A method for finding the location of fusion neutron incident point

Abstract

The present invention discloses a method for quickly finding the position of fusion neutron incident points in the field of pulsed fast neutron imaging with high precision. All measured signals are classified into several sets, and the mapping relationship of the Hough transform is used to reconstruct the tracks of all signals in each set, and the positions of all neutron incident points are found. The present invention first classifies and segments the signals, so that the amount of data required for analysis is greatly reduced. All signals in each signal set can be cyclically exhausted and their tracks can be extracted, so that the operating speed of the system is greatly improved. In addition, the present invention adopts an optimized Hough transform algorithm in the process of realizing track finding, and finally obtains the neutron incident point, which can realize efficient and high-precision neutron incident point finding.

Description

Technical Field

[0001] The present invention relates to the field of pulsed fast neutron imaging, and in particular to a method for finding the position of a fusion neutron incident point. Background Art

[0002] Fusion nuclear reactions produce a large number of 14MeV fast neutrons. Neutron measurements can be used to image the neutron source, thereby enabling imaging and diagnosis of the fusion region. Fast neutron imaging primarily utilizes pinhole or penumbra-encoded imaging technology, which primarily involves components such as the pinhole or penumbra aperture and a fast neutron imaging detector. The neutron's impact position is determined based on the signal characteristics of the neutron-induced recoil protons or ions, enabling imaging of the fusion region. During this process, the neutron impact position is determined by: 1) visually inspecting the signal image obtained to determine the location of the neutron impact point; and 2) reconstructing the track using a computer program, often employing the Hough transform algorithm.

[0003] However, it is difficult to accurately extract the locations of the incident points of a large number of neutrons using conventional Hough transform, and it is even more difficult to fully utilize large-area fast neutron detectors to perform neutron imaging in high-neutron-yield fusion regions.

[0004] The above defects are worth solving. Summary of the Invention

[0005] In order to overcome the deficiencies of the existing technology, the present invention provides a method for finding the position of the fusion neutron incident point.

[0006] The technical solution of the present invention is as follows:

[0007] A method for finding the position of fusion neutron incident points is characterized by classifying all measured signals into several sets, using the mapping relationship of Hough transform to reconstruct the tracks of all signals in each set, and finding the positions of all neutron incident points.

[0008] The present invention according to the above solution is characterized in that it includes the following steps:

[0009] S1. Obtain the distribution map of all signals recorded by the detector;

[0010] S2. Classify all signals in the signal distribution diagram to form several signal sets;

[0011] S3. Decompose each signal set using the Hough transform algorithm to obtain the track or point of each signal set;

[0012] S4. Determine the starting position of the track based on the ion energy deposition characteristics, thereby obtaining the position of the neutron incident point.

[0013] Furthermore, in step S1 , a capillary array detector is simulated to obtain a distribution diagram of all signals.

[0014] Furthermore, in step S2, all adjacent signals are determined to be a set.

[0015] Furthermore, in the signal distribution diagram, the signals are classified and divided according to whether their positions are adjacent, to obtain several signal sets.

[0016] Furthermore, in the same signal set, the difference between the pixel rows and columns of two adjacent signals is no more than 2.

[0017] Furthermore, in step S3, all signals of each set are transformed into the Hough space, and the peak values in the Hough space are extracted in sequence to obtain the corresponding straight line signals in the signal distribution diagram, and then the tracks of all signals are obtained.

[0018] Furthermore, it specifically includes:

[0019] Step S31: Perform Hough transform on all signals in each set to obtain a curve of each signal in the Hough space;

[0020] Step S32: Find the maximum value in the Hough space;

[0021] Step S33: determining a straight line corresponding to the maximum value in the signal distribution diagram, and extracting a track falling on the straight line;

[0022] Step S34: Repeat steps S32 to S33 until all signal traces are extracted.

[0023] Furthermore, in step S4, in the process of finding the position of the neutron incident point, the end of the track with a weak signal is determined to be the position of the neutron incident point based on the energy deposition characteristics of the generated neutrons.

[0024] The present invention, according to the above-described scheme, has the beneficial effect of first classifying and segmenting the signals, significantly reducing the amount of data required for analysis. It then loops through all signals in each signal set and extracts their tracks, significantly improving the system's operating speed. Furthermore, the present invention utilizes an optimized Hough transform algorithm during track finding to ultimately determine the neutron impact point, enabling efficient and high-precision neutron impact point finding. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of the Hough transform principle;

[0026] Figure 2 is a signal distribution image detected in a specific embodiment;

[0027] Figure 3 A sinusoidal graph obtained by Hough transforming a signal distribution image in a specific embodiment;

[0028] Figure 4 for Figure 3 The signal ensemble is formed by the combination of three recoil proton tracks;

[0029] Figures 5 to 7 A schematic diagram of changes in a Hough space sinogram for extracting the neutron incident point position in a preferred embodiment;

[0030] Figure 8 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0032] like Figures 1 to 7 As shown, in order to achieve rapid and high-precision search and confirmation of neutron incident position points, the present invention provides a method for finding the position of fusion neutron incident points, which classifies all measured signals into several sets, uses the mapping relationship of Hough transform to reconstruct the tracks of all signals in each set, and finds the positions of all neutron incident points.

[0033] The present invention classifies and segments signals by position, dividing all signals into sets of multiple signals, thereby reducing the complexity of the analyzed data; on this basis, by optimizing the Hough transform algorithm, tracking is performed in units of signal sets, and finally the positions of all neutron incident points are obtained. This is an efficient, high-precision method for automatically finding the positions of neutron incident points.

[0034] The Hough transform is widely used in image analysis and processing. The Hough transform transforms image information from the image coordinate space into the Hough parameter space, achieving image fitting within both spaces. Specifically, the Hough transform leverages the duality between points and lines, mapping discrete pixels on a line in image space into curves in Hough space using parametric equations. Furthermore, the intersections of multiple curves in Hough space are used as parameters of the line equations to map these lines back to the image space.

[0035] From this we can see that a point in the image space corresponds to a line in the Hough space; the points corresponding to the straight line passing through a point in the Hough space are on a straight line; a line in the image space corresponds to a point in the Hough space, and the corresponding straight lines of a collinear point in the Hough space intersect at a point.

[0036] like Figure 1As shown, a straight line in image space (xy coordinate space) with two points (x1, y1) and (x2, y2) on it is mapped to a point (r', θ') in Hough space (θ-r coordinate space). Conversely, a point (r', θ') in Hough space (θ-r coordinate space) is transformed to image space (xy coordinate space) to obtain a straight line. In Hough space, this is described by the following formula:

[0037] r=x×cosθ+y×sinθ

[0038] Where r is the distance from the origin to the normal of the line where the points (x1, y1) and (x2, y2) are located in the image space (xy coordinate space), and θ is the angle between the normal of the line where the points (x1, y1) and (x2, y2) are located in the image space (xy coordinate space) and the x-axis.

[0039] Figure 2 In a specific embodiment, the signal distribution of 200 fusion fast neutrons obtained after simulating a capillary array detector is shown. This signal distribution is then transformed into a sinusoidal image in Hough space (θ-r coordinate space). Finding all points representing straight lines in these sinusoidal images is extremely difficult. Therefore, the present invention uses signal classification and an optimized Hough algorithm to quickly and accurately determine the location of the neutron impact point.

[0040] Specifically, the method for finding the position of the fusion neutron incident point includes the following steps:

[0041] S1. Obtain a distribution map of all signals in the fusion reaction.

[0042] like Figure 2 As shown, in this step, image signals are acquired by a neutron imaging detector with micron resolution. Specifically, an analog capillary array detector can be used to obtain a distribution diagram of all signals in the image space (xy coordinate space), which can obtain higher position resolution results.

[0043] S2. Classify all signals in the signal distribution diagram into several signal sets.

[0044] like Figure 2 、 Figure 4 As shown in the figure, in the signal distribution diagram, the positions and energy values of all signals (i.e., the amplitudes of the corresponding signals) are first extracted; then, the signals are classified and segmented based on whether they are adjacent. Adjacent signals are taken as a set to obtain several signal sets.

[0045] In the same signal set, the criterion for determining whether two signals are adjacent is that the difference between the pixel rows and columns where the two adjacent signals are located is no more than 2.

[0046] After all adjacent signals are taken as a signal set, the signal set may be a set of one or more recoil protons, recoil carbon ions, or other particle signals. Figure 4 Shown is a signal composite consisting of three recoil proton tracks.

[0047] S3. Use the Hough transform algorithm to decompose each signal set to obtain the track or point of each signal set.

[0048] Transform all signals of each set into the Hough space, extract the peak values in the Hough space in sequence, obtain the corresponding straight line signals in the signal distribution diagram, and then obtain the tracks of all signals. This step specifically includes:

[0049] Step S31: Perform Hough transform on all signals in each set to obtain a curve of each signal in the Hough space.

[0050] Step S32: Find the maximum value in the Hough space.

[0051] Step S33: Determine the straight line corresponding to the maximum value in the signal distribution diagram, and extract the track falling on the straight line.

[0052] Step S34: Repeat steps S32 to S33 until all signal traces are extracted.

[0053] Preferably, due to the difference in linear energy deposition (LET), the signal caused by electrons is weaker than the ion signal (one order of magnitude difference), so before performing the Hough transform algorithm, a corresponding threshold is set for each pixel signal to remove the interference of gamma-induced electron tracks (considered as noise signals, interference signals).

[0054] like Figure 2 、 Figures 4 to 7 As shown, the following is a detailed description using a specific embodiment:

[0055] A. First, determine a signal set in the image space, such as Figure 2 、 Figure 4 Then, Figure 4 All signals in the signal set shown in the figure are subjected to conventional Hough transformation to obtain Figure 5 The sinusoidal image of the Hough space (θ-r coordinate space) shown is Figure 5 As shown in the middle number (1).

[0056] B. Find the first maximum value (peak), the brightest point, in the Hough space (θ-r coordinate space), such as Figure 5 The point indicated by number (2).

[0057] C. Determine the straight line in the image space (xy coordinate space) corresponding to the peak value shown by label (2) in the Hough space (θ-r coordinate space), extract the signal falling on the straight line, and obtain the following: Figure 4 The straight line signal is shown in (3). Since the tracks of various ions induced by neutrons are continuous, if there are non-adjacent signals, the longest continuous signal can be selected. Other non-adjacent signals belong to other tracks and are not considered here.

[0058] At this point, the first signal trace is obtained.

[0059] D. Return to Figure 5 The Hough space (θ-r coordinate space) shown in the figure removes the sinusoidal curve obtained by Hough transformation of the first signal trajectory, and the result is as follows Figure 6 The set of sinusoidal curves shown in (4).

[0060] E. Find the second maximum value (peak), the brightest point, in the Hough space (θ-r coordinate space), such as Figure 6 The point indicated by number (5) in the image space. This step returns to step B and continues to determine the straight line in the image space, that is, Figure 4 The straight line signal shown by the number (6).

[0061] F. And so on, until all the signal traces are extracted. Figure 6 The curve corresponding to the trajectory of the second signal in Figure 7 The set of sinusoidal curves in the label (7); according to the third peak in the Hough space (θ-r coordinate space) ( Figure 7 The point indicated by number (8) in the image space determines the third track in the image space, namely, Figure 4 The straight line signal shown by the number (9).

[0062] At this point, the tracks of all signals corresponding to the set in the image space (xy coordinate space) are extracted, namely, line (3), line (6), and line (9).

[0063] S4. Determine the starting position of the track based on the ion energy deposition characteristics, thereby determining the location of the neutron impact point. Specifically, based on the energy deposition characteristics of the recoil protons or other ions, a Bragg peak is generated at the end of the track. Therefore, the end opposite the Bragg peak (the end with the weaker signal) is determined to be the location of the neutron impact point.

[0064] In this step, the process of finding the position of the neutron incident point specifically includes: if the track is a straight line, it is judged to be a recoil proton track, and the starting point of the straight line track is the neutron incident point; if the track is a point or a short straight line (specifically, a signal of less than 5 pixels), it is judged to be an ion track, and the starting point of the point or short straight line is the neutron incident point.

[0065] Specifically, the characteristics of the Bragg peak are used to determine the starting point of the straight line track, that is, to determine the other end of the Bragg peak as the neutron incident point. Figure 4 In the image set shown, the positions circled by numbers (10), (11), and (12) are the starting points of the three tracks, that is, the incident points of the neutrons.

[0066] The present invention utilizes the above-mentioned method to classify and segment the signals, thereby greatly reducing the data that needs to be analyzed, and can greatly improve the operating speed when all signals in each signal set are exhausted in a loop; at the same time, by centrally judging the signals in the same set, it is possible to track the multiple straight lines and intersecting straight line tracks, and the accuracy of extracting the neutron incident point position can be improved.

[0067] like Figure 8 As shown, the present invention also provides a system for implementing the above-mentioned method of finding the location of the fusion neutron impact point, which includes an image signal acquisition module, a signal classification module, a Hough transform module, a track signal extraction module, and an impact point judgment module.

[0068] (1) Image signal acquisition module

[0069] The image signal acquisition module is connected to the signal detector to acquire a distribution diagram of the signal of the signal detector.

[0070] (2) Signal classification module

[0071] The signal classification module is used to classify all signals in the signal distribution map according to the signal classification judgment criteria, forming several signal sets. The classification judgment criteria of this signal classification module are: whether the difference between the row and column of the pixels where two adjacent signals are located is no more than 2. If the difference between the row and column of the pixels where two adjacent signals are located is no more than 2, the two signals are judged to belong to the same signal set.

[0072] Of course, in order to avoid interference from signals such as noise and interference, the signal classification module needs to first set a threshold for each pixel signal to ensure the accuracy of the final determined signal.

[0073] (3) Hough transform module

[0074] The Hough transform module embeds the Hough transform algorithm, which is used to perform Hough transform on all signal polarities in each signal set, converting the image signal in the image space (xy coordinate space) into a sinusoidal graph in the Hough space (θ-r coordinate space).

[0075] (4) Track signal extraction module

[0076] The track signal extraction module is used to sequentially find the tracks of multiple straight lines (or point signals) within the signal set. Specifically:

[0077] First, determine the first peak in the Hough space (θ-r coordinate space), and extract the straight line corresponding to the peak in the image space (xy coordinate space) to obtain the trajectory of the first signal;

[0078] Then, in the signal set after removing the first peak in the Hough space (θ-r coordinate space), the second peak is determined, and the straight line corresponding to the second peak is extracted in the image space (xy coordinate space) to obtain the trajectory of the second signal;

[0079] And so on, until the traces of all signals are extracted.

[0080] (5) Incident point judgment module

[0081] The incident point judgment module is used to judge the tracks of all signals extracted by the track signal extraction module and determine the position of the neutron incident point.

[0082] If the signal track is a straight line, then based on the existence principle of the Bragg peak, the neutron incident point is determined to be the starting point of the straight line track (that is, the other end of the Bragg peak); if the signal track is a point or a very short straight line, then it is judged that the track corresponds to the recoil carbon ion, or other ion signal, and the neutron incident point is the point where this track is located.

[0083] At this point, the system can quickly and accurately obtain the position of the neutron incident point of all image signals, and the result is highly accurate and acquired quickly.

[0084] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention.

[0085] The above is an exemplary description of the patent of the present invention in conjunction with the accompanying drawings. It is obvious that the implementation of the patent of the present invention is not limited to the above-mentioned method. As long as various improvements are made by adopting the method concept and technical solution of the patent of the present invention, or the concept and technical solution of the patent of the present invention are directly applied to other occasions without improvement, they are all within the scope of protection of the present invention.

Claims

1. A method for finding the position of a fusion neutron incident point, characterized in that: All measured signals are classified into several sets, and the trajectory of all signals in each set is reconstructed using the mapping relationship of the Hough transform, and the positions of all neutron incident points are found. The specific steps include: S1. Obtain the distribution map of all signals recorded by the detector; S2. Classify all signals in the signal distribution graph into several signal sets, and determine that all adjacent signals are a set; S3. Using the Hough transform algorithm, decompose each signal set to obtain the trajectory of each signal set; S4. Determine the starting position of the track based on the ion energy deposition characteristics, thereby obtaining the position of the neutron incident point; In step S3, all signals of each set are transformed into the Hough space, and the peak values in the Hough space are extracted in sequence to obtain the corresponding straight line signals in the signal distribution diagram, and then the tracks of all signals are obtained, which specifically includes: Step S31: Perform Hough transform on all signals in each set to obtain a curve of each signal in the Hough space; Step S32: Find the maximum value in the Hough space; Step S33: determining a straight line corresponding to the maximum value in the signal distribution diagram, and extracting a track falling on the straight line; Step S34: remove the sinusoidal curve corresponding to the extracted trajectory in the Hough space, and repeat steps S32 to S33 until the trajectory of all signals is extracted.

2. The method for finding the position of the fusion neutron incident point according to claim 1, characterized in that: In step S1 , a capillary array detector is simulated to obtain a distribution diagram of all signals.

3. The method for finding the position of the fusion neutron incident point according to claim 1, characterized in that: In the signal distribution diagram, the measured signals are classified and divided according to whether their positions are adjacent to each other, thereby obtaining several signal sets.

4. The method for finding the position of the fusion neutron incident point according to claim 3, characterized in that: In the same signal set, the difference between the pixel rows and columns of two adjacent signals is no more than 2.

5. The method for finding the position of the fusion neutron incident point according to claim 1, characterized in that: In step S4, in the process of searching for the position of the neutron incident point, the end of the track with a weak signal is determined to be the position of the neutron incident point based on the energy deposition characteristics of the generated neutrons.

Images