CN114137004A - A material identification method, device and storage medium - Google Patents

Abstract

The invention discloses a material identification method, device and storage medium. The method includes: acquiring the approximate scattering point and scattering angle corresponding to each particle in a plurality of radiation particles passing through a detection area where a material to be identified is deployed; It is divided into multiple sub-regions, and for each sub-region in the multiple sub-regions, from the multiple radiation particles, the radiation particles whose corresponding approximate scattering points are located in the sub-regions are selected as the corresponding radiation particles; for each of the multiple sub-regions There are several sub-regions, based on the scattering angles of the corresponding radiation particles, the corresponding scattering densities are determined, and multiple scattering densities corresponding to the multiple sub-regions are obtained; the preset material identification model is used to determine the material type of the material to be identified based on the multiple scattering densities. Identify, and obtain the material type identification result of the material to be identified. Through the above technical solutions, the efficiency of material identification is improved.

Description

A material identification method, device and storage medium

technical field

The present application relates to the technical field of material detection, and in particular, to a material identification method, device and storage medium.

Background technique

In recent years, the illegal transfer of nuclear materials, nuclear proliferation and other issues have always threatened homeland security, and nuclear material detection technology has gradually attracted the attention of countries around the world.

Compared with X-ray, the traditional detection method for nuclear materials, muons, as a natural radiation source, have no radiation hazards, are sensitive to high-Z substances, have strong penetrating ability, and have natural advantages in the application of nuclear material detection technology. , Due to the limited flux of natural muons, in order to improve the image quality of muon imaging, a long detection time is usually required. poor.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the embodiments of the present invention are expected to provide a material identification method, device and storage medium, which directly use a preset material identification model and combine the scattering density corresponding to the detection area to determine the material type of the material to be identified, thereby improving the performance of the material. Efficiency of material identification.

The technical scheme of the present invention is realized as follows:

The present invention provides a material identification method, the method comprising:

Obtain the approximate scattering point and scattering angle corresponding to each particle among the multiple radiation particles passing through the detection area where the material to be identified is deployed;

The detection area is divided into a plurality of sub-areas, and for each sub-area of the plurality of sub-areas, from the plurality of radiation particles, the radiation particles whose corresponding approximate scattering points are located in the sub-areas are selected as corresponding radiation particles;

For each sub-area in the plurality of sub-areas, based on the scattering angle of the corresponding radiation particle, determine the corresponding scattering density, and obtain a plurality of scattering densities corresponding to the plurality of sub-areas one-to-one;

Using a preset material identification model, the material type identification of the to-be-identified material is performed based on the plurality of scattering densities, and a material type identification result of the to-be-identified material is obtained.

In the above method, obtaining the approximate scattering point and scattering angle corresponding to each particle among the plurality of radiation particles passing through the detection area where the material to be identified is deployed includes:

acquiring an incident track and an exit track corresponding to each of the plurality of radiation particles;

For each particle in the plurality of radiation particles, the intersection of the extension line of the corresponding incident track and the extension line of the output track is determined as the corresponding approximate scattering point;

For each particle in the plurality of radiation particles, the included angle between the extension line of the corresponding incident track and the extension line of the output track is determined as the corresponding scattering angle.

In the above method, for each sub-region of the plurality of sub-regions, a corresponding scattering density is determined based on the scattering angle of the corresponding radiation particle, and a plurality of scattering densities corresponding to the plurality of sub-regions one-to-one are obtained, including :

For each sub-region in the plurality of sub-regions, calculate the variance of the scattering angle of the corresponding radiation particle to obtain the corresponding variance of the scattering angle;

For each sub-region in the plurality of sub-regions, the height of the corresponding sub-region is determined as the corresponding particle passing length;

For each sub-region in the plurality of sub-regions, the corresponding scattering angle variance and the ratio of the particle passing length are determined as the corresponding scattering density.

In the above method, using a preset material identification model to identify the material type of the material to be identified based on the multiple scattering densities, to obtain a material type identification result of the material to be identified, including:

From the plurality of scattering densities, selecting a scattering density that characterizes anomalies;

The selected scattering density is input into the preset material identification model to obtain the material type identification result.

In the above method, before the material type identification is performed on the material to be identified based on the plurality of scattering densities by using a preset material identification model, and the material type identification result of the material to be identified is obtained, the method further includes:

Acquiring a scattering density sample, and using the material identification model to be trained to identify the material type of the sample to be identified based on the scattering density sample, to obtain a material type identification result of the sample to be identified;

Calculate the loss information between the material type identification result of the to-be-identified sample and the target material type identification result preset for the scattering density sample to obtain the loss information;

Based on the loss information, model parameters are adjusted for the material identification model to be trained to obtain the preset material identification model.

The present invention provides a material identification device, comprising:

an acquisition module, configured to acquire the approximate scattering point and scattering angle corresponding to each particle among the plurality of radiation particles passing through the detection area where the material to be identified is deployed;

A selection module, configured to divide the detection area into a plurality of sub-areas, and for each sub-area in the plurality of sub-areas, from the plurality of radiation particles, select radiation particles whose corresponding approximate scattering points are located in the sub-areas , which is determined as the corresponding radiation particle;

a determining module, configured to, for each sub-region in the plurality of sub-regions, determine a corresponding scattering density based on the scattering angle of the corresponding radiation particle, and obtain a plurality of scattering densities corresponding to the plurality of sub-regions one-to-one;

The identification module is configured to use a preset material identification model to identify the material type of the material to be identified based on the plurality of scattering densities, and obtain a material type identification result of the material to be identified.

In the above device, the acquiring module is specifically configured to acquire the incident track and the outgoing track corresponding to each particle in the plurality of radiation particles; for each particle in the plurality of radiation particles, the corresponding incident track The intersection of the extension line of the track and the extension line of the exit track is determined as the corresponding approximate scattering point; for each particle in the plurality of radiation particles, the extension line of the corresponding incident track and the extension line of the exit track are determined as corresponding approximate scattering points; The included angle of the line is determined as the corresponding scattering angle.

In the above device, the determining module is specifically configured to, for each sub-region in the multiple sub-regions, calculate the variance of the scattering angle of the corresponding radiation particle to obtain the corresponding scattering angle variance; for each of the multiple sub-regions For each sub-region, the height of the corresponding sub-region is determined as the corresponding particle passing length; for each sub-region in the multiple sub-regions, the ratio of the corresponding scattering angle variance to the particle passing length is determined as the corresponding scattering density.

In the above device, the identification module is specifically configured to select a scattering density characterizing abnormality from the plurality of scattering densities; input the selected scattering density into the preset material identification model to obtain the material type Identify the results.

In the above-mentioned device, a model training module is further included, which is used to obtain a scattering density sample, and use a material identification model to be trained to identify the material type of the to-be-identified sample based on the scattering density sample, and obtain the material type identification of the to-be-identified sample. Result; calculate the loss information between the material type identification result of the sample to be identified and the target material type identification result preset for the scattering density sample, and obtain the loss information; based on the loss information, analyze the material to be trained The identification model adjusts model parameters to obtain the preset material identification model.

The invention provides a material identification device, comprising: a processor, a memory and a communication bus;

the communication bus for realizing the communication connection between the processor and the memory;

The processor is configured to execute the material identification program stored in the memory, so as to realize the above-mentioned material identification method.

The present invention provides a computer-readable storage medium, where one or more programs are stored in the computer-readable storage medium, and the one or more programs can be executed by one or more processors to implement the above-mentioned material identification method .

The present invention provides a material identification method, device and storage medium. The method includes: acquiring the approximate scattering point and scattering angle corresponding to each particle in a plurality of radiation particles passing through a detection area where a material to be identified is deployed; It is divided into multiple sub-regions, and for each sub-region in the multiple sub-regions, from the multiple radiation particles, the radiation particles whose corresponding approximate scattering points are located in the sub-regions are selected as the corresponding radiation particles; There are sub-regions, and the corresponding scattering densities are determined based on the scattering angles of the corresponding radiation particles, and multiple scattering densities corresponding to the multiple sub-regions are obtained; the preset material identification model is used to determine the material type of the material to be identified based on the multiple scattering densities. Identify, and obtain the material type identification result of the material to be identified. The technical solution provided by the present invention directly utilizes a preset material identification model and combines the scattering density corresponding to the detection area to determine the material type of the material to be identified, thereby improving the efficiency of material identification.

Description of drawings

1 is a schematic flowchart of a material identification method according to an embodiment of the present invention;

FIG. 2 is a schematic flowchart of an exemplary simulated detection environment provided by an embodiment of the present invention;

FIG. 3 is an exemplary schematic diagram of determining an approximate scattering point and a scattering angle according to an embodiment of the present invention;

4 is a schematic structural diagram of an exemplary preset material identification model provided by an embodiment of the present invention;

5 is a schematic diagram of an exemplary model accuracy varying with the number of iterations provided by an embodiment of the present invention;

6 is a schematic diagram of an exemplary loss function changing with the number of iterations provided by an embodiment of the present invention;

FIG. 7 is a schematic structural diagram 1 of a material identification device according to an embodiment of the present invention;

FIG. 8 is a second schematic structural diagram of a material identification device according to an embodiment of the present invention.

Detailed ways

The technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the related application, but not to limit the application. In addition, it should be noted that, for the convenience of description, only the parts related to the relevant application are shown in the drawings.

The present invention provides a material identification method, which is applied to a material identification device. FIG. 1 is a schematic flowchart of a material identification method provided by an embodiment of the present invention. As shown in Figure 1, it mainly includes the following steps:

S101. Acquire an approximate scattering point and a scattering angle corresponding to each particle among a plurality of radiation particles passing through the detection area where the material to be identified is deployed.

In the embodiment of the present invention, the material identification device may acquire the approximate scattering point and scattering angle corresponding to each particle among the plurality of radiation particles passing through the detection area where the material to be identified is deployed.

It should be noted that, in the embodiment of the present invention, the radiation particles may be muons. As a natural radiation source, the radiation particles have no radiation hazards, are sensitive to high-Z substances, and have strong penetrating ability. There are natural advantages in application.

FIG. 2 is a schematic flowchart of an exemplary simulated detection environment provided by an embodiment of the present invention. As shown in Fig. 2, a plurality of radiation particles first enter the detection area through the 2-layer detection array, then pass through the detection area, and finally scatter out through the 2-layer detection array.

Specifically, in the embodiment of the present invention, the material identification device obtains the approximate scattering point and scattering angle corresponding to each particle in a plurality of radiation particles passing through the detection area where the material to be identified is deployed, including: obtaining a plurality of radiation particles For each particle in the multiple radiation particles, the intersection of the extension line of the corresponding incident track and the extension line of the exit track is determined as the corresponding approximate scattering point ; For each particle in the plurality of radiation particles, the included angle between the extension line of the corresponding incident track and the extension line of the exit track is determined as the corresponding scattering angle.

It should be noted that, in the embodiment of the present invention, the material identification device will acquire the incident track and the output track corresponding to each particle in the plurality of radiation particles, and then extend the incident track and the output track, and connect the two straight lines. The intersection point is determined as the approximate scattering point, and the angle between the two straight lines is determined as the scattering angle.

FIG. 3 is an exemplary schematic diagram of determining an approximate scattering point and a scattering angle according to an embodiment of the present invention. As shown in Figure 3, the detection array located on the upper two layers of the detection area is used to record the position information of the incident radiation particles. The material identification device uses the upper two layers, namely 1# and 2#, to detect the two positions of the radiation particles recorded by the array. The information fits the incident track. The two layers of detection arrays located below the detection area are used to record the position information of the radiation particles. The material identification device uses the upper two layers, namely 3# and 4#, to detect the two positions of the radiation particles recorded by the array. The information is fitted to the outgoing track, then, the incoming and outgoing tracks are extended, the intersection γ of the two straight lines is determined as the approximate scattering point, and the angle θ between the two straight lines is determined as the scattering angle.

S102. Divide the detection area into a plurality of sub-areas, and for each sub-area in the plurality of sub-areas, select a radiation particle whose corresponding approximate scattering point is located in the sub-area from the plurality of radiation particles, and determine it as a corresponding radiation particle.

In the embodiment of the present invention, the material identification device divides the detection area into a plurality of sub-areas, and for each sub-area of the plurality of sub-areas, selects a radiation particle whose corresponding approximate scattering point is located in the sub-area from the plurality of radiation particles , determined as the corresponding radiation particles.

It should be noted that, in the embodiment of the present invention, the material identification device divides the detection area into multiple sub-areas. The specific division method is shown in Figure 3. The specific division size can be set according to actual needs and application scenarios. In this regard, the present invention is not limited.

It should be noted that, in the embodiment of the present invention, after dividing the detection area into a plurality of sub-areas, the material identification device determines each sub-area in the plurality of sub-areas according to the approximate scattering point corresponding to each particle in the plurality of radiation particles corresponding radiation particles.

S103. For each sub-region in the multiple sub-regions, determine the corresponding scattering density based on the scattering angle of the corresponding radiation particle, and obtain multiple scattering densities corresponding to the multiple sub-regions one-to-one.

In the embodiment of the present invention, the material identification device determines a corresponding scattering density for each sub-area of the plurality of sub-areas based on the scattering angle of the corresponding radiation particle, and obtains a plurality of scattering densities corresponding to the plurality of sub-areas one-to-one.

It should be noted that, in the embodiment of the present invention, the material identification device determines a scattering density for each sub-region in the multiple sub-regions by using the radiation angles corresponding to all the radiation particles corresponding to the sub-region.

Specifically, in the embodiment of the present invention, the material identification device determines the corresponding scattering density for each sub-area of the plurality of sub-areas based on the scattering angle of the corresponding radiation particle, and obtains a plurality of scattering densities corresponding to the plurality of sub-areas one-to-one. Density, including: for each sub-region in multiple sub-regions, calculating the variance of the scattering angle of the corresponding radiation particle to obtain the corresponding scattering angle variance; for each sub-region in the multiple sub-regions, determining the height of the corresponding sub-region as the corresponding For each sub-region in the multiple sub-regions, the ratio of the corresponding scattering angle variance and the particle-passing length is determined as the corresponding scattering density.

It should be noted that, in the embodiment of the present invention, the material identification device calculates the variance of the scattering angles corresponding to all the radiation particles corresponding to the region for each sub-region in the multiple sub-regions. The specific calculation method is shown in formula (1):

为第i个子区域对应的M个辐射粒子的散射角方差，σ_θ1,σ_θ2,…,σ_θN为第i个子区域对应的M个辐射粒子的散射角，σ_a为第i个子区域对应的M个辐射粒子的散射角均值。in,

is the scattering angle variance of the M radiation particles corresponding to the ith sub-region, σ _θ1 , σ _θ2 ,…,σ _θN is the scattering angle of the M radiation particles corresponding to the ith sub-region, σ _a is the corresponding The mean value of the scattering angles of the M radiation particles.

It should be noted that, in the embodiment of the present invention, the material identification device determines the height of the corresponding sub-region as the particle passing length for each sub-region in the multiple sub-regions, and the specific particle passing length is the radiation particle passing through length. The distance between the two parallel surfaces of , that is, the height of the subregion.

It should be noted that, in the embodiment of the present invention, after obtaining the particle passing length and scattering angle variance corresponding to each subregion in the multiple subregions, the material identification device compares the scattering angle variance corresponding to a certain subregion with the particle passing The length ratio is determined as the scattering angular density of the corresponding sub-region. The specific calculation formula is shown in formula (2):

为第i个子区域对应的散射角方差。Among them, λ _i is the scattering density corresponding to the ith sub-region, L _i is the height corresponding to the ith sub-region,

is the scattering angle variance corresponding to the ith subregion.

S104 , using a preset material identification model to identify the material type of the material to be identified based on multiple scattering densities, and obtain a material type identification result of the material to be identified.

In an embodiment of the present invention, the material identification device uses a preset material identification model to identify the material type of the material to be identified based on multiple scattering densities, and obtains a material type identification result of the material to be identified.

It should be noted that, in the embodiment of the present invention, the material type identification result may include the material type of the material to be identified, the accuracy of identifying the material type, the false alarm rate, or the overall accuracy, the specific material type identification The result can be set according to actual requirements and application scenarios, which are not limited in the present invention.

The calculation method of the recognition accuracy of a specific preset material recognition model for a certain material type can be as follows:

为材料x_i正确识别的样本数，

为材料x_i的总样本数。Among them, t _accuracy ( _xi ) is the accuracy of correct identification of material _xi ,

the number of samples correctly identified for material x _i ,

is the total number of samples of material x _i .

The specific preset material identification model misidentifies a certain material type as other materials. The calculation method of the misrecognition rate can be as follows:

Among them, t( _xi , x _j ) is the misrecognition rate of the material _xi being identified as x _j .

The formula for calculating the overall accuracy of the specific preset material identification model can be:

为Q种材料的正确识别样本数。具体的，在本发明的实施例中，材料识别装置利用预设材料识别模型，基于多个散射密度对待识别材料进行材料类型识别，得到待识别材料的材料类型识别结果，包括：从多个散射密度中，选取出表征异常的散射密度；将选取出的散射密度输入预设材料识别模型，得到材料类型识别结果。Among them, t _total is the overall accuracy of the preset material identification model, N _{total number of samples is the total number} of samples of Q materials,

Number of correctly identified samples for Q materials. Specifically, in the embodiment of the present invention, the material identification device uses a preset material identification model to identify the material type of the material to be identified based on multiple scattering densities, and obtains a material type identification result of the material to be identified, including: from multiple scattering densities. In the density, select the scattering density that characterizes the abnormality; input the selected scattering density into the preset material identification model to obtain the material type identification result.

It should be noted that, in the embodiment of the present invention, after acquiring multiple scattering densities, the material identification device will first screen the multiple scattering densities, and select the scattering densities corresponding to the sub-regions affected by the material to be identified. Then, the selected scattering density is used to determine the material type identification result of the material to be identified. Since only the scattering density corresponding to the sub-region affected by the material to be identified is processed, the accuracy and efficiency of material identification can be improved.

Specifically, in the embodiment of the present invention, the material identification device uses a preset material identification model to identify the material type of the material to be identified based on multiple scattering densities, and before obtaining the material type identification result of the material to be identified, the following steps may also be performed. : Obtain the scattering density sample, and use the material identification model to be trained to identify the material type of the sample to be identified based on the scattering density sample, and obtain the material type identification result of the sample to be identified; Loss information between preset target material type identification results is obtained; based on the loss information, model parameters are adjusted for the identification model of the material to be trained to obtain a preset material identification model.

It should be noted that, in the embodiment of the present invention, the material identification model to be trained may be a convolutional neural network model, and the material identification model may input the scattering density samples into the material identification model to be trained, and perform feature extraction on the scattering density samples , to obtain the identification result of the material type of the sample to be identified.

FIG. 4 is a schematic structural diagram of an exemplary identification model of a material to be identified according to an embodiment of the present invention. As shown in Figure 4, the material identification model inputs the scattering density samples into the preset material identification model, and the feature map output by the convolution layer is passed to the pooling layer for feature selection and information filtering, further reducing the data dimension and speeding up the model. Training speed. After multiple rounds of convolutional layer and pooling layer processing, the input scattering density is abstracted into high-order features, and the extracted features are nonlinearly combined by the fully connected layer to obtain the sample to be identified corresponding to the scattering density sample. The result of the material type identification.

It should be noted that, in the embodiment of the present invention, after obtaining the material type identification result of the sample to be identified, the material identification device will calculate the identification result of the material type of the sample to be identified and the target material type identification preset for the scattering density sample. The loss information between the results is obtained, that is, the weight that contributes the most to the loss is determined, and then based on the loss information, the model parameters are adjusted for the identification model of the material to be trained, and the preset material identification model is obtained. The specific parameter adjustment process is to first use the cross The entropy loss function adjusts the optimization weight to reduce the loss, the expression is shown in formula (6):

Among them, C is the cross loss function, N is the number of samples, y represents the target material type recognition result preset for the scattering density sample, x represents the material type recognition result of the sample to be recognized in the previous layer, and a represents the material type of the sample to be recognized. The identification result, the specific expression of a is shown in formula (7):

a=σ(z)(z=wx+b) (7)

Among them, w is the link weight, b is a parameter, and the cross entropy loss function is derived:

The corresponding parameter update formula is:

where η is the learning rate. It can be seen from equation (10) that the update speed of the weight is linearly related to the loss information (a-y). When the loss information is large, the weight update is fast, and when the loss information is small, the weight update is slow. Input multiple scattering density samples into the material identification model to be trained for iterative training.

FIG. 5 is a schematic diagram of an exemplary variation of model accuracy with iteration times according to an embodiment of the present invention. As shown in Figure 5, as the number of iterations increases, the accuracy gradually converges. FIG. 6 is a schematic diagram of an exemplary loss function changing with the number of iterations provided by an embodiment of the present invention. As shown in FIG. 6 , as the number of iterations increases, the loss function gradually converges. Finally, an optimal weight set is obtained, and a trained preset material recognition model is obtained.

It should be noted that, in the embodiment of the present invention, the scattering density samples obtained by the material identification device may be simulated data or actually measured data. Correspondingly, the trained model is only applicable to the corresponding environment. For example, a preset material identification model trained by the material identification device using simulated data can be applied to material identification in a simulated environment, and a preset material identification model trained by the material identification device using actual data can be applied to material identification in actual situations. Of course, the models corresponding to the length of the detection time may be different or the same, which is not limited in the present invention.

The present invention provides a material identification method. The method includes: acquiring the approximate scattering point and scattering angle corresponding to each particle in a plurality of radiation particles passing through a detection area where a material to be identified is deployed; and dividing the detection area into a plurality of sub-areas , and for each sub-region in the multiple sub-regions, from the multiple radiation particles, select the radiation particle whose corresponding approximate scattering point is located in the sub-region, and determine it as the corresponding radiation particle; for each sub-region in the multiple sub-regions, based on the corresponding According to the scattering angle of the radiation particles, the corresponding scattering densities are determined, and multiple scattering densities corresponding to multiple sub-regions are obtained; using the preset material identification model, based on the multiple scattering densities, the material type is identified for the material to be identified, and the identified material is obtained. The material type identification result for the material. The material identification method provided by the present invention directly utilizes a preset material identification model and determines the material type of the material to be identified in combination with the scattering density corresponding to the detection area, thereby improving the efficiency of material identification.

The present invention provides a material identification device, and FIG. 7 is a first structural schematic diagram of a material identification device provided by an embodiment of the present invention. As shown in Figure 7, including:

An acquisition module 701, configured to acquire the approximate scattering point and scattering angle corresponding to each particle among the plurality of radiation particles passing through the detection area where the material to be identified is deployed;

The selection module 702 is configured to divide the detection area into a plurality of sub-areas, and for each sub-area in the plurality of sub-areas, from the plurality of radiation particles, select the radiation whose corresponding approximate scattering point is located in the sub-area particle, determined as the corresponding radiation particle;

A determination module 703, configured to, for each sub-region of the plurality of sub-regions, determine a corresponding scattering density based on the scattering angle of the corresponding radiation particle, and obtain a plurality of scattering densities corresponding to the plurality of sub-regions one-to-one;

The identification module 704 is configured to use a preset material identification model to identify the material type of the material to be identified based on the plurality of scattering densities, and obtain a material type identification result of the material to be identified.

Optionally, the obtaining module 701 is specifically configured to obtain the incident track and the output track corresponding to each particle in the plurality of radiation particles; for each particle in the plurality of radiation particles, the corresponding incident track is obtained. The intersection of the extension line of the track and the extension line of the exit track is determined as the corresponding approximate scattering point; for each particle in the plurality of radiation particles, the extension line of the corresponding incident track and the extension line of the exit track are determined as corresponding approximate scattering points; The included angle of the line is determined as the corresponding scattering angle.

Optionally, the determining module 703 is specifically configured to, for each sub-region in the multiple sub-regions, calculate the variance of the scattering angle of the corresponding radiation particle to obtain the corresponding scattering angle variance; For each sub-region, the height of the corresponding sub-region is determined as the corresponding particle passing length; for each sub-region in the multiple sub-regions, the ratio of the corresponding scattering angle variance to the particle passing length is determined as the corresponding scattering density.

Optionally, the identification module 704 is specifically configured to select a scattering density representing anomalies from the plurality of scattering densities; input the selected scattering density into the preset material identification model to obtain the material type Identify the results.

Optionally, the material identification device further includes a model training module (not shown in the figure), which is used to obtain scattering density samples, and use the material identification model to be trained to identify the material type of the samples to be identified based on the scattering density samples. , obtain the identification result of the material type of the sample to be identified; calculate the loss information between the identification result of the material type of the sample to be identified and the identification result of the target material type preset for the scattering density sample, and obtain the loss information; For the loss information, model parameters are adjusted for the material identification model to be trained to obtain the preset material identification model.

The present invention provides a material identification device, and FIG. 8 is a second structural schematic diagram of a material identification device provided by an embodiment of the present invention. As shown in FIG. 8 , the material identification device includes: a processor 801, a memory 802 and a communication bus 803;

The communication bus 803 is used to realize the communication connection between the processor 801 and the memory 802;

The processor 801 is configured to execute the material identification program stored in the memory 802 to implement the above material identification method.

The invention provides a material identification device, which acquires the approximate scattering point and scattering angle corresponding to each particle in a plurality of radiation particles passing through a detection area where a material to be identified is deployed; the detection area is divided into a plurality of sub-areas, and the For each sub-region of the multiple sub-regions, from the multiple radiation particles, select the radiation particles whose corresponding approximate scattering points are located in the sub-region, and determine them as the corresponding radiation particles; for each sub-region in the multiple sub-regions, based on the corresponding radiation particles According to the scattering angle, the corresponding scattering densities are determined, and multiple scattering densities corresponding to multiple sub-regions are obtained; the preset material identification model is used to identify the material type of the material to be identified based on the multiple scattering densities, and the material of the material to be identified is obtained. Type identification result. The material identification device provided by the present invention directly utilizes a preset material identification model and determines the material type of the material to be identified in combination with the scattering density corresponding to the detection area, thereby improving the efficiency of material identification.

The present invention provides a computer-readable storage medium, where one or more programs are stored in the computer-readable storage medium, and the one or more programs can be executed by one or more processors to implement the above-mentioned material identification method . The computer-readable storage medium may be a volatile memory (volatile memory), such as a random-access memory (Random-Access Memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (Read Only Memory). -Only Memory, ROM), flash memory (flash memory), hard disk (Hard Disk Drive, HDD) or solid-state drive (Solid-State Drive, SSD); it can also be a respective device including one or any combination of the above memories, Such as mobile phones, computers, tablet devices, personal digital assistants, etc.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, optical storage, and the like.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application. , all should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. A material identification method, characterized in that the method comprises:

acquiring approximate scattering points and scattering angles corresponding to each of a plurality of radiation particles passing through a detection area where a material to be identified is deployed;

dividing the detection area into a plurality of sub-areas, and selecting radiation particles with corresponding approximate scattering points in the sub-areas from the plurality of radiation particles aiming at each sub-area in the plurality of sub-areas to determine the radiation particles as corresponding radiation particles;

for each sub-region in the plurality of sub-regions, determining a corresponding scattering density based on a scattering angle of the corresponding radiation particle, and obtaining a plurality of scattering densities corresponding to the plurality of sub-regions one to one;

and carrying out material type identification on the material to be identified based on the plurality of scattering densities by using a preset material identification model to obtain a material type identification result of the material to be identified.

2. The method of claim 1, wherein acquiring a plurality of radiation particles traversing a detection region in which a material to be identified is deployed, each particle corresponding to an approximate scatter point and scatter angle, comprises:

acquiring an incident track and an emergent track corresponding to each of the plurality of radiation particles;

determining, for each of the plurality of radiation particles, an intersection of an extension of the corresponding incident track and an extension of the exit track as a corresponding approximate scattering point;

for each of the plurality of radiation particles, determining an angle between an extension of the corresponding incident track and an extension of the exit track as a corresponding scattering angle.

3. The method of claim 1, wherein determining, for each of the plurality of sub-regions, a corresponding scattering density based on a scattering angle of a corresponding radiation particle, resulting in a plurality of scattering densities in one-to-one correspondence with the plurality of sub-regions comprises:

calculating the variance of the scattering angle of the corresponding radiation particle aiming at each sub-area in the plurality of sub-areas to obtain the corresponding variance of the scattering angle;

for each sub-region of the plurality of sub-regions, determining a height of the corresponding sub-region as a corresponding particle penetration length;

for each sub-region of the plurality of sub-regions, determining a ratio of a corresponding scatter angle variance and a particle pass length as a corresponding scatter density.

4. The method according to claim 1, wherein the performing material type recognition on the material to be recognized based on the plurality of scattering densities by using a preset material recognition model to obtain a material type recognition result of the material to be recognized comprises:

selecting a scattering density characterizing an anomaly from the plurality of scattering densities;

and inputting the selected scattering density into the preset material identification model to obtain the material type identification result.

5. The method according to claim 1, wherein before performing material type recognition on the material to be recognized based on the plurality of scattering densities by using a preset material recognition model to obtain a material type recognition result of the material to be recognized, the method further comprises:

acquiring a scattering density sample, and performing material type recognition on the sample to be recognized based on the scattering density sample by using a material recognition model to be trained to obtain a material type recognition result of the sample to be recognized;

calculating loss information between a material type identification result of the sample to be identified and a target material type identification result preset for the scattering density sample to obtain loss information;

and adjusting model parameters of the material recognition model to be trained based on the loss information to obtain the preset material recognition model.

6. A material identification device, comprising:

an acquisition module for acquiring an approximate scatter point and a scatter angle corresponding to each of a plurality of radiation particles passing through a detection region where a material to be identified is deployed;

a selecting module, configured to divide the detection region into a plurality of sub-regions, and select, for each sub-region in the plurality of sub-regions, a radiation particle whose corresponding approximate scattering point is located in the sub-region from the plurality of radiation particles, and determine the radiation particle as a corresponding radiation particle;

a determining module, configured to determine, for each of the multiple sub-regions, a corresponding scattering density based on a scattering angle of the corresponding radiation particle, so as to obtain multiple scattering densities corresponding to the multiple sub-regions one to one;

and the identification module is used for identifying the material type of the material to be identified based on the plurality of scattering densities by using a preset material identification model to obtain a material type identification result of the material to be identified.

7. The apparatus of claim 6,

the acquiring module is specifically configured to acquire an incident track and an exit track corresponding to each of the plurality of radiation particles; determining, for each of the plurality of radiation particles, an intersection of an extension of the corresponding incident track and an extension of the exit track as a corresponding approximate scattering point; for each of the plurality of radiation particles, determining an angle between an extension of the corresponding incident track and an extension of the exit track as a corresponding scattering angle.

8. The apparatus of claim 6,

the determining module is specifically configured to calculate, for each sub-region of the plurality of sub-regions, a variance of a scattering angle of the corresponding radiation particle, so as to obtain a corresponding scattering angle variance; for each sub-region of the plurality of sub-regions, determining a height of the corresponding sub-region as a corresponding particle penetration length; for each sub-region of the plurality of sub-regions, determining a ratio of a corresponding scatter angle variance and a particle pass length as a corresponding scatter density.

9. A material identification device, comprising: a processor, a memory, and a communication bus;

the communication bus is used for realizing communication connection between the processor and the memory;

the processor is configured to execute the material identification program stored in the memory to implement the material identification method according to any one of claims 1 to 5.

10. A computer-readable storage medium storing one or more programs, the one or more programs being executable by one or more processors to implement the material identification method of any one of claims 1-5.

Images