Material testing method, device, equipment and storage medium

Abstract

The present application provides a material detection method, device, equipment and storage medium, the method comprising: obtaining terahertz reflection waveform data of a material to be detected, extracting effective pulse features from the terahertz reflection waveform data and obtaining the scanning point index corresponding to the effective pulse features, generating an effective pulse feature set and a scanning point index set of the material to be detected; using a preset first self-organizing mapping network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set, generating a first cluster type set; according to the first cluster type set, using a preset second self-organizing mapping network to perform pulse type clustering processing on each scanning point in the scanning point index set, generating a second cluster type set; according to the second cluster type set, performing imaging processing on the material to be detected, generating an imaging result map of the material to be detected. The method can effectively reduce the loss of terahertz spectrum information, and the imaging reliability is high.

Description

Technical Field

[0001] The present application relates to the technical field of material detection, and in particular to a material detection method, device, equipment and storage medium. Background Art

[0002] With the development of high-tech, various high-performance new materials are widely used in equipment in various fields such as aerospace, electric power, etc. During the production, transportation and operation of the equipment, defects such as air gaps, holes, cracks, inclusions, etc. will appear inside the materials. These defects cannot be identified from the appearance and will affect the performance of the equipment, cause economic losses, and even casualties. Terahertz wave imaging technology is one of the most widely used intuitive and fast means for non-destructive detection of internal defects in specific materials. However, most of the existing terahertz imaging technologies use a single spectral characteristic parameter for imaging. This method will cause a lot of information loss and cannot effectively use the terahertz spectrum information in the entire frequency range for imaging. It is not enough to effectively image the actual structural defects of various and complex types, and the imaging reliability is poor. Summary of the invention

[0003] In view of this, the embodiments of the present application provide a material detection method, device, equipment and storage medium, which can effectively utilize the terahertz spectrum information in the entire frequency range for imaging, reduce the loss of terahertz spectrum information, and improve imaging reliability.

[0004] A first aspect of an embodiment of the present application provides a material detection method, comprising:

[0005] Acquire terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain scanning point indexes corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected;

[0006] Using a preset first self-organizing map network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set;

[0007] According to the first cluster type set, a preset second self-organizing map network is used to perform pulse type clustering processing on each scanning point in the scanning point index set to generate a second cluster type set;

[0008] According to the second cluster type set, imaging processing is performed on the material to be detected to generate an imaging result map of the material to be detected.

[0009] In combination with the first aspect, in a first possible implementation manner of the first aspect, the step of obtaining terahertz reflection waveform data of the material to be detected further includes:

[0010] Using a terahertz imaging device to perform terahertz wave scanning on the material to be detected to generate bipolar pulse data of the material to be detected;

[0011] Deconvolution processing is performed on the bipolar pulse data to generate corresponding unipolar pulse data, and the unipolar pulse data is acquired as terahertz reflection waveform data of the material to be detected.

[0012] In combination with the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the step of extracting effective pulse features from the terahertz reflection waveform data includes:

[0013] Extracting a pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and comparing the pulse peak value corresponding to each scanning point with a preset peak value threshold value, and if the pulse peak value is greater than the preset peak value threshold value, determining the pulse peak value as a valid pulse feature;

[0014] All pulse peaks determined as effective pulse features are collected to generate an effective pulse peak set, and based on the effective pulse peak set, the peak time corresponding to each pulse peak in the effective pulse peak set is extracted from the terahertz reflection waveform data, and the peak times corresponding to each pulse peak are collected to generate an effective pulse peak time set.

[0015] In combination with the first aspect, in a third possible implementation manner of the first aspect, according to the first cluster type set, using a preset second self-organizing map network to perform pulse type clustering processing on each scanning point in the scanning point index set to generate a second cluster type set includes:

[0016] According to the first cluster type set and the scanning point index set, the terahertz waveform information of each scanning point in the scanning point index set is encoded to generate a coding matrix, and the coding matrix is ​​input as an input vector into the second self-organizing mapping network for pulse type clustering processing to generate a second cluster type set.

[0017] In combination with the first aspect, in a fourth possible implementation manner of the first aspect, the step of performing imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected includes:

[0018] According to the second cluster type set, the pulse type corresponding to each scanning point is converted into an image pixel value, and the image pixel value corresponding to each scanning point is filled into the image matrix of the material to be detected according to the scanning point to generate an imaging result map of the material to be detected.

[0019] A second aspect of an embodiment of the present application provides a material detection device, the material detection device comprising:

[0020] An extraction module is used to obtain terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain the scanning point index corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected;

[0021] A first clustering module, used for performing feature information clustering processing on each effective pulse feature in the effective pulse feature set using a preset first self-organizing map network to generate a first cluster type set;

[0022] A second clustering module, configured to perform pulse type clustering processing on each scanning point in the scanning point index set using a preset second self-organizing map network according to the first cluster type set, so as to generate a second cluster type set;

[0023] An imaging module is used to perform imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected.

[0024] In combination with the second aspect, in a first possible implementation manner of the second aspect, the material detection device further includes:

[0025] A scanning submodule, used for performing terahertz wave scanning on the material to be detected by using a terahertz imaging device to generate bipolar pulse data of the material to be detected;

[0026] The deconvolution submodule is used to perform deconvolution processing on the bipolar pulse data to generate corresponding unipolar pulse data, and obtain the unipolar pulse data as terahertz reflection waveform data of the material to be detected.

[0027] In combination with the second aspect or the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the material detection device further includes:

[0028] A comparison submodule, used for extracting a pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and comparing the pulse peak value corresponding to each scanning point with a preset peak value threshold value, and if the pulse peak value is greater than the preset peak value threshold value, determining the pulse peak value as a valid pulse feature;

[0029] The extraction submodule is used to collect all pulse peak values ​​determined as effective pulse characteristics to generate an effective pulse peak value set, and according to the effective pulse peak value set, extract the peak time corresponding to each pulse peak value in the effective pulse peak value set from the terahertz reflection waveform data, collect the peak times corresponding to each pulse peak value, and generate an effective pulse peak time set.

[0030] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the electronic device, wherein when the processor executes the computer program, each step of the material detection method provided in the first aspect is implemented.

[0031] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, each step of the material detection method provided in the first aspect is implemented.

[0032] The material detection method, device, electronic device and storage medium provided in the embodiments of the present application have the following beneficial effects:

[0033] The present application obtains the terahertz reflection waveform data of the material to be detected, extracts the effective pulse feature from the terahertz reflection waveform data and obtains the scanning point index corresponding to the effective pulse feature, generates the effective pulse feature set and scanning point index set of the material to be detected; uses a preset first self-organizing mapping network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set; according to the first cluster type set, uses a preset second self-organizing mapping network to perform pulse type clustering processing on each scanning point in the scanning point index set to generate a second cluster type set; according to the second cluster type set, performs imaging processing on the material to be detected to generate an imaging result map of the material to be detected. Based on this method, two self-organizing mapping networks are used to perform unsupervised clustering processing on the pulse features and scanning points in succession, thereby improving the information utilization rate of terahertz imaging, effectively reducing the loss of terahertz spectrum information, and realizing the effective use of the entire frequency range of terahertz spectrum information for imaging, which can effectively image various and complex actual structural defects and improve imaging reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0035] Figure 1 A flow chart of a material detection method provided in an embodiment of the present application;

[0036] Figure 2 A schematic flow chart of a method for obtaining terahertz reflection waveform data of a material to be detected in a material detection method provided in an embodiment of the present application;

[0037] Figure 3 A comparison diagram of bipolar pulse and unipolar pulse;

[0038] Figure 4 A schematic flow chart of a method for extracting effective pulse features in a material detection method provided in an embodiment of the present application;

[0039] Figure 5 A schematic diagram of the structure of the coding matrix;

[0040] Figure 6 A basic structural block diagram of a material detection device provided in an embodiment of the present application;

[0041] Figure 7 A first detailed structural block diagram of a material detection device provided in an embodiment of the present application;

[0042] Figure 8 A second detailed structural block diagram of the material detection device provided in the embodiment of the present application;

[0043] Fig. 9 A basic structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] The material detection method provided in the embodiment of the present application aims to utilize the self-organizing map algorithm, through two self-organizing map networks, to successively perform unsupervised clustering of pulses and scanning points; a terahertz imaging method is proposed for the simultaneous presence of multiple internal defects, which can detect and evaluate the internal defects of the material and reduce the possibility of failures and accidents. Based on this material detection method, the imaging parameters can effectively reflect the information of the entire waveform and improve the imaging performance; various types of structural defects can be processed, which is universal and does not require manual analysis of the waveform data in advance, reducing the workload; using unsupervised learning methods, better results can be achieved without the need for a large amount of labeled data, saving workload; imaging of structural defects inside the material can be achieved, which is helpful for the detection and evaluation of the equipment status and enhancing the safety and stability of various types of equipment.

[0046] See also Figure 1 , Figure 1 A flow chart for implementing a material detection method provided in an embodiment of the present application.

[0047] Details are as follows:

[0048] S11: Acquire terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain scanning point indexes corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected.

[0049] In this embodiment, terahertz (THz) waves refer to electromagnetic waves with a frequency in the range of 0.1 to 10 THz, and a wavelength between 0.03 and 3 mm. The principle of terahertz imaging is to scan each scanning point of the material to be detected by using a terahertz incident wave, obtain the reflected wave of each scanning point of the material to be detected, and assign the spectral parameters of each scanning point according to the reflected wave, wherein the spectral parameters include pulse amplitude, pulse phase, time slice, spectrum parameters, etc., and then the imaging of the material to be detected is realized by the spectral parameters of each scanning point. In this embodiment, firstly, the material to be detected is scanned by using a terahertz imaging device, and the spectral parameters corresponding to each scanning point are obtained according to the scanning points. The spectral parameters corresponding to each scanning point are the terahertz reflection waveform data of the material to be detected. After obtaining the terahertz reflection waveform data of the material to be detected, extract the effective pulse features from the terahertz reflection waveform data and obtain the scanning point index corresponding to each effective pulse feature, collect the extracted effective pulse features to generate the effective pulse feature set of the material to be detected, and collect the scanning point index corresponding to each effective pulse feature to generate a scanning point index set. Specifically, in this embodiment, the pulse feature includes a pulse peak value and a peak time. For the pulse peak feature, a corresponding effective pulse peak value set can be obtained. For the peak time feature, a corresponding effective pulse peak time set can be obtained.

[0050] In some embodiments of this application, please refer to Figure 2 , Figure 2 A schematic flow chart of a method for obtaining terahertz reflection waveform data of a material to be detected in a material detection method provided in an embodiment of the present application. The details are as follows:

[0051] S21: using a terahertz imaging device to perform terahertz wave scanning on the material to be detected, to generate bipolar pulse data of the material to be detected;

[0052] S22: performing deconvolution processing on the bipolar pulse data to generate corresponding unipolar pulse data, and acquiring the unipolar pulse data as terahertz reflection waveform data of the material to be detected.

[0053] In this embodiment, please refer to Figure 3 , Figure 3 This is a comparison diagram of bipolar pulses and unipolar pulses. Figure 3As shown, the pulses emitted by the terahertz imaging device for scanning are generally bipolar pulses. In this embodiment, in order to facilitate the subsequent data clustering processing, when obtaining the terahertz reflection waveform data of the material to be detected, the terahertz imaging device is used to perform terahertz wave scanning on the material to be detected. After generating the bipolar pulse data of the material to be detected, the obtained bipolar pulse data can be deconvoluted to generate corresponding unipolar pulse data, and the unipolar pulse data can be obtained as the terahertz reflection waveform data of the material to be detected.

[0054] In some embodiments of this application, please refer to Figure 4 , Figure 4 A schematic flow chart of a method for extracting effective pulse features in the material detection method provided in the embodiment of the present application. The details are as follows:

[0055] S41: extracting a pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and comparing the pulse peak value corresponding to each scanning point with a preset peak threshold value, and if the pulse peak value is greater than the preset peak threshold value, determining the pulse peak value as a valid pulse feature;

[0056] S42: All pulse peaks determined as effective pulse features are collected to generate an effective pulse peak set, and based on the effective pulse peak set, the peak time corresponding to each pulse peak in the effective pulse peak set is extracted from the terahertz reflection waveform data, and the peak times corresponding to each pulse peak are collected to generate an effective pulse peak time set.

[0057] In this embodiment, the effective pulse feature set of the material to be detected includes an effective pulse peak value set and an effective pulse peak time set. Due to the presence of background noise in the scanning process of the material to be detected, many "false pulses" with small amplitudes may be generated in the terahertz reflection waveform data. In this embodiment, the effective pulse feature can be obtained from the terahertz reflection waveform data by threshold screening. Exemplarily, a peak threshold is first set, and then the pulse peak value corresponding to each scanning point is extracted from the terahertz reflection waveform data, and the pulse peak value corresponding to each scanning point is compared with the peak threshold value one by one. For the pulse peak value corresponding to each scanning point, if the pulse peak value is less than the peak threshold value, it is characterized that the pulse peak value is a "false pulse" caused by noise, and the pulse peak value is screened out at this time. If the pulse peak value is greater than the peak threshold value, it is characterized that the pulse peak value is not a "false pulse" caused by noise, and the pulse peak value is determined as an effective pulse feature. Then, all the pulse peak values ​​determined as effective pulse features obtained by screening are collected to generate an effective pulse peak value set. Since each pulse amplitude in the terahertz reflection waveform data has corresponding time information, after obtaining the effective pulse peak value set, the peak time corresponding to each pulse peak in the effective pulse peak value set can be extracted from the terahertz reflection waveform data, and then the peak time corresponding to each pulse peak is aggregated to generate an effective pulse peak time set. Finally, the generated effective pulse peak value set and effective pulse peak time set are configured as the effective pulse feature set of the material to be detected.

[0058] S12: using a preset first self-organizing map network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set.

[0059] In this embodiment, the self-organizing map (SOM) is an unsupervised learning method for generating a low-dimensional map of an input vector by learning the features of the data in the training set. The self-organizing map network specifically includes a two-layer structure of an input layer and a competition layer, wherein the number of neurons in the input layer is the same as the dimension of the input vector, and the size of the competition layer depends on the target of the data analysis. Each neuron in the competition layer is connected to the input layer through a weight vector. In this embodiment, the training process of the self-organizing map network includes: first, the training set is used as an input vector to input into the self-organizing map network for training, the Euclidean distance between each input vector and the weight vectors of all competition layer neurons is calculated, and then the neuron in the competition layer with the closest Euclidean distance to the current input vector is used as the best matching unit (BMU); and then the weight vector of the neuron in the domain range of the best matching unit is updated according to the neuron with the closest Euclidean distance obtained in each iteration, until the iteration limit, that is, the self-organizing map network is trained to a convergence state. Exemplarily, in this embodiment, the formula for updating the weight vector of the neuron in the domain range of the best matching unit is expressed as:

[0060] W v (s+1)=W v (s)+θ(u,v,s)·α(s)·(D(t)-W v (s))

[0061] Among them, W v is the weight vector of neuron v, s is the current iteration number, u is the index number corresponding to the BMU, D is the input vector set, and t is the index number of the current input vector. θ(u, v, s) is the domain function, and α(s) is the learning rate. It should be noted that during the training process of the self-organizing map network, the domain function θ(u, v, s) and the learning rate α(s) decrease as the number of iterations increases.

[0062] In this embodiment, the preset first self-organizing map network is a network that is trained to a convergence state with a pulse feature set as an input vector, and is used to cluster all pulses in the pulse feature set according to their respective pulse feature information. In this embodiment, after obtaining the effective pulse feature set and the scanning point index set of the material to be detected, the pulse feature data recorded in the effective pulse feature set of the material to be detected is input as an input vector into the preset first self-organizing map network, and the first self-organizing map network is used to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set, in which a number of cluster types obtained by clustering are recorded, and the number of cluster types is determined according to the clustering result of the first self-organizing map network.

[0063] S13: According to the first cluster type set, a preset second self-organizing map network is used to perform pulse type clustering processing on each scanning point in the scanning point index set to generate a second cluster type set.

[0064] In this embodiment, the preset second self-organizing map network is a network trained to a convergent state with the encoding matrix as the input vector, and is used to cluster each scanning point in the scanning point index set according to the pulse type to achieve classification of the scanning points. Figure 5 , Figure 5 is a schematic diagram of the structure of the coding matrix. Figure 5 As shown, after the first self-organizing map network generates the first cluster type set, it is assumed that the first cluster type set is C1(i) ​​(i=1, 2, ..., m), that is, the set contains m cluster types. Assume that the scanning point index set is N P (i) (i = 1, 2, ..., n), that is, the set contains n scanning points. At this time, according to the first cluster type set and the scanning point index set, by encoding the terahertz waveform information of each scanning point, an m*n encoding matrix can be constructed. After obtaining the encoding matrix, the encoding matrix is ​​input as an input vector into the second self-organizing map network to perform pulse type clustering processing on each scanning point in the scanning point index set of the second self-organizing map network, and generate a second cluster type set, which records the pulse type corresponding to each scanning point obtained by clustering. It can be understood that in this embodiment, m = S X1 ·S Y1 , n=N x ·N y Among them, S X1 and S Y1 are the two size parameters of the competition layer in the first self-organizing map network, N x and N y It is the number of scanning steps in two scanning directions during two-dimensional scanning.

[0065] Exemplarily, when constructing the coding matrix, if the reflected waveform of the scanning point k contains a valid pulse of type j in the first self-organizing map network, the corresponding element in the coding matrix is ​​assigned a value of 1, otherwise, it is assigned a value of 0. The specific mathematical expression is as follows:

[0066]

[0067] In this embodiment, after the encoding matrix is ​​obtained, the encoding matrix can be expressed as a set D2(i)=[P(1, i), P(2, i), ..., P(m, i)] T (i=1, 2, ..., n), the set includes the valid pulse types contained in all scanning points.

[0068] S14: performing imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected.

[0069] In this embodiment, after the second cluster type set is obtained, the second cluster type set records the pulse type corresponding to each scanning point. In this embodiment, by converting the pulse type corresponding to each scanning point into an image pixel value of an image matrix, when imaging the material to be detected, by filling the image matrix of the material to be detected with corresponding pixels according to the scanning points, and filling the image pixel value corresponding to each scanning point into the corresponding scanning point position in the image matrix of the material to be detected, an imaging result map of the material to be detected can be generated.

[0070] It is understandable that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0071] In some embodiments of this application, please refer to Figure 6 , Figure 6 This is a basic structural block diagram of a material detection device provided in an embodiment of the present application. The various units included in the device in this embodiment are used to perform the various steps in the above method embodiment. For details, please refer to the relevant description in the above method embodiment. For ease of explanation, only the parts related to this embodiment are shown. Figure 6 As shown, the material detection device includes: an extraction module 61, a first clustering module 62, a second clustering module 63 and an imaging module 64. Among them: the extraction module 61 is used to obtain the terahertz reflection waveform data of the material to be detected, extract the effective pulse feature from the terahertz reflection waveform data and obtain the scanning point index corresponding to the effective pulse feature, and generate the effective pulse feature set and scanning point index set of the material to be detected. The first clustering module 62 is used to use a preset first self-organizing map network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set. The second clustering module 63 is used to use a preset second self-organizing map network to perform pulse type clustering processing on each scanning point in the scanning point index set according to the first cluster type set to generate a second cluster type set. The imaging module 64 is used to perform imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected.

[0072] In some embodiments of this application, please refer to Figure 7 , Figure 7This is a first detailed structural block diagram of the material detection device provided in the embodiment of the present application. Figure 7 As shown, the material detection device also includes a scanning submodule 71 and a deconvolution submodule 72. The scanning submodule 71 is used to use a terahertz imaging device to perform terahertz wave scanning on the material to be detected to generate bipolar pulse data of the material to be detected. The deconvolution submodule 72 is used to perform deconvolution processing on the bipolar pulse data to generate corresponding unipolar pulse data, and obtain the unipolar pulse data as terahertz reflection waveform data of the material to be detected.

[0073] In some embodiments of this application, please refer to Figure 8 , Figure 8 This is a second detailed structural block diagram of the material detection device provided in the embodiment of the present application. Figure 8 As shown, the material detection device also includes a comparison submodule 81 and an extraction submodule 82. The comparison submodule 81 is used to extract the pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and compare the pulse peak value corresponding to each scanning point with the preset peak threshold value. If the pulse peak value is greater than the preset peak threshold value, the pulse peak value is determined as a valid pulse feature. The extraction submodule 82 is used to collect all pulse peak values ​​determined as valid pulse features to generate a valid pulse peak value set, and according to the valid pulse peak value set, extract the peak time corresponding to each pulse peak in the valid pulse peak value set from the terahertz reflection waveform data, collect the peak time corresponding to each pulse peak value, and generate a valid pulse peak time set.

[0074] It should be understood that the above-mentioned material detection device corresponds one-to-one with the above-mentioned material detection method, and will not be described in detail here.

[0075] In some embodiments of this application, please refer to Fig. 9 , Fig. 9 This is a basic structural block diagram of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device 9 of this embodiment includes: a processor 91, a memory 92, and a computer program 93 stored in the memory 92 and executable on the processor 91, such as a program for a material detection method. When the processor 91 executes the computer program 93, the steps in each embodiment of the above-mentioned material detection method are implemented. Alternatively, when the processor 91 executes the computer program 93, the functions of each module in the embodiment corresponding to the above-mentioned material detection device are implemented. Please refer to the relevant description in the embodiment for details, which will not be repeated here.

[0076] Exemplarily, the computer program 93 may be divided into one or more modules (units), which are stored in the memory 92 and executed by the processor 91 to complete the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 93 in the electronic device 9. For example, the computer program 93 may be divided into:

[0077] An extraction module is used to obtain terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain the scanning point index corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected;

[0078] A first clustering module, used for performing feature information clustering processing on each effective pulse feature in the effective pulse feature set using a preset first self-organizing map network to generate a first cluster type set;

[0079] A second clustering module, configured to perform pulse type clustering processing on each scanning point in the scanning point index set using a preset second self-organizing map network according to the first cluster type set, so as to generate a second cluster type set;

[0080] An imaging module is used to perform imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected.

[0081] The electronic device may include, but is not limited to, a processor 91 and a memory 92. Those skilled in the art will appreciate that Fig. 9 It is only an example of the electronic device 9 and does not constitute a limitation of the electronic device 9. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0082] The processor 91 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0083] The memory 92 may be an internal storage unit of the electronic device 9, such as a hard disk or memory of the electronic device 9. The memory 92 may also be an external storage device of the electronic device 9, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 9. Further, the memory 92 may also include both an internal storage unit of the electronic device 9 and an external storage device. The memory 92 is used to store the computer program and other programs and data required by the electronic device. The memory 92 may also be used to temporarily store data that has been output or is to be output.

[0084] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0085] The embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented. In this embodiment, the computer-readable storage medium can be non-volatile or volatile.

[0086] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0087] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0088] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0089] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0090] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A material detection method, characterized in that: include: Acquire terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain scanning point indexes corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected; Using a preset first self-organizing map network to perform feature information clustering processing on each effective pulse feature in the effective pulse feature set to generate a first cluster type set; According to the first cluster type set, a preset second self-organizing map network is used to perform pulse type clustering processing on each scanning point in the scanning point index set to generate a second cluster type set; According to the second cluster type set, imaging processing is performed on the material to be detected to generate an imaging result map of the material to be detected; The step of performing pulse type clustering processing on each scanning point in the scanning point index set using a preset second self-organizing map network according to the first cluster type set to generate a second cluster type set comprises: According to the first cluster type set and the scanning point index set, encoding processing is performed on the terahertz waveform information of each scanning point in the scanning point index set to generate a coding matrix, and the coding matrix is ​​input as an input vector into the second self-organizing map network to perform pulse type clustering processing to generate a second cluster type set; The step of performing imaging processing on the material to be detected according to the second cluster type set to generate an imaging result map of the material to be detected comprises: According to the second cluster type set, the pulse type corresponding to each scanning point is converted into an image pixel value, and the image pixel value corresponding to each scanning point is filled into the image matrix of the material to be detected according to the scanning point to generate an imaging result map of the material to be detected.

2. The material detection method according to claim 1, characterized in that: The step of obtaining terahertz reflection waveform data of the material to be detected also includes: Using a terahertz imaging device to perform terahertz wave scanning on the material to be detected to generate bipolar pulse data of the material to be detected; Deconvolution processing is performed on the bipolar pulse data to generate corresponding unipolar pulse data, and the unipolar pulse data is acquired as terahertz reflection waveform data of the material to be detected.

3. The material detection method according to claim 1 or 2, characterized in that: The step of extracting effective pulse features from the terahertz reflection waveform data comprises: Extracting a pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and comparing the pulse peak value corresponding to each scanning point with a preset peak value threshold value, and if the pulse peak value is greater than the preset peak value threshold value, determining the pulse peak value as a valid pulse feature; All pulse peaks determined as effective pulse features are collected to generate an effective pulse peak set, and based on the effective pulse peak set, the peak time corresponding to each pulse peak in the effective pulse peak set is extracted from the terahertz reflection waveform data, and the peak times corresponding to each pulse peak are collected to generate an effective pulse peak time set.

4. A material detection device, characterized in that: The material detection device comprises: An extraction module is used to obtain terahertz reflection waveform data of the material to be detected, extract effective pulse features from the terahertz reflection waveform data and obtain the scanning point index corresponding to the effective pulse features, and generate an effective pulse feature set and a scanning point index set of the material to be detected; A first clustering module, used for performing feature information clustering processing on each effective pulse feature in the effective pulse feature set using a preset first self-organizing map network to generate a first cluster type set; A second clustering module, configured to perform pulse type clustering processing on each scanning point in the scanning point index set using a preset second self-organizing map network according to the first cluster type set, so as to generate a second cluster type set; An imaging module, used for performing imaging processing on the material to be detected according to the second cluster type set, and generating an imaging result map of the material to be detected; The second clustering module is further used to encode the terahertz waveform information of each scanning point in the scanning point index set according to the first cluster type set and the scanning point index set to generate a coding matrix, and input the coding matrix as an input vector into the second self-organizing map network to perform pulse type clustering processing to generate a second cluster type set; The imaging module is also used to convert the pulse type corresponding to each scanning point into an image pixel value according to the second cluster type set, and fill the image pixel value corresponding to each scanning point into the image matrix of the material to be detected according to the scanning point, so as to generate an imaging result map of the material to be detected.

5. The material detection device according to claim 4, characterized in that: The material detection device also includes: A scanning submodule, used for performing terahertz wave scanning on the material to be detected by using a terahertz imaging device to generate bipolar pulse data of the material to be detected; The deconvolution submodule is used to perform deconvolution processing on the bipolar pulse data to generate corresponding unipolar pulse data, and obtain the unipolar pulse data as terahertz reflection waveform data of the material to be detected.

6. The material detection device according to claim 4 or 5, characterized in that: The material detection device also includes: A comparison submodule, used for extracting a pulse peak value corresponding to each scanning point from the terahertz reflection waveform data, and comparing the pulse peak value corresponding to each scanning point with a preset peak value threshold value, and if the pulse peak value is greater than the preset peak value threshold value, determining the pulse peak value as a valid pulse feature; The extraction submodule is used to collect all pulse peak values ​​determined as effective pulse characteristics to generate an effective pulse peak value set, and according to the effective pulse peak value set, extract the peak time corresponding to each pulse peak value in the effective pulse peak value set from the terahertz reflection waveform data, collect the peak times corresponding to each pulse peak value, and generate an effective pulse peak time set.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images