CN109727295A - Electromagnetic image extracting method, device, computer equipment and storage medium - Google Patents
Electromagnetic image extracting method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109727295A CN109727295A CN201811542900.3A CN201811542900A CN109727295A CN 109727295 A CN109727295 A CN 109727295A CN 201811542900 A CN201811542900 A CN 201811542900A CN 109727295 A CN109727295 A CN 109727295A
- Authority
- CN
- China
- Prior art keywords
- frequency point
- determinand
- value matrix
- electromagnetic
- transmitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Processing (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
This application involves a kind of electromagnetic image extracting method, device, computer equipment and storage mediums.The electromagnetic field raw image data that determinand surface obtains is scanned by near-field surface the described method includes: obtaining, electromagnetic field raw image data is that near-field surface scans transmitted value of the determinand surface under each frequency point；By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to the transmitting value matrix of each frequency point respectively according to frequency point；Standard deviation is asked respectively to the transmitting value matrix of each frequency point；It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand in the transmitting value matrix of specific frequency point；The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.It is automatically extracted using the raw image data that this method can obtain scanning, without checking that specific frequency point extracts manually.
Description
Technical field
This application involves near-field surface scanning technique fields, more particularly to a kind of electromagnetic image extracting method, device, meter
Calculate machine equipment and storage medium.
Background technique
Near-field surface scanning is currently used for carrying out one kind of electromagnetic compatibility test to plate electrode product and device level product often
With technology, test method is the magnetic distribution obtained in the near-field region of determinand surface by emf probe, be can be used for
Carry out electromagnetic interference positioning and evaluation etc..
Near-field surface sweep test generally pass through frequency spectrograph in certain regional scope stationary point collect determinand not
With the electromagnetic emission signal under frequency point, until all scanning elements are gone through time, but collected signal is in addition to required for test
It further include ambient noise outside the intentional electromagnetic emission signal of determinand, it is therefore desirable to by the intentional electromagnetism of required determinand
Transmitting signal extraction comes out.
Since the Electromagnetic Launching of determinand only occurs on specific frequency point, other frequency points are ambient noise signal, tradition
Method be that these specific frequency points are found by way of checking manually, extract the data under these specific frequency points.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of electromagnetic image extracting method, device, computer and set
Standby and storage medium.
A kind of electromagnetic image extracting method, method include:
It obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to each frequency point respectively according to frequency point
Transmitting value matrix；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
It is screened in one of the embodiments, according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Include:
From the transmitting value matrix of each frequency point, the transmitting value matrix that standard deviation is greater than standard deviation threshold method is filtered out.
Electromagnetic image extracting method in one of the embodiments, further include:
Transmitting value matrix to determinand in each specific frequency point carries out dimension-reduction treatment respectively；
Transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point clusters, and obtains determinand electromagnetic image
Cluster result.
The transmitting value matrix to determinand in each specific frequency point carries out dimension-reduction treatment respectively in one of the embodiments,
Include:
Transmitting value matrix by determinand in all specific frequency points is reassembled as input matrix；
Morphological segment processing is carried out to input matrix, obtains binaryzation matrix；
Dimension-reduction treatment is carried out to binaryzation matrix.
Carrying out dimension-reduction treatment to the transmitting value matrix of the specific frequency point of determinand in one of the embodiments, includes:
Transmitting value matrix using Principal Component Analysis to determinand in each specific frequency point carries out dimension-reduction treatment.
Carrying out cluster to the transmitting value matrix of the specific frequency point of determinand after dimensionality reduction in one of the embodiments, includes:
Transmitting value matrix using K-Means clustering procedure to determinand in each specific frequency point clusters.
Electromagnetic image extracting method in one of the embodiments, further include:
Electromagnetism test model is established using the cluster result of the electromagnetic image of determinand.
A kind of electromagnetic image extraction element, which is characterized in that device includes:
Raw image data obtains module, scans the electromagnetic field original that determinand surface obtains by near-field surface for obtaining
Beginning image data, electromagnetic field raw image data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
Data division module, for being stored according to frequency point by the transmitted value of frequency point each in electromagnetic field raw image data
For the transmitting value matrix of corresponding frequency point；
Standard deviation processing module seeks standard deviation for the transmitting value matrix to each frequency point respectively；
Screening module is extracted for being screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Transmitting value matrix of the determinand in specific frequency point out；
Image generation module, for generating the magnetic chart of determinand in the transmitting value matrix of specific frequency point according to determinand
Picture.
A kind of computer equipment, including memory and processor, memory are stored with computer program, and processor executes meter
It is performed the steps of when calculation machine program
It obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to each frequency point respectively according to frequency point
Transmitting value matrix；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
A kind of computer readable storage medium is stored thereon with computer program, when computer program is executed by processor
It performs the steps of
It obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to each frequency point respectively according to frequency point
Transmitting value matrix；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
Above-mentioned electromagnetic image extracting method, device, computer equipment and storage medium, by the way that near field surface scan is to be measured
The electromagnetic field raw image data that object surface obtains is recombinated according to frequency, the transmitted value square of each frequency point in counterweight group data set
Battle array seeks standard deviation respectively, due to the electromagnetic image that intentional electromagnetic emission signal is regular pattern, the transmitting Distribution value of pixel compared with
At random, standard deviation is larger, and ambient noise is the electromagnetic image of pattern-free, and the transmitting Distribution value of pixel is relatively concentrated, standard deviation phase
It is smaller, therefore screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, ambient noise can be removed,
Determinand is extracted in the intentional electromagnetic emission signal image data of specific frequency point, the raw image data that can be obtained to scanning
It is automatically extracted, without checking that specific frequency point extracts manually.
Detailed description of the invention
Fig. 1 is the flow diagram of electromagnetic image extracting method in one embodiment；
Fig. 2 is the flow diagram of electromagnetic image extracting method in another embodiment；
Fig. 3 is the flow diagram of electromagnetic image extracting method dimension-reduction treatment method in one embodiment；
Fig. 4 is the structural block diagram of electromagnetic image extraction element in one embodiment；
Fig. 5 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
In one embodiment, as shown in Figure 1, providing a kind of electromagnetic image extracting method, including step 110~150:
Step 110, it obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field is scanned by near-field surface
Raw image data is that near-field surface scans transmitted value of the determinand surface under each frequency point.
Near-field surface scanning be by carrying out electromagnetic field near-field scan to PCB circuit board or electronic component, with obtain to
Survey the magnetic distribution situation of object.
Specifically, determinand is scanned using near-field surface scanner, obtains the scanning of near-field surface scanner
The electromagnetic signal emitting value of each frequency point of the obtained determinand surface in certain frequency point ranges.
Step 120, it by the transmitted value of frequency point each in electromagnetic field raw image data, is stored as corresponding to respectively according to frequency point
The transmitting value matrix of each frequency point.
Due to scanning the value that the transmitted value directly obtained is single pixel, handles for the ease of carrying out and finally can
The electromagnetic image that is distributed according to frequency point is obtained, the transmitted value of all acquisitions is divided according to frequency point, and by each frequency point pair
The whole transmitted value answered are stored as the transmitting value matrix of the frequency point.
The transmitted value of each single frequency point is stored as matrix Xm,
Wherein, XmRefer to the transmitting value matrix scanned under m-th of frequency point, xijFor the electromagnetism of the i-th row jth column pixel
Signal transmitted value.
Step 120, standard deviation is asked respectively to the transmitting value matrix of each frequency point.
Intentional electromagnetic emission signal distribution of the determinand under specific frequency point is more at random, and dispersion degree is high, therefore standard deviation
It is larger, and the transmitting signal distributions of ambient noise are more concentrated, dispersion degree is lower, therefore standard deviation is smaller, can be by pre-
If standard deviation threshold method screened, allow to carry out extraction process screening when data set become smaller, only need to be to each transmitted value square
The standard deviation of battle array is screened, and is screened without the data to each transmitting value matrix itself, is shortened the time of extraction.
Specifically, if the transmitted value standard deviation of m-th of frequency point is σm,
Wherein, μmFor the mean value of the transmitted value of all pixels point of m-th of frequency point, n is the sum of pixel, i.e. n=i ×
j。
Step 140, it is screened, is extracted to be measured according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Transmitting value matrix of the object in specific frequency point.
The standard deviation of transmitted value under all frequencies is compared with preset standard deviation threshold method, standard deviation is less than threshold value
Transmitting value matrix belong to background noise, standard deviation reach threshold value transmitting value matrix be the specific frequency point of determinand transmitted value
Matrix, the as corresponding data of intentional electromagnetic emission signal of determinand.
Step 150, the electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
The electromagnetic image of determinand can be generated according to the transmitting value matrix extracted, the electromagnetic image of determinand can be straight
The magnetic distribution situation of observation determinand.
In above-mentioned electromagnetic image extracting method, pass through the electromagnetic field original graph for obtaining near field surface scan determinand surface
As data, electromagnetic field raw image data is stored as transmitting value matrix to the transmitting value matrix of each frequency point according to frequency point respectively
Standard deviation is sought respectively, and due to the electromagnetic image that intentional electromagnetic emission signal is regular pattern, the transmitting Distribution value of pixel relatively dissipates
Disorderly, standard deviation is larger, and ambient noise is the electromagnetic image of pattern-free, and the transmitting Distribution value of pixel is relatively concentrated, and standard deviation is compared
It is small, therefore screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, ambient noise can be removed, is mentioned
Take out determinand specific frequency point intentional electromagnetic emission signal image data, can to scanning obtain raw image data into
Row automatically extracts, without checking that specific frequency point extracts manually.
It is screened in one of the embodiments, according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Step includes:
From the transmitting value matrix of each frequency point, the transmitting value matrix that standard deviation is greater than standard deviation threshold method is filtered out.
The standard deviation of the transmitting value matrix of each frequency point is compared with preset standard deviation threshold method, standard deviation is greater than mark
The transmitting value matrix of quasi- difference threshold value, corresponding frequency point is specific frequency point, and transmitted value is determinand under specific frequency point
Intentional electromagnetic emission signal value, therefore filter out this part transmitting value matrix, standard deviation be less than standard deviation threshold method transmitted value
Transmitted value in matrix is the electromagnetic emission signal of ambient noise.
In one of the embodiments, as shown in Fig. 2, electromagnetic image extracting method further includes step 160~170:
Step 160, the transmitting value matrix to determinand in each specific frequency point carries out dimension-reduction treatment respectively.
First the determinand extracted is subjected to dimensionality reduction in the transmitting value matrix of each specific frequency point respectively, reduces cluster and need
The data volume of processing improves treatment effeciency, moreover it is possible to which reducing interference reduces error, improves the accuracy rate of cluster.
Step 170, the transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point clusters, and obtains determinand
The cluster result of electromagnetic image.
With the presence or absence of connection between the electromagnetic image extracted at present for analysis, processing method is had no, it is if desired sharp
Modeling test is carried out with electromagnetic image, accurate mould could be established by needing to carry out modeling using the data of all electromagnetic images
Type, workload is very big, higher for the time investment tested.
Cluster can analyze the similitude between determinand electromagnetic image, can be by the higher electricity of similarity by cluster
Magnetic image is divided into one kind, only need to extract one to two electromagnetic images from one kind in subsequent test or assessment and be divided
Analysis, the data that can be obtained the characteristic for representing this kind of electromagnetic images are assessed or are modeled.
Clustering method can use K-Means clustering procedure, mean shift clustering method, based on density in one embodiment
Clustering method (DBSCAN) is rolled into a ball with greatest hope (EM) clustering procedure of gauss hybrid models (GMM), Agglomerative Hierarchical Clustering method or figure
Body detection method (Graph Community Detection) etc., those skilled in the art, which can according to need, to be selected.
In one of the embodiments, as shown in figure 3, to determinand each specific frequency point transmitting value matrix respectively into
Row dimension-reduction treatment includes step 161~163:
Step 161, the transmitting value matrix by determinand in each specific frequency point is reassembled as input matrix；
Specifically, if determinand m-th of frequency point transmitting value matrix be Ym, by YmExpand into a vector Am, then:
Wherein, yijFor the determinand electromagnetic image that extracts the i-th row jth column pixel transmitted value.
The transposition for the vector that transmitting value matrix under each specific frequency point extracted is unfolded is reassembled as input square
Battle array E, input matrix contain the transmitting value matrix under m frequency point, then:
Step 162, it treats input matrix and carries out morphological segment processing, obtain binaryzation matrix；
Step 163, dimension-reduction treatment is carried out to binaryzation matrix.
Morphological segment is the profile in order to protrude determinand electromagnetic image, specifically, by the drafting department in electromagnetic image
Point it is converted by morphological segment algorithm completely black, remainder Quan Bai, use 1 indicates black, and 0 expression is white, and two-value can be obtained
Change matrix, the corresponding image generated of binaryzation matrix enables to Euclidean distance of the point of identical patterns in feature space to become
Smaller, Euclidean distance of the point of different pattern in feature space becomes much larger, so that being more convenient for clustering processing.
The transmitting value matrix to the specific frequency point of determinand or binaryzation matrix carry out at dimensionality reduction in one of the embodiments,
Reason can be using Principal Component Analysis (PCA), t- distribution neighborhood embedded mobile GIS (t-SNE), multidimensional scaling (MDS) or SVD square
The methods of battle array decomposition method, those skilled in the art, which can according to need, to be selected.
Wherein, Principal Component Analysis (PCA) is a kind of common dimensionality reduction statistical method, may by one group using orthogonal transformation
Relevant variable input data is converted to one group of value for being referred to as the linear uncorrelated variables of principal component.Since PCA is only simple
Input data is converted, so it can both be used in classification problem, regression problem can also be used in.Nonlinear situation can
To use kernel method kernelized PCA, but since PCA has the speed of feature space after good mathematical property, discovery conversion
The ability mutually converted between feature after degree and original again and transformation, in dimensionality reduction feature extraction in other words, PCA can be with
Meet most of situation.Those skilled in the art, which can according to need, to be selected
Given luv space, PCA can find the Linear Mapping for arriving more low dimensional space.Since it is desired that making all samples
Projection it is as separated as possible, then need to maximize the variance of subpoint.The variance that PCA retains is the largest, and final reconstruct
The error of (original case is returned to after transformation) is the smallest.
Specific processing method are as follows:
In order to measure the dispersion degree between each dimension of features vector (each pixel as in electromagnetic image), calculate
Covariance between two dimensions A, B is for example shown below:
Enable cpq=cov (p, q) represents the pth row of covariance matrix C and q is arranged, and covariance matrix is defined as follows:
According to the definition of feature vector:
WhereinRefer to the vector of the different λ of n, λ is the corresponding characteristic value of n feature vector, and this n feature to
The eigenvectors matrix V for measuring composition is for example shown below:
V=(v1 v2 …vn)
For single feature vector vi, i meet 1≤i≤n and be an integer.Therefore, above-mentioned definition can be transformed into:
C×vi=λ vi
N such feature vector equations are solved, it is available to contain n different feature value vectorsAnd n spy
Levy the matrix V of vector composition.Since feature vector is all unitization (i.e. length is 1), and it is mutually orthogonal, therefore these are special
Sign vector can be used as to dimensionality reduction data (the transmitting value matrix of the specific frequency point of the determinand extracted, the i.e. electricity of determinand
Magnetic image) substrate.Obtain C V andLater, the transmitting value matrix of the specific frequency point of the determinand extracted can be dropped
Dimension.By the feature vector v in Vi(i.e. each column) press individual features value λiDescending arrangement, k principal component is as extraction before then taking
The transmitted value matrix stack E of the specific frequency point of determinand out1The basis matrix V ' that can be projected is as follows:
V '=(v '1 v′2… v′k)
Wherein v '1、v′2……v′kIt refers to the preceding k feature vector after arranging by characteristic value descending, while being also former square
The principal component of battle array C.By and large, the principal component in V ' contains the main information of the covariance matrix C of script, then can be with
Use V ' Lai Zhidai C.
The final step of PCA is that the input matrix E after equalization is multiplied with basis matrix V ' after projection, that is, will be equal
Input matrix after value is projected as in principal component and the substrate chosen, that is, has following projection matrix:
P=V 'T×Emean-adjusted T
It can derive:
Wherein fm kSubscript m refer to the element be located at projection matrix P m column, represent the column be transmitting value matrix
Concentrate the data of m-th of frequency；fm kSubscript k refer to that the element is located at the row k of projection matrix P, it is main for represent the row
Ingredient projection, that is, a data dimension remaining after dimensionality reduction.
By above-mentioned processing, data are down to k dimension from the dimension of script, and too many information will not be lost, when guaranteeing reconstruct
Error is minimum.
Each column of projection matrix P are clustered after dimensionality reduction, the relatively high classification knot of an accuracy rate can be obtained
Fruit.
It is poly- using K-Means to the transmitting value matrix of the specific frequency point of determinand after dimensionality reduction in one of the embodiments,
Class method is clustered.
In another embodiment, the binaryzation matrix after dimensionality reduction is clustered using K-Means clustering procedure.
K-Means clustering procedure has the characteristics that calculating is easy, can effectively improve calculating speed.
Specific sorting procedure are as follows:
Step 171, using every a line of projection matrix P as the frequency point in feature space, k mass center of random initializtion is (i.e.
Central point).
Step 172, each frequency point is calculated to the distance of each mass center, and it is most short that each frequency point is referred to gained distance respectively
Mass center.
Step 173, for each mass center, the mean value for being referred to all frequency points of the mass center is calculated separately, and will be resulting
Mean value is updated to new mass center.
Step 172~173 are repeated, until variance within clusters converge to a minimum value.
Variance within clusters are as follows:
Wherein, pjFor frequency point, ciFor class representated by corresponding mass center i, (1,2 ... k), μ by i ∈iFor mass center.
Electromagnetic image extracting method in one of the embodiments, further include:
Step 180, electromagnetism test model is established using the cluster result of the electromagnetic image of determinand.
Electromagnetism test model is established, if there are z classes for cluster result, a part of electromagnetic image data are extracted from each class
For modeling to such product, there is specific aim, and reduce workload.
It should be understood that although each step in the flow chart of Fig. 1-3 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 1-3
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 4, providing a kind of electromagnetic image extraction element, comprising: raw image data
Module 210, data division module 220, standard deviation processing module 230, screening module 240 and image generation module 250 are obtained,
In:
Raw image data obtains module 210, scans the electromagnetism that determinand surface obtains by near-field surface for obtaining
Field raw image data, electromagnetic field raw image data are that near-field surface scans transmitting of the determinand surface under each frequency point
Value；
Data division module 220, for being deposited according to frequency point by the transmitted value of frequency point each in electromagnetic field raw image data
Storage is the transmitting value matrix of corresponding frequency point；
Standard deviation processing module 230 seeks standard deviation for the transmitting value matrix to each frequency point respectively；
Screening module 240 is mentioned for being screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Determinand is taken out in the transmitting value matrix of specific frequency point；
Image generation module 250, for generating the electromagnetism of determinand in the transmitting value matrix of specific frequency point according to determinand
Image.
Electromagnetic image extraction element in one of the embodiments, further include: dimension-reduction treatment module and cluster module,
In:
Dimensionality reduction module carries out dimension-reduction treatment for the transmitting value matrix to determinand in each specific frequency point respectively；
Cluster module is clustered for the transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point, is obtained
The cluster result of determinand electromagnetic image.
Dimension-reduction treatment module includes: recombination module, binarization block and dimensionality reduction operation mould in one of the embodiments,
Block, in which:
Recombination module is reassembled as input matrix for the transmitting value matrix by determinand in each specific frequency point；
Binarization block carries out morphological segment processing for the transmitting value matrix to the specific frequency point of determinand, obtains two
Value matrix；
Dimensionality reduction computing module, for carrying out dimension-reduction treatment to binaryzation matrix.
Electromagnetic image extraction element in one of the embodiments, further include:
Modeling module, the cluster result for the electromagnetic image using the determinand establish electromagnetism test model.
Specific about electromagnetic image extraction element limits the limit that may refer to above for electromagnetic image extracting method
Fixed, details are not described herein.Modules in above-mentioned electromagnetic image extraction element can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 5.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The network interface of machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor with
Realize a kind of electromagnetic image extracting method.
It will be understood by those skilled in the art that structure shown in Fig. 5, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, is stored in memory
Computer program, the processor perform the steps of when executing computer program
It obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to each frequency point respectively according to frequency point
Transmitting value matrix；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
In one embodiment, it is also performed the steps of when processor executes computer program
Transmitting value matrix to determinand in each specific frequency point carries out dimension-reduction treatment respectively；
Transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point clusters, and obtains determinand electromagnetic image
Cluster result.
In one embodiment, it is also performed the steps of when processor executes computer program
Transmitting value matrix by determinand in each specific frequency point is reassembled as input matrix；
Morphological segment processing is carried out to input matrix, obtains binaryzation matrix；
Dimension-reduction treatment is carried out to binaryzation matrix.
In one embodiment, it is also performed the steps of when processor executes computer program
Electromagnetism test model is established using the cluster result of the electromagnetic image of determinand.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
It obtains and the electromagnetic field raw image data that determinand surface obtains, electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans transmitted value of the determinand surface under each frequency point；
By the transmitted value of frequency point each in electromagnetic field raw image data, it is stored as corresponding to each frequency point respectively according to frequency point
Transmitting value matrix；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of determinand is generated in the transmitting value matrix of specific frequency point according to determinand.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Transmitting value matrix to determinand in each specific frequency point carries out dimension-reduction treatment respectively；
Transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point clusters, and obtains determinand electromagnetic image
Cluster result.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Transmitting value matrix by determinand in each specific frequency point is reassembled as input matrix；
Morphological segment processing is carried out to input matrix, obtains binaryzation matrix；
Dimension-reduction treatment is carried out to binaryzation matrix.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Electromagnetism test model is established using the cluster result of the electromagnetic image of determinand.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of electromagnetic image extracting method, which is characterized in that the described method includes:
It obtains and the electromagnetic field raw image data that determinand surface obtains, the electromagnetic field original image is scanned by near-field surface
Data are that near-field surface scans electromagnetic signal emitting value of the determinand surface under each frequency point；
By the electromagnetic signal emitting value of each frequency point in the electromagnetic field raw image data, it is stored as corresponding to respectively according to frequency point
The transmitting value matrix of each frequency point；
Standard deviation is asked respectively to the transmitting value matrix of each frequency point；
It is screened according to transmitting value matrix of the preset standard deviation threshold method to each frequency point, extracts the determinand specific
The transmitting value matrix of frequency point；
The electromagnetic image of the determinand is generated in the transmitting value matrix of specific frequency point according to the determinand.
2. electromagnetic image extracting method according to claim 1, which is characterized in that according to preset standard deviation threshold method to each
The transmitting value matrix of a frequency point carries out screening
From the transmitting value matrix of each frequency point, the transmitting value matrix that standard deviation is greater than the standard deviation threshold method is filtered out.
3. electromagnetic image extracting method according to claim 1, which is characterized in that further include:
Transmitting value matrix to the determinand in each specific frequency point carries out dimension-reduction treatment respectively；
Transmitting value matrix to the determinand after dimensionality reduction in each specific frequency point clusters, and obtains the determinand electromagnetism
The cluster result of image.
4. electromagnetic image extracting method according to claim 3, which is characterized in that the determinand in each specific frequency
The transmitting value matrix of point carries out dimension-reduction treatment respectively and includes:
Transmitting value matrix by the determinand in each specific frequency point is reassembled as input matrix；
Morphological segment processing is carried out to the input matrix, obtains binaryzation matrix；
Dimension-reduction treatment is carried out to the binaryzation matrix.
5. electromagnetic image extracting method according to claim 3, which is characterized in that the hair of the specific frequency point of the determinand
Penetrating value matrix progress dimension-reduction treatment includes:
Transmitting value matrix using Principal Component Analysis to the determinand in each specific frequency point carries out dimension-reduction treatment.
6. electromagnetic image extracting method according to claim 3, which is characterized in that specific to the determinand after dimensionality reduction
The transmitting value matrix of frequency point carries out cluster
Transmitting value matrix using K-Means clustering procedure to the determinand in each specific frequency point clusters.
7. electromagnetic image extracting method according to claim 6, which is characterized in that further include:
Electromagnetism test model is established using the cluster result of the electromagnetic image of the determinand.
8. a kind of electromagnetic image extraction element, which is characterized in that described device includes:
Raw image data obtains module, scans the electromagnetic field original graph that determinand surface obtains by near-field surface for obtaining
As data, the electromagnetic field raw image data is that near-field surface scans transmitting of the determinand surface under each frequency point
Value；
Data matrix module, for being deposited according to frequency point by the transmitted value of each frequency point in the electromagnetic field raw image data
Storage is the transmitting value matrix of corresponding frequency point；
Standard deviation processing module seeks standard deviation for the transmitting value matrix to each frequency point respectively；
Screening module extracts institute for screening according to transmitting value matrix of the preset standard deviation threshold method to each frequency point
Determinand is stated in the transmitting value matrix of specific frequency point；
Image generation module, for generating the electromagnetism of the determinand in the transmitting value matrix of specific frequency point according to the determinand
Image.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 7 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811542900.3A CN109727295B (en) | 2018-12-17 | 2018-12-17 | Electromagnetic image extraction method, electromagnetic image extraction device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811542900.3A CN109727295B (en) | 2018-12-17 | 2018-12-17 | Electromagnetic image extraction method, electromagnetic image extraction device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109727295A true CN109727295A (en) | 2019-05-07 |
| CN109727295B CN109727295B (en) | 2021-06-29 |
Family
ID=66297650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811542900.3A Active CN109727295B (en) | 2018-12-17 | 2018-12-17 | Electromagnetic image extraction method, electromagnetic image extraction device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109727295B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561821A (en) * | 2020-12-17 | 2021-03-26 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Chip surface electromagnetic data noise reduction method based on near field scanning |
| CN112749668A (en) * | 2021-01-18 | 2021-05-04 | 上海明略人工智能(集团)有限公司 | Target image clustering method and device, electronic equipment and computer readable medium |
| CN113781332A (en) * | 2021-08-24 | 2021-12-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Method and device for reducing noise of electromagnetic data, computer equipment and storage medium |
| CN112561821B (en) * | 2020-12-17 | 2024-05-17 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Near field scanning-based chip surface electromagnetic data noise reduction method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030089899A1 (en) * | 2000-08-22 | 2003-05-15 | Lieber Charles M. | Nanoscale wires and related devices |
| CN104584326A (en) * | 2012-05-09 | 2015-04-29 | 杜克大学 | Metamaterial devices and methods of using the same |
| CN105026970A (en) * | 2013-01-08 | 2015-11-04 | 麻省理工学院 | Optical phased arrays |
-
2018
- 2018-12-17 CN CN201811542900.3A patent/CN109727295B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030089899A1 (en) * | 2000-08-22 | 2003-05-15 | Lieber Charles M. | Nanoscale wires and related devices |
| CN104584326A (en) * | 2012-05-09 | 2015-04-29 | 杜克大学 | Metamaterial devices and methods of using the same |
| CN105026970A (en) * | 2013-01-08 | 2015-11-04 | 麻省理工学院 | Optical phased arrays |
Non-Patent Citations (2)
| Title |
|---|
| 宋喆，吴仕飞，李卫东，张彦: "一种新的自适应宽带扫频方法及其在分层微带结构全波分析中的应用", 《中国科学》 * |
| 鲁文波，蒋伟康，潘思伟，向上: "基于近场声全息声像图纹理特征的机械故障诊断方法", 《振动工程学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561821A (en) * | 2020-12-17 | 2021-03-26 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Chip surface electromagnetic data noise reduction method based on near field scanning |
| CN112561821B (en) * | 2020-12-17 | 2024-05-17 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Near field scanning-based chip surface electromagnetic data noise reduction method |
| CN112749668A (en) * | 2021-01-18 | 2021-05-04 | 上海明略人工智能(集团)有限公司 | Target image clustering method and device, electronic equipment and computer readable medium |
| CN113781332A (en) * | 2021-08-24 | 2021-12-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Method and device for reducing noise of electromagnetic data, computer equipment and storage medium |
| CN113781332B (en) * | 2021-08-24 | 2023-09-29 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Noise reduction method and device for electromagnetic data, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109727295B (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bengua et al. | Efficient tensor completion for color image and video recovery: Low-rank tensor train | |
| Riaz et al. | Texture classification using rotation-and scale-invariant gabor texture features | |
| Jia et al. | Gabor feature-based collaborative representation for hyperspectral imagery classification | |
| Chen et al. | Hyperspectral image classification using dictionary-based sparse representation | |
| Dev et al. | Categorization of cloud image patches using an improved texton-based approach | |
| Song et al. | Unsupervised classification of polarimetric SAR imagery using large-scale spectral clustering with spatial constraints | |
| CN104866524A (en) | Fine classification method for commodity images | |
| WO2022078191A1 (en) | Method and apparatus for identifying device type, computer device, and storage medium | |
| Zhang et al. | Maximum mean and covariance discrepancy for unsupervised domain adaptation | |
| Sun et al. | Orthogonal tubal rank-1 tensor pursuit for tensor completion | |
| Xu et al. | Discriminative analysis for symmetric positive definite matrices on lie groups | |
| CN105868711B (en) | Sparse low-rank-based human behavior identification method | |
| CN109727295A (en) | Electromagnetic image extracting method, device, computer equipment and storage medium | |
| Li et al. | Breast cancer histopathological image classification based on deep second-order pooling network | |
| Deng et al. | Citrus disease recognition based on weighted scalable vocabulary tree | |
| CN109784233B (en) | Phase-locked value weighted spatial filtering method and device based on Lp-norm | |
| CN111680579A (en) | Remote sensing image classification method for adaptive weight multi-view metric learning | |
| CN114004998B (en) | Unsupervised polarization SAR image ground object classification method based on multi-vision tensor product diffusion | |
| Le et al. | Exploring the asynchronous of the frequency spectra of gan-generated facial images | |
| CN106056131A (en) | Image feature extraction method based on LRR-LDA | |
| Fursov et al. | Thematic classification with support subspaces in hyperspectral images | |
| Zhang et al. | Group-wise learning for aurora image classification with multiple representations | |
| Li et al. | Outlier identification via randomized adaptive compressive sampling | |
| Meng et al. | Robust discriminant projection via joint margin and locality structure preservation | |
| Frohnert et al. | Explainable representation learning of small quantum states |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 511300 No.78, west of Zhucun Avenue, Zhucun street, Zengcheng District, Guangzhou City, Guangdong Province Applicant after: CHINA ELECTRONIC PRODUCT RELIABILITY AND ENVIRONMENTAL TESTING RESEARCH INSTITUTE ((THE FIFTH ELECTRONIC RESEARCH INSTITUTE OF MIIT)(CEPREI LABORATORY)) Address before: 510610 No. 110 Zhuang Road, Tianhe District, Guangdong, Guangzhou, Dongguan Applicant before: CHINA ELECTRONIC PRODUCT RELIABILITY AND ENVIRONMENTAL TESTING RESEARCH INSTITUTE ((THE FIFTH ELECTRONIC RESEARCH INSTITUTE OF MIIT)(CEPREI LABORATORY)) |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |