Laser damage evaluation method and system based on remote multispectral imaging temperature measurement

Abstract

The application discloses a laser damage evaluation method and system based on remote multispectral imaging temperature measurement, belongs to the technical field of laser countermeasure and target damage efficiency evaluation, and solves the technical problem that laser weapons are difficult to perform real-time and accurate laser damage evaluation in complex environments such as strong background light, target movement and atmospheric interference when attacking a long-distance target. Multispectral radiation images of a target to be measured are acquired in real time; key parameters of a laser weapon system and a real temperature field of the target to be measured are acquired in real time by using a control and interconnection system, the attitude of an optoelectronic turntable is adjusted in real time according to the acquired key parameters and the real temperature field, the imaging optical axis direction of a long-focus optical system is made consistent with the laser irradiation direction, the real temperature field of the target to be measured is obtained by inversion, damage state evaluation is performed based on the current time characteristics and space characteristics of the target to be measured, and a damage state evaluation result is obtained. The application is used for realizing real-time and accurate long-distance small target damage evaluation.

Description

Technical Field

[0001] This invention relates to the field of laser countermeasures and target damage effectiveness assessment technology, specifically a laser damage assessment method and system based on long-range multispectral imaging thermometry. Background Technology

[0002] With the rapid development of high-energy laser weapon technology, laser damage effectiveness has become a key technical indicator in laser equipment demonstration, weapon system calibration, operational application evaluation, and anti-drone countermeasures systems.

[0003] The laser-induced damage process is essentially a multi-physics transient coupling process spanning optics, thermodynamics, materials science, mechanics, and fluid dynamics. It involves a series of high-speed evolutionary phenomena, including energy absorption and heat transfer, material phase transformation and structural degradation, thermal ablation, and gas backflow. The damage effectiveness is highly dependent on material properties, laser power density, interaction time, and atmospheric transport conditions. Simultaneously, weapon systems have an increasingly urgent need for the quantitative determination of key parameters such as "damage efficiency," "damage threshold," "target residual capability assessment," and "disability status determination." Obtaining these key parameters is predicated on the assessment and prediction of the target material's ablation degree. Currently, targets in external environments are often in complex states such as long distances, small sizes, motion disturbance resistance, and strong background light, making real-time observation of the laser damage process, temperature field reconstruction, and threat status assessment extremely challenging scientific and engineering problems. Therefore, a long-range, multi-modal, and quantitative laser damage assessment technology system capable of stable operation in real-world scenarios is urgently needed.

[0004] Previous studies by scholars both domestically and internationally have yielded significant progress in theoretical research and close-range experiments on laser damage effects. For instance, Chinese patent document CN119375163A discloses a "high-speed long-range laser effect evaluation system and method," but this technical solution heavily relies on a series of six beam splitters, resulting in high optical path loss, complex structure, and decreased signal-to-noise ratio at long distances. Furthermore, multi-point photoelectric detection is extremely sensitive to background light, target size, and atmospheric attenuation. This solution does not address radiative transmission processing at long distances (hundreds of meters to kilometers), and temperature inference still requires the assumption of known emissivity or reliance on spectrometer peak values, which is difficult to meet in practical engineering. In addition, this solution is more geared towards close-range ablation monitoring in the laboratory, and its engineering feasibility and ability to truly assess long-range, high-speed damage remain insufficient.

[0005] In the prior art, Chinese patent document CN110926533A discloses a "device and method for real-time measurement of multiple parameters in laser damage". However, this scheme has a high degree of equipment dependence, complex device structure, and experimental conditions are limited to indoor near-field target ranges, making it unsuitable for long-distance or actual combat environments. The measurement principle is highly dependent on the homogenization assumption of the integrating sphere, which may lead to distortion when facing high temperature, strong radiation, and high-speed debris. In addition, the device is large in size, has high requirements for cooling and optical path calibration, and has limited engineering promotion and adaptability to the field.

[0006] In summary, existing technologies have the technical problem of making it difficult to conduct real-time and accurate laser damage assessment when laser weapons strike long-range targets in complex environments such as strong background light, target movement, and atmospheric interference. Summary of the Invention

[0007] This invention solves the technical problem in the prior art that it is difficult to conduct real-time and accurate laser damage assessment when laser weapons strike long-range targets in complex environments such as strong background light, target movement, and atmospheric interference.

[0008] The laser damage assessment system based on long-distance multispectral imaging temperature measurement described in this invention includes:

[0009] The long-range multispectral damage field acquisition module is used to acquire multispectral radiation images of the target under test during laser irradiation.

[0010] The long-range multispectral damage field acquisition module includes a long focal length optical system and a control and interconnection system. The long focal length optical system is used for imaging the target under test.

[0011] The control and interconnection system is used to acquire key parameters of the laser weapon system in real time, and synchronously adjust the attitude of the photoelectric turntable according to the key parameters so that the imaging optical axis of the long focal length optical system is consistent with the laser irradiation direction. Based on the damage assessment results, the system controls the start-up and shutdown of the laser weapon system and the adjustment of key parameters in real time.

[0012] The temperature and emissivity joint inversion module is used to invert the true temperature field of the target under test based on the current multispectral radiation image;

[0013] The laser damage status assessment module is used to assess the damage status in real time based on the actual temperature field of the target at the current moment, and transmit the damage status assessment results to the control and interconnection system.

[0014] Laser weapon systems are used to generate high-energy laser beams and irradiate the surface of the target.

[0015] Optoelectronic turntables are used to support long focal length optical systems.

[0016] Furthermore, in one embodiment of the present invention, the long-distance multispectral damage field acquisition module further includes a broadband modulation optical window and a multispectral modulation detector;

[0017] The broadband modulation optical window is used to adjust the transmittance of different bands to achieve suppression of high-temperature target radiation peaks, separability of sensitivity between channels, and cutoff of laser bands.

[0018] The multispectral modulation detector is used to acquire multi-channel radiometric images of the target under a single-frame exposure.

[0019] Furthermore, in one embodiment of the present invention, the system also utilizes a control and interconnection system to obtain the real temperature field of the target under test at the current moment, and performs fine-tuning compensation on the attitude of the photoelectric turntable.

[0020] Furthermore, in one embodiment of the present invention, the system further includes an external computing processing unit, a SOC preprocessing unit, a serial port channel, and a time-series multispectral data transmission interface;

[0021] The SOC preprocessing unit is used to preprocess multispectral radiometric images;

[0022] Serial port channel is used to connect control and interconnect systems to laser weapon systems;

[0023] A time-series multispectral data transmission interface is used to transmit preprocessed data to an external computing unit;

[0024] An external computing unit is used to perform inversion and solution on the preprocessed data to obtain the true temperature field of the target.

[0025] The laser damage assessment method based on long-distance multispectral imaging thermometry described in this invention, which is implemented based on any of the laser damage assessment systems described above, includes the following steps:

[0026] Step 1: Acquire multispectral radiation images of the target under test during laser irradiation in real time;

[0027] Step 2: Based on the current multispectral radiation image, retrieve the true temperature field of the target under test;

[0028] The key parameters of the laser weapon system and the real temperature field of the target under test are acquired in real time by the control and interconnection system. The attitude of the photoelectric turntable is adjusted synchronously according to the acquired key parameters and real temperature field so that the imaging optical axis of the long focal length optical system is consistent with the laser irradiation direction.

[0029] Step 3: Extract the temporal and spatial characteristics of the current target's real temperature field, and perform damage assessment based on the temporal and spatial characteristics to obtain the damage assessment results.

[0030] Furthermore, in one embodiment of the present invention, the damage status assessment based on temporal and spatial characteristics in step 3 specifically includes:

[0031] Based on temporal and spatial characteristics, and combined with corresponding preset thresholds, a coarse assessment of the current damage state is made. Temperature time-series trend matching is performed using DTW, and the damage state of the target under test during laser radiation is calculated using HMM. The coarse assessment result of the damage state, the temperature time-series trend matching result, and the damage state of the target under test during laser radiation are comprehensively assessed to obtain the final damage state evaluation result.

[0032] Furthermore, in one embodiment of the present invention, the time feature includes at least the regional average temperature. Maximum temperature Temperature rise rate and duration of high temperature ;

[0033] The spatial features include at least the area of ​​the high-temperature region. Changes, boundary morphology Fragmentation degree, percentage of internal low-temperature cavities Changes in the position of the centroid of hotspots .

[0034] Furthermore, in one embodiment of the present invention, the coarse judgment of the current damage state is to construct an intermediate feature vector and a process vector based on time and space features, and to determine the damage stage based on the relationship between the intermediate feature vector and the process vector and a preset threshold.

[0035] Furthermore, in one embodiment of the present invention, the use of DTW for temperature time-series trend matching specifically includes:

[0036] DTW is used to calculate the matching distance between the actual temperature field of the target under test and the preset temperature curves for normal heating, stable ablation, and rapid drop in perforation, respectively, to obtain the corresponding similarity index. Based on the similarity index, the category to which the current temperature evolution trend belongs is determined.

[0037] Furthermore, in one embodiment of the present invention, the step of using Hidden Markov Models (HMMs) to calculate the damage state of the target under test during laser radiation specifically involves:

[0038] The damage state sequence of the target under test during laser irradiation is established based on Hidden Markov Model (HMM). The damage state includes the heating stage, the stable ablation stage, and the perforation stage. The damage state of the target under test during laser radiation is calculated based on the observation probability of the intermediate feature vector.

[0039] This invention solves the technical problem of existing technologies where laser weapons struggle to perform real-time and accurate laser damage assessment in complex environments such as strong background light, target movement, and atmospheric interference when striking long-range targets. Specific beneficial effects include:

[0040] 1. This invention proposes a laser damage assessment system based on long-range multispectral imaging temperature measurement, and constructs a complete closed-loop structure of "laser action → optical alignment → temperature inversion → damage assessment". It not only solves the problems of optical axis misalignment, timing inconsistency and target loss in traditional systems, but also significantly improves the damage identification capability of the system in long-range and high-noise environments. It enables the system to make continuous and highly reliable judgments on damage stages such as heating, ablation and perforation in complex combat scenarios such as strong background light, target movement and atmospheric interference.

[0041] 2. This invention proposes a laser damage assessment system based on long-range multispectral imaging temperature measurement. To ensure spatial consistency between the laser irradiation point, the optical imaging area, and the temperature inversion results, a linkage control mechanism is constructed among the laser weapon pointing information, the optical acquisition turntable, and the multispectral imaging system. This linkage mechanism enables synchronous tracking and automatic alignment of the laser action process, which is key to achieving long-range damage assessment of small targets.

[0042] 3. This invention proposes a laser damage assessment method based on long-distance multispectral imaging temperature measurement. By extracting the temporal and spatial features of the temperature field in each frame of the real temperature field sequence, a time-space joint assessment method for laser damage processes is achieved. This method is used for real-time identification and classification of the heating, ablation, and perforation processes of targets under long-distance conditions. The method is designed around the dynamic evolution characteristics of the temperature field sequence, using multiple frames of inverted temperature images as input. It focuses on temporal variation trends, spatial morphological evolution, and intermediate state behavior as core features to achieve automated and quantitative determination of laser damage events. Attached Figure Description

[0043] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0044] Figure 1 This is a schematic diagram of the laser damage assessment system based on long-distance multispectral imaging temperature measurement as described in Implementation Method 1;

[0045] Figure 2 This is a schematic diagram of the laser damage assessment method based on long-distance multispectral thermometry as described in Implementation Method 8;

[0046] Figure 3 This is a schematic diagram of the laser damage assessment execution steps based on long-distance multispectral thermometry as described in Implementation Method 8;

[0047] Figure 4 This is a schematic diagram illustrating the linkage principle between the laser weapon and the damage assessment system as described in Implementation Method 8;

[0048] Figure 5 This is a time-series temperature field result diagram of laser ablation of carbon fiber as described in Embodiment 8;

[0049] Figure 6 These are the morphological images of the ABS plastic after laser ablation and the temperature distribution diagram for determining ablation, as described in Embodiment 8.

[0050] Figure 7 It is the damage state temperature diagram described in Implementation Method 3;

[0051] Figure 8 These are multispectral images captured as described in Embodiment 3;

[0052] Figure 9 These are schematic diagrams of the three types of "standard curves" described in Implementation Method Six;

[0053] Figure 10 This is a schematic diagram of the stable ablation mode and perforation mode stage curves and temperature field images using carbon fiber material as an example, as described in Embodiment 6.

[0054] Figure label explanation: 1 External computing processing unit, 2 Laser weapon system, 3 Target under test, 4- Photoelectric turntable, 5- Broadband modulation optical window, 6 Long focal length optical system, 7 Multispectral modulation detector, 8-SOC preprocessing unit, 9-serial port channel, 10-time-series multispectral data transmission interface, 11-control and interconnection system. Detailed Implementation

[0055] Various embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. The embodiments described with reference to the drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0056] Implementation Method 1: Existing laser damage assessment technologies are generally characterized by complex structures, large volumes, and the need for multi-level optical path separation or multi-source sensor collaboration. They can often only operate under close-range, controlled experimental conditions, making it difficult to deploy on mobile platforms. They also cannot maintain stable output in real environments such as strong background light, weak target signals, and atmospheric disturbances. Their assessment methods mostly rely on single-band brightness, shape changes, or near-field point detection, and cannot obtain continuous and quantitative temperature field information under long-range conditions. Furthermore, they lack the ability to invert the temporal evolution and spatial distribution of material ablation processes. Therefore, it is difficult to achieve high frame rate and high-definition damage monitoring of lightweight targets such as UAVs at real combat distances.

[0057] To address the aforementioned technical problems, this embodiment proposes a laser damage assessment system based on long-distance multispectral imaging temperature measurement, such as... Figure 1 As shown, the laser damage assessment system proposed in this embodiment has a compact structure and is easy to install on a platform. It achieves high frame rate acquisition by relying on the multispectral modulation detector 7, and is compatible with strong light and weak light scenes through broadband modulation and dual background adaptation mechanism. At the same time, the system has the ability to interconnect with the laser weapon system 2 in real time. It can perform quantitative damage assessment based on the real temperature field time series information and spatial distribution information retrieved by multispectral inversion under long-distance conditions. It fundamentally overcomes the shortcomings of existing technologies, such as difficulty in engineering deployment, difficulty in adapting to complex environments, and inability to accurately characterize the laser action process. It has significant engineering application value and research necessity.

[0058] The laser damage assessment system based on long-distance multispectral imaging thermometry includes:

[0059] The long-distance multispectral damage field acquisition module is used to acquire multispectral radiation images of the target 3 during laser irradiation.

[0060] The long-range multispectral damage field acquisition module includes a long focal length optical system 6 and a control and interconnection system 11. The long focal length optical system 6 is used for imaging the target 3 under test.

[0061] The long focal length optical system 6 and the multispectral modulation detector 7 have matched spectral range, pixel size and field of view height, which can maintain stable imaging quality and radiation flux in the range of tens of meters to thousands of meters, enabling the system to be compatible with near-field experiments and long-distance target damage monitoring.

[0062] The long focal length optical system 6 is responsible for imaging distant targets and may include multiple lenses, a zoom assembly, and a focusing mechanism. Its focal length range can be switched from medium focal length to super telephoto, for example, 50 mm–1500 mm. The front end of the long focal length optical system 6 receives the radiated light from the target 3 and focuses it onto the broadband modulation optical window 5.

[0063] The control and interconnection system 11 is used to acquire key parameters of the laser weapon system 2 in real time, and synchronously adjust the attitude of the photoelectric turntable 4 according to the key parameters so that the imaging optical axis of the long focal length optical system 6 is consistent with the laser irradiation direction. Based on the damage assessment results, it controls the start-up and shutdown of the laser weapon system 2 and the adjustment of key parameters in real time.

[0064] The control and interconnection system 11 is used to realize the zoom and focus adjustment, exposure adjustment, channel selection and imaging parameter synchronous control of the long focal length optical system 6, and is connected to the laser weapon system 2 through the serial port channel 9 to acquire key parameters such as laser power, light emission time and aiming angle in real time, so as to form the basis for the correspondence analysis of laser input and damage response.

[0065] The temperature and emissivity joint inversion module is used to invert the true temperature field of the target 3 based on the current multispectral radiation image.

[0066] The laser damage status assessment module is used to assess the damage status in real time based on the actual temperature field of the target 3 at the current moment, and transmit the damage status assessment results to the control and interconnection system 11.

[0067] The laser damage assessment system receives the target temperature field results from a multispectral thermometer or an external temperature inversion module. The temperature field consists of multiple frames of two-dimensional temperature distribution data arranged in chronological order, which are used as input information for the laser damage status assessment module.

[0068] The laser damage assessment module is used to extract the target's heating rate, temperature peak location, ablation zone expansion direction, thermal diffusion characteristics, and spatial temperature gradient changes based on the reconstructed real temperature field. It combines this with laser power density, application time, and irradiation trajectory to identify the damage stage and determine the damage status. The system can output the damaged area, damage level, structural risk warnings, and residual performance status in real time, providing quantitative data for laser weapon effectiveness assessment.

[0069] Furthermore, in one embodiment of this implementation, the laser damage state assessment module constructs a temporal-spatial joint feature description model based on the multi-frame temperature field sequence to characterize the dynamic thermal response and structural evolution behavior of the target during laser irradiation.

[0070] The laser damage assessment module employs an event detection mechanism based on physical threshold rules. By setting thresholds for sudden temperature drop, high-temperature area growth, boundary fragmentation, and void area ratio, it can quickly identify key damage events such as precursors to perforation, ablation splitting, or abrupt changes in material structure.

[0071] Laser weapon system 2 is used to generate a high-energy laser beam and irradiate the surface of the target 3;

[0072] The laser weapon system 2 generates a high-energy laser beam and irradiates the target surface to induce heating, ablation, or perforation of the material. The laser weapon system 2 sends real-time information such as laser emission status, firing timestamp, laser power, elevation angle, and azimuth angle to the system via a communication interface. This information is used to drive the optoelectronic turntable 4, adjust the optical axis, and synchronize the damage time sequence.

[0073] The target 3 to be tested can be a drone shell, carbon fiber composite plate, metal structural component, or any material object that requires laser damage assessment. During laser irradiation, the target surface undergoes behaviors such as temperature rise, material erosion, cracking, or perforation, and the system captures its radiation changes through long-distance imaging.

[0074] The photoelectric turntable 4 is used to support the long-focal-length optical system 6 and the multispectral modulation detector 7, and to perform precise rotation in the pitch and azimuth directions. The function of the photoelectric turntable 4 is to ensure that the optical axis always points to the laser irradiation position, thereby keeping the temperature inversion region consistent with the damaged region.

[0075] This embodiment establishes a real-time communication interface with the laser weapon system 2 to acquire parameters such as laser emission status, emission timestamp, laser elevation angle, and azimuth angle, and uses these as inputs for the control of the long focal length optical system 6 and the photoelectric turntable 4. The dual-axis or three-axis turntable in the evaluation system automatically adjusts its attitude according to the laser pointing information, so that the imaging optical axis of the long focal length optical system 6 follows the laser irradiation direction, keeping the target within the optimal imaging field of view at all times.

[0076] Implementation Method 2: The difference between this implementation method and Implementation Method 1 is that the long-distance multispectral damage field acquisition module further includes a broadband modulation optical window 5 and a multispectral modulation detector 7.

[0077] The broadband modulation optical window 5, made of a high-transmittance optical material, protects the internal detector and provides a stable spectral response in the visible to near-infrared range. The window can be coated with multiple layers as needed to suppress infrared overheating or enhance transmittance in specific bands. By designing transmittance ratios for different bands, reasonable suppression near the radiation peaks of high-temperature targets, separability of sensitivity between channels, and cutoff of the laser band can be achieved, thereby improving the system's dynamic range and temperature inversion stability in high-temperature and strong background ranges.

[0078] The multispectral modulation detector 7 is an n×n (typically n=2, 3, or 4) pixel-level filter array structure, capable of realizing nine-channel multispectral synchronous imaging. Each channel has different transmittance in different bands, used to acquire high frame rate multi-channel broadband radiation images of the target under laser irradiation conditions in a single-frame exposure, covering the transient radiation changes of the target in each stage of heating, softening, melting, and ablation, and is a key unit for temperature inversion;

[0079] Furthermore, the temperature and emissivity joint inversion module includes a camera system-on-a-chip (SOC) preprocessing unit 8, an external computing processing unit 1, and a time-series multispectral data transmission interface 10.

[0080] The SOC preprocessing unit 8 is used to complete the denoising, shadow correction, channel registration and high-dimensional spectral image segmentation of the multispectral radiometric image, and transmits the processed multispectral radiometric data to the external computing and processing unit 1 via the time-series multispectral data transmission interface 10.

[0081] The external computing unit 1 is based on the broadband modulation response model and the spectral emissivity model of the target material in the visible band. By constructing a temperature-emissivity joint solution set, it realizes the inversion solution of the real temperature field under unknown emissivity conditions. In the inversion process, spatial consistency constraints and time series smoothing constraints are added to ensure that the temperature field results remain continuous and physically consistent under the conditions of high-speed heating of the target, attitude change and background interference.

[0082] In one embodiment of this method, temperature inversion is constructed jointly based on a broadband modulation response model and an emissivity spectral basis expansion model. After obtaining a nine-channel broadband radiometric response in a single exposure, the multimodal spectral chip performs modulation array channel segmentation, dark field correction, pixel-level non-uniformity correction, and cross-channel spatial registration on the RAW image data, so that the data of each channel forms a multi-channel radiometric dataset that can be used for inversion under a unified spatial coordinate system. Subsequently, the computing platform normalizes the amplitude of each channel and obtains the spectral transmission function of each channel by combining the broadband modulation database, converting the measured signal into an equivalent radiometric quantity.

[0083] Based on the calculation of radiance, the response of each channel is expressed as a combination of the expansions of the broadband modulation function, the blackbody radiation function, and the emissivity spectral basis function. A multi-channel nonlinear radiation equation system is constructed. By jointly solving the temperature parameters and the emissivity expansion coefficients, the synchronous inversion of temperature and emissivity is achieved. The solution process adopts an iterative optimization method to minimize the residual between the calculated radiance and the measured radiance of each channel, thereby obtaining the temperature value of each pixel and forming an initial temperature field.

[0084] To ensure the physical consistency of the inversion results, structural consistency constraints are introduced during the solution process to keep the spatial boundary of the temperature field consistent with the gradient of the target structure image, thus avoiding temperature drift caused by noise or strong background. At the same time, continuity constraints are introduced in the time dimension to ensure that the temperature meets the second-order smooth evolution law and the emissivity meets the first-order slow change law, thereby suppressing non-physical jumps in high-speed dynamic scenes.

[0085] Through the comprehensive inversion framework of broadband response modeling, joint solution and spatial and temporal consistency constraints, we obtain continuous and stable real temperature field results under long-distance, high-dynamic and complex background conditions.

[0086] Furthermore, the external computing processing unit 1 is also used to receive nine-channel RAW data from the camera, perform multispectral splitting, flat field correction, broadband modulation model solving, temperature inversion, and damage assessment algorithms (HMM + DTW + threshold rules). The external computing processing unit 1 can adopt a high-performance CPU, GPU, or SoC platform, and can display temperature pseudo-color images and damage status levels in real time.

[0087] Furthermore, the system also utilizes the control and interconnection system 11 to obtain the real temperature field of the target 3 at the current moment, and performs fine-tuning compensation on the attitude of the photoelectric turntable 4.

[0088] This embodiment further utilizes the location of the hot spot centroid or high-temperature region in the temperature field as a visual feedback signal to fine-tune and compensate the attitude of the photoelectric turntable 4, ensuring that the optical axis remains accurately aligned with the laser irradiation point even at long distances, under atmospheric disturbances, or when the target is moving.

[0089] Implementation Method 3: A laser damage assessment method based on long-distance multispectral imaging thermometry, wherein the method is implemented based on any one of the systems described in Implementation Methods 1 and 2, and includes the following steps:

[0090] Step 1: Acquire multispectral radiation images of the target 3 under test during laser irradiation in real time, such as... Figure 8 As shown;

[0091] The control and interconnection system 11 acquires the key parameters of the laser weapon system 2 and the real temperature field of the target 3 in real time. Based on the acquired key parameters and real temperature field, the attitude of the photoelectric turntable 4 is adjusted synchronously so that the imaging optical axis of the long focal length optical system 6 is consistent with the laser irradiation direction.

[0092] Step 2: Based on the current multispectral radiation image, the true temperature field of the target 3 is retrieved, such as... Figure 7 As shown, the central dark spot represents the result of laser-induced damage and perforation detection.

[0093] First, obtain multiple consecutive frames of broadband multimodal temperature fields, and denote the temperature field of the k-th frame as:

[0094] ;

[0095] Where (x,y)∈Ω is the selected evaluation region.

[0096] Step 3: For each frame of temperature field, the system automatically extracts the temporal and spatial features of the real temperature field of the target 3 under test, and performs damage state assessment based on the temporal and spatial features to obtain the damage state assessment result.

[0097] To adapt to different operating distances of laser weapons, the long-focal-length optical system 6 in this embodiment is coupled with the optoelectronic turntable 4 for control, enabling automatic zooming and focusing, thus ensuring image clarity and temperature inversion accuracy. Simultaneously, a time reference is established between laser emission and multispectral camera acquisition using a unified clock or pulse synchronization signal, ensuring strict alignment of laser action events and temperature change sequences in the time dimension. This provides a stable physical basis for subsequent damage assessment algorithms based on HMM, DTW, and threshold rules.

[0098] Implementation Method Four: The difference between this implementation method and Implementation Method Three lies in the fact that the damage status assessment based on time and spatial characteristics is specifically as follows:

[0099] Based on temporal and spatial characteristics, and combined with corresponding preset thresholds, a coarse assessment of the current damage state is made. Temperature time-series trend matching is performed using DTW, and the damage state of the target 3 under test during laser radiation is calculated using HMM. The coarse assessment result of the damage state, the temperature time-series trend matching result, and the damage state of the target 3 under test during laser radiation are comprehensively assessed to obtain the final damage state evaluation result.

[0100] The time characteristic includes at least the regional average temperature. Maximum temperature Temperature rise rate and duration of high temperature ;

[0101] The regional average temperature is defined as:

[0102] ;

[0103] in, For the first Frame image.

[0104] The maximum temperature in the area is:

[0105] ;

[0106] To characterize the overall heating or cooling trend, the system calculates the first-order difference of the average temperature:

[0107] ;

[0108] To measure the duration of a high-temperature state, a high-temperature threshold is set. Define the indicator function:

[0109] ;

[0110] The duration of high temperature can then be expressed as:

[0111] ;

[0112] in, This represents the total number of frames captured in the image.

[0113] Calculate the maximum drop in average temperature:

[0114] ;

[0115] When the maximum drop in average temperature greater than the preset temperature drop threshold At that time, it was determined that a sudden drop in temperature had occurred.

[0116] The spatial features include at least the area of ​​the high-temperature region. Changes, boundary morphology Fragmentation degree, percentage of internal low-temperature cavities Changes in the position of the centroid of hotspots .

[0117] To describe the spatial morphology of the ablation region, the system first uses a temperature threshold. Extracting the mask from the high-temperature region:

[0118] ;

[0119] Further obtain the area of ​​the high-temperature zone:

[0120] ;

[0121] The boundary of the high-temperature region is obtained through edge detection or contour extraction, and its perimeter P is calculated. k This further yields the shape factor (roundness factor):

[0122] ;

[0123] when A value significantly greater than 1 indicates that the edges of the high-temperature zone are rough and the degree of fragmentation is high. At the continuity level, for By marking the connected components, the number of high-temperature connected components can be obtained:

[0124] ;

[0125] when When the value suddenly increases from 1 to more than 1, it indicates that the ablation zone is splitting or multiple hot spots have appeared, reflecting behaviors such as material edge fracture and erosion.

[0126] To detect potential central cryogenic voids, the system further sets a cryogenic threshold within the high-temperature region. Construct an internal low-temperature assembly:

[0127] ;

[0128] By filtering out connected components completely surrounded by high-temperature regions using geometric topological relationships, the effective void area is obtained. The percentage of void area is defined as:

[0129] ;

[0130] When the area of ​​internal low-temperature cavities accounts for When the area of ​​voids exceeds the preset threshold, it is determined that there are obvious void structures in the local area of ​​the material, corresponding to a perforated or near-perforated state.

[0131] To assess the drift of the ablation spot location, the system calculates the centroid of the high-level region:

[0132] ;

[0133] And define the centroid displacement between adjacent frames:

[0134] ;

[0135] It is used to detect whether hotspot locations are experiencing rapid drift or overall structural shift.

[0136] Based on temporal and spatial characteristics, an intermediate feature vector is constructed. This is used to characterize the state evolution of a target during laser irradiation. Taking the k-th frame as an example, a vector can be constructed:

[0137] ;

[0138] At the process level, a compressed process vector can be constructed:

[0139] ;

[0140] ;

[0141] The aforementioned vector contains intensity features representing "how hot," duration features representing "how long it has been hot," and spatial morphological features representing "what the ablation site looks like." According to... or The damage stage is determined by the relationship between the damage and a preset threshold.

[0142] According to or The damage stage is determined based on the relationship between the vector components and preset thresholds, using logical judgment based on the combination relationship between the aforementioned vector components and multiple sets of thresholds. For example:

[0143] (1) Stage of no significant damage or slight heating

[0144] ;

[0145] ;

[0146] in, The critical temperature used to characterize the initiation of significant thermal damage or ablation in a material. Used to characterize whether a stable and sustainable ablation zone forms in a high-temperature region.

[0147] (2) Stable ablation stage

[0148] When the process vector is satisfied, the target is determined to have entered the stable ablation stage, indicating that the material has withstood high temperatures for a relatively long period of time but has not yet shown obvious structural damage.

[0149] ;

[0150] ;

[0151] ;

[0152] ;

[0153] in, To stabilize the ablation duration threshold, The threshold value is for near-circular boundaries.

[0154] (3) Hollow-type perforation or penetrating damage stage

[0155] ;

[0156] ;

[0157] ;

[0158] in, The severe ablation time threshold characterizes the duration of high-intensity ablation. The threshold temperature for severe ablation. The perforation threshold is the proportion of voids that can be stably distinguished.

[0159] This implementation method constructs a multi-source decision model by integrating threshold rule output, DTW pattern similarity, and HMM state probability. It comprehensively considers multi-dimensional information such as sudden events, time-series trends, and state evolution to achieve real-time, continuous, and highly reliable determination of damage stages such as heating, ablation, and perforation, and generates quantitative damage levels.

[0160] The physical threshold rule is used to detect key events such as sudden temperature drops, abrupt changes in high-temperature regions, and void formation; the temperature time-series trend matching is used to identify the similarity between temperature evolution and preset damage patterns; the state transition model is used to constrain the damage stages to evolve in the physical order of heating, ablation, structural damage, and perforation, so as to avoid non-physical state jumps.

[0161] Compared with traditional single-threshold methods, this implementation has the ability to perform joint temporal and spatial analysis, strong adaptability to extremely weak signals and long-distance environments, high robustness of the three complementary models, stable operation under high background light, atmospheric disturbance, and smoke and dust obstruction, and extreme sensitivity to pre-perforation signs (cavity expansion + peak collapse), which are not available in existing laser damage assessment technologies.

[0162] Implementation Method Five: The difference between this implementation method and Implementation Method Four is that the use of DTW for temperature time series trend matching is as follows:

[0163] DTW was used to calculate the matching distance between the actual temperature field of the target 3 and the preset temperature curves for normal heating, stable ablation, and rapid drop in perforation, respectively, and the corresponding similarity index was obtained. Based on the similarity index, the category to which the current temperature evolution trend belongs was determined.

[0164] To overcome the problem of "misjudgment caused by short-term fluctuations," this implementation uses Trend-Driven Weather (DTW) to perform trend matching on the temperature time series. The DTW input structure takes a fixed window length m for the maximum temperature series, forming:

[0165] ;

[0166] like Figure 9 As shown, three types of "standard curves" were established using a database constructed by the materials experiment institute: normal heating mode, stable ablation mode, and perforation mode. Figure 10 As shown, the curves and temperature field images of the stable ablation mode and perforation mode stages are presented using carbon fiber materials as an example.

[0167] Distances of the current sequence to the three patterns:

[0168] ;

[0169] Even with different heating rates and time scales, DTW can align key trend points (peaks, drop segments, etc.), thus enabling it to identify "perforation trends" rather than relying on single-frame mutations.

[0170] This implementation uses the Dynamic Time Warping (DTW) algorithm to perform pattern matching on the temperature peak sequence. By calculating the matching distance between the current temperature time series and the preset "normal heating mode", "stable ablation mode" and "sudden drop before perforation mode", a pattern similarity index characterizing the temperature evolution trend is obtained, which is used to improve the stability of damage stage determination under noise interference and time scale differences.

[0171] Implementation Method Six: The difference between this implementation method and Implementation Method Four is that the calculation of the damage state of the target 3 under test during the laser radiation process using Hidden Markov Model (HMM) is specifically as follows:

[0172] The damage state sequence of the target 3 under test during laser irradiation is established based on the Hidden Markov Model (HMM). The damage state includes the heating stage, the stable ablation stage, and the perforation stage. The damage state of the target 3 under test during laser irradiation is calculated based on the observation probability of the intermediate feature vector.

[0173] This implementation performs damage stage inference based on a Hidden Markov Model (HMM), which is the "core logic layer" of the system and is used to identify the various stages of laser damage:

[0174] The input to an HMM is a state vector: ;

[0175] HMM output: ;

[0176] in, This represents the most likely state of damage. For each possible state ( , , ), This is the heating stage. To stabilize the ablation stage, This is the perforation stage.

[0177] State transition models reflect the physical laws of materials: .

[0178] The fusion decision of the three models: "Jumping back" to lower stages is not allowed (materials will not recover on their own), and... Direct jump (must go through an ablation stage).

[0179] The observation model is determined by features and jointly judged by multiple frames. It suppresses noise and HMM can avoid misjudging the short-term temperature drop caused by smoke and dust obstruction as a perforation due to brightness flicker in a certain frame. HMM is the highest level of stability guarantee mechanism.

[0180] This implementation method establishes the implicit state sequence of the target material during laser irradiation based on a Hidden Markov Model (HMM), including the heating stage, the stable ablation stage, and the perforation stage, and then uses eigenvectors... The observation probability is used to calculate the most likely damage state in the current frame. The physical evolution order of the damage process is constrained by the state transition matrix to avoid non-physical transitions caused by instantaneous noise, thereby improving the robustness of damage stage identification.

[0181] Implementation Method Seven: This implementation method consists of three specific embodiments based on Implementation Methods One through Six.

[0182] Example 1: Laser Damage Assessment Process Based on Multispectral Temporal Images

[0183] like Figure 2 As shown, this embodiment provides a laser damage assessment process based on multispectral time-series images.

[0184] After the evaluation begins, the system acquires multiple consecutive frames of multispectral modulated images and automatically selects the laser-affected area based on the radiation distribution. Subsequently, features of the temperature field are extracted in both the time and spatial domains.

[0185] The time domain is used to calculate the average temperature, maximum temperature, and rate of temperature change of the region, which is used to identify abnormal temperature rises or sudden temperature drops.

[0186] The spatial domain extracts the area, shape factor, edge evolution, and internal voids of the high-temperature region to determine signs of ablation propagation or structural rupture.

[0187] The aforementioned time-domain and spatial-domain features are fused to construct an intermediate feature vector. Based on this feature vector, the damage stages of the target material are determined, including the heating stage, the stable ablation stage, the suspected fracture stage, and the perforation stage.

[0188] When a sudden drop in temperature occurs or a low-temperature cavity forms inside a high-temperature zone, the frame is marked as a suspected perforation or material peeling feature to enhance the reliability of the stage judgment.

[0189] The system ultimately fuses temperature anomaly events, spatial morphological features, and stage determination results from multiple sources to form a real-time damage level assessment of the target, outputting data including damage stage, perforation risk, and evolution trend of high-temperature regions. This process achieves automated and intelligent laser damage recognition based on multispectral time-series images, and can be used for risk control and effect evaluation during laser strikes.

[0190] Example 2: A Damage Stage Identification Method Based on Threshold-DTW-HMM

[0191] like Figure 5 and Figure 6 As shown, this embodiment proposes a damage stage identification method that integrates temporal-spatial temperature features. The system first extracts multi-dimensional features from the continuously inverted temperature field, including maximum temperature, average temperature, heating rate, area of ​​high-temperature regions, edge fragmentation, proportion of void area, and hotspot centroid drift, to construct a feature vector characterizing material heating, ablation, and structural changes. Subsequently, based on a physical threshold model, key events such as sudden temperature drops, rapid area expansion, cavity formation, and edge fragmentation are rapidly detected, enabling early identification of suspected perforation or significant ablation behavior.

[0192] This embodiment proposes a damage stage identification method that integrates temporal-spatial temperature features. The system first extracts multi-dimensional features from the continuously retrieved temperature field, including maximum temperature, average temperature, heating rate, area of ​​high-temperature regions, edge fragmentation, void area ratio, and hotspot centroid drift, to construct a feature vector characterizing material heating, ablation, and structural changes. Subsequently, based on a physical threshold model, key events such as sudden temperature drops, rapid area expansion, cavity formation, and edge fragmentation are rapidly detected, enabling early identification of suspected perforation or significant ablation behavior.

[0193] To characterize the evolution trend of different laser-induced processes, this embodiment uses the Dynamic Time Warping (DTW) algorithm to align with three typical temperature curves: normal heating, stable ablation, and rapid drop in perforation, respectively, to obtain corresponding similarity indices, which are used to reflect the consistency between the temperature evolution trajectory and the typical patterns.

[0194] Simultaneously, a Hidden Markov Model (HMM) was constructed, incorporating three implicit states: heating, stable ablation, and perforation. The physical unidirectionality of the damage process was constrained by the state transition matrix, and the most probable damage stage at the current moment was determined using a forward algorithm. Time-series temperature field images were also provided throughout the evaluation process.

[0195] Example 3: Turntable linkage and weapon closed-loop control

[0196] like Figure 3 and Figure 4As shown in this embodiment, the laser weapon system outputs real-time information such as laser emission status, operating mode, pointing angle (pitch / azimuth), laser power, and emission timestamp, which is transmitted to the temperature measurement system via the linkage control module. The linkage control module analyzes the above status variables, generates an optical axis pointing command, and sends it to the photoelectric turntable system, enabling the turntable to perform pitch adjustment, azimuth tracking, and attitude synchronization correction. Through this process, the imaging optical path can maintain real-time consistency before and after the laser weapon's action, achieving "where the weapon points, the imaging points" optical axis follow-up linkage.

[0197] After the telephoto optical imaging system focuses the target radiation onto the multispectral detection camera, the nine-channel image generated by the camera is sent to the temperature inversion and damage module. This module extracts features such as hotspot locations, spot centroids, and temperature center regions, and calculates the optical axis offset. The linkage control module then sends this offset as a feedback signal back to the photoelectric turntable system, causing the turntable to perform secondary fine-tuning correction. Through this image feedback alignment mechanism, the system can maintain a high degree of overlap between the laser application point and the imaging center even with external disturbances, target movement, or turntable errors, thereby ensuring the stability and accuracy of temperature inversion and damage analysis.

[0198] After the system completes temperature inversion and damage assessment, if it determines that the target has entered a high-risk state of severe ablation or perforation, this embodiment can transmit the corresponding results back to the laser weapon system as a control basis for power adjustment or termination of firing. Through this closed-loop mechanism, the accuracy of the assessment can be guaranteed while avoiding excessive action, improving the safety and effectiveness control capability of the laser strike process, and realizing a coordinated closed loop between weapon pointing, optical axis monitoring, and damage assessment.

[0199] The laser damage assessment method and system based on long-distance multispectral imaging temperature measurement proposed in this invention have been described in detail above. Specific examples have been used to illustrate the principle and implementation of this invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A laser damage assessment system based on long-distance multispectral imaging temperature measurement, characterized in that, include: The long-distance multispectral damage field acquisition module is used to acquire the multispectral radiation image of the target (3) during laser irradiation. The long-range multispectral damage field acquisition module includes a long focal length optical system (6) and a control and interconnection system (11), wherein the long focal length optical system (6) is used to image the target (3) to be tested; The control and interconnection system (11) is used to acquire key parameters of the laser weapon system (2) in real time, and adjust the attitude of the photoelectric turntable (4) synchronously according to the key parameters so that the imaging optical axis of the long focal length optical system (6) is consistent with the laser irradiation direction. Based on the damage status assessment results, the system controls the start-up and shutdown of the laser weapon system (2) and the adjustment of key parameters in real time. The temperature and emissivity joint inversion module is used to invert the true temperature field of the target (3) based on the current multispectral radiation image; The laser damage status assessment module is used to assess the damage status in real time based on the real temperature field of the target (3) at the current moment, and transmit the damage status assessment results to the control and interconnection system (11). A laser weapon system (2) is used to generate a high-energy laser beam and irradiate the surface of the target (3) to be tested; Optical turntable (4) is used to support long focal length optical system (6). A laser damage assessment method based on long-distance multispectral imaging thermometry, the method being implemented using a laser damage assessment system, includes the following steps: Step 1: Real-time acquisition of multispectral radiation images of the target (3) during laser irradiation; Step 2: Based on the current multispectral radiation image, the true temperature field of the target (3) to be measured is obtained by inversion; The key parameters of the laser weapon system (2) and the real temperature field of the target (3) are acquired in real time using the control and interconnection system (11). The attitude of the photoelectric turntable (4) is adjusted synchronously according to the acquired key parameters and real temperature field so that the imaging optical axis of the long focal length optical system (6) is consistent with the laser irradiation direction. Step 3: Extract the time and space features of the real temperature field of the current target (3), and perform damage assessment based on the time and space features to obtain the damage assessment results; Step 3, which assesses the damage status based on temporal and spatial characteristics, specifically involves: Based on time and space characteristics, combined with the corresponding preset thresholds, a coarse judgment is made on the current damage state. DTW is used for temperature time series trend matching, and HMM is used to calculate the damage state of the target (3) during the laser radiation process. The coarse judgment result of the damage state, the temperature time series trend matching result and the damage state of the target (3) during the laser radiation process are comprehensively judged to obtain the final damage state assessment result.

2. The laser damage assessment system based on long-distance multispectral imaging temperature measurement according to claim 1, characterized in that, The long-distance multispectral damage field acquisition module also includes a broadband modulation optical window (5) and a multispectral modulation detector (7). The broadband modulation optical window (5) is used to adjust the transmittance of different bands, thereby suppressing the radiation peak of high-temperature targets, separating the sensitivity between channels, and cutting off the laser band. The multispectral modulation detector (7) is used to acquire multi-channel radiation images of the target under a single-frame exposure.

3. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The system also uses the control and interconnection system (11) to obtain the real temperature field of the target (3) at the current moment and to fine-tune and compensate the attitude of the photoelectric turntable (4).

4. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The system also includes an external computing processing unit (1), a SOC preprocessing unit (8), a serial port channel (9), and a time-series multispectral data transmission interface (10). The SOC preprocessing unit (8) is used to preprocess the multispectral radiation image; A serial port channel (9) is used to connect the control and interconnection system (11) to the laser weapon system (2); A time-series multispectral data transmission interface (10) is used to transmit the preprocessed data to an external computing unit (1). An external computing unit (1) is used to perform inversion and solution on the preprocessed data to obtain the true temperature field of the target (3) to be measured.

5. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The time characteristic includes at least the regional average temperature. Maximum temperature Temperature rise rate and duration of high temperature ; The spatial features include at least the area of ​​the high-temperature region. Changes, boundary morphology Fragmentation degree, percentage of internal low-temperature cavities Changes in the position of the centroid of hotspots .

6. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The coarse assessment of the current damage state involves constructing intermediate feature vectors and process vectors based on temporal and spatial characteristics, and determining the damage stage based on the relationship between the intermediate feature vectors, process vectors, and preset thresholds.

7. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The use of DTW for temperature time-series trend matching is specifically as follows: DTW was used to calculate the matching distance between the actual temperature field of the target (3) and the preset normal heating, stable ablation and perforation drop temperature curves, respectively, and the corresponding similarity index was obtained. Based on the similarity index, the category to which the current temperature evolution trend belongs was determined.

8. The laser damage assessment system based on long-distance multispectral imaging thermometry according to claim 1, characterized in that, The calculation of the damage state of the target (3) during laser radiation using Hidden Markov Model (HMM) is as follows: The damage state sequence of the target (3) under test during laser irradiation is established based on HMM. The damage state includes heating stage, stable ablation stage and perforation stage. The damage state of the target (3) under test during laser irradiation is calculated based on the observation probability of the intermediate feature vector.

Images