Space target pBRDF model construction method based on improved SBA algorithm

Abstract

The invention discloses a space target pBRDF model construction method based on an improved SBA algorithm, and relates to the technical field of space target polarization measurement and modeling, and the method comprises the steps: preliminarily building a pBRDF model of a target material corresponding to a to-be-measured sample according to a pBRDF theory, a light elastic scattering theory and a blackbody radiation theory; the method comprises the following steps of: representing the actually measured # imgabs0 # value of a to-be-measured sample at each angle in a simulated atmospheric environment by acquiring a polarization characteristic measurement result of the to-be-measured sample; an improved SBA algorithm is adopted to perform parameter inversion according to the measured # imgabs 1 # value; and substituting a parameter inversion result into the pBRDF model to calculate an error between a simulation value and a measured value of the model, and selecting whether to output the pBRDF model or not according to error precision to obtain a final pBRDF model. According to the invention, the accuracy of polarization detection of the space target can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of space target polarization measurement and modeling, and particularly to a method for constructing a space target pBRDF model based on an improved SBA algorithm. Background Art

[0002] The research results of a large number of space target polarization detections show that the polarization detection technology has advantages that cannot be compared with traditional optical detection methods such as intensity and spectrum in enhancing the detection and recognition ability of dim space targets, diagnosing the state of targets in the space environment, and reducing the influence of atmospheric effects on space observations. It is very likely to become a new means for effective monitoring and recognition of future spacecraft and space debris. Polarization detection of space targets requires obtaining the polarization characteristics of target materials at various angles in advance, and pBRDF (p-Bidirectional Reflectance Distribution Function) is very suitable for studying the multi-angle polarization characteristics of target materials.

[0003] However, the existing polarization measurement devices do not fully consider the influence of the atmospheric environment on the measurement accuracy; at the same time, the existing parameter inversion methods based on pBRDF have problems of over-reliance on initial values and poor robustness. The multi-angle polarization degree (Degree Of Polarization, DOP) data and parameter inversion methods have a great influence on the accuracy of the pBRDF model, resulting in a low accuracy of the pBRDF model established by the existing methods, and further affecting the accuracy of space target polarization detection. Summary of the Invention

[0004] The purpose of the present application is to provide a method for constructing a space target pBRDF model based on an improved SBA algorithm, which can improve the accuracy of measuring the polarization characteristics of space targets.

[0005] To achieve the above purpose, the present application provides the following solutions.

[0006] In the first aspect, the present application provides a method for constructing a space target pBRDF model based on an improved SBA algorithm, and the method for constructing a space target pBRDF model based on an improved SBA algorithm includes the following steps.

[0007] Preliminarily construct a pBRDF model, and obtain the measurement results of the polarization characteristics of the sample to be measured; the measurement results of the polarization characteristics include the values of the sample to be measured at various angles.

[0008] According to the polarization characteristic measurement results, an improved SBA algorithm is used for parameter inversion to obtain parameter inversion results; in the improved SBA algorithm, the Tent chaotic mapping algorithm is used to initialize the positions of the mother plants, and the simulated annealing algorithm is used to update the positions of the stolons of the mother plants.

[0009] Substitute the parameter inversion results into the pBRDF model to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error precision to obtain the final pBRDF model; wherein, the final pBRDF model is used to analyze the infrared polarization characteristics of the sample to be measured at multiple angles.

[0010] Optionally, a pBRDF model is initially constructed, which specifically includes the following steps.

[0011] According to the pBRDF theory, the elastic light scattering theory, and the blackbody radiation theory, a pBRDF model of the target material is initially established and the formula is determined; the target material refers to the material corresponding to the sample to be measured.

[0012] Optionally, according to the polarization characteristic measurement results, an improved SBA algorithm is used for parameter inversion to obtain parameter inversion results, which specifically includes the following steps.

[0013] Define the initial parameters of the SBA algorithm; the initial parameters include the number of inversion parameters, the initial mother plant population, the maximum iteration times of the strawberry optimization algorithm and the Tent chaotic mapping algorithm, and the maximum iteration times of the simulated annealing algorithm.

[0014] According to the initial parameters, use the Tent chaotic mapping algorithm to initialize the positions of the mother plants to obtain the initialized positions of the mother plants.

[0015] According to the initialized positions of the mother plants, randomly generate the daughter roots and stolons of the mother plants to obtain a plant propagation matrix; in each mother plant in the plant propagation matrix, a daughter root and a stolon are randomly generated.

[0016] According to the plant propagation matrix, use the simulated annealing algorithm to update the positions of the stolons to obtain the updated positions of the stolons.

[0017] Based on the updated positions of the stolons, perform fitness calculation with the goal of minimizing the objective function solution, and use the calculated fitness values and the roulette wheel algorithm to select the mother plant population for the next iteration.

[0018] Determine whether to output the parameter inversion results according to whether any of the termination conditions is met; the parameter inversion results include the optimal inversion values of multiple parameters to be inverted, and the multiple parameters to be inverted include σ, n, k, k s 、km and k v , where σ is the roughness, n is the real part of the complex refractive index, k is the imaginary part of the complex refractive index, k s is the specular reflectance, k m is the diffuse reflectance, k v is the volume scattering rate.

[0019] The inversion accuracy is represented by the average relative error between the inversion values and the measured values of the parameters σ, n, and k to be inverted.

[0020] Optionally, substitute the parameter inversion result into the pBRDF model to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy to obtain the final pBRDF model, which specifically includes the following steps.

[0021] Substitute the optimal inversion value corresponding to each parameter to be inverted in the parameter inversion result into the formula to calculate the simulation value of the pBRDF model.

[0022] Use an evaluation function to evaluate the simulation value of the pBRDF model, and determine the final pBRDF model according to the evaluation result.

[0023] Optionally, use an evaluation function to evaluate the simulation value of the pBRDF model, and determine the final pBRDF model according to the evaluation result, which specifically includes the following steps.

[0024] Use the MAE function to calculate the MAE value between the simulation value of the pBRDF model and the value obtained from experimental measurement; the value obtained from experimental measurement refers to the value in the measured polarization characteristic measurement result.

[0025] Determine the final pBRDF model according to the MAE value and a preset threshold.

[0026] Optionally, determine the final pBRDF model according to the MAE value and a preset threshold, which specifically includes the following steps.

[0027] When the MAE value is less than 5%, it is determined that the simulation value of the pBRDF model is within a reasonable threshold range, and the current pBRDF model is used as the final pBRDF model.

[0028] When the MAE value is greater than or equal to 5%, it is determined that the simulation of the pBRDF model value is not within a reasonable threshold range. At this time, improve the pBRDF model and return to the step of "initializing the position of the mother plant using the Tent chaotic mapping algorithm according to the initial parameters to obtain the initialized position of the mother plant" until the MAE value is less than 5%.

[0029] Optionally, obtain the measurement results of the polarization characteristics of the sample to be measured, which specifically includes the following steps.

[0030] Measure the sample to be measured using a space target infrared polarization characteristic measurement device to obtain the measurement results of the polarization characteristics of the sample to be measured.

[0031] The space target infrared polarization characteristic measurement device includes a gas generation system, a closed movement system, an electric stage, a light source system, a polarization acquisition system, a control system, and a computer.

[0032] The gas generation system is internally connected to the closed movement system through a pipeline. The electric stage is arranged inside the closed movement system. The light source system and the polarization acquisition system are arranged inside the closed movement system. The gas generation system, the electric stage, the light source system, and the polarization acquisition system are all connected to the control system, and the control system is also connected to the computer.

[0033] The gas generation system is used to generate a mixed gas and fill the mixed gas into the closed movement system through the pipeline; the mixed gas includes aerosol and water vapor.

[0034] The electric stage is used to place the sample to be measured and rotate the sample to be measured.

[0035] The light source system is used to emit an infrared beam to the sample to be measured.

[0036] The polarization acquisition system is used to acquire the polarization image of the sample to be measured.

[0037] The closed movement system is used to provide a closed environment for the mixed gas and move the light source system in the closed environment to achieve multi-angle emission of the infrared beam, and at the same time move the polarization acquisition system in the closed environment to achieve multi-angle acquisition of the polarization image.

[0038] The control system is used to receive the instructions issued by the computer and respectively control the working states of the gas generation system, the electric stage, the light source system, and / or the polarization acquisition system according to the instructions; and acquire the polarization image and transmit it to the computer.

[0039] The computer is used to analyze and process the polarization image to obtain the measurement result of the polarization characteristics of the sample to be measured.

[0040] Optionally, the enclosed moving system includes a semi-circular sealed box and a semi-circular guide rail.

[0041] The center of the top of the semi-circular sealed box is communicated with the gas generation system through the pipeline, and the electric stage is arranged directly below the center of the top of the semi-circular sealed box.

[0042] The semi-circular guide rail is laid on the inner surface of the semi-circular sealed box, and the semi-circular guide rail is connected to the control system.

[0043] The light source system and the polarization acquisition system are slidably arranged on the semi-circular guide rail.

[0044] Optionally, the semi-circular guide rail includes a first guide rail and a second guide rail.

[0045] The light source system is slidably arranged on the first guide rail, and the polarization acquisition system is slidably arranged on the second guide rail.

[0046] First angle marks are arranged on both the first guide rail and the second guide rail. The first angle marks are 0° to 90°, and the graduation value of the first angle marks is 1°.

[0047] The first angle marks of 0° on the first guide rail and the second guide rail are both located at one end close to the top of the semi-circular sealed box.

[0048] Optionally, the electric stage is provided with a second angle mark. The second angle mark is 0° to 360°, the graduation value of the second angle mark is 1°, and the second angle mark of 0° on the electric stage faces the first guide rail.

[0049] According to the specific embodiments provided by the present application, the following technical effects are disclosed in the present application.

[0050] The present application provides a method for constructing a spatial target pBRDF model based on an improved SBA algorithm. First, a pBRDF model is initially constructed and the measurement result of the polarization characteristics of the sample to be measured is obtained, including the Values; then, according to the polarization characteristic measurement results, an improved SBA algorithm is used for parameter inversion to obtain the parameter inversion results. Among them, in the improved SBA algorithm, the Tent chaotic mapping algorithm is used to initialize the positions of the parent plants, and the simulated annealing algorithm is used to update the positions of the stolons of the parent plants. Then, the parameter inversion results are substituted into the preliminarily constructed pBRDF model to calculate the error between the model simulation value and the measured value, and whether to output the pBRDF model is selected according to the error accuracy to obtain the final pBRDF model. This final pBRDF model is used to analyze the long-wave infrared polarization characteristics of the sample to be measured at multiple angles, which can improve the accuracy of polarization detection of space targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 FIG. is a schematic structural diagram of a space target infrared polarization characteristic measurement device provided by an embodiment of the present application.

[0053] Figure 2 FIG. is a schematic flowchart of a method for constructing a space target pBRDF model based on an improved SBA algorithm provided by an embodiment of the present application.

[0054] Figure 3 FIG. is a schematic flowchart of parameter inversion using the improved SBA algorithm provided by an embodiment of the present application.

[0055] Figure 4 FIG. is a schematic flowchart of the working process of a model creation module provided by an embodiment of the present application.

[0056] Figure 5 FIG. is a schematic flowchart of the working process of an image signal acquisition module provided by an embodiment of the present application.

[0057] Figure 6 FIG. is a schematic flowchart of initializing the positions of the parent plants provided by an embodiment of the present application.

[0058] Figure 7 FIG. is a schematic flowchart of generating a new parent population provided by an embodiment of the present application.

[0059] Figure 8 FIG. is a three-dimensional coordinate diagram of the DOLP generated by the pBRDF model provided by an embodiment of the present application with respect to the azimuth angle and the zenith angle.

[0060] Figure 9The polar plot generated by the pBRDF model provided by an embodiment of the present application.

[0061] Reference numerals: 1 - Gas generation system; 2 - Semi-circular closed box; 3 - Semi-circular guide rail; 4 - Electric stage; 5 - Light source system; 6 - Long-wave infrared polarization camera; 7 - Control system; 8 - Computer. Detailed implementation manners

[0062] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0063] Currently, existing polarization measurement devices do not consider the influence of the atmospheric environment on the measurement accuracy, and the measured multi-angle values are the basic data for parameter inversion and constructing the pBRDF model. Moreover, traditional parameter inversion methods mainly include the nonlinear least squares method and the particle swarm optimization algorithm. Among them, the nonlinear least squares method is sensitive to the initial value, requires a large number of experimental data, and has a large amount of calculation. While the particle swarm optimization algorithm may not be accurate enough when dealing with some specific problems, and the performance and results of the algorithm may be affected by random factors, resulting in low robustness.

[0064] Based on this, this embodiment aims to provide a method for constructing a pBRDF model of a space target based on an improved SBA (Strawberry optimization) algorithm. Among them, the pBRDF model is established based on the pBRDF theory, the theory of elastic scattering of light, and the blackbody radiation theory. As an infrared polarization characteristic model of a space target, the pBRDF model can be used to further analyze the infrared polarization characteristics of a space target at multiple angles. Infrared refers to the band of 8 - 14 μm. First, use a space target infrared polarization characteristic measurement device to accurately measure the infrared polarization characteristics of a space target in a simulated atmospheric environment, obtain more accurate, reliable, and multi-angle polarization characteristic measurement results, and at the same time construct a preliminary pBRDF model. Then, apply the polarization characteristic measurement results to the process of improving the pBRDF model. Through parameter inversion and evaluation verification, the final pBRDF model is obtained. This method can effectively solve the problems of traditional parameter inversion methods that rely on the initial value and have poor robustness, and improve the disadvantages of the traditional SBA algorithm with slow convergence speed and easy to fall into local optimal solutions, and can improve the accuracy of parameter inversion for metal objects.

[0065] To make the above objects, features, and advantages of the present application more apparent and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] As Figure 1 shown, this embodiment provides a device for measuring the infrared polarization characteristics of a space target. The device for measuring the infrared polarization characteristics of a space target includes a gas generation system 1, a closed movement system, an electric stage 4, a light source system 5, a polarization acquisition system 6, a control system 7, and a computer 8.

[0067] Among them, the gas generation system 1 is connected to the inside of the closed movement system through a pipeline. The electric stage 4 is arranged inside the closed movement system. The light source system 5 and the polarization acquisition system 6 are arranged inside the closed movement system. The gas generation system 1, the electric stage 4, the light source system 5, and the polarization acquisition system 6 are all connected to the control system 7, and the control system 7 is also connected to the computer 8.

[0068] In this embodiment, the gas generation system 1 is used to generate a mixed gas and fill the mixed gas into the inside of the closed movement system through the pipeline. The mixed gas includes aerosol and water vapor. The gas generation system 1 adopts the HRF-4B series gas generation system of Suzhou Hongrui Purification Technology Co., Ltd. to generate mixed gases of different proportions of aerosol and water vapor.

[0069] In this embodiment, the light source system 5 is used to emit an infrared beam to the sample to be measured. The light source system 5 is a blackbody light source. The blackbody light source adopts the SLS303 series light source of Thorlabs Company, which can generate a beam of 550nm - 15μm, the output beam power is 4.5w, and the output beam diameter is 50mm.

[0070] In this embodiment, the polarization acquisition system 6 is used to acquire the polarization image of the sample to be measured. The polarization acquisition system is a long-wave infrared polarization camera. The long-wave infrared polarization camera adopts the GWPL0318X2A model polarization camera of North Guangwei. It is a split focal plane type polarization camera, and the response band is 8 - 14μm, which can be used to obtain the long-wave infrared polarization information of an object.

[0071] In this embodiment, the closed movement system is used to provide a closed environment for the mixed gas and make the light source system 5 move in the closed environment to achieve multi-angle emission of the infrared beam, and at the same time make the polarization acquisition system 6 move in the closed environment to achieve multi-angle acquisition of the polarization image.

[0072] In this embodiment, the control system 7 is configured to receive instructions sent by the computer 8 and respectively control the operating states of the gas generation system 1, the electric stage 4, the light source system 5, and / or the polarization acquisition system 6 according to the instructions; and acquire the polarization image and transmit it to the computer 8.

[0073] In this embodiment, the computer 8 is configured to analyze and process the polarization image to obtain the measurement result of the polarization characteristics of the sample to be measured. Among them, the measurement result of the polarization characteristics includes the values at various angles of the sample to be measured. In this embodiment, the measurement result of the infrared polarization characteristics of the space target obtained by the space target infrared polarization characteristic measurement device can be used for parameter inversion and constructing and outputting a pBRDF model to further analyze the infrared polarization characteristics of the space target at multiple angles in detail.

[0074] In this embodiment, the computer 8 is built-in with contrast software for extracting the degree of polarization of the polarization image.

[0075] In this embodiment, the enclosed moving system includes a semi-circular sealed box 2 and a semi-circular guide rail 3. Among them, the center of the top of the semi-circular sealed box 2 is communicated with the gas generation system 1 through the pipeline, and the electric stage 4 is arranged directly below the center of the top of the semi-circular sealed box 2. The semi-circular guide rail 3 is laid on the inner surface of the semi-circular sealed box 2, and the semi-circular guide rail 3 is connected to the control system 7. The light source system 5 and the polarization acquisition system 6 are slidably arranged on the semi-circular guide rail 3.

[0076] In this embodiment, the semi-circular guide rail 3 is a customized model, which is divided into left and right sides. The left semi-circular guide rail 3, that is, the first guide rail, is used for dynamically measuring the angles between 0° and 90°. The right semi-circular guide rail 3, that is, the second guide rail, is used for dynamically measuring the angles between 0° and 90°.

[0077] Among them, the light source system 5 is slidably arranged on the first guide rail, that is, the light source system 5 can slide along the first guide rail; the polarization acquisition system 6 is slidably arranged on the second guide rail, that is, the polarization acquisition system 6 can slide along the second guide rail.

[0078] In this embodiment, first angle markings are provided on both the first guide rail and the second guide rail. The first angle markings are 0° to 90°, and the graduation value of the first angle markings is 1°. Among them, the 0° first angle markings of the first guide rail and the second guide rail are both located at one end close to the top of the semi-circular sealed box 2.

[0079] In this embodiment, the electric stage 4 is used to place the sample to be measured and rotate the sample to be measured. The electric stage 4 adopts the DDR100 series electric stage of Thorlabs, measures the angle between 0° and 360° dynamically, and is provided with a second angle mark. The second angle mark is 0° to 360°, and the graduation value of the second angle mark is 1°. Among them, the second angle mark of 0° is directly opposite to the first guide rail.

[0080] In this embodiment, the control system 7 includes a gas generation control system, a light source control system, an imaging device control system, and an electric turntable control system. Among them, the gas generation control system is used to control the working state and working power of the gas generation system 1, etc. The light source control system is used to control the working state and working power of the light source system 5, etc. The imaging device control system is used to control the working state and working power of the long-wave infrared polarization camera 6, etc. The electric turntable control system is used to control the working state and rotation angle of the electric stage 4, etc.

[0081] In this embodiment, the sample to be measured is a circular wafer made of three common metal materials for spacecraft, namely aluminum alloy, titanium alloy, and stainless steel. Its diameter is equal to the beam aperture, both of which are 50 mm.

[0082] In an exemplary embodiment, a method for constructing a pBRDF model of a space target based on an improved SBA algorithm is provided. The method for constructing a pBRDF model of a space target based on an improved SBA algorithm can apply the above-mentioned space target infrared polarization characteristic measurement device to measure the polarization characteristic measurement result, and perform parameter inversion and verification based on the polarization characteristic measurement result, so as to determine the final pBRDF model. As Figure 2 shown, the method for constructing a pBRDF model of a space target based on an improved SBA algorithm mainly includes the following steps.

[0083] Step S1: Initially construct a pBRDF model and obtain the polarization characteristic measurement result of the sample to be measured.

[0084] In this embodiment, the space target infrared polarization characteristic measurement device can be used to measure the polarization characteristic measurement result of the sample to be measured, and obtain the value at each angle of the sample to be measured.

[0085] Step S2: According to the polarization characteristic measurement result, use the improved SBA algorithm to perform parameter inversion to obtain the parameter inversion result.

[0086] In this embodiment, compared with the traditional SBA algorithm, the main improvement of the improved SBA algorithm is to use the Tent chaotic mapping algorithm to initialize the position of the mother plant and use the simulated annealing algorithm to update the position of the stolon of the mother plant.

[0087] Step S3: Substitute the parameter inversion result into the pBRDF model to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy to obtain the final pBRDF model. The final pBRDF model can be used to further comprehensively and detailedly analyze the infrared polarization characteristics of the sample to be measured at multiple angles.

[0088] In this embodiment, step S1 initially constructs a pBRDF model, which specifically includes the following steps.

[0089] According to the pBRDF theory, the elastic light scattering theory, and the blackbody radiation theory, initially establish the pBRDF model of the target material and determine the formula; the target material refers to the material corresponding to the sample to be measured.

[0090] In this embodiment, step S2 performs parameter inversion using an improved SBA algorithm according to the polarization characteristic measurement result to obtain a parameter inversion result, which specifically includes the following steps.

[0091] Step S21: Define the initial parameters of the SBA algorithm; the initial parameters include the number of inversion parameters, the initial number of the parent plant population, the maximum number of iterations of the strawberry optimization algorithm and the Tent chaos mapping algorithm, and the maximum number of iterations of the simulated annealing algorithm.

[0092] Step S22: According to the initial parameters, use the Tent chaos mapping algorithm to initialize the position of the mother plant to obtain the initialized position of the mother plant.

[0093] Step S23: According to the initialized position of the mother plant, randomly generate the daughter roots and stolons of the mother plant to obtain a plant propagation matrix; for each mother plant in the plant propagation matrix, randomly generate a daughter root and a stolon.

[0094] Step S24: According to the plant propagation matrix, use the simulated annealing algorithm to update the position of the stolon to obtain the updated position of the stolon.

[0095] Step S25: Based on the updated position of the stolon, perform fitness calculation with the goal of minimizing the solution of the objective function, and at the same time use the fitness value and the roulette wheel algorithm to select the new parent population for the next iteration.

[0096] Step S26: Determine whether to output the parameter inversion result according to whether any termination condition is satisfied. The parameter inversion result includes the optimal inversion values of multiple parameters to be inverted. The multiple parameters to be inverted include σ, n, k, k s 、k m and k v ,where σ is the roughness, n is the real part of the complex refractive index, k is the imaginary part of the complex refractive index, ks is the specular reflectivity, k m is the diffuse reflectivity, k v is the volume scattering rate. Among them, the termination conditions are as follows.

[0097] (1) All three algorithms, namely the Tent chaos mapping, the strawberry optimization algorithm, and the simulated annealing algorithm, reach the maximum number of iterations.

[0098] (2) A set of optimal solutions appears such that the fitness value of the objective function is greater than 20.

[0099] Step S27: Express the inversion accuracy according to the average relative error between the inversion values and the measured values of the parameters to be inverted σ, n, and k. The relative error is the ratio of the absolute error value to the true value. Among them, the absolute error value is the absolute value of the difference between the measured value and the inversion value of the parameter to be inverted, and the true value is the measured value of the parameter to be inverted.

[0100] In this embodiment, in step S3, the parameter inversion result is substituted into the pBRDF model to calculate the error between the model simulation value and the measured value, and whether to output the pBRDF model is selected according to the error accuracy to obtain the final pBRDF model, which specifically includes the following steps.

[0101] Step S31: Substitute the optimal inversion values corresponding to each parameter to be inverted in the parameter inversion result into the formula to calculate the simulation value of the pBRDF model.

[0102] Step S32: Evaluate the simulation value of the pBRDF model using an evaluation function, and determine the final pBRDF model according to the evaluation result.

[0103] In this embodiment, in step S32, an evaluation function is used to evaluate the simulation value of the pBRDF model, and the final pBRDF model is determined according to the evaluation result, which specifically includes the following steps.

[0104] Step S321: Use the MAE (Mean Absolute Error) function to calculate the MAE value between the simulation value of the pBRDF model and the value obtained by experimental measurement. Among them, the value obtained by experimental measurement refers to the value in the measured polarization characteristic measurement result. For example, the value in the polarization characteristic measurement result of the sample to be measured using the above-mentioned space target infrared polarization characteristic measurement device.

[0105] Step S322: Determine the final pBRDF model according to the MAE value and a preset threshold.

[0106] In this embodiment, step S322 determines the final pBRDF model according to the MAE value and a preset threshold, which specifically includes the following two cases.

[0107] (1) When the MAE value is less than 5%, it is determined that the simulation value of the pBRDF model is within a reasonable threshold range, and the current pBRDF model is used as the final pBRDF model. When the MAE value is less than 5%, it is determined that the simulation value of the pBRDF model is within a reasonable threshold range, and the current pBRDF model is used as the final pBRDF model.

[0108] (2) When the MAE value is greater than or equal to 5%, it is determined that the simulation value of the pBRDF model is not within a reasonable threshold range. At this time, the model needs to be further improved and then return to step S22 "According to the initial parameters, use the Tent chaotic mapping algorithm to initialize the position of the mother plant to obtain the initialized position of the mother plant", re-initialize the position of the mother plant and obtain the corresponding parameter inversion result, and according to the newly obtained parameter inversion result, bring the parameter inversion result into the pBRDF model again to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy until its corresponding MAE value is less than 5%. When the MAE value is greater than or equal to 5%, it is determined that the simulation value of the pBRDF model is not within a reasonable threshold range. At this time, the model needs to be further improved and then return to step S22 "According to the initial parameters, use the Tent chaotic mapping algorithm to initialize the position of the mother plant to obtain the initialized position of the mother plant", re-initialize the position of the mother plant and obtain the corresponding parameter inversion result, and according to the newly obtained parameter inversion result, bring the parameter inversion result into the pBRDF model again to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy until its corresponding MAE value is less than 5%.

[0109] To make the technical solution of this embodiment clearer, the following will take an example to illustrate the specific structure of the device and the implementation steps of the method in detail.

[0110] A method for constructing a spatial target pBRDF model based on an improved SBA algorithm proposed in this embodiment, as Figure 3 shown, this method is divided into a polarization information acquisition stage, a parameter inversion stage and a verification stage. Among them, the polarization information acquisition stage uses a model creation module and an image signal acquisition module. Among them, as Figure 4 shown, the model creation module is mainly used to determine the pBRDF theory based on the specular reflection component, diffuse reflection component and volume scattering component, and initially construct the pBRDF model by comprehensively considering the pBRDF theory, the theory of elastic scattering of light and the theory of blackbody radiation, and obtain the formula. As Figure 5As shown in the figure, the image signal acquisition module is mainly used to start inflating with the gas generation system 1 and generate a light beam with a blackbody light source. The control system 7 is used to set the activity range and stepping angle of the long-wave infrared light source, the long-wave infrared polarization camera, and the electric stage 4. The light source on the first rail slides within the range of 20° to 60° at a stepping angle of 10°. Each time the light source steps, the electric stage 4 rotates one week at a stepping angle of 120°. Each time the light source and the electric stage 4 step, the long-wave infrared polarization camera on the second rail slides from 20° to 60° at a stepping angle of 2°, and at the same time takes pictures of the sample to be measured to collect polarization signals. Then, the polarization degree of the pictures is extracted using contrast software, and the polarization degree data is sorted out, that is, the value obtained by experimental measurement.

[0111] In this embodiment, the polarization information acquisition stage specifically includes the following steps.

[0112] Step 1: The gas generation system 1 starts to inflate. The control system 7 is used to set the ratio of aerosol and water vapor in the mixed gas generated by the gas generation system 1, and then starts to inflate the semi-circular closed box 2.

[0113] Among them, the main components of the inflated gas are water vapor and a small amount of aerosol, because these two components in the atmosphere have the greatest impact on the long-wave infrared polarization detection of space targets.

[0114] Step 2: The blackbody light source generates a light beam. The control system 7 is used to set the activity range and stepping angle of the long-wave infrared light source, the long-wave infrared polarization camera, and the electric stage 4. The long-wave infrared light source is located on the first rail, and the long-wave infrared polarization camera is located on the second rail. The light source on the first rail, that is, the long-wave infrared light source, and the long-wave infrared polarization camera on the second rail both slide within the range of the zenith angle of 20° to 60°, and the electric stage 4 rotates within 0° to 360°; among them, the 0° of the electric stage 4 is the position facing the first rail; the stepping angle of the long-wave infrared light source is 10°, the stepping angle of the long-wave infrared polarization camera 6 is 2°, and the stepping angle of the electric stage 4 is 120°.

[0115] Step 3: Control the long-wave infrared polarization camera 6 to start shooting. Each time the long-wave infrared light source steps, the electric stage 4 rotates one week. Each time the long-wave infrared light source and the electric stage 4 step, the long-wave infrared polarization camera 6 steps from 20° to 60°. A total of 5×4×21 = 420 polarization images can be obtained in one round of acquisition, numbered 001-420.

[0116] Step 4: Use the contrast software to extract the polarization degree of the photos. Input all the polarization images in the dataset into the computer 8, and obtain the polarization degree of the sample to be measured in each polarization image through the contrast software. Finally, obtain the polarization degree data of the sample to be measured at different zenith angles and different azimuth angles under different simulated atmospheric environments.

[0117] Compared with the traditional nonlinear least squares method and particle swarm optimization algorithm, under the same number of groups of experimental data, the improved SBA algorithm has higher accuracy and better robustness for the parameter inversion of metal objects, and is very suitable for parameter inversion of space targets, which are mainly made of metal materials and need to consider the influence of the atmospheric environment on their polarization characteristics. The SBA algorithm is a heuristic algorithm that describes how strawberry plants migrate from one place to another and produce multiple daughter roots and stolons. Strawberry plants use roots and stolons for local and global searches to discover life resources. Whether a location is resource-rich means the quality of the solution of the objective function. Moving the roots and stolons to a new location with rich resources to produce daughter plants is regarded as finding an optimal solution. Through continuous iteration, the optimal solution of the objective function can be obtained. The basic idea of the SBA algorithm in this embodiment is that the process of each mother root generating new daughter roots and stolons represents the inversion of a set of parameters, and the position of each new root and new stolon represents a set of inverted parameter values. The mother plant required for the next iteration is selected by combining the fitness value of the objective function and the roulette wheel algorithm. By continuously iterating until the predetermined termination condition is met, the final optimal parameter solution is obtained. At the same time, this embodiment introduces the Tent chaos mapping algorithm and simulated annealing algorithm to improve the traditional SBA algorithm, and improves the problems of slow convergence speed and easy to fall into local optimal solutions of the traditional SBA algorithm.

[0118] In this embodiment, a parameter inversion module is used in the parameter inversion stage, and the working process of the parameter inversion module specifically includes the following steps.

[0119] The parameters to be inverted in this embodiment include roughness σ, the real part n of the complex refractive index, the imaginary part k of the complex refractive index, the specular reflectivity k s , the diffuse reflectivity k m and the volume scattering rate k v .

[0120] Among them, in the parameter inversion algorithm, the position solutions of the daughter roots and stolons represent a set of unknown parameters to be inverted. Each mother plant simultaneously inverses 6 parameters to be inverted, and the optimal solution obtained is the optimal inversion value of the parameters to be inverted finally inverted.

[0121] Step 1: First, initially construct a pBRDF model according to the pBRDF theory, the theory of elastic scattering of light, and the theory of blackbody radiation, and then calculate the polarization degree , The formula is as follows.

[0122] 。

[0123] Among them, the known parameters are as follows: The Mueller matrix representing the mixture of aerosols and water vapor in the atmosphere, is the irradiance on the surface of the object, is the incident zenith angle, is the detection zenith angle, is the detection azimuth angle, is the Mueller matrix of the specular reflection component, 、 、 are all elements in, and represent the diffuse reflection component and the volume scattering component respectively, is the blackbody radiation formula of the object, is the shadow masking function, is the probability distribution function of the surface normal, σ is the roughness, n is the real part of the complex refractive index, and k is the imaginary part of the complex refractive index. The unknown parameters are: the specular reflectivity k s 、the diffuse reflectivity k m 、the volume scattering rate k v 。Among them, σ, n, k, k s 、k m 、k v are the parameters to be inverted.

[0124] Among them, the pBRDF theory describes the reflection characteristics of the target material at each observation angle; the elastic scattering theory of light includes Rayleigh scattering and Mie scattering, mainly describing the interaction between light and atmospheric particles; the blackbody radiation theory mainly describes the spontaneous radiation of objects.

[0125] Step 2: First, set the initial parameters of the SBA algorithm, including the number m of inversion parameters, the initial number N of the parent plant population, the Tent chaotic mapping algorithm, and the maximum number of iterations t of the simulated annealing algorithm max1 , the maximum number of iterations t of the strawberry optimization algorithm max2 , the number m of inversion parameters is 6, N is generally taken between 30 and 50. Since there are many inversion parameters, so N = 50. The number of iterations is generally between 50 and 500. Since N has a large value, more searches can be performed in each iteration. Therefore, the number of iterations can be appropriately reduced. t max1 = 50, t max2 = 200.

[0126] Step 3: Use the Tent chaotic mapping algorithm to initialize the position of the mother plant.

[0127] In the traditional SBA algorithm, the positions of the mother plants are initialized using a random operator, which easily encounters problems such as uneven distribution of the mother plant positions, weak global search ability, and low population diversity. As a result, it is prone to falling into local optima, affecting the overall optimization efficiency. Therefore, in this embodiment, a Tent chaotic mapping algorithm is introduced into the SBA algorithm to initialize the mother population. The Tent chaotic mapping algorithm adds a random number rand(0, 1) / N, which maintains the randomness, ergodicity, and regularity of the Tent chaotic mapping algorithm and can effectively avoid iterative falling into small periodic points and unstable periodic points. The Tent chaotic mapping sequence is defined as follows.

[0128] 。

[0129] 。

[0130] Among them, represents the (t + 1)-th iteration value of the chaotic mapping, represents the t-th iteration value of the chaotic mapping, 0 < t < t max1 , rand(0, 1) is a random number uniformly distributed between 0 and 1; N is the number of initial mother plant populations; X lb is the lower bound of the position variable, X ub is the upper bound of the position variable, represents the position of the i-th mother plant after the t-th iteration.

[0131] When this embodiment uses the Tent chaotic mapping algorithm to initialize the positions of the mother plants, as Figure 6 shown, first calculate the positions of the initial population, then generate a random number sequence rand(0, 1) in the interval [0, 1], then use the Tent mapping formula to generate the chaotic distribution values of each individual in the interval [0, 1], and then convert the chaotic distribution values of each individual into actual position parameters, and finally generate the initial population.

[0132] Step 4: Each mother plant randomly generates a sub-root and a stolon.

[0133] The initial mother plant population is represented by an m×N matrix. Then, each mother plant randomly generates a relatively close root and a relatively far stolon according to the initial population in each iteration, thereby generating new roots and stolons. In order to obtain the optimal solution, the stolons search around. When the stolon exactly reaches the local minimum, the algorithm will obtain a faster speed and better global search ability. Through continuous iteration, after the (t + 1)-th iteration, each variable can search for a possible optimal solution 。This process can be expressed as follows: .

[0134] Among them, represents the plant propagation matrix at the (t + 1)-th iteration, where 0 < t < t max2 , and its dimension is m × 2N, where m is the number of stolons and N is the number of initial maternal plant populations; and represent the positions of the roots and stolons at the (t + 1)-th iteration respectively, and their dimensions are both m × N; is the optimal solution of the root and stolon positions at the t-th iteration, and its dimension is also m × N; d root and d runner are two scalars, representing the distance from the root to the mother plant and the distance from the stolon to the mother plant respectively. Usually, there is d runner > d root ; r1 and r2 are random matrices, and their elements are uniformly distributed in the range [-0.5, 0.5].

[0135] Step 5: Use the simulated annealing algorithm to update the position of the stolon.

[0136] In this embodiment, the simulated annealing algorithm combines the Gaussian distribution and the Cauchy distribution, and gradually changes from global search to local search as the number of iterations increases. In the initial stage, due to the dominance of global search, the algorithm can explore a larger solution space; while in the later stage, local search dominates, which helps to find a more accurate solution in a smaller range. In this way, the algorithm can effectively avoid falling into local optimal solutions and ensure finding better solutions in the convergence stage. The following formula is used to calculate the updated position of the stolon.

[0137] .

[0138] .

[0139] .

[0140] Among them, represents the position of the stolon obtained at the (t + 1)-th iteration, where 0 < t < t max1 , and are the new positions generated according to the Gaussian distribution and the Cauchy distribution at the t-th iteration respectively, is the scale parameter of the Gaussian distribution, is the scale parameter of the Cauchy distribution; N(0, 1) and C(0, 1) are the standard normal distribution and the standard Cauchy distribution respectively, Indicates the position of the stolon at the t-th iteration. Among them, the Gaussian distribution is used for local search, and a smaller step size is required. Therefore, the scale parameter of the Gaussian distribution usually takes a value between 0.01 and 0.1, while the Cauchy distribution is used for global search, and a larger step size is required. Therefore, the scale parameter of the Cauchy distribution usually takes a value between 0.1 and 1.0. In this embodiment = 0.05, = 0.5. This stolon position update formula can avoid falling into local optimal solutions as much as possible and ensure finding better solutions during the convergence stage.

[0141] Step 6: Calculation of fitness value.

[0142] In this embodiment, the fitness values of all optimization solutions based on the objective function can be expressed as the following formula.

[0143] .

[0144] Among them, a is an adjustable parameter used to adjust the calculation method of the fitness value. Generally, it takes 0, is the objective function, and its expression is as follows.

[0145] .

[0146] Among them, σ, n, k, k s , k m and k v are all parameters to be inverted. σ is the roughness, n is the real part of the complex refractive index, k is the imaginary part of the complex refractive index, k s is the specular reflectivity, k m is the diffuse reflectivity, k v is the volume scattering rate, is the value obtained from experimental measurement, is the value substituted with the optimal solution, is the incident zenith angle, is the detection zenith angle, is the difference between the incident zenith angle and the detection zenith angle.

[0147] Step 7: Select the new parent population for the next iteration.

[0148] First, sort the calculated fitness values of the objective function in ascending order, and select the optimization solutions with the first N / 2 smaller fitness values. Then, use the roulette wheel algorithm to select the remaining N / 2 optimal solutions. Finally, take the N optimal solutions combined by the two selections as the new parent population for the next iteration.

[0149] In this embodiment, the expression of the roulette wheel algorithm is as follows.

[0150] 。

[0151] 。

[0152] Wherein, represents the probability that the individual i is selected, represents the individual i 's fitness, is the sum of the fitness values of all individuals, that is, the cumulative probability.

[0153] Then randomly generate an array m, the elements in the array range from 0 to 1, and sort them in ascending order. If the cumulative probability F total is greater than the element m[i] in the array, then the individual x(i) is selected. If it is less than m[i], then compare the next individual x(i + 1) until an individual is selected. Repeat this step N / 2 times to obtain the remaining N / 2 optimal solutions.

[0154] When this embodiment uses the fitness value and the roulette wheel algorithm to select the new parent population for the next iteration, as Figure 7 shown, first calculate the fitness value, then sort the calculated fitness values in ascending order, and select the first N / 2 smaller fitness values. At the same time, use the roulette wheel algorithm to calculate the sum of the fitness values of all individuals in the population, use the roulette wheel algorithm to calculate the selection probability of each individual, so as to generate random numbers and select individuals. After repeating N / 2 times, N / 2 fitness values can be obtained. Combining with the first N / 2 smaller fitness values selected, a total of N fitness values are obtained. Finally, the obtained N fitness values are used as the positions of the mother plants for the next iteration.

[0155] Step 8: Loop or output the final solution.

[0156] In this embodiment, when any one of the following two termination conditions occurs, the loop is terminated and the optimal inversion values of the 6 parameters to be inverted are output.

[0157] (1) The three algorithms of the Tent chaos mapping algorithm, the strawberry optimization algorithm, and the simulated annealing algorithm all reach the maximum number of iterations.

[0158] (2) A set of optimal solutions appears, such that the fitness value of the objective function is greater than the fitness threshold. In this embodiment, the fitness threshold is set to 20.

[0159] Step 9: Calculate the accuracy of the inversion parameters.

[0160] In this embodiment, the relative errors of the parameters σ, n, and k are calculated separately first, and then the average value of the relative errors of the three parameters is used to characterize the inversion result. The relative error expression is as follows.

[0161] 。

[0162] Where δ is the relative error value; Δ is the absolute error value, that is, the absolute value of the difference between the measured value and the inversion value of the parameter to be inverted; L is the true value, that is, the measured value of the parameter to be inverted.

[0163] In this embodiment, the verification stage uses a test module. The working process of the test module specifically includes the following steps.

[0164] First, after obtaining the optimal inversion values of the six parameters to be inverted, namely σ, n, k, k s , k m and k v , substitute the optimal inversion values of these six parameters to be inverted into the formula to obtain the simulation value of the pBRDF model, which characterizes the polarization degree data simulated by the pBRDF model. Then, it is judged whether the simulation value is within a reasonable threshold range through an evaluation function. In this embodiment, the evaluation function adopts MAE, and its expression is as follows.

[0165] 。

[0166] Where is the simulation value of the pBRDF model, is the value obtained from experimental measurement, and n is the total number of iterations of the inversion algorithm, n = 300.

[0167] Then, it is judged whether the MAE value between the polarization degree data simulated by the model (the simulation value of the pBRDF model) and the polarization degree data obtained from experimental measurement (the value obtained from experimental measurement) is less than 5%. When the MAE value is less than 5%, it means that the simulation value of the pBRDF model is within a reasonable threshold range, and thus the final pBRDF model is obtained. When the MAE value is greater than or equal to 5%, the pBRDF model needs to be further corrected, and then jump to step S22 "According to the initial parameters, use the Tent chaotic mapping algorithm to initialize the position of the mother plant to obtain the initialized position of the mother plant", re-initialize the position of the mother plant and obtain the corresponding parameter inversion result, and according to the newly obtained parameter inversion result, re-construct the pBRDF model and simulation The value is calculated and evaluated until its corresponding MAE value is less than 5%.

[0168] Figure 8 The three-dimensional coordinate diagram of DOLP (degree of linear polarization) generated by the pBRDF model and the detection azimuth angle and detection zenith angle. It can be seen from the figure that at an azimuth angle of 180°, the degree of polarization first increases and then decreases with the change of the zenith angle, and reaches the maximum at an observation zenith angle of 45°; Figure 9 Polar coordinate diagram generated for the pBRDF model, where the radial direction is the observation zenith angle and the circumference is the observation azimuth angle. It describes the polarization characteristics of the aluminum plate material in the entire hemisphere under the simulated atmospheric environment, and its polarization characteristic distribution is symmetrical about the azimuth angle of 0°-180°. Figure 8 and Figure 9 The darker the color, the lower the polarization degree, and the whiter the color, the higher the polarization degree; the azimuth angle of the light source is set to 0° and the zenith angle is set to 45°.

[0169] In view of the problems that the current parameter inversion method based on pBRDF has dependence on initial values ​​and poor robustness, as well as the problems that the traditional SBA algorithm has slow convergence speed and is easy to fall into local optimal solutions, this embodiment firstly establishes a pBRDF model of the target material based on pBRDF theory, elastic scattering theory of light and blackbody radiation theory, and obtains Formula, and then use the space target infrared polarization characteristic measurement device to collect polarization images and obtain the multi-angle polarization characteristic measurement results of the sample to be tested; at the same time, the improved SBA algorithm is coupled to invert the roughness, complex refractive index and other inversion parameters, and finally the optimal inversion value of the parameter to be inverted is substituted into After the formula is obtained, the simulation of the target material can be obtained. The value is calculated and the mean absolute error is used to determine whether the accuracy of the established pBRDF model meets the requirements and determine the final pBRDF model. This method can effectively solve the shortcomings of traditional parameter inversion methods. Its inversion accuracy for metal objects can be improved by about 20% compared with traditional parameter inversion methods such as particle swarm optimization algorithm, thereby obtaining a more accurate pBRDF model.

[0170] The pBRDF model is used to analyze the long-wave infrared polarization characteristics of the sample material under test at multiple angles. For long-wave infrared polarization detection, which has low resolution and large background noise interference, the accuracy of polarization detection of space targets can be improved by comparing the polarization data results obtained by the pBRDF model. In addition, it can also provide a reference for the construction of polarization characteristic models in different environments in other fields.

[0171] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0172] In this specification, specific examples are used to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for constructing a spatial target pBRDF model based on an improved SBA algorithm, characterized in that The method for constructing the pBRDF model of spatial target based on the improved SBA algorithm includes: Preliminarily construct a pBRDF model and obtain the measurement results of the polarization characteristics of the sample to be measured; the measurement results of the polarization characteristics include the values at various angles of the sample to be measured; According to the measured polarization characteristics results, use the improved SBA algorithm for parameter inversion to obtain the parameter inversion results; in the improved SBA algorithm, the Tent chaos mapping algorithm is used to initialize the position of the mother plant, and the simulated annealing algorithm is used to update the position of the stolons of the mother plant; Substitute the parameter inversion results into the pBRDF model to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy to obtain the final pBRDF model; wherein, the final pBRDF model is used to analyze the infrared polarization characteristics of the sample to be measured at multiple angles.

2. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 1, wherein, Preliminarily construct the pBRDF model, specifically including: According to the pBRDF theory, the theory of elastic light scattering, and the blackbody radiation theory, a pBRDF model of the target material is preliminarily established and the formula is determined; the target material refers to the material corresponding to the sample to be measured.

3. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 2, wherein According to the measured polarization characteristics results, use the improved SBA algorithm for parameter inversion to obtain the parameter inversion results, specifically including: Define the initial parameters of the SBA algorithm; the initial parameters include the number of inversion parameters, the initial mother plant population, the maximum iteration times of the strawberry optimization algorithm and the Tent chaos mapping algorithm, and the maximum iteration times of the simulated annealing algorithm; According to the initial parameters, use the Tent chaos mapping algorithm to initialize the position of the mother plant to obtain the initialized position of the mother plant; According to the initialized position of the mother plant, randomly generate the daughter roots and stolons of the mother plant to obtain the plant propagation matrix; each mother plant in the plant propagation matrix randomly generates a daughter root and a stolon; According to the plant propagation matrix, use the simulated annealing algorithm to update the position of the stolons to obtain the updated position of the stolons; Based on the updated position of the stolons, perform fitness calculation with the goal of minimizing the objective function solution, and use the calculated fitness value and roulette algorithm to select the mother plant population for the next iteration; Determine whether to output the parameter inversion result according to whether any of the termination conditions is satisfied; the parameter inversion result includes the optimal inversion values of multiple parameters to be inverted, and the multiple parameters to be inverted include σ, n, k, k s , k m and k v , where σ is the roughness, n is the real part of the complex refractive index, k is the imaginary part of the complex refractive index, k s is the specular reflectivity, k m is the diffuse reflectivity, k v is the volume scattering rate; Use the average relative error between the inversion values and the measured values of the parameters σ, n, and k to be inverted to represent the inversion accuracy.

4. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 3, characterized in that Substitute the parameter inversion results into the pBRDF model to calculate the error between the model simulation value and the measured value, and select whether to output the pBRDF model according to the error accuracy to obtain the final pBRDF model, specifically including: Substitute the optimal inversion values corresponding to each parameter to be inverted in the parameter inversion result into the formula to calculate the simulation value of the pBRDF model; Evaluate the simulation of the pBRDF model using an evaluation function value, and determine the final pBRDF model according to the evaluation result.

5. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 4, characterized in that, Evaluating the simulation of the pBRDF model using an evaluation function Evaluating the value, and determining the final pBRDF model according to the evaluation result, specifically including: Using the MAE function, calculate the simulation of the pBRDF model value and the MAE value between the value obtained from experimental measurement; the value obtained from experimental measurement refers to the value in the measured polarization characteristic measurement results; Determine the final pBRDF model according to the MAE value and the preset threshold.

6. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 5, wherein Determine the final pBRDF model according to the MAE value and the preset threshold, specifically including: When the MAE value is less than 5%, it is determined that the simulation of the pBRDF model value is within a reasonable threshold range, and the current pBRDF model is used as the final pBRDF model; When the MAE value is greater than or equal to 5%, it is determined that the simulation of the pBRDF model value is not within a reasonable threshold range. At this time, improve the pBRDF model and return to the step of "initializing the position of the mother plant using the Tent chaotic mapping algorithm according to the initial parameters to obtain the initialized position of the mother plant" until the MAE value is less than 5%.

7. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 1, wherein Obtain the measured polarization characteristics results of the sample to be measured, specifically including: Use the spatial target infrared polarization characteristics measurement device to measure the sample to be measured to obtain the measured polarization characteristics results of the sample to be measured; The spatial target infrared polarization characteristics measurement device includes a gas generation system, a closed movement system, an electric stage, a light source system, a polarization acquisition system, a control system and a computer; The gas generation system is connected to the interior of the enclosed movement system through a pipeline. The electric stage is arranged inside the enclosed movement system. The light source system and the polarization acquisition system are arranged inside the enclosed movement system. The gas generation system, the electric stage, the light source system, and the polarization acquisition system are all connected to the control system, and the control system is also connected to the computer; The gas generation system is used to generate a mixed gas and fill the mixed gas into the interior of the enclosed movement system through the pipeline; the mixed gas includes aerosol and water vapor; The electric stage is used to place the sample to be measured and rotate the sample to be measured; The light source system is used to emit an infrared beam towards the sample to be measured; The polarization acquisition system is used to acquire the polarization image of the sample to be measured; The enclosed movement system is used to provide a sealed environment for the mixed gas and move the light source system in the sealed environment to emit the infrared beam at multiple angles, and at the same time move the polarization acquisition system in the sealed environment to acquire the polarization image at multiple angles; The control system is used to receive the instructions issued by the computer and respectively control the working states of the gas generation system, the electric stage, the light source system, and / or the polarization acquisition system according to the instructions; and acquire the polarization image and transmit it to the computer; The computer is used to analyze and process the polarization image to obtain the measurement result of the polarization characteristics of the sample to be measured.

8. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 7, characterized in that The enclosed movement system includes a semi-circular sealed box and a semi-circular guide rail; The center of the top of the semi-circular sealed box is connected to the gas generation system through the pipeline, and the electric stage is arranged directly below the center of the top of the semi-circular sealed box; The semi-circular guide rail is laid on the inner surface of the semi-circular sealed box, and the semi-circular guide rail is connected to the control system; The light source system and the polarization acquisition system are slidably arranged on the semi-circular guide rail.

9. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 8, wherein The semi-circular guide rail includes a first guide rail and a second guide rail; The light source system is slidably arranged on the first guide rail, and the polarization acquisition system is slidably arranged on the second guide rail; First angle marks are arranged on both the first guide rail and the second guide rail. The first angle marks are from 0° to 90°, and the graduation value of the first angle marks is 1°; The 0° first angle marks of the first guide rail and the second guide rail are both located at one end close to the top of the semi-circular sealed box.

10. The method for constructing a spatial target pBRDF model based on the improved SBA algorithm according to claim 9, wherein, Second angle marks are arranged on the electric stage. The second angle marks are from 0° to 360°, and the graduation value of the second angle marks is 1°. The 0° second angle mark of the electric stage faces the first guide rail.

Images