Three-dimensional smoke computed tomography method and system

Abstract

The invention discloses a three-dimensional smoke computed tomography method and system, and the method comprises the following steps: S1, obtaining a to-be-reconstructed smoke image, constructing a corresponding reconstruction region, and carrying out the calculation to obtain a light list corresponding to the pixels of the to-be-reconstructed smoke image; s2, creating a reconstructed sub-region list according to the light list, performing collision detection on elements of the reconstructed sub-region list, and calculating a weight matrix of smoke to be reconstructed; and S3, solving the weight matrix of the to-be-reconstructed smoke, and outputting a reconstruction result of the to-be-reconstructed smoke.

Description

Technical Field

[0001] The present invention belongs to the field of computer tomography (CT) technology, and in particular relates to a three-dimensional smoke computer tomography method and system. Background Art

[0002] Generally, three-dimensional imaging methods can be achieved through computational imaging or scanning plane imaging. Computational imaging specifically includes light field imaging and volume tomography. Light field imaging can achieve instantaneous volume imaging with a single camera, but at the cost of reduced spatial resolution or a reduced field of view. Volume tomography can overcome the above shortcomings of other methods by using multiple cameras to simultaneously capture the projection of emitted or spontaneous emission, but this method is obtained by iterative inversion based on the initial model. Therefore, the degree of closeness between the initial model and the real structure is directly related to whether the imaging results can accurately reflect the objective material properties. How to reasonably and accurately describe the initial model is crucial.

[0003] Scanning plane imaging is to emit sheet lasers through laser emitters and polarizers, and use displacement mechanisms to move the light-emitting mechanism, so as to achieve layer-by-layer scanning of the measured physical field, and combine the data of each layer to complete the three-dimensional reconstruction of scanning plane imaging. However, each layer of data obtained by this method is obtained at a different time, so strictly speaking, the reconstructed three-dimensional image cannot represent the actual situation of the physical field at a certain moment.

[0004] At present, there is no three-dimensional smoke reconstruction algorithm on the market that is efficient, accurate, and saves computing time and computing resources. Summary of the invention

[0005] In view of the technical problems in the prior art, the present application proposes a three-dimensional smoke computed tomography scanning method and system.

[0006] According to one aspect of the present invention, a three-dimensional smoke computer tomography method is provided, comprising:

[0007] S1, obtaining a smoke image to be reconstructed, constructing a corresponding reconstruction area, and calculating a light list corresponding to pixels of the smoke image to be reconstructed;

[0008] S2, creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating a weight matrix of the smoke to be reconstructed;

[0009] S3. Solve the weight matrix of the smoke to be reconstructed, and output the reconstruction result of the smoke to be reconstructed.

[0010] Preferably, S2 specifically includes:

[0011] S201, dividing the reconstruction area into a plurality of voxels according to the number of rays in the ray list;

[0012] S202, taking a beam of light to be measured from the light list, and creating a corresponding reconstruction sub-region list, with the reconstruction region being the first element of the reconstruction sub-region list;

[0013] S203, performing collision detection on the first element in the list of the light to be measured and the reconstructed sub-region, and if a collision occurs, determining whether the first element is composed of only one voxel: if so, calculating the weight of the voxel; if not, decomposing the first element and adding it to the end of the first element;

[0014] S204, removing the first element in the reconstructed sub-region list, and then determining whether there are any remaining elements in the reconstructed sub-region list: if so, repeating S203; if not, sorting all weights of the light to be measured according to the position of the voxel to obtain a weight matrix of the light to be measured;

[0015] S205, repeating S202-S204 for all the lights to be measured in the light list to obtain a weight matrix of the smoke to be reconstructed.

[0016] By using a collision algorithm to sequentially detect whether all rays in the ray list collide with the voxels in the reconstructed sub-area, the accuracy of reconstructing and restoring three-dimensional smoke can be improved.

[0017] Further preferably, obtaining the weight matrix of the light to be measured specifically includes: using a Monte Carlo algorithm for the voxels that collide to calculate the weight of the voxels, sorting all the weights of the light to be measured according to the positions of the voxels, and obtaining the weight matrix of the light to be measured. The weights of the colliding voxels are calculated using the Monte Carlo algorithm, and the obtained weight matrix digitally represents the shape of the three-dimensional smoke to be reconstructed.

[0018] Preferably, solving the weight matrix of the smoke to be reconstructed specifically includes: calculating the projection error between the weight matrix of the smoke to be reconstructed and the smoke image to be reconstructed based on an algebraic reconstruction method and correcting it accordingly to obtain a corrected smoke concentration value, and outputting the reconstruction result of the smoke to be reconstructed. Correcting the projection error between the weight matrix and the smoke image to be reconstructed by the algebraic reconstruction method, and then redistributing the error to each voxel according to the weight value and correcting it in turn can improve the reconstruction accuracy of the three-dimensional smoke.

[0019] Preferably, the weight matrix of the smoke to be reconstructed and the projection error of the smoke image to be reconstructed are calculated based on the synchronous algebraic reconstruction method and corrected accordingly, and processed using the TV regularization algorithm to obtain a corrected smoke concentration value, and output the reconstruction result of the smoke to be reconstructed. The projection error between the weight matrix and the smoke image to be reconstructed is corrected by the synchronous algebraic reconstruction method, and then the error is redistributed to each voxel according to the weight value and corrected in sequence, and then iterated by TV regularization, which can reduce the projection error while reducing the impact of image noise and improve the reconstruction accuracy of three-dimensional smoke.

[0020] Preferably, solving the weight matrix of the smoke to be reconstructed specifically includes: performing singular value decomposition on the weight matrix of the smoke to be reconstructed based on the truncated singular value decomposition method, selecting the optimal number of singular values ​​and the number of singular vectors based on the ridge regression method to solve the weight matrix of the smoke to be reconstructed, and outputting the reconstruction result of the smoke to be reconstructed. The truncated singular value decomposition method has the effect of fast calculation, and combining the ridge regression method to select the optimal number of singular values ​​can achieve higher accuracy.

[0021] Preferably, S1 specifically includes constructing a reconstruction area corresponding to the smoke to be reconstructed and the light list according to the parameters of the smoke image to be reconstructed.

[0022] According to one aspect of the present invention, a three-dimensional smoke computer tomography system is provided, comprising:

[0023] Preprocessing module: obtaining the smoke image to be reconstructed, constructing the corresponding reconstruction area, and calculating a list of rays corresponding to the pixels of the smoke image to be reconstructed;

[0024] Collision detection module: creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating the weight matrix of the smoke to be reconstructed;

[0025] Matrix solving module: solves the weight matrix of the smoke to be reconstructed, and outputs the reconstruction result of the smoke to be reconstructed.

[0026] According to one aspect of the present invention, a computer program product is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to any one of the first aspects is implemented.

[0027] According to one aspect of the present invention, a computing system is provided, comprising a processor and a memory, wherein the processor is configured to execute the method as described in any one of the first aspects.

[0028] Compared with the prior art, the present invention is beneficial in that:

[0029] By detecting whether the voxels in each reconstructed sub-area collide, the shape of the 3D smoke to be reconstructed is represented digitally. The real shape of the 3D smoke at a certain moment is reconstructed by solving the weight matrix to improve the reconstruction accuracy.

[0030] It combines multiple algorithms to solve the weight matrix, which can save computing time and resources while improving calculation accuracy. Users can choose according to the actual application scenario to meet the usage needs in various scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated into and constitute a part of this specification. The accompanying drawings illustrate the embodiments and are used together with the description to explain the principles of the present invention. It will be easy to recognize other embodiments and many expected advantages of the embodiments because they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale with each other. The same reference numerals refer to corresponding similar parts.

[0032] Figure 1 A flow chart of a method for three-dimensional smoke computed tomography scanning according to the present invention is shown;

[0033] Figure 2 shows a smoke image to be reconstructed according to an embodiment of the present invention;

[0034] Figure 3 A flow chart of performing collision detection according to the method proposed in the present invention is shown;

[0035] Figure 4 A flow chart of solving the weight matrix of smoke to be reconstructed according to the method proposed by the present invention is shown;

[0036] Figure 5 The reconstructed smoke image obtained by the method proposed in the present invention is shown;

[0037] Figure 6 A system structure diagram of a three-dimensional smoke computer tomography according to the present invention is shown;

[0038] Figure 7 A schematic diagram of the computer system structure of an electronic device suitable for implementing an embodiment of the present application is shown. DETAILED DESCRIPTION

[0039] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0041] Figure 1 A flow chart of a method for three-dimensional smoke computed tomography scanning according to an embodiment of the present invention is shown, comprising:

[0042] S1, obtaining a smoke image to be reconstructed, constructing a corresponding reconstruction area, and calculating a light list corresponding to pixels of the smoke image to be reconstructed;

[0043] S2, creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating a weight matrix of the smoke to be reconstructed;

[0044] S3. Solve the weight matrix of the smoke to be reconstructed, and output the reconstruction result of the smoke to be reconstructed.

[0045] In a specific embodiment, there is no restriction on the angle, quantity and size of the input photo, and there is no restriction on the size of the reconstruction area and the method of dividing the voxels. S1 specifically includes constructing a reconstruction area corresponding to the smoke to be reconstructed according to the smoke image to be reconstructed, and obtaining the light list according to the parameters (size, quantity, projection angle) of the smoke image to be reconstructed.

[0046] According to the projection angle of any smoke image to be reconstructed, the local coordinate system of the image can be transformed into the global coordinate system, that is, the directions of the x-axis (pixel rows remain unchanged, columns change), y-axis (pixel columns remain unchanged, rows change), and z-axis (perpendicular to the image plane) in the global coordinate system of the plane where the image is located are obtained. The center point of the image is placed at the origin of the global coordinate system, and the spatial coordinates of the center point of the pixel can be obtained according to the relative position of any pixel in the image and the center pixel. With the coordinate as the center, the length of each pixel in the x- and y-axis directions is the length and width, and the larger length in the z-axis direction (throughout the entire calculation domain) is the height. A rectangular parallelepiped can be obtained, which is the light corresponding to the pixel in the calculation domain. The above operations are performed on all images and pixels to obtain the light list.

[0047] In an alternative embodiment, if Figure 2 As shown in the figure, there are 5 smoke images to be reconstructed, each with a size of 81*81 pixels. Based on this, the reconstruction area is constructed as a cube with a side length of 82.5 pixels, and the area is discretized into 35937 cube-shaped voxels with a side length of 2.5 pixels. The normal of the projection surface of each photo is known, so 32805 cuboid rays can be obtained.

[0048] like Figure 3As shown, these rectangular rays are put into a ray list L, a beam of light to be measured is taken out from the ray list L, and a corresponding reconstruction sub-region list A is created, and the reconstruction region is included in the reconstruction sub-region list as its first reconstruction sub-region, that is, the first element. The light to be measured and the first element of the reconstruction sub-region list A are subjected to collision detection by a collision detection algorithm.

[0049] In one embodiment, the collision detection algorithm uses the obb directional bounding box and its 15 separating axes for calculation.

[0050] If a collision occurs, it is determined whether the first element consists of only one voxel. When the first element consists of only one voxel, the weight of the voxel is calculated by the Monte Carlo algorithm to obtain the weight matrix of the light to be measured. If the first element does not consist of one voxel, the first element is decomposed and added to the end of the reconstructed sub-region list A. Figure 3 As shown, the first element is decomposed into two sub-areas. It can be understood that the decomposition method is not limited to the dichotomy method and can be adjusted according to actual needs.

[0051] The Monte Carlo algorithm calculates the weight value of each position by counting random points. The number of random points in the Monte Carlo algorithm is 50,000 to 2 million. The larger the set number, the more accurate the calculated weight matrix is, and the calculation time increases.

[0052] Regardless of whether a collision occurs, after the collision detection is completed, the first element in the reconstructed sub-region list A needs to be removed, and then it is determined whether there are any remaining elements in the reconstructed sub-region list A.

[0053] If there is, then take out the new first element in the reconstructed sub-region list A in order, and perform the above collision detection step with the light to be tested again. Otherwise, all weights of the light are sorted according to the position of the voxel, and the weight matrix calculation of the light is completed. Perform the above operation on all the rays in the ray list to obtain a complete weight matrix.

[0054] The ray list can be distributed to different threads for calculation through multi-threaded parallel computing.

[0055] After obtaining the weight matrix of the smoke to be reconstructed, the weight matrix is ​​solved.

[0056] like Figure 4 As shown, the solution calculation method can select one of the algebraic reconstruction method (ART algorithm), truncated singular value decomposition method, synchronous algebraic reconstruction method (SART algorithm) combined with TV regularization algorithm according to specific circumstances such as computing time and resources, computing accuracy, whether to use an iterative algorithm, and whether to use a noise reduction operation.

[0057] In one embodiment, the weight matrix of the smoke to be reconstructed is solved, specifically including: performing singular value decomposition on the weight matrix of the smoke to be reconstructed based on the truncated singular value decomposition method. Theoretically, the singular value decomposition method can decompose multiple singular values ​​at a time and select a certain number for calculation during calculation, but it is found in use that only selecting part of the singular value decomposition for calculation will lead to large errors. For example, the error of selecting 1000 for calculation after performing 2000 singular value decompositions is much greater than the error of only decomposing 1000 singular values ​​and using 1000 for calculation. Therefore, when the singular value decomposition method is calculated, the number of singular values ​​gradually increases from small to large, and each singular value number is subjected to a singular value decomposition and equation solving until the optimal singular value number is found. Due to the defects of this method in application, it is recommended to give priority to iterative algorithms in actual use.

[0058] Therefore, the present invention selects the optimal number of singular values ​​and the number of singular vectors based on the truncated singular value decomposition method in combination with the ridge regression method, thereby solving the weight matrix of the smoke to be reconstructed and outputting the reconstruction result of the smoke to be reconstructed.

[0059] The specific operation of the ridge regression method is as follows: select multiple singular values, use the singular values ​​of this number to solve respectively, statistically analyze the second norm of the solution result ||x||2 and the second norm of the back projection error ||Ax-P||2, draw a graph with ||x||2 as the horizontal coordinate and ||Ax-P||2 as the vertical coordinate, and select the number of singular values ​​corresponding to the point with the largest curvature in the image as the optimal number of singular values.

[0060] In one embodiment, the weight matrix of the smoke to be reconstructed is solved, specifically including: initializing the smoke concentration values ​​in all voxels, calculating the projection error between the weight matrix of the smoke to be reconstructed and the smoke image to be reconstructed based on the algebraic reconstruction method and correcting it accordingly to obtain the corrected smoke concentration value, and outputting the reconstruction result of the smoke to be reconstructed.

[0061] The calculation principle of the algebraic reconstruction method is to calculate the projection error between the weight matrix of the smoke to be reconstructed and the smoke image to be reconstructed row by row and feed back the correction calculation result. Its physical meaning is to calculate the projection of the current reconstructed area at a certain pixel point, compare it with the smoke image to be reconstructed, and obtain the projection error. Then, the error is redistributed to each voxel according to the weight value. Finally, each voxel is corrected according to the error value assigned to it.

[0062] The calculation formula is: in, is the value in the jth voxel after the kth iteration, is the value after the k+1th iteration, w i，j is the weight of the ith pixel for the jth voxel, p iis the value in the i-th pixel, N is the number of columns in the matrix and also the number of voxels. λ is the relaxation factor, ranging from 0 to 2, which is used to control the convergence speed and calculation accuracy. The typical value range is 0 to 1. The larger the number, the faster the calculation, but the error will be greater when the calculation terminates; the smaller the value, the slower the convergence, but the calculation accuracy is generally higher. The method for determining the convergence criterion is to calculate the 2-norm δ of the change between two adjacent iterative solutions k To judge whether it converges, the calculation formula is: δ k is the 2-norm of the change between two adjacent iteration results, F k is the calculation result of the kth iteration, N is the number of voxels, and the convergence criterion ξ is a constant ranging from 0 to 1. For example, 1e -6 .

[0063] In one embodiment, the weight matrix of the smoke to be reconstructed and the projection error of the smoke image to be reconstructed are calculated based on the synchronous algebraic reconstruction method and corrected accordingly, and processed using the TV regularization algorithm to obtain the corrected smoke concentration, and the reconstruction result of the smoke to be reconstructed is output.

[0064] The calculation principle of the synchronous algebraic reconstruction method is that the weight matrix of the smoke to be reconstructed is divided into blocks by row according to the smoke image to be reconstructed, and the projection errors of all rows in a block (all pixels in a smoke image to be reconstructed) are calculated at the same time, and then redistributed to each voxel according to the weight in the weight matrix, and finally each voxel is iterated according to the error value to complete the synchronous algebraic reconstruction method. Its calculation formula is in, is the value in the jth voxel after the kth iteration, is the value after the k+1th iteration, w i，j is the weight of the ith pixel for the jth voxel, p i is the value in the i-th pixel, P φ Refers to any smoke image to be reconstructed, N is the number of columns in the matrix and also the number of voxels, λ is the relaxation factor. The relaxation factor λ of the synchronous algebraic reconstruction method is 0-2, and the convergence criterion ξ ranges from 0 to 1.

[0065] The calculation principle of the TV regularization algorithm is to calculate the total variation of all voxels after obtaining the results of the synchronous algebraic reconstruction method, and then differentiate the total variation for each voxel and iterate according to the steepest descent method to reduce noise. In addition, after each iteration in the process of solving the equation, the negative values ​​in the voxels need to be set to zero to ensure their non-negativity.

[0066] The calculation steps are as follows: (1) Calculate the total variation: Where i, j, k are the positions of the voxels, x, y, z can be any number between -1, 0, and 1, determined by the selected adjacent voxels, and ε is a small constant used to ensure that μ i，j，k Not 0, for example, it can be 1e -8 , μ of all voxels i，j，k The sum gives the expression for total variation ||F|| TV =∑ i，j，k μ i，j，k ; (2) Calculate the gradient of the total variation (3) Calculation using the steepest descent method in

[0067] The value of η in step (3) is calculated in real time during the iteration process, and the formula is: The proportionality coefficient α is obtained by performing geometric similarity calculations without using TV regularization or is given based on experience. For example, α can be the minimum point in the error curve corresponding to the error when only the SART algorithm is used without using TV regularization. The proportional coefficient with Nλ is determined by the characteristics of the example itself. In actual use, it can be given based on experience. Since the error curve drops faster and rises slower, α can be made smaller to delay the effective time of TV regularization.

[0068] Both the algebraic reconstruction method and the synchronous algebraic reconstruction method combined with the TV regularization algorithm can distribute the weight matrix to different threads through multi-thread distribution for iterative calculation until the calculation results meet the convergence conditions and then output them.

[0069] During the calculation process, when you need to quickly obtain an overview of the smoke distribution, select the ART algorithm. When you need a higher precision result, use the SART and TV regularization algorithms. The ART algorithm is simple to operate and requires fewer parameters to be set during calculation. You only need to input the image and weight matrix and set the relaxation factor and convergence conditions to start the calculation. These parameters can be fixed, and the values ​​can be large to achieve a fast calculation effect. They are mainly used to obtain the approximate shape of the distribution in a short time. The operation of the SART and TV regularization algorithms is more complicated. In order to achieve higher precision, the values ​​when setting the relaxation factor and convergence conditions are often relatively small. Therefore, the calculation time using the SART and TV regularization algorithms will be longer. They are suitable for conditions that allow long-term calculations to obtain high precision.

[0070] When the basic shape of smoke distribution needs to be obtained in a short time, the algebraic reconstruction method can be used and the relaxation factor and convergence condition can be set to a relatively large value. When a high-precision result is required and a long calculation period is allowed, the synchronous algebraic reconstruction method combined with the TV regularization algorithm can be selected and the parameters can be set to a relatively small value. The truncated singular value decomposition method can achieve the same fast calculation effect as ART by setting a small number of singular values ​​and performing only one calculation. By selecting multiple singular values ​​from small to large for calculation and using the ridge regression method to select the optimal number of singular values, a higher-precision result can be achieved.

[0071] Figure 5 Shown is based on Figure 2 The smoke image to be reconstructed is calculated using the synchronous algebraic reconstruction method combined with the TV regularization algorithm. Figure 5 The positions and numbers of voxels with a concentration greater than 50 are given, and voxels with a concentration less than 50 in the computational domain are hidden. Figure 5 and Figure 2 By comparison, it can be seen that the method provided by the present invention achieves the calculation purpose of three-dimensional reconstruction of smoke.

[0072] like Figure 6 As shown, according to one aspect of the present invention, a three-dimensional smoke computer tomography system is proposed, comprising:

[0073] Preprocessing module 601: obtaining a smoke image to be reconstructed, constructing a corresponding reconstruction area, and calculating a list of rays corresponding to pixels of the smoke image to be reconstructed;

[0074] Collision detection module 602: creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating a weight matrix of the smoke to be reconstructed;

[0075] Matrix solving module 603: solves the weight matrix of the smoke to be reconstructed, and outputs the reconstruction result of the smoke to be reconstructed.

[0076] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 suitable for implementing an electronic device of an embodiment of the present application. Figure 7 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0077] like Figure 7As shown, the computer system 700 includes a central processing unit (CPU) 701, which performs various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 709 into a random access memory (RAM) 704. In the RAM 704, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, the ROM 703, and the RAM 704 are connected to each other via a bus 705. An input / output (I / O) interface 706 is also connected to the bus 705.

[0078] The following components are connected to the I / O interface 706: an input section 707 including a keyboard, a mouse, etc.; an output section 708 including a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 709 including a hard disk, etc.; and a communication section 710 including a network interface card such as a LAN card, a modem, etc. The communication section 710 performs communication processing via a network such as the Internet. A drive 711 is also connected to the I / O interface 706 as needed. A removable medium 712, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 711 as needed so that a computer program read therefrom is installed into the storage section 709 as needed.

[0079] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart is implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program is downloaded and installed from the network through the communication part 710, and / or installed from the removable medium 712. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the method of the present application are executed.

[0080] It should be noted that the computer-readable storage medium of the present application is a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium is, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media 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 of the above. In the present application, a computer-readable storage medium is any tangible medium containing or storing a program that is used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium includes a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal takes a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium is any computer-readable storage medium other than a computer-readable storage medium that sends, propagates or transmits a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable storage medium is transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0081] Computer program code for performing the operations of the present application is written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code is executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer is connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or is connected to an external computer (e.g., via the Internet using an Internet service provider).

[0082] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram represents 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 implementations as replacements, the functions marked in the box also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession are actually executed substantially in parallel, and they are sometimes also 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, are implemented by a dedicated hardware-based system that performs the specified function or operation, or are implemented by a combination of dedicated hardware and computer instructions.

[0083] The modules involved in the embodiments of the present application are implemented by software and hardware.

[0084] As another aspect, the present application also provides a computer-readable storage medium, which is included in the electronic device described in the above embodiment; it also exists independently and is not assembled into the electronic device. The above computer-readable storage medium carries one or more programs. When the above one or more programs are executed by the electronic device, the electronic device: S1. Obtain the smoke image to be reconstructed, construct the corresponding reconstruction area, and calculate the light list corresponding to the pixels of the smoke image to be reconstructed; S2. Create a reconstruction sub-area list according to the light list, perform collision detection on the elements of the reconstruction sub-area list, and calculate the weight matrix of the smoke to be reconstructed; S3. Solve the weight matrix of the smoke to be reconstructed, and output the reconstruction result of the smoke to be reconstructed.

[0085] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application 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 invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A three-dimensional smoke computed tomography method, characterized in that: The following steps are involved: S1, obtaining a smoke image to be reconstructed, constructing a corresponding reconstruction area, and calculating a list of rays corresponding to pixels of the smoke image to be reconstructed; S2, creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating a weight matrix of the smoke to be reconstructed; S3. Solve the weight matrix of the smoke to be reconstructed, and output a reconstruction result of the smoke to be reconstructed.

2. The three-dimensional smoke computed tomography method according to claim 1, characterized in that: S2 specifically includes: S201, dividing the reconstruction area into a plurality of voxels according to the number of rays in the ray list; S202, taking a beam of light to be measured from the light list, and creating a corresponding reconstruction sub-region list, with the reconstruction region being the first element of the reconstruction sub-region list; S203, performing collision detection on the first element in the list of the light to be measured and the reconstructed sub-region, and if a collision occurs, determining whether the first element is composed of only one voxel: if so, calculating the weight of the voxel; if not, decomposing the first element and adding it to the end of the first element; S204, removing the first element in the reconstructed sub-region list, and then determining whether there are any remaining elements in the reconstructed sub-region list: if so, repeating S203; if not, sorting all weights of the light to be measured according to the position of the voxel to obtain a weight matrix of the light to be measured; S205, repeating S202-S204 for all the lights to be measured in the light list to obtain a weight matrix of the smoke to be reconstructed.

3. The three-dimensional smoke computed tomography method according to claim 2, characterized in that: Obtaining the weight matrix of the light to be measured specifically includes: using a Monte Carlo algorithm on the collided voxels to calculate the weights of the voxels, and sorting all the weights of the light to be measured according to the positions of the voxels to obtain the weight matrix of the light to be measured.

4. The method for three-dimensional smoke computed tomography according to claim 1, characterized in that: Solving the weight matrix of the smoke to be reconstructed specifically includes: calculating the projection error of the weight matrix of the smoke to be reconstructed and the smoke image to be reconstructed based on an algebraic reconstruction method and correcting it accordingly to obtain a corrected smoke concentration value, and outputting a reconstruction result of the smoke to be reconstructed.

5. The method for three-dimensional smoke computed tomography according to claim 1, characterized in that: The weight matrix of the smoke to be reconstructed is solved, specifically including: calculating the weight matrix of the smoke to be reconstructed and the projection error of the smoke image to be reconstructed based on the synchronous algebraic reconstruction method and correcting it accordingly, and processing it with the TV regularization algorithm to obtain the corrected smoke concentration value, and outputting the reconstruction result of the smoke to be reconstructed.

6. The method for three-dimensional smoke computed tomography according to claim 1, characterized in that: Solving the weight matrix of the smoke to be reconstructed specifically includes: performing singular value decomposition on the weight matrix of the smoke to be reconstructed based on a truncated singular value decomposition method, selecting the optimal number of singular values ​​and the number of singular vectors based on a ridge regression method to solve the weight matrix of the smoke to be reconstructed, and outputting a reconstruction result of the smoke to be reconstructed.

7. The three-dimensional smoke computed tomography method according to claim 1, characterized in that: S1 specifically includes constructing a reconstruction area corresponding to the smoke to be reconstructed and the light list according to the parameters of the smoke image to be reconstructed.

8. A three-dimensional smoke computer tomography system, characterized in that: include: Preprocessing module: obtaining the smoke image to be reconstructed, constructing the corresponding reconstruction area, and calculating a list of rays corresponding to the pixels of the smoke image to be reconstructed; Collision detection module: creating a reconstruction sub-region list according to the light list, performing collision detection on the elements of the reconstruction sub-region list, and calculating the weight matrix of the smoke to be reconstructed; Matrix solving module: solves the weight matrix of the smoke to be reconstructed, and outputs the reconstruction result of the smoke to be reconstructed.

9. A computer program product having a computer program stored thereon, characterized in that: The computer program implements the method according to any one of claims 1 to 7 when executed by a processor.

10. A computing system comprising a processor and a memory, wherein the processor is configured to execute the method according to any one of claims 1 to 7.

Images