CN113835111B - Unmanned aerial vehicle-based radioactive source positioning method and device - Google Patents
Unmanned aerial vehicle-based radioactive source positioning method and device Download PDFInfo
- Publication number
- CN113835111B CN113835111B CN202111418212.8A CN202111418212A CN113835111B CN 113835111 B CN113835111 B CN 113835111B CN 202111418212 A CN202111418212 A CN 202111418212A CN 113835111 B CN113835111 B CN 113835111B
- Authority
- CN
- China
- Prior art keywords
- positions
- detection
- grid
- neural network
- detection area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
- G01T1/00—Measuring X-radiation, gamma radiation, corpuscular radiation, or cosmic radiation
- G01T1/16—Measuring radiation intensity
- G01T1/18—Measuring radiation intensity with counting-tube arrangements, e.g. with Geiger counters
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64C—AEROPLANES; HELICOPTERS
- B64C39/00—Aircraft not otherwise provided for
- B64C39/02—Aircraft not otherwise provided for characterised by special use
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Aviation & Aerospace Engineering (AREA)
- High Energy & Nuclear Physics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Measurement Of Radiation (AREA)
Abstract
The invention relates to the technical field of radiation detection, and provides a radioactive source positioning method and a device based on an unmanned aerial vehicle, wherein the unmanned aerial vehicle carries a radiation detector, and the method comprises the following steps: carrying out grid division on the detection area, and setting a number for each grid; determining the detection efficiency of the radiation detector at a plurality of second positions of the detection area respectively when the radiation source is positioned at each first position of the detection area through simulation calculation; training a neural network; measuring, by the radiation detector, the count rates at a plurality of second locations, respectively; processing the counting rates of the plurality of measuring points and inputting the processed counting rates into the trained neural network so that the trained neural network outputs grid numbers; and determining the target position of the radioactive source to be detected in the detection area according to the grid number. Therefore, the method can reduce the measured data volume and reduce the calculation cost while ensuring the positioning accuracy, thereby accelerating the positioning speed of the radioactive source and having the advantage of simple operation.
Description
Technical Field
The invention relates to the technical field of radiation detection, in particular to a radioactive source positioning method based on an unmanned aerial vehicle and a radioactive source positioning device based on the unmanned aerial vehicle.
Background
In the radiometric field, under the circumstances that the radiation source is lost, adopt unmanned aerial vehicle automatic source seeking technique usually, unmanned aerial vehicle surveys and seeks the radiation source in the air promptly, can remove the radiation hazard that the artificial source seeking probably caused the human body from to unmanned aerial vehicle is free activity in the air, and detection range is wide, receives the topography restriction less.
However, when positioning the radioactive source by adopting the unmanned aerial vehicle source searching technology, the phenomenon of huge measurement data amount often exists, thereby causing the problems of low positioning efficiency, high calculation cost and complex operation.
Disclosure of Invention
The invention provides the following technical scheme for solving the problems of low positioning efficiency, high calculation cost and complex operation caused by huge measurement data quantity.
The embodiment of the first aspect of the invention provides a radioactive source positioning method based on an unmanned aerial vehicle, wherein the unmanned aerial vehicle carries a radiation detector, and the method comprises the following steps: determining a detection area corresponding to a radioactive source to be detected, carrying out grid division on the detection area, and setting grid numbers for each grid; determining a plurality of first locations in the detection region; determining the grid where each first position is located so as to determine the grid number corresponding to each first position; determining the detection efficiency of the radiation detector at a fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each first position of the detection area through simulation calculation so as to obtain a group of detection efficiencies corresponding to each first position; training a neural network based on the grid numbers corresponding to the first positions and the multiple groups of detection efficiency; measuring, by the radiation detector at the fixed height, count rates at the plurality of second locations of the detection zone, respectively, to obtain a count rate at each second location; processing the counting rates of the plurality of measuring points and inputting the processed counting rates into a trained neural network so that the trained neural network outputs grid numbers corresponding to the positions of the radioactive sources to be measured; and determining the target position of the radioactive source to be detected in the detection area according to the grid number corresponding to the position of the radioactive source to be detected.
In addition, the unmanned aerial vehicle-based radioactive source positioning method according to the above embodiment of the present invention may further have the following additional technical features.
According to one embodiment of the invention, the plurality of second locations comprises: the vertex and the center point of the detection region.
According to an embodiment of the present invention, the set of detection efficiencies corresponding to each of the first locations includes: the detection efficiency E1 corresponding to the first vertex, the detection efficiency E2 corresponding to the second vertex, the detection efficiency E3 corresponding to the third vertex, the detection efficiency E4 corresponding to the fourth vertex and the detection efficiency E5 corresponding to the central point; the count rates at the plurality of second locations include: the counting rate C1 corresponding to the first vertex, the counting rate C2 corresponding to the second vertex, the counting rate C3 corresponding to the third vertex, the counting rate C4 corresponding to the fourth vertex and the counting rate C5 corresponding to the central point.
According to an embodiment of the present invention, training a neural network based on the grid numbers and the detection efficiency corresponding to each of the first positions includes: according to the detection efficiency corresponding to each first position, sequentially determining four first ratios corresponding to the first positions, wherein the four first ratios comprise: E1/E5, E2/E5, E3/E5 and E4/E5; and taking the four first ratios as input, taking the grid number at the first position corresponding to the input as target output, and training the neural network by adopting a gradient descent mode.
According to an embodiment of the present invention, the processing the plurality of count rates and inputting the processed count rates into the trained neural network, so that the trained neural network outputs the grid number corresponding to the position of the radiation source to be measured, includes: sequentially determining four second ratios at each second position according to the counting rates at the plurality of second positions, wherein the four second ratios comprise: C1/C5, C2/C5, C3/C5 and C4/C5; and inputting the four second ratios into the trained neural network so that the trained neural network outputs the grid number corresponding to the position of the radiation source to be detected.
According to an embodiment of the present invention, the unmanned aerial vehicle-based radioactive source positioning method further includes: when measuring during the count rate, control unmanned aerial vehicle is in respectively a plurality of second positions stop for preset time, so that be located fixed height the radiation detector measures the count rate of a plurality of second positions.
According to an embodiment of the present invention, after determining the target position of the radiation source to be measured, the method further includes: and determining the activity of the radioactive source to be detected.
According to one embodiment of the invention, the detection zone is square in shape.
According to one embodiment of the invention, the radiation detector is a G-M (Geiger-Muller) count tube.
An embodiment of a second aspect of the present invention provides a radioactive source positioning device based on an unmanned aerial vehicle, where the unmanned aerial vehicle carries a radiation detector, and the device includes: the dividing module is used for determining a detection area corresponding to the radioactive source to be detected, carrying out grid division on the detection area and setting grid numbers for each grid; a first determination module for determining a plurality of first locations in the detection area; a second determining module, configured to determine a grid where each first location is located, so as to determine a grid number corresponding to each first location; the simulation module is used for determining the detection efficiency of the radiation detector at a fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each first position of the detection area through simulation calculation so as to obtain a group of detection efficiencies corresponding to each first position; the training module is used for training the neural network based on the grid numbers corresponding to the first positions and the multiple groups of detection efficiency; a measuring module for measuring a count rate at the plurality of second positions of the detection area by the radiation detector at the fixed height, respectively, to obtain a count rate at each of the second positions; the processing module is used for processing the plurality of counting rates and inputting the processed counting rates into a trained neural network so that the trained neural network outputs grid numbers corresponding to the positions of the radioactive sources to be detected; and the third determining module is used for determining the target position of the radioactive source to be detected in the detection area according to the grid number corresponding to the position of the radioactive source to be detected.
According to the technical scheme of the embodiment of the invention, the training sample of the neural network is obtained by simulating that the radioactive source is positioned in the detection area, the trained neural network is obtained based on the training of the training sample, the counting rate of the detection area is measured, and the counting rate is input into the trained neural network to obtain the position of the radioactive source, so that the radioactive source is positioned. Therefore, the method can reduce the measured data volume while ensuring the positioning accuracy, thereby reducing the calculation cost, accelerating the positioning speed of the radioactive source and having the advantage of simple operation.
Drawings
Fig. 1 is a flowchart of a method for positioning a radioactive source based on an unmanned aerial vehicle according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of meshing a detection region according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of the positions of corresponding measurement points when five count rates are measured in the detection region according to an embodiment of the present invention.
Fig. 4 is a block diagram of a positioning device of a radioactive source based on a drone according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of a method for positioning a radioactive source based on an unmanned aerial vehicle according to an embodiment of the present invention.
It should be noted that the main executing body of the unmanned aerial vehicle-based radioactive source positioning method in the embodiment of the present invention may be an electronic device, and the electronic device may send a control instruction to the unmanned aerial vehicle in the embodiment of the present invention, and the unmanned aerial vehicle and the electronic device may perform data transmission with each other. Specifically, the electronic device may be, but is not limited to, a computer and a mobile terminal, the unmanned aerial vehicle is equipped with a radiation detector, and the radiation detector may be a G-M (Geiger-Muller) counter tube.
As shown in fig. 1, the unmanned aerial vehicle-based radioactive source positioning method includes the following steps S1 to S8.
And S1, determining a detection area corresponding to the radioactive source to be detected, performing grid division on the detection area, and setting a grid number for each grid.
In the embodiment of the present invention, a radiation source to be searched, for example, a radiation source that has been lost, may be referred to as a radiation source to be detected, and a corresponding search area may be referred to as a detection area.
Specifically, when the radioactive source needs to be detected, a detection area corresponding to the radioactive source is determined, square grid division is performed on the detection area, and a unique number is set for each grid. Wherein the shape of the detection area may be square, for example square.
As shown in fig. 2, when the detection region has a square shape, n can be obtained by dividing the detection region in the horizontal direction and the vertical direction2A number of grids (n rows and n columns), each of which is numbered, for example n =4, the 4 grids of the first row may be numbered in turn: 1. 2, 3 and 4, the 4 grids of the second row may be numbered sequentially as: 5. 6, 7 and 8, the 4 grids of the third row may be numbered sequentially as: 9. 10, 11 and 12, the 4 grids of the fourth row may be numbered sequentially as: 13. 14, 15 and 16, thus dividing the detection area into 16 grids with unique numbers.
S2, a plurality of first locations in the detection region are determined.
In particular, a plurality of different first positions in the detection area may be determined arbitrarily.
And S3, determining the grid where each first position is located so as to determine the grid number corresponding to each first position.
Since the whole detection area is divided and numbered in a grid mode, each determined first position has the grid where the first position is located, and the grid number is the grid number corresponding to the first position.
And S4, determining the detection efficiency of the radiation detector at the fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each first position of the detection area through simulation calculation so as to obtain a group of detection efficiencies corresponding to each first position.
It should be noted that step S4 is to assume that the radiation source is located at the first position a of the detection area, and determine the detection efficiency of the radiation detector at the plurality of second positions B by calculation when the radiation source is assumed to be located at the first position a of the detection area, so this step may be referred to as a simulation step. Wherein position a and position B may be different.
Specifically, after determining the plurality of first positions in the detection area, it may be assumed that there are radioactive sources at the plurality of first positions a of the detection area, respectively, and when determining that a radioactive source is located at each a, detection efficiencies of the radiation detector at a fixed height at a plurality of second positions B of the detection area, respectively, are obtained, so as to obtain a plurality of sets of detection efficiencies, where each set of detection efficiencies corresponds to one first position a, and each set of detection efficiencies includes detection efficiencies at a plurality of second positions B. The height may refer to a distance from the ground of the radiation detector located above the detection area, and the distances from the ground of the radiation detector located at the second positions B are the same, that is, the radiation detector is located at a fixed height above the detection area.
And S5, training the neural network based on the grid numbers and the detection efficiency corresponding to the first positions.
The neural network may be a Back Propagation (BP) neural network.
Specifically, after obtaining a plurality of groups of detection efficiencies through the simulation step, a neural network may be constructed, and the neural network may be subjected to learning training based on the plurality of groups of detection efficiencies to obtain a trained neural network. That is, in one example, a training sample of the BP neural network is obtained through the simulation step, and then the trained BP neural network is obtained through training based on the training sample, and the trained BP neural network may output a corresponding grid number according to its input.
S6, measuring the count rate at a plurality of second positions of the detection area respectively by said radiation detector at a fixed height to obtain the count rate at each second position.
It should be noted that, in step S6, the radiation signals at a plurality of second positions are measured in the upper space of the detection region to obtain the count rate of the radiation signals at each second position, and this step may be referred to as an actual measurement step.
Specifically, after obtaining the neural network after training, can be fixed in the overhead fixed height of detection zone with unmanned aerial vehicle to make unmanned aerial vehicle measure the count rate in a plurality of second position B departments respectively, when measuring at a plurality of second positions, unmanned aerial vehicle is fixed height apart from the distance on ground, thereby obtains the count rate that every measuring point also is second position department.
And S7, processing the plurality of counting rates and inputting the processed counting rates into the trained neural network so that the trained neural network outputs the grid number corresponding to the position of the radiation source to be detected.
Specifically, after a plurality of count rates (a plurality of count rates correspond to a plurality of second positions one to one) are obtained through measurement, the plurality of count rates can be processed, the processed count rates are input into the trained neural network, then the trained neural network outputs a grid number according to the input count rates, and the grid number is a number corresponding to the position of the radiation source to be measured.
And S8, determining the target position of the radioactive source to be detected in the detection area according to the grid number corresponding to the position of the radioactive source to be detected.
Specifically, after the grid numbers are obtained, the positions of the grids in the detection area can be obtained according to the number rules or grid coordinates, and the positions are the target positions of the radioactive source to be detected in the detection area, so that the radioactive source can be positioned by unmanned aerial vehicle detection.
Based on the above description, it can be seen that the embodiment of the present invention converts the problem of source location reconstruction into the problem of finding a source by using a count rate, and the grid number output by the trained neural network represents the source location, so that the problem is simplified, a neural network with a simpler structure can be used, the training speed is increased, and the difficulty in establishing a data set is reduced.
The embodiment of the invention aims at positioning the position of the outdoor radioactive source by carrying the radiation detector by the unmanned aerial vehicle under the condition that the outdoor radioactive source is lost, realizes the positioning of the lost radioactive source, considers the timeliness and the simplicity while realizing the positioning, and can be used for aiming at the scene of an unknown radioactive source (the position is unknown and the activity is unknown), and under the scene, the position of the radioactive source can be quickly obtained by measuring the fixed working condition point of the radiation detector and predicting the trained neural network. Therefore, the measurement data volume is reduced, the calculation cost is greatly reduced, the speed of positioning the source position is accelerated, and the method is suitable for quickly and accurately searching the radioactive source in the field.
According to the unmanned aerial vehicle-based radioactive source positioning method, training samples of the neural network are obtained by simulating that the radioactive source is located in the detection area, the neural network is trained based on training of the training samples, the counting rate of the detection area is measured, and the counting rate is input into the trained neural network to obtain the position of the radioactive source, so that the radioactive source is positioned. Therefore, the method can reduce the measured data volume while ensuring the positioning accuracy, thereby reducing the calculation cost, accelerating the positioning speed of the radioactive source and having the advantage of simple operation.
It should be noted that the plurality of second positions in the embodiment of the present invention may include the vertex and the center point of the detection region, wherein the number of vertices may be determined by the shape of the detection region, for example, when the shape of the detection region is a triangle, the number of vertices is 3; when the shape of the detection region is a square, the number of vertices is 4, the shape of the detection region is not limited to a triangle and a square, and when the shape is a square, the set of detection efficiencies corresponding to each first position may include detection efficiencies of four vertices and a detection efficiency of a center point.
That is, in an embodiment of the present invention, the plurality of second positions in the steps S4 and S6 may include: the vertices and the center point of the detection region.
Accordingly, when the detection area is square, the set of detection efficiencies corresponding to each first position may include: the detection efficiency E1 corresponding to the first vertex, the detection efficiency E2 corresponding to the second vertex, the detection efficiency E3 corresponding to the third vertex, the detection efficiency E4 corresponding to the fourth vertex, and the detection efficiency E5 corresponding to the center point.
Specifically, after a plurality of first positions are determined, when the radiation source is located at each first position of the detection area, the detection efficiency of the radiation detector at a fixed height at the first vertex, the second vertex, the third vertex and the fourth vertex of the detection area is determined through simulation calculation, that is: detecting efficiency E1 corresponding to a first vertex, detecting efficiency E2 corresponding to a second vertex, detecting efficiency E3 corresponding to a third vertex and detecting efficiency E4 corresponding to a fourth vertex, and determining detecting efficiency E5 of a central point of a radiation detector at a fixed height in a detecting area to obtain multiple groups of detecting efficiencies, wherein each group of detecting efficiencies comprises E1, E2, E3, E4 and E5, and each first position corresponds to one group of detecting efficiencies.
In one example, the step S5 is: training the neural network based on the grid numbers and the detection efficiency corresponding to the first positions may include: according to the detection efficiency corresponding to each first position, sequentially determining four first ratios corresponding to the first positions, wherein the four first ratios comprise: E1/E5, E2/E5, E3/E5 and E4/E5; and taking the four first ratios as input, taking the grid numbers at the first positions corresponding to the input as target output, and training the neural network in a gradient descending manner.
Wherein, E1/E5 refers to a ratio between detection efficiency E1 corresponding to the first vertex and detection efficiency E5 of the detection area center point, E2/E5 refers to a ratio between detection efficiency E2 corresponding to the second vertex and detection efficiency E5 of the detection area center point, E3/E5 refers to a ratio between detection efficiency E3 corresponding to the third vertex and detection efficiency E5 of the detection area center point, and E4/E5 refers to a ratio between detection efficiency E4 corresponding to the fourth vertex and detection efficiency E5 of the detection area center point.
Specifically, after obtaining a plurality of groups of detection efficiencies, for each group of detection efficiencies, four first ratios E1/E5, E2/E5, E3/E5, and E4/E5 are calculated, and the four first ratios are input to the trained neural network, and the grid numbers at the first positions corresponding to the four first ratios (i.e., the grid numbers where the simulated radiation source is located) are output as targets, and the neural network is trained in a gradient descent manner to obtain the trained neural network, so that the trained neural network can output the grid numbers according to the input of the grid numbers, and it can be understood that the grid numbers can represent the positions where the grids are located in the detection area.
The previous work before positioning the reflection source is completed through the above steps S1 to S5, and then the steps S6, S7 and S8 are executed, that is, the positioning of the radiation source is realized through the unmanned aerial vehicle and the trained neural network.
In one embodiment of the present invention, as shown in FIG. 3, the count rates at the plurality of second locations may include: the counting rate C1 corresponding to the first vertex, the counting rate C2 corresponding to the second vertex, the counting rate C3 corresponding to the third vertex, the counting rate C4 corresponding to the fourth vertex and the counting rate C5 corresponding to the central point.
In this embodiment, in step S7, the following steps are performed: the neural network after processing a plurality of count rates and input training to the neural network after making training output the mesh number that the radiation source that awaits measuring belongs to the position correspondence, can include: sequentially determining four second ratios corresponding to each second position according to the counting rates at the plurality of second positions, wherein the four second ratios comprise C1/C5, C2/C5, C3/C5 and C4/C5; and inputting the four second ratios into the trained neural network so that the trained neural network outputs the grid number corresponding to the position of the radiation source to be detected.
Wherein, C1/C5 is a ratio between a count rate C1 corresponding to the first vertex and a count rate C5 corresponding to the central point, C2/C5 is a ratio between a count rate C2 corresponding to the second vertex and a count rate C5 corresponding to the central point, C3/C5 is a ratio between a count rate C3 corresponding to the third vertex and a count rate C5 corresponding to the central point, and C4/C5 is a ratio between a count rate C4 corresponding to the fourth vertex and a count rate C5 corresponding to the central point.
Specifically, as shown in fig. 3, the unmanned aerial vehicle may be fixed at a fixed height above the detection area, and count rates of the unmanned aerial vehicle may be measured at a first vertex, a second vertex, a third vertex, a fourth vertex, and a center point of the detection area, respectively, so as to obtain a count rate C1 corresponding to the first vertex, a count rate C2 corresponding to the second vertex, a count rate C3 corresponding to the third vertex, a count rate C4 corresponding to the first vertex, and a count rate C5 corresponding to the center point.
After the counting rates C1, C2, C3, C4 and C5 are obtained, four second ratios C1/C5, C2/C5, C3/C5 and C4/C5 are calculated, when energy is fixed for a fixed nuclide, the counting rate ratios are equal to the detection efficiency ratio, the four second ratios are input into the trained neural network, the trained neural network outputs grid numbers corresponding to the four second ratios, and the grid numbers corresponding to the four second ratios are the grid numbers of the position of the radioactive source to be detected.
That is to say, the embodiment of the invention positions the radioactive source by using the neural network technology, and can realize the positioning of the common lost radioactive sources such as Co-60 and Cs-137: firstly, controlling an unmanned aerial vehicle to detect counting rates at five second position points in a detection area, processing the five counting rates to be used as input of a trained neural network, and obtaining output which is a target position of a radioactive source to be detected, wherein the neural network is trained by using a data set obtained by simulation in advance.
In an embodiment of the present invention, the unmanned aerial vehicle-based radioactive source positioning method may further include: when the counting rate is measured, the unmanned aerial vehicle is controlled to stay at the second positions for a preset time respectively, so that the radiation detector at the fixed height measures the counting rate of the second positions.
Particularly, when measuring a plurality of second positions of the detection area, for example, the counting rates of four vertexes and a central point, the unmanned aerial vehicle can be controlled to stay at each position for a preset time, wherein the preset time can be determined according to actual conditions until an accurate and stable counting rate is detected, and therefore, the stability and the accuracy of the counting rate of the radiation detector can be ensured.
In an embodiment of the present invention, the unmanned aerial vehicle-based radioactive source positioning method may further include: after the target position of the radioactive source to be detected is determined, the activity of the radioactive source to be detected is determined.
That is, the embodiment of the present invention can position the radiation source under the condition that the position of the radiation source is unknown and the activity is unknown, and can quickly obtain the target position where the radiation source is located through the measurement of the fixed operating point of the radiation detector and the prediction of the trained neural network, and then can determine the activity of the radiation source to be detected, thereby implementing the activity detection of the radiation source.
It should be noted that, in the embodiment of the present invention, after the grid division, the unmanned aerial vehicle may stay at each grid point to obtain measurement data, and the grid where the radiation source is located is determined through comparative analysis of the data.
In conclusion, the embodiment of the invention is suitable for quickly and accurately searching the radioactive source in the field, and the position of the source is positioned by utilizing the neural network technology, so that the measured data volume is reduced, the calculation cost is greatly reduced, and the speed of positioning the source position is accelerated. The embodiment of the invention has great expansibility, and the position and activity detection of a plurality of sources can be carried out in a larger area only by correspondingly adjusting the structures of the data set and the network and the number of the test points.
Corresponding to the unmanned aerial vehicle radioactive source positioning method of the embodiment, the invention further provides a radioactive source positioning device based on the unmanned aerial vehicle.
Fig. 4 is a block diagram of a positioning device of a radioactive source based on a drone according to an embodiment of the present invention.
The unmanned aerial vehicle in this embodiment carries a radiation detector, and the radiation detector is a G-M count tube.
As shown in fig. 4, the drone-based radiation source positioning device 100 includes: a partitioning module 10, a first determining module 20, a second determining module 30, a simulation module 40, a training module 50, a measurement module 60, a processing module 70, and a third determining module 80.
The dividing module 10 is configured to determine a detection area corresponding to a radiation source to be detected, perform mesh division on the detection area, and set a mesh number for each mesh; a first determination module 20 for determining a plurality of first locations in the detection area; the second determining module 30 is configured to determine a grid where each first location is located, so as to determine a grid number corresponding to each first location; the simulation module 40 is configured to determine, through simulation calculation, detection efficiencies of the radiation detector at a fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each of the first positions of the detection area, so as to obtain a set of detection efficiencies corresponding to each of the first positions; the training module 50 is configured to train the neural network based on the grid numbers and the detection efficiency corresponding to each of the first positions; a measuring module 60 for measuring, by said radiation detector at said fixed height, the count rate at said plurality of second positions of said detection area, respectively, to obtain the count rate at each of said second positions; the processing module 70 is configured to process the plurality of count rates and input the processed count rates into a trained neural network, so that the trained neural network outputs a grid number corresponding to a position where the radiation source to be detected is located; the third determining module 80 is configured to determine a target position of the radiation source to be detected in the detection area according to the grid number corresponding to the position of the radiation source to be detected.
In one embodiment of the invention, the plurality of second locations comprises: the vertex and the center point of the detection region.
In one embodiment of the present invention, the set of detection efficiencies corresponding to each of the first locations comprises: the detection efficiency E1 corresponding to the first vertex, the detection efficiency E2 corresponding to the second vertex, the detection efficiency E3 corresponding to the third vertex, the detection efficiency E4 corresponding to the fourth vertex and the detection efficiency E5 corresponding to the central point; the count rates at the plurality of second locations include: the counting rate C1 corresponding to the first vertex, the counting rate C2 corresponding to the second vertex, the counting rate C3 corresponding to the third vertex, the counting rate C4 corresponding to the fourth vertex and the counting rate C5 corresponding to the central point.
In one embodiment of the present invention, training module 50 may include: a first determining unit, configured to sequentially determine four first ratios corresponding to each first position according to detection efficiency corresponding to the first position, where the four first ratios include: E1/E5, E2/E5, E3/E5 and E4/E5; and the training unit is used for taking the four first ratios as input, taking the grid numbers at the first positions corresponding to the input as target output, and training the neural network in a gradient descending mode.
In one embodiment of the present invention, the processing module 70 may include: a second determining unit, configured to sequentially determine four second ratios corresponding to each of the second positions according to the count rates at the plurality of second positions, where the four second ratios include: C1/C5, C2/C5, C3/C5 and C4/C5; and the input unit is used for inputting the four second ratios into the trained neural network so as to enable the trained neural network to output the grid number corresponding to the position of the radioactive source to be detected.
In an embodiment of the present invention, the drone-based radiation source positioning device 100 may further include: and the control module is used for measuring the counting rate, controlling the unmanned aerial vehicle to stay at the second positions for preset time respectively so as to enable the unmanned aerial vehicle to be positioned at the fixed height, and measuring the counting rate of the second positions by the radiation detector.
In an embodiment of the present invention, the drone-based radiation source positioning device 100 may further include: and the activity determination module is used for determining the activity of the radioactive source to be detected after the target position of the radioactive source to be detected is determined.
In one embodiment of the invention, the detection zone is square in shape.
It should be noted that, for the specific implementation and implementation principle of the radiation source positioning device based on the unmanned aerial vehicle, reference may be made to the specific implementation of the radiation source positioning method based on the unmanned aerial vehicle, and in order to avoid redundancy, detailed description is omitted here.
According to the radioactive source positioning device based on the unmanned aerial vehicle, the training sample of the neural network is obtained by simulating that the radioactive source is located in the detection area, the neural network is trained based on the training sample, the counting rate of the detection area is measured, and the counting rate is input into the trained neural network to obtain the position of the radioactive source, so that the radioactive source is positioned. Therefore, the method can reduce the measured data volume while ensuring the positioning accuracy, thereby reducing the calculation cost, accelerating the positioning speed of the radioactive source and having the advantage of simple operation.
In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The meaning of "plurality" is two or more unless specifically limited otherwise.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments. In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (9)
1. A method for positioning a radioactive source based on an unmanned aerial vehicle, wherein the unmanned aerial vehicle carries a radiation detector, the method comprising:
determining a detection area corresponding to a radioactive source to be detected, carrying out grid division on the detection area, and setting grid numbers for each grid;
determining a plurality of first locations in the detection region;
determining the grid where each first position is located so as to determine the grid number corresponding to each first position;
determining, by simulation calculation, detection efficiencies of the radiation detector at a fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each of the first positions of the detection area, so as to obtain a set of detection efficiencies corresponding to each of the first positions, where the plurality of second positions include: the vertex and the center point of the detection area;
training a neural network based on the grid numbers and the detection efficiency corresponding to the first positions;
measuring, by the radiation detector at the fixed height, count rates at the plurality of second locations of the detection zone, respectively, to obtain a count rate at each of the second locations;
processing the plurality of counting rates and inputting the processed counting rates into a trained neural network so that the trained neural network outputs grid numbers corresponding to the positions of the radioactive sources to be detected;
and determining the target position of the radioactive source to be detected in the detection area according to the grid number corresponding to the position of the radioactive source to be detected.
2. The drone-based radiation source positioning method of claim 1, wherein the set of detection efficiencies for each of the first locations includes: the detection efficiency E1 corresponding to the first vertex, the detection efficiency E2 corresponding to the second vertex, the detection efficiency E3 corresponding to the third vertex, the detection efficiency E4 corresponding to the fourth vertex and the detection efficiency E5 corresponding to the central point;
the count rates at the plurality of second locations include: the counting rate C1 corresponding to the first vertex, the counting rate C2 corresponding to the second vertex, the counting rate C3 corresponding to the third vertex, the counting rate C4 corresponding to the fourth vertex and the counting rate C5 corresponding to the central point.
3. The unmanned aerial vehicle-based radioactive source positioning method of claim 2, wherein training the neural network based on the mesh number and the detection efficiency corresponding to each of the first positions comprises:
according to the detection efficiency corresponding to each first position, sequentially determining four first ratios corresponding to the first positions, wherein the four first ratios comprise: E1/E5, E2/E5, E3/E5 and E4/E5;
and taking the four first ratios as input, taking the grid number at the first position corresponding to the input as target output, and training the neural network in a gradient descending manner.
4. The unmanned aerial vehicle-based radioactive source positioning method of claim 3, wherein the counting rates are processed and then input to a trained neural network, so that the trained neural network outputs a grid number corresponding to a position of the radioactive source to be detected, and the method comprises the following steps:
according to the counting rates of the plurality of second positions, sequentially determining four second ratios corresponding to each second position, wherein the four second ratios comprise: C1/C5, C2/C5, C3/C5 and C4/C5;
and inputting the four second ratios into the trained neural network so that the trained neural network outputs the grid number corresponding to the position of the radiation source to be detected.
5. The drone-based radiation source positioning method of claim 1, further comprising:
when measuring during the count rate, control unmanned aerial vehicle is in respectively a plurality of second positions stop for preset time, so that be located fixed height the radiation detector measures the count rate of a plurality of second positions.
6. The unmanned aerial vehicle-based radioactive source positioning method according to claim 1, further comprising, after determining the target position of the radioactive source to be measured:
and determining the activity of the radioactive source to be detected.
7. The drone-based radiation source positioning method of claim 2, wherein the detection zone is square in shape.
8. The drone-based radiation source positioning method of claim 1, wherein the radiation detector is a G-M counting tube.
9. The utility model provides a radiation source positioner based on unmanned aerial vehicle, its characterized in that, unmanned aerial vehicle carries on radiation detector, the device includes:
the dividing module is used for determining a detection area corresponding to the radioactive source to be detected, carrying out grid division on the detection area and setting grid numbers for each grid;
a first determination module for determining a plurality of first locations in the detection area;
a second determining module, configured to determine a grid where each first location is located, so as to determine a grid number corresponding to each first location;
a simulation module, configured to determine, through simulation calculation, detection efficiencies of the radiation detector at a fixed height at a plurality of second positions of the detection area respectively when the radiation source is located at each of the first positions of the detection area, so as to obtain a set of detection efficiencies corresponding to each of the first positions, where the plurality of second positions include: the vertex and the center point of the detection area;
the training module is used for training the neural network based on the grid numbers and the detection efficiency corresponding to the first positions;
a measuring module for measuring a count rate at the plurality of second positions of the detection area by the radiation detector at the fixed height, respectively, to obtain a count rate at each of the second positions;
the processing module is used for processing the plurality of counting rates and inputting the processed counting rates into the trained neural network so that the trained neural network outputs grid numbers corresponding to the positions of the radioactive sources to be detected;
and the third determining module is used for determining the target position of the radioactive source to be detected in the detection area according to the grid number corresponding to the position of the radioactive source to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111418212.8A CN113835111B (en) | 2021-11-26 | 2021-11-26 | Unmanned aerial vehicle-based radioactive source positioning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111418212.8A CN113835111B (en) | 2021-11-26 | 2021-11-26 | Unmanned aerial vehicle-based radioactive source positioning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113835111A CN113835111A (en) | 2021-12-24 |
| CN113835111B true CN113835111B (en) | 2022-02-11 |
Family
ID=78971636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111418212.8A Active CN113835111B (en) | 2021-11-26 | 2021-11-26 | Unmanned aerial vehicle-based radioactive source positioning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113835111B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114994778B (en) * | 2022-05-12 | 2025-03-21 | 中国原子能科学研究院 | Nuclear radiation source tracing method, tracing device, tracing equipment and storage medium |
| CN115657102B (en) * | 2022-12-08 | 2025-05-06 | 重庆建安仪器有限责任公司 | A radiation source positioning method, a radiation source positioning device and a control method |
| CN117991320B (en) * | 2024-01-10 | 2024-11-08 | 北京方鸿智能科技有限公司 | Method, device, equipment and storage medium for positioning radioactive source |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8401621B2 (en) * | 2006-05-11 | 2013-03-19 | Forimtech S.A. | Method and device of detecting, locating and/or analyzing a radioactive source(s) in a material, e.g. a biological tissue |
| US8304737B2 (en) * | 2009-10-02 | 2012-11-06 | Ut-Battelle, Llc | Apparatus and method to achieve high-resolution microscopy with non-diffracting or refracting radiation |
| CN107884812B (en) * | 2017-11-09 | 2020-09-04 | 中国工程物理研究院材料研究所 | Radioactive source positioning and tracking method based on scintillator detector |
| CN110780333A (en) * | 2019-09-30 | 2020-02-11 | 南京航空航天大学 | A dual-coded plate-based gamma camera and method for radioactive material localization |
| CN112258597B (en) * | 2020-12-18 | 2021-04-20 | 成都理工大学 | Fast imaging method and device based on neural network positioning algorithm |
| CN113536679B (en) * | 2021-07-20 | 2024-09-24 | 南京航空航天大学 | Point source dose rate correction method based on artificial neural network |
-
2021
- 2021-11-26 CN CN202111418212.8A patent/CN113835111B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113835111A (en) | 2021-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113835111B (en) | Unmanned aerial vehicle-based radioactive source positioning method and device | |
| KR102116824B1 (en) | Positioning system based on deep learnin and construction method thereof | |
| CN115481558B (en) | A method, system and device for inverting atmospheric pollution emissions based on machine learning | |
| CN104811968B (en) | A kind of bearing calibration of propagation model and device | |
| EP3769091A1 (en) | Wind flow sensing system and method for determining velocity fields of wind flow | |
| CN108716918A (en) | A kind of RSSI indoor positioning algorithms based on Grid Clustering | |
| CN108681487B (en) | Distributed system and method for adjusting and optimizing sensor algorithm parameters | |
| CN109191408B (en) | Rapid circulation ground weather fusion method and device and server | |
| CN103017655B (en) | Method and system for extracting floor area of multi-floor building | |
| CN108242962A (en) | A Calculation Method and Device for Indoor Signal Propagation Loss Based on Measurement Report | |
| CN113822892A (en) | Evaluation method, device and equipment of simulated radar and computer program product | |
| CN113125635A (en) | Atmospheric pollution early warning method and device and readable storage medium | |
| CN109903367A (en) | Construct the method, apparatus and computer readable storage medium of map | |
| CN114040336A (en) | Beacon deployment method based on epitome decision positioning method and particle swarm optimization algorithm | |
| CN117991331B (en) | Method for tracking multiple rays in two-dimensional complex model based on seismic monitoring | |
| KR102714799B1 (en) | Apparatus and method of cyclonic circulation center detection based on radar data | |
| CN114663854A (en) | Lane line evaluation method and device and storage medium | |
| WO2019128018A1 (en) | Method for determining wind resource, and apparatus | |
| CN116796455B (en) | A method to characterize the occurrence of rock mass fissures | |
| CN117094167A (en) | Landslide whole-process surface three-dimensional deformation Beidou radar surface scene data simulation method | |
| CN110648280A (en) | Data processing method for splicing large-scale karst cave mass point cloud data | |
| CN117131998A (en) | Distributed low-altitude monitoring radar site selection area planning method based on Internet of things | |
| Pohl et al. | A combination of the sound particle simulation method and the radiosity method | |
| CN116341231A (en) | Method and device for simulating artificial precipitation catalyst diffusion track, server and storage medium | |
| CN110968656A (en) | Topographic map processing method and processing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |