secondary damage. Therefore, we proposed a CMOS camera damage perception system and method based on damaged plasma spectrum information to solve the above problems. Summary of the Invention

[0008] (1) Technical problems solved

[0009] In view of the deficiencies in the prior art, the present invention provides a CMOS camera damage perception system and method based on damaged plasma spectrum information, which solves the problems raised in the above background technology.

[0010] (2) Technical solution

[0011] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0012] The CMOS camera damage perception system based on damaged plasma spectrum information includes a laser source module, an optical module, a CMOS detector module, a plasma spectrum acquisition module, a laser and spectrum signal synchronization control module, a data processing and model building module, a thermal imaging system, and a positioning and adjustment device. The optical module includes a focusing lens, a beam splitter prism, and an optical fiber. The data processing and model building module includes a high-speed data acquisition card and a computer.

[0013] The laser source module is located at the starting point of the optical path, and the laser beam is accurately irradiated to the CMOS detector module through the optical module. The laser source module is connected to the laser and spectrum signal synchronization control module; the optical module is installed between the laser source module and the plasma spectrum acquisition module, and the focusing lens is immediately connected to the output end of the laser source module, and the laser beam is focused by this lens; the beam splitter prism is located between the focusing lens and the CMOS detector module, forming a 45° angle with the optical path; the optical fiber is located at the end of the reflective light path of the beam splitter prism, and the optical fiber is connected to the plasma spectrum acquisition module, and the plasma spectrum acquisition module is connected to the laser and spectrum signal synchronization control module;

[0014] The CMOS detector module is located at the end of the transmission light path of the beam splitter prism, with the laser focusing point on the target surface; the data processing and model building module is connected to the plasma spectrum acquisition module; a high-speed data acquisition card receives digital signals from the plasma spectrum acquisition module and the thermal imaging system; the computer is connected to the high-speed data acquisition card; the thermal imaging system is installed in an independent light path and aligned with the surface of the CMOS detector module; the positioning and adjustment device integrates a translation stage and a rotating bracket for mechanically supporting the CMOS detector module.

[0015] Furthermore, the laser source module has a wavelength of 1064nm or 532nm, a pulse width adjustable from nanoseconds to femtoseconds, an energy density that needs to exceed the CMOS damage threshold, and a repetition frequency of 1Hz-1kHz (single or continuous pulse mode).

[0016] Furthermore, the focal length of the focusing mirror is 50-200mm, and the short focal length is used for high energy density focusing. The material is fused quartz, AR coating, and resistance to laser damage; the splitting ratio of the beam splitter is 90% transmission and 10% reflection, and the working band covers plasma ultraviolet-visible light radiation; the numerical aperture (NA) of the optical fiber is 0.22-0.39, ensuring efficient coupling and ultraviolet-near-infrared broadband transmission.

[0017] Furthermore, the target surface material of the CMOS detector module is silicon-based CMOS; the translation stage of the positioning and adjustment device has a travel range of 0-100 mm (XYZ three-axis) with an accuracy of ±1μm, and the pitch of the rotating bracket is ±30° with an accuracy of ±0.1°.

[0018] Furthermore, the plasma spectrum acquisition module has a resolution of 0.1-1nm, can identify the characteristic peak of the Si element, and has a wavelength range of 200-800nm, covering the main plasma radiation; the laser and spectrum signal synchronization control module has a synchronization accuracy of ±1ns, ensuring that the spectrum acquisition is strictly synchronized with the laser pulse, and the trigger interface is TTL / PECL to be compatible with the laser source module and the plasma spectrum acquisition module; the thermal imaging system has a thermal sensitivity of ≤20mK, can detect tiny temperature rises, and has a frame rate of ≥100fps to capture transient heat diffusion.

[0019] Furthermore, the high-speed data acquisition card has a sampling rate of ≥1GS / s, captures transient spectral signals, has 2-4 channels, and synchronously acquires spectral and thermal imaging data; the computer runs a damage assessment algorithm (such as PLS regression, neural network), associates spectral features with damage levels, and outputs damage location, depth, and material phase change information.

[0020] A method for a CMOS camera damage perception system based on damaged plasma spectrum information includes the following steps:

[0021] S1: The position of the CMOS detector module is adjusted through the positioning and adjustment device so that the target surface is in the laser focal plane. The thermal imaging system monitors the CMOS surface in real time to ensure that there are no local thermal anomalies at the beginning. The computer sets the laser parameters. The laser and spectral signal synchronization control module configures the trigger timing. The CMOS is positioned using a high-precision three-dimensional displacement platform to ensure that the measurement position is consistent with the plasma spectrum acquisition position.

[0022] S2: The laser source module emits a single or continuous pulse, which is focused by a focusing lens to a spot size of 5-50μm and a focal length of 50-200mm. The beam splitter transmits 90% of the energy to the CMOS target surface, where the silicon-based material absorbs the energy and melts and vaporizes. High-temperature plasma is generated instantly when the CMOS surface is damaged, radiating ultraviolet-visible light with characteristic spectral lines such as Si.

[0023] S3: The beam splitter reflects 10% of the plasma radiation to the optical fiber, which then enters the plasma spectrum acquisition module. The plasma spectrum acquisition module captures characteristic peaks with a resolution of 0.1-1nm. The ICCD detector has a gating time of ≤10ns and is strictly synchronized with the laser pulse. The high-speed data acquisition card records transient spectral signals at a sampling rate of ≥1GS / s.

[0024] S4: The thermal imaging system captures the damaged area at a frame rate of ≥100 fps to capture the temperature rise distribution. The data is synchronously transmitted to a computer via a high-speed data acquisition card. Simultaneously, offline measurements are performed using an automatic zoom 3D surface measuring instrument to scan the damaged area and obtain the cross-sectional profile of the damage.

[0025] S5: The computer correlates the elemental composition and intensity of the spectral data with the temperature rise and thermal diffusion of the thermal imaging data. A PLS regression or neural network algorithm is run to analyze the mapping relationship between the spectral characteristics and the damage level. The output results include damage location, damage depth, and material phase change. The damage location is determined by the laser spot coordinates and the position of the translation stage. The damage depth is inverted based on the Si characteristic peak intensity and thermal diffusion rate. Material phase changes, such as amorphization and lattice destruction, are determined through multi-peak fitting. The offline measured CMOS depth data is compared with the plasma spectrum calculation results to establish an error correction model.

[0026] S6: Based on the damage assessment results, the computer feeds back to the laser source module to adjust the subsequent pulse energy or scanning path; the positioning and adjustment device automatically moves the CMOS to the new detection location; a secondary verification is performed on areas with large errors, and the algorithm parameters are iteratively optimized.

[0027] Furthermore, in S5, after laser-induced CMOS damage, the system uses plasma spectroscopy and thermal imaging data for damage assessment. Data preprocessing begins with baseline correction and normalization of the collected plasma emission spectra to eliminate noise interference, extract key characteristic peaks, analyze the temperature distribution in the damaged area, calculate the maximum temperature rise and thermal diffusion range, and fit the temperature field to estimate the heat-affected depth. Second, parameters such as characteristic peak intensity, peak width, and peak position shift are extracted from the spectra to determine changes in material composition. The temperature rise rate and thermal diffusion radius are extracted from the thermal imaging data to assist in assessing the depth and extent of the damage. Partial least squares regression (PLS) is then used to establish a quantitative relationship between spectral features and damage level, predicting damage severity, ranging from mild ablation to deep melting. For complex data, a deep learning model is employed to fuse spectral and thermal imaging features to improve damage classification and location accuracy. Finally, the model calculation results are integrated to output damage location, depth, and material phase transition type. This is then combined with thermal imaging for verification to ensure the reliability of the assessment results, which are then fed back to the control system to adjust subsequent inspection parameters.

[0028] (3) Beneficial effects

[0029] Compared with the existing technology, the present invention provides a CMOS camera damage perception system and method based on damaged plasma spectrum information, which has the following beneficial effects:

[0030] The present invention uses laser-induced plasma generation on the CMOS target surface, combined with high-resolution spectral acquisition and simultaneous thermal imaging monitoring, to achieve real-time, non-contact, and precise perception of the damage process. Efficient signal coupling between a beam splitter and optical fiber ensures rapid capture and transmission of the plasma spectrum. High-speed data acquisition cards and computer model construction enable multimodal fusion analysis of spectral characteristics and thermal imaging data, significantly improving the accuracy and reliability of damage assessment. Furthermore, the micron-level precision control of the positioning and adjustment device and the laser-spectral synchronous triggering technology enable the device to adapt to remote detection needs in complex environments, providing an efficient and automated solution for online health monitoring and failure analysis of CMOS cameras. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0032] Figure 2 A schematic diagram of plasma generation after damage to the detector of the present invention;

[0033] Figure 3 This is a schematic diagram of the three-dimensional morphology of the detector after damage.

[0034] In the figure: 1. Laser source module; 2. Optical module; 3. Focusing lens; 4. Beam splitter; 5. Optical fiber; 6. Plasma spectrum acquisition module; 7. High-speed data acquisition card; 8. Computer; 9. Data processing and model building module; 10. Laser and spectrum signal synchronization control module; 11. CMOS detector module; 12. Thermal imaging system; 13. Positioning and adjustment device. DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0036] Example

[0037] like Figure 1-3As shown, a CMOS camera damage perception system based on damaged plasma spectrum information proposed in one embodiment of the present invention includes a laser source module 1, an optical module 2, a CMOS detector module 11, a plasma spectrum acquisition module 6, a laser and spectrum signal synchronization control module 10, a data processing and model building module 9, a thermal imaging system 12, and a positioning and adjustment device 13; the optical module 2 includes a focusing lens 3, a beam splitter prism 4, and an optical fiber 5; the data processing and model building module 9 includes a high-speed data acquisition card 7 and a computer 8;

[0038] The laser source module 1 is located at the starting point of the optical path, and the laser beam is accurately irradiated to the CMOS detector module 11 through the optical module 2. The laser source module 1 is connected to the laser and spectrum signal synchronization control module 10; the optical module 2 is installed between the laser source module 1 and the plasma spectrum acquisition module 6, and the focusing lens 3 is immediately connected to the output end of the laser source module 1, and the laser beam is focused by this lens; the beam splitter prism 4 is located between the focusing lens 3 and the CMOS detector module 11, forming a 45° angle with the optical path; the optical fiber 5 is located at the end of the reflected light path of the beam splitter prism 4, and the optical fiber 5 is connected to the plasma spectrum acquisition module 6, and the plasma spectrum acquisition module 6 is connected to the laser and spectrum signal synchronization control module 10;

[0039] The CMOS detector module 11 is located at the end of the transmission light path of the beam splitter prism 4, with the laser focusing point on the target surface; the data processing and model building module 9 is connected to the plasma spectrum acquisition module 6; the high-speed data acquisition card 7 receives the digital signals of the plasma spectrum acquisition module 6 and the thermal imaging system 12; the computer 8 is connected to the high-speed data acquisition card 7; the thermal imaging system 12 is installed in an independent light path and is aligned with the surface of the CMOS detector module 11; the positioning and adjustment device 13 integrates a translation stage and a rotating bracket for mechanically supporting the CMOS detector module 11.

[0040] Among them, the laser source module 1 has a wavelength of 1064nm or 532nm, a pulse width adjustable from nanoseconds to femtoseconds, an energy density that needs to exceed the CMOS damage threshold, and a repetition frequency of 1Hz-1kHz (single or continuous pulse mode); the focusing mirror 3 has a focal length of 50-200mm, and the short focal length is used for high energy density focusing. The material is fused quartz, AR coating, and anti-laser damage; the splitting ratio of the beam splitter prism 4 is 90% transmission and 10% reflection, and the working band covers plasma ultraviolet-visible light radiation; the numerical aperture (NA) of the optical fiber 5 is 0.22-0.39, ensuring efficient coupling and ultraviolet-near-infrared broadband transmission; the target surface material of the CMOS detector module 11 is silicon-based CMOS; the positioning and adjustment device 13 translation stage has a travel of 0-100 mm (XYZ three-axis) with an accuracy of ±1μm, and the pitch of the rotating bracket is ±30° with an accuracy of ±0.1°.

[0041] The plasma spectrum acquisition module 6 has a resolution of 0.1-1nm and can identify the characteristic peak of the Si element. The band range is 200-800nm, covering the main plasma radiation. The synchronization accuracy of the laser and spectral signal synchronization control module 10 is ±1ns, ensuring that the spectrum acquisition is strictly synchronized with the laser pulse. The trigger interface is TTL / PECL to be compatible with the laser source module 1 and the plasma spectrum acquisition module 6. The thermal imaging system 12 has a thermal sensitivity of ≤20mK and can detect tiny temperature rises. The frame rate is ≥100fps and captures transient thermal diffusion. The high-speed data acquisition card 7 has a sampling rate of ≥1GS / s, captures transient spectral signals, has 2-4 channels, and synchronously acquires spectral and thermal imaging data. The computer 8 runs a damage assessment algorithm (such as PLS regression and neural network), associates spectral characteristics with damage levels, and outputs damage location, depth, and material phase change information.

[0042] A method for a CMOS camera damage perception system based on damaged plasma spectrum information includes the following steps:

[0043] S1: The position of the CMOS detector module 11 is adjusted through the positioning and adjustment device 13 so that the target surface is in the laser focal plane; the thermal imaging system 12 monitors the CMOS surface in real time to ensure that there are no local thermal anomalies initially; the computer 8 sets the laser parameters; the laser and spectral signal synchronization control module 10 configures the trigger timing; a high-precision three-dimensional displacement platform is used to position the CMOS to ensure that the measurement position is consistent with the plasma spectrum acquisition position.

[0044] S2: The laser source module 1 emits a single or continuous pulse, which is focused by the focusing mirror 3 to a spot size of 5-50μm and a focal length of 50-200mm; the beam splitter 4 transmits 90% of the energy to the CMOS target surface, and the silicon-based material melts and vaporizes after absorbing the energy; high-temperature plasma is generated instantly when the CMOS surface is damaged. The plasma generates Figure 2 As shown, it radiates ultraviolet-visible light and contains characteristic spectral lines such as Si;

[0045] S3: Spectroscopic prism 4 reflects 10% of the plasma radiation to optical fiber 5, which then enters plasma spectrum acquisition module 6. This module captures characteristic peaks with a resolution of 0.1-1 nm. The ICCD detector has a gating time of ≤10 ns, strictly synchronized with the laser pulse. A high-speed data acquisition card 7 records transient spectral signals at a sampling rate of ≥1 GS / s.

[0046] S4: The thermal imaging system 12 shoots the damaged area at a frame rate of ≥100 fps to capture the temperature rise distribution; the data is synchronously transmitted to the computer 8 through the high-speed data acquisition card 7; and offline measurement is performed simultaneously, using an automatic zoom three-dimensional surface measuring instrument to scan the damaged area to obtain the damage cross-sectional profile. The damage morphology is as follows: Figure 3 shown.

[0047] S5: Computer 8 correlates the elemental composition and intensity of the spectral data with the temperature rise and thermal diffusion of the thermal imaging data; runs the PLS regression or neural network algorithm to analyze the mapping relationship between the spectral characteristics and the damage level. The output results include the damage location, damage depth and material phase change. The damage location is determined by the laser spot coordinates and the position of the translation stage. The damage depth is inverted based on the Si characteristic peak intensity and thermal diffusion rate. Material phase changes, such as amorphization and lattice destruction, are judged through multi-peak fitting; the offline measured CMOS depth data is compared with the plasma spectrum calculation results to establish an error correction model.

[0048] S6: Based on the damage assessment results, the computer 8 feeds back to the laser source module 1 to adjust the subsequent pulse energy or scanning path; the positioning and adjustment device 13 automatically moves the CMOS to a new detection position; a secondary verification is performed on areas with large errors, and the algorithm parameters are iteratively optimized.

[0049] It should be further explained here that in S5, after laser-induced CMOS damage, the system uses plasma spectra and thermal imaging data to perform damage assessment. First, data preprocessing is performed to perform baseline correction and normalization on the collected plasma emission spectra, eliminate noise interference, extract key characteristic peaks, analyze the temperature distribution in the damaged area, calculate the maximum temperature rise and thermal diffusion range, and fit the temperature field to assess the heat-affected depth. Secondly, characteristic peak intensity, peak width, peak position shift and other parameters are extracted from the spectrum to determine changes in material composition, and the temperature rise rate and thermal diffusion radius are extracted from the thermal imaging data to assist in assessing the depth and range of the damage.

[0050] Then, partial least squares regression (PLS) is used to establish a quantitative relationship between spectral features and damage level to predict the degree of damage, such as slight ablation and deep melting. If the data is complex, a deep learning model is used to fuse spectral and thermal imaging features to improve the accuracy of damage classification and positioning. Finally, the comprehensive model calculation results are used to output the damage location, depth, and material phase change type, combined with thermal imaging verification to ensure the reliability of the evaluation results, and feedback is given to the control system to adjust subsequent detection parameters.

[0051] This invention uses laser-induced plasma spectroscopy to remotely monitor and accurately assess the damage status of CMOS detectors. This method uses laser pulses to induce plasma in the CMOS detector, collecting its spectral information in real time. Combined with a thermal imaging system, it simultaneously monitors the temperature distribution and establishes a mapping relationship between damage signature parameters and spectral characteristics. This innovative fusion of spectral analysis and thermal radiation measurement ensures highly sensitive and accurate identification of CMOS detector damage, providing new insights into remote intelligent sensing and dynamic assessment.

[0052] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A CMOS camera damage sensing system based on damaged plasma spectrum information, characterized by: It comprises a laser source module (1), an optical module (2), a CMOS detector module (11), a plasma spectrum acquisition module (6), a laser and spectrum signal synchronization control module (10), a data processing and model building module (9), a thermal imaging system (12), and a positioning and adjustment device (13); the optical module (2) comprises a focusing lens (3), a beam splitter prism (4), and an optical fiber (5); the data processing and model building module (9) comprises a high-speed data acquisition card (7) and a computer (8); The laser source module (1) is located at the starting point of the optical path, and the laser beam is accurately irradiated to the CMOS detector module (11) through the optical module (2). The laser source module (1) is connected to the laser and spectrum signal synchronization control module (10); the optical module (2) is installed between the laser source module (1) and the plasma spectrum acquisition module (6), and the focusing mirror (3) is immediately connected to the output end of the laser source module (1), and the laser beam is focused by this mirror; the beam splitter prism (4) is located between the focusing mirror (3) and the CMOS detector module (11), and forms an angle of 45° with the optical path; the optical fiber (5) is located at the end of the reflection optical path of the beam splitter prism (4), and the optical fiber (5) is connected to the plasma spectrum acquisition module (6), and the plasma spectrum acquisition module (6) is connected to the laser and spectrum signal synchronization control module (10); The CMOS detector module (11) is located at the end of the transmission light path of the beam splitter prism (4), with the laser focusing point being the target surface; the data processing and model building module (9) is connected to the plasma spectrum acquisition module (6); the high-speed data acquisition card (7) receives digital signals from the plasma spectrum acquisition module (6) and the thermal imaging system (12); the computer (8) is connected to the high-speed data acquisition card (7); the thermal imaging system (12) is installed in an independent light path and is aligned with the surface of the CMOS detector module (11); the positioning and adjustment device (13) integrates a translation stage and a rotation bracket for mechanically supporting the CMOS detector module (11).

2. The CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 1, characterized in that: The laser source module (1) has a wavelength of 1064 nm or 532 nm, a pulse width adjustable from nanoseconds to femtoseconds, an energy density that exceeds the CMOS damage threshold, and a repetition frequency of 1 Hz-1 kHz.

3. The CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 1, characterized in that: The focusing mirror (3) has a focal length of 50-200 mm, and the short focal length is used for high energy density focusing. The material is fused quartz, AR coating, and anti-laser damage. The splitting ratio of the beam splitter (4) is 90% transmission and 10% reflection, and the working band covers the plasma ultraviolet-visible light radiation. The numerical aperture (NA) of the optical fiber (5) is 0.22-0.39, ensuring efficient coupling and ultraviolet-near infrared broadband transmission.

4. The CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 1, characterized in that: The target surface material of the CMOS detector module (11) is silicon-based CMOS; the translation stage of the positioning and adjustment device (13) has a travel range of 0-100 mm (XYZ three axes) with an accuracy of ±1 μm, and the pitch of the rotating bracket is ±30° with an accuracy of ±0.1°.

5. The CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 1, characterized in that: The plasma spectrum acquisition module (6) has a resolution of 0.1-1 nm and can identify the characteristic peak of the Si element. The wavelength range is 200-800 nm, covering the main plasma radiation. The laser and spectrum signal synchronization control module (10) has a synchronization accuracy of ±1 ns, ensuring that the spectrum acquisition is strictly synchronized with the laser pulse. The trigger interface is TTL / PECL to be compatible with the laser source module (1) and the plasma spectrum acquisition module (6). The thermal imaging system (12) has a thermal sensitivity of ≤20 mK and can detect small temperature rises. The frame rate is ≥100 fps and can capture transient heat diffusion.

6. The CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 1, characterized in that: The high-speed data acquisition card (7) has a sampling rate of ≥1GS / s, captures transient spectral signals, has 2-4 channels, and synchronously acquires spectral and thermal imaging data; the computer (8) runs a damage assessment algorithm, associates spectral features with damage levels, and outputs damage location, depth, and material phase change information.

7. The method of a CMOS camera damage sensing system based on damaged plasma spectrum information according to any one of claims 1 to 6, characterized in that: The steps include: S1: The position of the CMOS detector module (11) is adjusted by the positioning and adjustment device (13) so that the target surface is in the laser focal plane; the thermal imaging system (12) monitors the CMOS surface in real time to ensure that there is no local thermal anomaly initially; the computer (8) sets the laser parameters; the laser and spectral signal synchronization control module (10) configures the trigger timing; a high-precision three-dimensional displacement platform is used to position the CMOS to ensure that the measurement position is consistent with the plasma spectrum acquisition position; S2: The laser source module (1) emits a single or continuous pulse, which is focused by the focusing mirror (3) to a spot size of 5-50μm and a focal length of 50-200mm; the beam splitter (4) transmits 90% of the energy to the CMOS target surface, and the silicon-based material melts and vaporizes after absorbing the energy; high-temperature plasma is generated instantly when the CMOS surface is damaged, radiating ultraviolet-visible light, including characteristic spectral lines such as Si; S3: The spectroscopic prism (4) reflects 10% of the plasma radiation to the optical fiber (5), which then enters the plasma spectrum acquisition module (6); the plasma spectrum acquisition module (6) captures the characteristic peak with a resolution of 0.1-1 nm, and the ICCD detector gate time is ≤10 ns, which is strictly synchronized with the laser pulse; the high-speed data acquisition card (7) records the transient spectrum signal with a sampling rate of ≥1 GS / s; S4: The thermal imaging system (12) captures the damaged area at a frame rate of ≥100 fps to capture the temperature rise distribution; the data is synchronously transmitted to the computer (8) via a high-speed data acquisition card (7); and offline measurements are performed simultaneously, using an automatic zoom three-dimensional surface measuring instrument to scan the damaged area and obtain the damage cross-sectional profile; S5: Computer (8) associates the elemental composition and intensity of the spectral data with the temperature rise and thermal diffusion of the thermal imaging data; runs PLS regression or neural network algorithm to analyze the mapping relationship between spectral characteristics and damage level, and outputs the damage location, damage depth and material phase change. The damage location is determined by the laser spot coordinates and the translation stage position. The damage depth is inverted based on the Si characteristic peak intensity and thermal diffusion rate. Material phase change, such as amorphization and lattice destruction, is determined by multi-peak fitting; the offline measured CMOS depth data is compared with the plasma spectrum calculation results to establish an error correction model; S6: Based on the damage assessment results, the computer (8) feeds back to the laser source module (1) to adjust the subsequent pulse energy or scanning path; the positioning and adjustment device (13) automatically moves the CMOS to a new detection position; a secondary verification is performed on the area with large errors, and the algorithm parameters are iteratively optimized.

8. The method of the CMOS camera damage sensing system based on damaged plasma spectrum information according to claim 7, characterized in that: In the aforementioned S5, after laser-induced CMOS damage, the system performs damage assessment using plasma spectrum and thermal imaging data. First, data preprocessing is performed to perform baseline correction and normalization on the collected plasma emission spectrum, eliminate noise interference, extract key characteristic peaks, analyze the temperature distribution of the damaged area, calculate the maximum temperature rise and thermal diffusion range, and fit the temperature field to assess the heat-affected depth. Second, characteristic peak intensity, peak width, peak position shift and other parameters are extracted from the spectrum to determine changes in material composition. The temperature rise rate and thermal diffusion radius are extracted from the thermal imaging data to assist in assessing the depth and range of the damage. Then, partial least squares regression (PLS) is used to establish a quantitative relationship between spectral characteristics and damage level to predict the degree of damage, such as slight ablation or deep melting. If the data is complex, a deep learning model is used to fuse spectral and thermal imaging features to improve the accuracy of damage classification and positioning. Finally, the model calculation results are integrated to output the damage location, depth, and material phase change type. Combined with thermal imaging verification, the assessment results are reliable and fed back to the control system to adjust subsequent detection parameters.

Images