Flame three-dimensional shape reconstruction method and system based on physical constraint and electronic equipment

Abstract

The invention discloses a flame three-dimensional shape reconstruction method and system based on physical constraint and electronic equipment, and relates to the technical field of flame shape reconstruction, and the method comprises the steps: calculating a visual boundary of a target flame in a three-dimensional space according to a plurality of flame images of the target flame at a target moment, and calculating the visual boundary of the target flame; constructing a radiation transfer equation set facing all pixels in each flame image of the target flame at the target moment, and performing adaptive adjustment iteration solution on the radiation transfer equation set to obtain three-dimensional distribution data of the radiation intensity at the target moment until three-dimensional distribution data of the radiation intensity at each preset moment is obtained; and constructing a three-dimensional shape of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment. According to the method, the number of unknown quantities can be reduced, the calculation amount can be reduced, the convergence speed of iterative calculation can be increased, and the calculation efficiency and accuracy of solution can be improved through visual boundary demarcation, adaptive adjustment of iterative solution and time sequence information transmission.

Description

Background Art

[0002] Vision is the main way for humans to obtain information. At least 80% of external information is obtained through vision. The three-dimensional morphology of the flame is determined by the thermal radiation emitted by the flame. It is one of the important properties of the flame. It can intuitively reflect the distribution trend of the flame and provide support for further research on the combustion conditions of the flame. It is of great significance for understanding the laws of combustion. Therefore, reconstructing the three-dimensional morphology of the flame is one of the important research directions in the field of combustion diagnosis. The three-dimensional morphology of the flame can have many different forms depending on the experimental conditions.

[0003] The flame three-dimensional morphology reconstruction method is usually based on the deduction of physical laws and mathematical calculations to obtain the three-dimensional distribution of flame radiation intensity, and then construct the flame three-dimensional morphology. The calculation of the three-dimensional distribution of flame radiation intensity is usually based on the flame radiation image, that is, based on the flame image obtained by the optical camera. Specifically, by dividing the flame into multiple layers, the radiation transfer equation group is listed through the optical transfer process, and the equation group is inverted and solved to obtain the three-dimensional distribution of flame radiation intensity, which is also the key to the flame three-dimensional morphology reconstruction process. For most cases, the coefficients of the obtained radiation transfer equation group are non-full rank matrices and lack a definite solution. The inversion process will inevitably encounter the inverse problem solution. This task is non-positive definite. The ill-posedness of the inverse problem makes it difficult to obtain the exact value, which is also an ill-posed problem. Therefore, it is necessary to solve the non-positive definite problem of this task and obtain the optimal solution by optimizing the inverse problem solution process.

[0004] In the field of flame 3D morphology reconstruction, the commonly used inverse problem solving methods mainly include the optimal criterion method, iterative method, regularization method and other types. Specifically:

[0005] 1) The optimal criterion method is to specify or design a criterion to determine the optimal solution under the criterion from all possible solutions to the ill-posed problem. Commonly used optimal criteria include smoothing criteria, Bayesian criteria, least squares criteria, maximum entropy criteria, etc. The most representative of the optimal criterion methods is the least squares QR decomposition (LSQR, LeastSquare QR-factorization) algorithm, which transforms the inverse problem solution into a least squares problem and uses QR decomposition to obtain the optimal solution. The disadvantages are:

[0006] When the criteria are reasonable, the optimal criterion method can obtain a very ideal optimal solution, but finding a suitable criterion is a very difficult task. Therefore, the optimal criterion method is difficult to put into practical use in actual scenarios.

[0007] 2) The inverse problem solution method of the iterative method is: gradually adjust the predicted value through repeated iterative calculations to gradually approach the final result. The iterative method and its related optimization methods are diverse and are the most commonly used algebraic operation inverse problem solution methods in this field. A series of methods such as FBP and ART are typical iterative methods. In addition, there are also iterative methods such as gradient methods represented by Gauss-Newton gradient, fastest gradient, conjugate gradient, etc., Landweber algorithm, etc. The optimization methods of the iterative method also include ant colony algorithm, simulated annealing method, Levenberg-Marquardt iterative method based on Tikhonov regularization optimization, etc. The disadvantages are:

[0008] The iterative method is relatively powerful and can be applied to many situations. The solution effect is relatively good in most cases, but iteration will bring a relatively large amount of calculation. The solution effect is often more dependent on the number of input flame image angles. The iterative solution process may also lead to results that do not conform to actual physical laws.

[0009] 3) Regularization method for solving inverse problems:

[0010] Regularization method is to reduce the pathological nature of the problem by imposing additional regularization constraints, so that the problem without a definite solution becomes a method with a definite solution or an optimal solution. The most commonly used regularization method is Tikhonov regularization, which has the following disadvantages:

[0011] Appropriate regularization can further improve the accuracy and stability of the solution, but similar to the optimal criterion method, the regularization method also needs to find a regularization constraint suitable for the target flame, which is more difficult in actual implementation.

[0012] In summary, the optimal criterion method, the iterative method and the regularization method can solve the pathological problems in the reconstruction of the three-dimensional morphology of flames to a certain extent. However, no matter the optimal criterion method, the iterative method or the regularization method, although they have their own strengths, they also have some shortcomings: the computational overhead is large and the solution efficiency is poor; in order to facilitate analysis and calculation, it is often assumed that the conditions are relatively ideal. The selection method of the optimal solution is disconnected from the actual physical meaning, and the physical information characteristics and laws of the flame itself cannot be fully utilized, which may produce solutions that do not conform to the actual physical laws. Summary of the invention

[0013] The technical problem to be solved by the present invention is to address the deficiencies of the prior art and specifically provide a method, system and electronic device for reconstructing three-dimensional flame morphology based on physical constraints, as follows:

[0014] 1) In the first aspect, the present invention provides a method for reconstructing three-dimensional flame morphology based on physical constraints, and the specific technical scheme is as follows:

[0015] The visible boundary of the target flame in the three-dimensional space is calculated based on multiple flame images of the target flame at the target time, wherein the multiple flame images of the target flame at the target time are obtained by shooting at each preset shooting angle;

[0016] Constructing a set of radiation transfer equations for all pixels in each flame image at a target time facing the target flame;

[0017] Combined with the visible boundary of the target flame at the target time and the initial data of the three-dimensional distribution of the radiation intensity, the radiation transfer equation group is adaptively adjusted and iteratively solved to obtain the three-dimensional distribution data of the radiation intensity at the target time, until the three-dimensional distribution data of the radiation intensity at each preset time is obtained, wherein the target time is any preset time, when the target time is the first preset time, the initial data of the three-dimensional distribution of the radiation intensity is determined in a pre-set manner, and when the target time is not the first preset time, the three-dimensional distribution data of the radiation intensity at the previous preset time of the target time is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target time;

[0018] The three-dimensional morphology of the target flame is constructed according to the three-dimensional distribution data of the radiation intensity at each preset moment.

[0019] The beneficial effects of the flame three-dimensional morphology reconstruction method based on physical constraints provided by the present invention are as follows:

[0020] On the one hand, by determining the visible boundary, only the flame area can be calculated, which reduces the amount of calculation and the number of unknown quantities. On the other hand, when the target moment is the first preset moment, the initial data of the three-dimensional distribution of the radiation intensity is determined in a pre-set manner. When the target moment is not the first preset moment, the three-dimensional distribution data of the radiation intensity at the previous preset moment of the target moment is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target moment. By utilizing the similarity of adjacent moments of the flame, the convergence speed of the iterative calculation is accelerated, and the overall reconstruction efficiency is improved.

[0021] On the basis of the above scheme, the flame three-dimensional morphology reconstruction method based on physical constraints of the present invention can also be improved as follows.

[0022] Furthermore, based on the multiple flame images of the target flame at the target time, the visible boundary of the target flame in the three-dimensional space is calculated, including: dividing the multiple flame images of the target flame at the target time into a flame area and a non-flame area, and projecting and superimposing them to calculate the visible boundary of the target flame in the three-dimensional space.

[0023] Further, a plurality of flame images of the target flame at the target time are divided into a flame area and a non-flame area, including:

[0024] Obtain and determine the threshold value corresponding to the flame image according to the maximum value among the R value, G value and B value of all pixels of any flame image at the target time, judge whether the maximum value among the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, judge the pixels with a judgment result of yes as the pixel point of the non-flame area, and judge the pixels with a judgment result of no as the pixel point of the flame area, until each flame image of the target flame at the target time is divided into the flame area and the non-flame area.

[0025] Furthermore, a set of radiation transfer equations for all pixels in each flame image facing the target flame at the target time is constructed, including:

[0026] According to the discretized radiation transfer equation, along the shooting direction corresponding to each preset shooting angle, a group of radiation transfer equations for all pixels in each flame image facing the target flame at the target time is constructed.

[0027] Furthermore, the process of obtaining the discretized radiation transfer equation includes:

[0028] Based on the flame radiation transfer law, the radiation transfer equation is constructed;

[0029] The target space of the radiation transfer process is discretized and combined with the radiation transfer equation to obtain the discretized radiation transfer equation.

[0030] Furthermore, it also includes: introducing an energy constraint function when adaptively adjusting and iteratively solving the radiation transfer equations.

[0031] Furthermore, it also includes: visually rendering the three-dimensional appearance of the target flame.

[0032] 2) In the second aspect, the present invention also provides a flame three-dimensional morphology reconstruction system based on physical constraints, and the specific technical scheme is as follows:

[0033] It includes a visible boundary acquisition module, an equation group construction module, a solution module and a 3D shape reconstruction module;

[0034] The visible boundary acquisition module is used to calculate the visible boundary of the target flame in the three-dimensional space according to multiple flame images of the target flame at the target time, wherein the multiple flame images of the target flame at the target time are obtained by shooting at each preset shooting angle;

[0035] The equation group building module is used to: build a radiation transfer equation group for all pixels in each flame image facing the target flame at the target time;

[0036] The solution module is used to: combine the visible boundary of the target flame at the target time and the initial data of the three-dimensional distribution of the radiation intensity, adaptively adjust and iteratively solve the radiation transfer equation group, obtain the three-dimensional distribution data of the radiation intensity at the target time, until the three-dimensional distribution data of the radiation intensity at each preset time is obtained, wherein the target time is any preset time, when the target time is the first preset time, the initial data of the three-dimensional distribution of the radiation intensity is determined by a preset method, when the target time is not the first preset time, the three-dimensional distribution data of the radiation intensity at the previous preset time of the target time is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target time;

[0037] The three-dimensional morphology reconstruction module is used to construct the three-dimensional morphology of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment.

[0038] On the basis of the above scheme, the flame three-dimensional morphology reconstruction system based on physical constraints of the present invention can also be improved as follows.

[0039] Furthermore, the visible boundary acquisition module is used to calculate the visible boundary of the target flame in the three-dimensional space based on multiple flame images of the target flame at the target time, including: dividing the multiple flame images of the target flame at the target time into a flame area and a non-flame area, and projecting and superimposing them to calculate the visible boundary of the target flame in the three-dimensional space.

[0040] Furthermore, the visible boundary acquisition module is also specifically used for:

[0041] Obtain and determine the threshold value corresponding to the flame image according to the maximum value among the R value, G value and B value of all pixels of any flame image at the target time, judge whether the maximum value among the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, judge the pixels with a judgment result of yes as the pixel point of the non-flame area, and judge the pixels with a judgment result of no as the pixel point of the flame area, until each flame image of the target flame at the target time is divided into the flame area and the non-flame area.

[0042] Furthermore, the equation group building module is specifically used to:

[0043] According to the discretized radiation transfer equation, along the shooting direction corresponding to each preset shooting angle, a group of radiation transfer equations for all pixels in each flame image facing the target flame at the target time is constructed.

[0044] Furthermore, the equation group building module is also specifically used for:

[0045] Based on the flame radiation transfer law, the radiation transfer equation is constructed;

[0046] The target space of the radiation transfer process is discretized and combined with the radiation transfer equation to obtain the discretized radiation transfer equation.

[0047] Furthermore, the solution module is also used to introduce an energy constraint function when adaptively adjusting and iteratively solving the radiation transfer equations.

[0048] Furthermore, it also includes a visualization rendering module, which is used to: perform visualization rendering on the three-dimensional appearance of the target flame.

[0049] 3) In a third aspect, the present invention further provides an electronic device, comprising a processor, the processor being coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor, so that the electronic device implements any of the above-mentioned flame three-dimensional morphology reconstruction methods based on physical constraints.

[0050] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, any of the above-mentioned methods for reconstructing three-dimensional flame morphology based on physical constraints is implemented.

[0051] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments of the present invention:

[0053] Figure 1 A schematic diagram of a flow chart of a flame three-dimensional morphology reconstruction method based on physical constraints according to an embodiment of the present invention;

[0054] Figure 2 The schematic diagram of the projection principle and the calculated visible boundary;

[0055] Figure 3 Schematic diagram of the discretization of the radiation transfer process;

[0056] Figure 4 It is a schematic diagram of timing information transmission;

[0057] Figure 5 Schematic diagram of the results of the experimental verification for visual boundary demarcation;

[0058] Figure 6 Schematic diagram of the comparison of the input angles of the two methods;

[0059] Figure 7 This is a schematic diagram of the comparison of the non-input angle results of the two methods;

[0060] Figure 8 A schematic diagram of the verification result of timing information transfer;

[0061] Fig. 9 Three-angle images of the space science combustion experiment;

[0062] Fig.10 The results of three-dimensional flame morphology reconstruction for space science combustion experiments;

[0063] Fig.11 It is a structural schematic diagram of a flame three-dimensional morphology reconstruction system based on physical constraints according to an embodiment of the present invention;

[0064] Fig.12 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0066] like Figure 1 As shown, a flame three-dimensional morphology reconstruction method based on physical constraints according to an embodiment of the present invention comprises the following steps:

[0067] S1. calculating a visible boundary of a target flame in a three-dimensional space according to a plurality of flame images of a target flame at a target time, wherein the plurality of flame images of the target flame at a target time are obtained by shooting at each preset shooting angle;

[0068] S2, constructing a set of radiation transfer equations for all pixels in each flame image at a target time facing the target flame;

[0069] S3. Adaptively adjust and iteratively solve the radiation transfer equations in combination with the visible boundary of the target flame at the target time and the initial data of the three-dimensional distribution of the radiation intensity, and obtain the three-dimensional distribution data of the radiation intensity at the target time, until the three-dimensional distribution data of the radiation intensity at each preset time is obtained, wherein the target time is any preset time, and when the target time is the first preset time, the initial data of the three-dimensional distribution of the radiation intensity is determined in a preset manner, and when the target time is not the first preset time, the three-dimensional distribution data of the radiation intensity at the previous preset time of the target time is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target time;

[0070] S4. Constructing a three-dimensional morphology of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment.

[0071] Optionally, in S1, a visible boundary of the target flame in the three-dimensional space is calculated based on multiple flame images of the target flame at the target time, including: dividing the multiple flame images of the target flame at the target time into a flame area and a non-flame area, and projecting and superimposing them to calculate the visible boundary of the target flame in the three-dimensional space.

[0072] Optionally, dividing a plurality of flame images of a target flame at a target time into a flame area and a non-flame area includes:

[0073] Obtain and determine the threshold value corresponding to the flame image according to the maximum value among the R value, G value and B value of all pixels of any flame image at the target time, judge whether the maximum value among the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, judge the pixels with a judgment result of yes as the pixel point of the non-flame area, and judge the pixels with a judgment result of no as the pixel point of the flame area, until each flame image of the target flame at the target time is divided into the flame area and the non-flame area.

[0074] Optionally, in S2, a radiation transfer equation group for all pixels in each flame image of the target flame at the target time is constructed, including:

[0075] According to the discretized radiation transfer equation, along the shooting direction corresponding to each preset shooting angle, a group of radiation transfer equations for all pixels in each flame image facing the target flame at the target time is constructed.

[0076] Optionally, the process of obtaining the discretized radiation transfer equation includes:

[0077] Based on the flame radiation transfer law, the radiation transfer equation is constructed;

[0078] The target space of the radiation transfer process is discretized and combined with the radiation transfer equation to obtain the discretized radiation transfer equation.

[0079] Optionally, the method further includes: introducing an energy constraint function when adaptively adjusting and iteratively solving the radiation transfer equations.

[0080] Optionally, it also includes:

[0081] S5. Visually render the three-dimensional morphology of the target flame.

[0082] The present invention is described by the following examples, which specifically include:

[0083] S10. Verify all flame images:

[0084] All flame images include: multiple flame images of the target flame at each preset moment, and the multiple flame images at each preset moment are obtained by shooting at each preset shooting angle.

[0085] Each flame image is parsed, the integrity of the image information in each flame image is verified, and the complete data of the image is read. When the integrity of all flame images is verified, S11 is executed.

[0086] S11, calculating the visible boundary of the target flame in the three-dimensional space where the target flame is located in each flame image at each moment.

[0087] The entire flame image is divided into flame area and non-flame area from the spatial dimension, and the flame area of ​​the multi-angle image is projected to calculate the visible boundary of the target flame in three-dimensional space, and the subsequent calculation is focused within the visible boundary to avoid the influence of the non-flame area in the calculation process and improve the calculation accuracy and efficiency. The details are as follows:

[0088] S110, dynamic calculation of threshold value: for different flame images, the threshold value used for demarcating the visible boundary is different, which is obtained by dynamic calculation of the current image. After multiple experimental comparisons and analyses, it is determined that 10% of the maximum value of the three RGB channels of each pixel in the flame image is the lower threshold value, that is, the threshold value corresponding to the flame image;

[0089] S111, visible boundary calculation:

[0090] Based on the calculated lower threshold, traverse each pixel of the flame image and take the maximum of the three-channel values ​​of RGB of each pixel to determine whether the pixel belongs to the flame area. Specifically:

[0091] It is determined whether the maximum value of the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, and the pixels with a judgment result of yes are determined as pixels in the non-flame area, and the pixels with a judgment result of no are determined as pixels in the flame area, until each flame image of the target flame at the target time is divided into a flame area and a non-flame area.

[0092] Among them, the target flame can be a flame generated by the space science combustion experiment hardware device, or it can be other flames, for example, the flame generated by the burning of a candle, etc. Multiple flame images of the target flame at each preset moment can be obtained by taking pictures with a high-speed camera in the space science combustion experiment hardware device. The space science combustion experiment hardware device includes three high-speed cameras with shooting angles of 0°, 135°, and 225°, respectively. Three flame images at different shooting angles will be obtained at each preset moment. The projection method is used to project each flame area along its respective shooting angle, and the common area obtained by the projection is regarded as the flame space range to be solved. The outer edge of the flame area is the visible boundary of the target flame in three-dimensional space. The time interval between two adjacent preset moments can be set according to actual conditions, such as 10ms or 20ms, etc. The principle of calculating the visible boundary by the projection method and the calculated visible boundary are as follows. Figure 2 shown.

[0093] S12. Construct the radiation transfer equation:

[0094] The process of flame radiation from emission to exit is called radiation transfer. Based on the flame radiation transfer law, the radiation transfer equation is constructed. The expression of the radiation transfer equation is:

[0095]

[0096] Where l is the outgoing direction, H is the local radiation source term, I(l) is the radiation intensity along the outgoing direction, I(l′) is the radiation intensity transmitted into the volume element from the spherical angle Ω, and Φ(l,l′) is the scattering coefficient from the spherical angle Ω at the volume element toward the outgoing direction. κ e is the absorption coefficient, κ s is the scattering coefficient.

[0097] The research object (target flame) of the present invention is a hydrocarbon flame, and the final combustion products are carbon dioxide and water, so the absorption effect and scattering effect of the hydrocarbon flame can be ignored. Assuming no background radiation, the integral form of the radiation transfer equation is:

[0098] I(l)=∫H·dl

[0099] S13, get the discretized radiation transfer equation:

[0100] In order to facilitate analysis and calculation, the target space of the radiation transfer process is discretized, such as Figure 3 As shown:

[0101] The discretized radiation transfer equation is:

[0102]

[0103] Where I(I) is the final total radiation intensity along the outgoing direction, Δl i is the radiation source term H i The width of the volume element in the outgoing direction.

[0104] Discretization of the radiation transmission process can transform continuous space into discrete volume elements, and the transmission formula can be transformed from integral form to summation form. This makes the number of unknown quantities finite, which is of great significance for the subsequent description and calculation of unknown quantities.

[0105] S14. Construct the radiation transfer equations:

[0106] According to the discretized radiation transfer equation, along the shooting direction corresponding to each preset shooting angle, the radiation transfer equation group for all pixels in each flame image facing the target flame at the target time is constructed:

[0107]

[0108] S15. Adaptively adjust and iteratively solve the radiation transfer equations:

[0109] For the radiation transfer equations constructed from the flame area data, the radiation information of the flame image is used to guide the adaptive iterative solution of the three-dimensional distribution of radiation intensity, which accelerates the convergence speed and reduces the computational overhead. The solution of the three-dimensional distribution of radiation intensity is a key step in determining the accuracy of the flame three-dimensional morphology reconstruction results. The specific implementation process is as follows:

[0110] 1) When the target moment is the first preset moment, the initial data of the three-dimensional distribution of the radiation intensity is determined by a preset method, specifically:

[0111] When iteratively solving the equations, a three-dimensional distribution data initialized with a smaller value is used as the three-dimensional distribution initial data (iterative initial value), denoted as H 0 , based on solving all flame images at the first preset moment, where H 0 Assign a preset value to the default.

[0112] When the target moment is not the first preset moment, the three-dimensional distribution data of the radiation intensity at the previous preset moment of the target moment is used as the three-dimensional distribution initial data of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target moment.

[0113] If it is necessary to reconstruct the three-dimensional morphology of flames at multiple consecutive moments, considering that flame combustion is a continuous process in time, its changes are also continuous, and the flames at adjacent moments captured by a high-speed camera have great similarity. Therefore, using the timing information of the flame can greatly reduce the total reconstruction cost. Figure 4 As shown in the figure, except for the first moment, each moment uses the 3D distribution calculation result of the previous moment as the initial value of the frame iteration process at the current moment. The transmission of timing information can greatly reduce the number of iterations required for the iterative process to converge at subsequent moments, thereby accelerating the convergence speed and improving the overall reconstruction efficiency.

[0114] 2) Adaptively adjust the initial data H of the three-dimensional distribution 0 , generate a predicted value, denoted as H′: When calculating H′, traditionally, methods such as random walk are often used. Such methods are relatively random, have poor convergence speed, and require a large amount of calculation. The present invention proposes an adaptive adjustment method, which uses the radiation information of the flame image to guide the adjustment process of the initial data of the three-dimensional distribution, accelerates the convergence speed, and reduces the computational overhead. At the same time, additional constraints are imposed based on the physical properties of the radiation intensity, limiting the elements in H′ to be in the non-negative interval to prevent the occurrence of unrealistic negative solutions and improve the accuracy of the reconstruction results. Assuming that there are m light paths passing through a certain volume element, and the point of the light path on its corresponding projection surface is p, then:

[0115]

[0116] in, I pi is the radiation intensity value of the i-th light path at the corresponding point on the projection surface, H ij is the radiation intensity emitted by the jth volume element on the i-th optical path passing through the target volume element, Δl ij H ij The width of the volume element in the outgoing direction. α is the relaxation factor, which is responsible for controlling the step size of each adjustment.

[0117] When the values ​​of more corresponding points on the projection surface are greater than or less than the calculated value, it means that the error of the target volume element is likely to have a more obvious impact on the whole, so the error may be larger; when the sum of the total deviations of the points greater than and less than are close, it means that the error of the target volume element is likely to be less obvious on the whole, so the error may be smaller. Although it cannot be guaranteed that the adjustment direction of each point in a single iteration round is correct, as long as the step size of each adjustment is reasonable, the overall solution result will definitely change towards a more accurate trend. When calculating the adjustment amount, the deviations of all points related to the target volume element on the projection surface are averaged, which can effectively take the above characteristics into account, thereby achieving the effect of improving the accuracy of the adjustment result. At the same time, it can also accelerate convergence and improve operation efficiency.

[0118] 3) Calculate the energy constraint value of H': The core step of the iterative solution is to introduce an energy constraint function into the equation group, and iteratively solve the equation group to minimize the energy constraint function. This solution is regarded as the optimal solution to the three-dimensional distribution problem of flame radiation intensity. The energy constraint function designed by the present invention is as follows:

[0119]

[0120] Among them, the first term on the right side of the equal sign is the error constraint term, the second term is the smoothness constraint term, I is the total radiation intensity, H is the local radiation source term, ΔL is the width of the volume element, is the Laplace operator, which represents the divergence of the target gradient. x, y and z represent the three-dimensional coordinates of the corresponding solution space. e and α s is a weighting coefficient, and its value depends on the specific input data and is determined through experiments. The Laplace operator can effectively extract the distribution characteristics of the target change trend and is suitable for calculating the smoothness constraint value. The introduction of the smoothness constraint term enables the energy constraint function to reflect the continuity of the solution result, thereby reducing the possibility of numerical jumps and making the solution result more realistic.

[0121] 4) Minimum value judgment: judge the obtained energy constraint value. When the specified number of iterations is reached or the energy constraint value is less than a certain allowable range, the iteration ends. Otherwise, return to the adaptive adjustment H. 0 In the step of generating the predicted value H', the solution is iterated in this way until the condition for ending the iteration is reached.

[0122] 5) Output the solution result H′ of the equation group: After the iteration is completed, the final predicted value H′ is output as the solution result of the equation group.

[0123] S16. Construct a three-dimensional morphology of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment, and perform visual rendering. Specifically:

[0124] In order to observe the flame intuitively and study the distribution law of flame structure, it is necessary to model and render the three-dimensional light morphology of the flame. Since the flame is a phenomenon of light and heat generated when the gas burns, it does not have a clear boundary like a rigid body, and the flame radiation source does not reflect external light like an opaque body. Therefore, it is necessary to design a targeted modeling method to achieve a high-fidelity three-dimensional flame morphology simulation. Considering that the main component of the hydrocarbon flame in the space science combustion experiment is transparent gas, the light intensity received during observation should be the result of the superposition of the radiation emitted by all volume elements of the flame on the path of sight. Therefore, the three-dimensional distribution of radiation intensity cannot be simply displayed as the three-dimensional morphology of the flame. Assuming that the scattering and absorption effects of the medium in the hydrocarbon flame are negligible, a non-uniform transparency method simulating physical laws is proposed to model the three-dimensional morphology of the flame. The material of the volume element uses a self-luminous material, the color grayscale is determined by the obtained radiation intensity, and the transparency ratio is obtained according to the carbon soot volume fraction of the volume element, which is used to simulate the radiation emission process. The brightness of the light emitted by multiple transparent luminous material volume elements is superimposed and rendered along the observation direction to simulate the radiation transmission process. Finally, the volume rendering method is used in the Unity 3D engine to perform visual rendering of the three-dimensional morphology and interactive operations.

[0125] The technical effects of the present invention are verified as follows:

[0126] Relative error is a commonly used indicator to evaluate the similarity between two values. It is the absolute value of the ratio of the difference between the two values ​​and the true value. Assume that the true value is I T , the solution value is I R , then the relative error expression is:

[0127]

[0128] When verifying the space science combustion experiment data, the 3D morphology reconstruction result is projected to the input direction and then the error is calculated pixel by pixel with the input image and averaged. The average relative error (abbreviated as error) is used as the evaluation index of the accuracy of 3D morphology reconstruction.

[0129] The experimental verification process is as follows:

[0130] 1) Visual boundary demarcation:

[0131] The visual boundary demarcation is experimentally verified and its effect is analyzed. Figure 5The flame area projection and error distribution calculation of the three-dimensional morphological reconstruction results with and without visible boundaries are shown. The comparison results show that the error of the flame internal area after the visible boundary is drawn is significantly smaller than the error of the flame internal area after solving the whole space. This is because after the visible boundary is drawn, the algorithm can shield the dilution effect of the background area during the solution process, focusing on reducing the reconstruction error of the flame internal area, and at the same time greatly alleviate the problem of the solution result diffusing into the background area, thereby improving the reconstruction accuracy of the flame area.

[0132] 2) The algebraic operation inversion solution method based on the LSQR algorithm is compared with the adaptive adjustment iterative method proposed in the present invention for verification. The LSQR algorithm is a method proposed by Paige et al. for solving large sparse matrices. Its core idea is to follow the least squares criterion and find the optimal solution in the space defined by the equation. The basis of the LSQR algorithm is the least squares criterion, and the main process is iterative solution. Therefore, the LSQR algorithm can embody the characteristics of the optimal criterion method and the iterative method to a certain extent. Table 1 shows the error comparison of the three-dimensional morphology reconstruction results under different numbers of input angles. Overall, the reconstruction error of the adaptive adjustment iterative method is smaller than the reconstruction error of the LSQR algorithm, which is particularly obvious when the input angles are small.

[0133] Table 1:

[0134]

[0135] Figure 6 (a) shows the projection effect and error distribution of the reconstruction results of the LSQR algorithm in three shooting directions. Figure 6 (b) shows the projection effect and error distribution of the reconstruction results of the adaptive adjustment iterative method in three shooting directions. Figure 6 It can be seen that the error distribution of the projection of the reconstruction results of the two methods is relatively close, and it can be considered that there is no substantial difference in the projection effect of the two in the shooting direction. However, after comparing the solution results at non-input angles, the difference between the LSQR-based algebraic operation inversion solution method and the adaptive adjustment iterative method was found. Figure 7 The non-input angle comparison results of the two methods are shown. Specifically, Figure 7 As shown in (a), the reconstruction result is projected in the horizontal rotation direction of 30°. It can be seen that compared with the result of the adaptive adjustment iterative method, the result of the LSQR algorithm is less smooth, and there is a part of the red channel that is too high in the brighter area in the lower middle part. Figure 7 The comparison of flame morphology rendering results in (b) can more clearly reflect this point.

[0136] Obviously, this red highlight phenomenon and the obvious mutation of the internal light intensity distribution are not in line with the actual situation of the target flame. This problem occurs because the traditional algebraic operation inverse problem solving method represented by LSQR, as a pure mathematical method, does not take into account the practical significance of physical quantities. In order to pursue the highest possible accuracy at the mathematical level, these methods are likely to produce results that do not conform to the actual situation, such as local mutations and negative values. These solution deviations may make the error calculation value smaller when calculating the input angle projection, but they deviate from the normal law in the actual physical sense. The adaptive adjustment iteration method is guided by the brightness of the actual picture, performs adaptive adjustments in the iteration process, and imposes non-negative constraints and smoothness constraints, which can avoid the above-mentioned unrealistic situations caused by pure mathematical operations as much as possible. In summary, it can be concluded that the adaptive adjustment iteration method has a better effect on the target task, so the inverse problem solving method with the adaptive adjustment iteration method as the core is effective.

[0137] 3) Timing information transmission:

[0138] The experiment used 10 sets of flame images at consecutive moments for reconstruction. In one set of experiments, the reconstruction results of the previous moment were transferred as the initial value between reconstructions at different moments. In the other set of experiments, the initial value of each reconstruction started from the same initial value. The overall running time of the algorithm and the running time of the iteration part were recorded at each moment. The results are shown in the figure. Figure 8 shown.

[0139] Depend on Figure 8 It can be intuitively concluded that the transmission of timing information can effectively reduce the reconstruction time of subsequent moments in the continuous reconstruction process, and the main function of the transmission of timing information is to speed up the iterative solution process of subsequent moments. In the process of continuous moment reconstruction, the transmission of timing information can avoid that each iteration starts from a small random initial value. Under the premise that the flame change has temporal continuity, using the result of the previous moment as the initial value is equivalent to starting the iteration from a distribution that is relatively close to convergence, thereby greatly reducing the number of iterations required for convergence and speeding up the reconstruction speed.

[0140] Engineering applications are as follows:

[0141] The flame data obtained from the space science combustion experiment was used to reconstruct the three-dimensional morphology using the method of the present invention. In the space science combustion experiment, the shooting angles of the three high-speed cameras are 0°, 135°, and 225° respectively. Therefore, the input data of the method is the image data of the three angles. The three-angle image at a random moment is Fig. 9 shown.

[0142] The method of the present invention is used to reconstruct the three-dimensional shape of the flame. The results are as follows: Fig.10 shown. Fig.10 (a) is the effect of observing from the front at 0°. Fig.10 (b) is the effect of rotating 15° horizontally and vertically. Fig.10 (c) is the effect of rotating 30° horizontally and vertically. Fig.10 It can be seen that the realism of the overall reconstruction result is quite impressive, but it is still different from the reconstructed image obtained by projection calculation. This is because the 3D rendering engine obtains the observation result after modeling and displaying all the voxels, rather than directly performing mathematical calculations and summing them up. Therefore, the difference between the two is foreseeable.

[0143] The beneficial effects of the present invention are as follows:

[0144] 1) Generally speaking, the range of valid data in the flame image usually only occupies part of the whole image, and there will be a certain background part in the image space range, so the proportion of the non-flame area in the three-dimensional space in the total space will be larger than the proportion of the background in the two-dimensional image in the total range. Inverse problem solving usually directly solves the unknown quantity matrix, so that the area outside the flame will be included in the solution process. Too many unknown quantities will not only increase the computational overhead, but also have a dilution effect on the flame area when calculating the overall error, resulting in the overall error value not accurately reflecting the deviation between the solution value and the actual value in the key area, thereby reducing the accuracy of the solution. The present invention proposes a method for visual boundary delineation, which divides the target space into a flame area and a non-flame area. When solving, only the flame area is calculated, which plays a role in reducing the number of unknown quantities, and reduces the influence of the background area in the subsequent iterative solution process, and improves the reconstruction accuracy and calculation efficiency.

[0145] 2) The iterative method for solving the inverse problem will bring about a relatively large amount of calculations, and the effect of the solution is often more dependent on the number of angles of the input flame image. The process of iteratively solving the optimal value may also lead to results that do not conform to the actual physical laws. The present invention proposes an inverse problem solving method with adaptive adjustment iteration, which uses the radiation information of the flame to adaptively adjust the iterative solution process, accelerates the convergence speed of the iterative process, and reduces the calculation cost. At the same time, additional constraints are imposed according to the physical properties of the flame radiation to improve the accuracy of the solution results.

[0146] 3) Flame combustion is a continuous process in time, and its changes are also continuous. The flames of adjacent time frames captured by a high-speed camera have great similarity. The present invention proposes an accelerated solution method for the three-dimensional distribution of flame radiation intensity using time series information transmission. By utilizing the similarity of adjacent moments of the flame, the convergence speed of iterative calculation is accelerated, and the overall reconstruction efficiency is improved.

[0147] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0148] like Fig.11 As shown, a flame three-dimensional morphology reconstruction system 200 based on physical constraints according to an embodiment of the present invention includes a visible boundary acquisition module 201, an equation group construction module 202, a solution module 203 and a three-dimensional morphology reconstruction module 204;

[0149] The visible boundary acquisition module 201 is used to calculate the visible boundary of the target flame in the three-dimensional space according to a plurality of flame images of the target flame at the target time, wherein the plurality of flame images of the target flame at the target time are obtained by shooting at each preset shooting angle;

[0150] The equation group construction module 202 is used to: construct a radiation transfer equation group for all pixels in each flame image facing the target flame at the target time;

[0151] The solution module 203 is used to: combine the visible boundary of the target flame at the target time and the initial data of the three-dimensional distribution of the radiation intensity, perform adaptive adjustment and iterative solution on the radiation transfer equation group, obtain the three-dimensional distribution data of the radiation intensity at the target time, until the three-dimensional distribution data of the radiation intensity at each preset time is obtained, wherein the target time is any preset time, when the target time is the first preset time, the initial data of the three-dimensional distribution of the radiation intensity is determined in a preset manner, when the target time is not the first preset time, the three-dimensional distribution data of the radiation intensity at the previous preset time of the target time is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target time;

[0152] The three-dimensional shape reconstruction module 204 is used to construct the three-dimensional shape of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment.

[0153] Optionally, in the above technical solution, the visible boundary acquisition module 201 is used to: calculate the visible boundary of the target flame in the three-dimensional space based on multiple flame images of the target flame at the target time, including: dividing the multiple flame images of the target flame at the target time into a flame area and a non-flame area, and projecting and superimposing them to calculate the visible boundary of the target flame in the three-dimensional space.

[0154] Optionally, in the above technical solution, the visible boundary acquisition module 201 is further specifically used for:

[0155] Obtain and determine the threshold value corresponding to the flame image according to the maximum value among the R value, G value and B value of all pixels of any flame image at the target time, judge whether the maximum value among the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, judge the pixels with a judgment result of yes as the pixel point of the non-flame area, and judge the pixels with a judgment result of no as the pixel point of the flame area, until each flame image of the target flame at the target time is divided into the flame area and the non-flame area.

[0156] Optionally, in the above technical solution, the equation group building module 204 is specifically used for:

[0157] According to the discretized radiation transfer equation, along the shooting direction corresponding to each preset shooting angle, a group of radiation transfer equations for all pixels in each flame image facing the target flame at the target time is constructed.

[0158] Optionally, in the above technical solution, the equation group building module 204 is further specifically used for:

[0159] Based on the flame radiation transfer law, the radiation transfer equation is constructed;

[0160] The target space of the radiation transfer process is discretized and combined with the radiation transfer equation to obtain the discretized radiation transfer equation.

[0161] Optionally, in the above technical solution, the solving module 203 is further used to introduce an energy constraint function when adaptively adjusting and iteratively solving the radiation transfer equations.

[0162] Optionally, the above technical solution further includes a visualization rendering module, which is used to perform visualization rendering on the three-dimensional appearance of the target flame.

[0163] It should be noted that the beneficial effects of the flame three-dimensional morphology reconstruction system 200 based on physical constraints provided in the above embodiment are the same as the beneficial effects of the flame three-dimensional morphology reconstruction method based on physical constraints, which will not be described in detail here. In addition, when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be described in detail here.

[0164] Among them, the flame three-dimensional morphology reconstruction system based on physical constraints of the present invention can be a computer program (including program code) running in a computer device. For example, the flame three-dimensional morphology reconstruction system based on physical constraints of the present invention is an application software that can be used to execute the corresponding steps in the flame three-dimensional morphology reconstruction method based on physical constraints of the present invention.

[0165] In some embodiments, the flame three-dimensional morphology reconstruction system based on physical constraints of the present invention can be implemented by a combination of software and hardware. As an example, the flame three-dimensional morphology reconstruction system based on physical constraints of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the flame three-dimensional morphology reconstruction method based on physical constraints of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0166] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The name of a module does not limit the module itself in some cases.

[0167] An electronic device according to an embodiment of the present invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any one of the above-mentioned methods for reconstructing the three-dimensional shape of flame based on physical constraints is implemented. That is to say, an electronic device according to an embodiment of the present invention may include but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the method for reconstructing the three-dimensional shape of flame based on physical constraints shown in any one of the embodiments of the present invention by calling the computer program.

[0168] In an alternative embodiment, an electronic device is provided, such as Fig.12 As shown, Fig.12The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0169] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0170] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.12 In the figure, only one thick line is used to represent the bus 4002, but this does not mean that there is only one bus or one type of bus.

[0171] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0172] The memory 4003 is used to store the application code (computer program) for executing the solution of the present invention, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0173] Among them, the electronic device can also be a terminal device, and the terminal device can be any device that can install applications, including at least one of a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a smart TV, and a smart car device.

[0174] It should be noted that Fig.12 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0175] A computer-readable storage medium according to an embodiment of the present invention stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned flame three-dimensional morphology reconstruction methods based on physical constraints is implemented.

[0176] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0177] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes any of the above-mentioned flame three-dimensional morphology reconstruction methods based on physical constraints.

[0178] It should be understood that the flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0179] The computer-readable storage medium provided by the embodiment of the present invention may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or component.

[0180] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0181] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

[0182] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0183] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0184] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A flame three-dimensional morphology reconstruction method based on physical constraints, characterized in that: include: According to a plurality of flame images of the target flame at the target time, a visible boundary of the target flame in the three-dimensional space is calculated, wherein the plurality of flame images of the target flame at the target time are obtained by shooting at each preset shooting angle; Constructing a set of radiation transfer equations for all pixels in each flame image at a target time facing the target flame; In combination with the visible boundary of the target flame at the target moment and the initial data of the three-dimensional distribution of the radiation intensity, the radiation transfer equation group is adaptively adjusted and iteratively solved to obtain the three-dimensional distribution data of the radiation intensity at the target moment, until the three-dimensional distribution data of the radiation intensity at each preset moment is obtained, wherein the target moment is any preset moment, when the target moment is the first preset moment, the initial data of the three-dimensional distribution of the radiation intensity is determined in a preset manner, and when the target moment is not the first preset moment, the three-dimensional distribution data of the radiation intensity at the previous preset moment of the target moment is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target moment; The three-dimensional morphology of the target flame is constructed according to the three-dimensional distribution data of the radiation intensity at each preset moment.

2. The flame three-dimensional morphology reconstruction method based on physical constraints according to claim 1 is characterized in that: The method comprises: dividing the multiple flame images of the target flame at the target moment into a flame area and a non-flame area, projecting and superimposing the images, and calculating the visible boundary of the target flame in the three-dimensional space.

3. The flame three-dimensional morphology reconstruction method based on physical constraints according to claim 1 is characterized in that: The multiple flame images of the target flame at the target time are divided into flame area and non-flame area, including: Obtain and determine the threshold value corresponding to the flame image according to the maximum value among the R value, G value and B value of all pixels of any flame image at the target time, judge whether the maximum value among the R value, G value and B value of each pixel in the flame image is less than the threshold value corresponding to the flame image, judge the pixels with a judgment result of yes as the pixel point of the non-flame area, and judge the pixels with a judgment result of no as the pixel point of the flame area, until each flame image of the target flame at the target time is divided into the flame area and the non-flame area.

4. The flame three-dimensional morphology reconstruction method based on physical constraints according to claim 1 is characterized in that: A set of radiation transfer equations for all pixels in each flame image of the target flame at the target time is constructed, including: According to the discretized radiation transfer equation, a radiation transfer equation group for all pixels in each flame image facing the target flame at the target time is constructed along the shooting direction corresponding to each preset shooting angle.

5. The method for reconstructing three-dimensional flame morphology based on physical constraints according to claim 4, characterized in that: The process of obtaining the discretized radiation transfer equation includes: Based on the flame radiation transfer law, the radiation transfer equation is constructed; The target space of the radiation transfer process is discretized and combined with the radiation transfer equation to obtain the discretized radiation transfer equation.

6. The flame three-dimensional morphology reconstruction method based on physical constraints according to claim 1 is characterized in that: Also includes: When adaptively adjusting and iteratively solving the radiation transfer equations, an energy constraint function is introduced.

7. A flame three-dimensional morphology reconstruction method based on physical constraints according to any one of claims 1 to 5, characterized in that: Also includes: The three-dimensional morphology of the target flame is visualized and rendered.

8. A flame three-dimensional morphology reconstruction system based on physical constraints, characterized in that: It includes a visible boundary acquisition module, an equation group construction module, a solution module and a 3D shape reconstruction module; The visible boundary acquisition module is used to calculate the visible boundary of the target flame in the three-dimensional space according to a plurality of flame images of the target flame at the target time, wherein the plurality of flame images of the target flame at the target time are obtained by shooting at each preset shooting angle; The equation group building module is used to: build a radiation transfer equation group for all pixels in each flame image of the target flame at the target time; The solution module is used to: combine the visible boundary of the target flame at the target time and the initial data of the three-dimensional distribution of the radiation intensity, adaptively adjust and iteratively solve the radiation transfer equation group, obtain the three-dimensional distribution data of the radiation intensity at the target time, until the three-dimensional distribution data of the radiation intensity at each preset time is obtained, wherein the target time is any preset time, when the target time is the first preset time, the initial data of the three-dimensional distribution of the radiation intensity is determined in a preset manner, when the target time is not the first preset time, the three-dimensional distribution data of the radiation intensity at the previous preset time of the target time is used as the initial data of the three-dimensional distribution of the radiation intensity for solving the three-dimensional distribution data of the radiation intensity at the target time; The three-dimensional shape reconstruction module is used to construct the three-dimensional shape of the target flame according to the three-dimensional distribution data of the radiation intensity at each preset moment.

9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a method for reconstructing three-dimensional flame morphology based on physical constraints as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for reconstructing three-dimensional flame morphology based on physical constraints as described in any one of claims 1 to 7 is implemented.

Images