CN110458228A - A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation - Google Patents
A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation Download PDFInfo
- Publication number
- CN110458228A CN110458228A CN201910732627.9A CN201910732627A CN110458228A CN 110458228 A CN110458228 A CN 110458228A CN 201910732627 A CN201910732627 A CN 201910732627A CN 110458228 A CN110458228 A CN 110458228A
- Authority
- CN
- China
- Prior art keywords
- signal
- scattering parameter
- matrix
- source
- unknown
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2134—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/05—Recognition of patterns representing particular kinds of hidden objects, e.g. weapons, explosives, drugs
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a kind of hazardous material detection methods separated based on information source number time-varying and self-adaptive blind source, comprising the following steps: S1. standard database is built;S2. scattering parameter model construction;S3. dynamic Sources number estimation;S4. self-adaptive blind source separates;S5. dangerous material identify.The present invention is based on information source number time-varying and self-adaptive blind source to separate, and realizes effective detection to liquid dangerous material, and can detect to multiple and different liquid form products, saves the time that the public crosses safety check, improves the safety check efficiency of public place;Meanwhile Sources number estimation is first carried out, blind source separating, precision with higher are carried out further according to estimated result, the later period only needs the S parameter for increasing sample newly in the database that the extension of detection classification can be realized, has good scalability.
Description
Technical field
The present invention relates to hazardous material detections, more particularly to a kind of danger separated based on information source number time-varying and self-adaptive blind source
Dangerous product detection method.
Background technique
With the development of national cause, country increasingly payes attention to public safety, puts into terms of public safety guarantee
A large amount of manpower, material resources and financial resources.Nevertheless, public place safety accident happens occasionally.Therefore public place safety inspection seems
It is particularly important.Common detecting instrument of the X ray tester as the link of public place safety check at this stage not only greatly improves case effect
Rate also reduces the frequency of safety accident.But for liquid dangerous material, check that work is relatively cumbersome and liquid
Dangerous material are easy leakage, are all the problem in safety check link all the time.So if can propose one kind to liquid hazardous material detection
Safe and feasible and efficient method, this undoubtedly checks that work is of great significance to safety of China;
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide one kind to be based on information source number time-varying and self-adaptive blind source
Isolated hazardous material detection method has advantage high-efficient, with high accuracy.
The purpose of the present invention is achieved through the following technical solutions: one kind being based on information source number time-varying and self-adaptive blind source
Isolated hazardous material detection method, comprising the following steps:
S1. standard database is built: using common liquid dangerous material and the safe product of liquid as sample, being acquired different samples and is existed
Scattering parameter and corresponding samples pictures in 8GHz-18GHz frequency range, build standard database;
S2. scattering parameter model construction: when being identified to several unknown liquid form products, first to liquid form product into
The acquisition of row scattering parameter, obtains the observation signal X (t) of t moment, and construct scattering parameter model:
Assuming that the source signal of t moment scattering parameter is S (t)=[S1(t),S2(t),S3(t),…,Sn(t)]T, S (t) be by
The set of n unknown and independent mean value signal composition each other source signals, each unknown liquid form product is a scattering
The information source of parameter corresponds to a mean value signal in source signal S (t);Sk(t) indicate k-th of unknown liquid form product in t moment
Scattering parameter mean value signal, k=0,1 ..., n;
Since the scattering parameter mean value signal of different unknown liquid form products can carry out mutually unknown linear hybrid, obtain most
Whole observation signal, therefore source signal S (t) is assumed after unknown Linear Hybrid Systems A (t) is handled, obtained t moment
Observation signal X (t) and source signal S (t) relationship are as follows:
X (t)=A (t) S (t);
S3. dynamic Sources number estimation: before carrying out blind source separating, it must be determined that the number information of information source, in order to dynamic
The quantity for estimating source signal, using estimating information source number, the detailed process of Sources number estimation based on the method for cross validation
It is as follows:
S301. the covariance matrix C (t) of calculating observation signal X (t):
Calculate the observation vector with zero-mean
It calculates intermediary matrix Δ X (t):
The covariance matrix C (t) of calculating observation signal:
Wherein, Δ X (t)TFor the transposition of Δ X (t);
S302. m diagonal matrix ψ is calculated(1), ψ(2)..., ψ(m);
For i-th of diagonal matrix ψ(i), i=1,2 ..., m, calculation is as follows:
Wherein,ΛiFor Matrix C (t) preceding i eigenvalue cluster at diagonal matrix;UiFor matrix ΛiFeature
It is worth the column vector of character pair vector composition;It indicatesTransposition, diag () expression seek diagonal matrix;
S303. the estimated value of information source number n is estimated
In formula,Representing matrix seeks all set for correspond to i when the highest of track in mark operation,
Representing matrix seeks all set for correspond to i when track is minimum in mark operation;
S4. self-adaptive blind source separates:
S401. information source number is enabledIf separation matrix W (t) includes n column vector, the i-th column vector wi(t) for m dimension to
Amount, i=1,2 ..., n;
S402. by wi(t) it is multiplied with observation signal, obtains separation signal yi(t):
yi(t)=wi(t)TX(t);
S403. w is calculatedi(t) optimal cost function wi,opt(t):
Wherein, | | * | | it is Frobenius norm, | K (yi) | it is yi(t) absolute value of kurtosis,Table
Show and seeks yi(t) corresponding y when kurtosis maximum absolute valuei(t) value;
It is w by the i-th column vector in separation matrix W (t)i(t) corresponding optimal cost function w is replaced withi,opt(t);
S404. in i=1,2 ..., n, step S402~S403 is repeated, by each column vector in separation matrix W (t)
Corresponding optimal cost function is replaced with, optimal separation matrix W ' (t) is obtained;
S405. the estimation signal of source signal S (t) is found out
Wherein, S 'k(t) indicate isolated k-th of unknown liquid form product in the scattering parameter mean value signal of t moment,
Wherein: k=0,1 ..., n;
S5. dangerous material identify: the estimation signal that blind source separating is obtainedCompared with the S parameter in standard database
Compared with identifying unknown liquid product type.
The step S5 includes:
S501. for estimating signalIn include any scattering parameter mean value signal S 'k(t), k=0,1 ..., n,
By S 'k(t) it is compared respectively with the scattering parameter of each sample in standard database:
As S 'k(t) when being less than given threshold with the scattering parameter T difference of a certain sample in standard database, S 'k(t) corresponding
Unknown liquid product type it is identical as liquid form product type corresponding to scattering parameter T;At this point, if corresponding to scattering parameter T
Liquid form product be liquid dangerous material, then scattering parameter S 'k(t) corresponding unknown liquid form product is also same liquid dangerous material,
Scattering parameter T counter sample picture is shown;If liquid form product corresponding to scattering parameter T is the safe product of liquid;
S502. in the case where k=0 1 ..., n, step S501 is repeated, identifies each scattering parameter mean value signal
Corresponding unknown liquid product type.
The beneficial effects of the present invention are: the present invention is based on information source number time-varying and self-adaptive blind source to separate, realize to liquid
Effective detection of dangerous material, and multiple and different liquid form products can be detected, the time that the public crosses safety check is saved, is mentioned
The safety check efficiency of high public place;Meanwhile first carry out Sources number estimation, further according to estimated result carry out blind source separating, have compared with
High precision, later period only need the S parameter for increasing sample newly in the database that the extension of detection classification can be realized, have good
Scalability.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawing, but protection scope of the present invention is not limited to
It is as described below.
Since bus, subway station, railway station are stepped on, flow of the people in public place is big, and passenger's belongings are many and diverse, when detection
The data extracted are complicated, increase the difficulty of hazardous material detection.Meanwhile traditional detection instrument is deposited on liquid hazardous material detection
In certain limitation.In view of this situation, the invention proposes a kind of liquid dangerous material at low cost, high-efficient, with high accuracy
Detection method.
The present invention is mainly that research is unfolded to the blind source separation algorithm based on isolated component (ICA), utilizes different independences point
Amount analysis (ICA) model and the independent characteristic based on these models are extracted according to different ICA analysis methods using computer
The independent characteristic of different models out.Specific the, in actually detected environment, the type and quantity for being detected liquid product are to change at random
, at this point, the scattering parameter mean value signal of different unknown liquid form products can carry out mutually unknown linear hybrid, obtain final
Observation signal leads to not the type that unknown liquid product are recognized accurately;Therefore the present invention is based on cross validation skill using a kind of
The method of art estimates the source signal number of dynamic change, increases the adaptivity of algorithm, and obtain not based on blind source separating
It is specific as follows with the scattering parameter of information source:
As shown in Figure 1, a kind of hazardous material detection method separated based on information source number time-varying and self-adaptive blind source, including it is following
Step:
S1. standard database is built: using common liquid dangerous material and the safe product of liquid as sample, being acquired different samples and is existed
Scattering parameter (S parameter) and corresponding samples pictures in 8GHz-18GHz frequency range, build standard database;Under normal circumstances,
The S parameter of common liquid dangerous material and safe product sample in 8GHz-18GHz frequency range is adopted using using wave beam focusing anteena
Collection;Unknown liquid form product is equally acquired with wave beam focusing anteena;
S2. scattering parameter model construction: when being identified to several unknown liquid form products, first to liquid form product into
The acquisition of row scattering parameter, obtains the observation signal X (t) of t moment, and construct scattering parameter model:
Assuming that the source signal of t moment scattering parameter is S (t)=[S1(t),S2(t),S3(t),…,Sn(t)]T, S (t) be by
The set of n unknown and independent mean value signal composition each other source signals, each unknown liquid form product is a scattering
The information source of parameter corresponds to a mean value signal in source signal S (t);Sk(t) indicate k-th of unknown liquid form product in t moment
Scattering parameter mean value signal, k=0,1 ..., n;
Since the scattering parameter mean value signal of different unknown liquid form products can carry out mutually unknown linear hybrid, obtain most
Whole observation signal, therefore source signal S (t) is assumed after unknown Linear Hybrid Systems A (t) is handled, obtained t moment
Observation signal X (t) and source signal S (t) relationship are as follows:
X (t)=A (t) S (t);
S3. dynamic Sources number estimation: before carrying out blind source separating, it must be determined that the number information of information source, in order to dynamic
The quantity for estimating source signal, using estimating information source number, the detailed process of Sources number estimation based on the method for cross validation
It is as follows:
S301. the covariance matrix C (t) of calculating observation signal X (t):
Calculate the observation vector with zero-mean
It calculates intermediary matrix Δ X (t):
The covariance matrix C (t) of calculating observation signal:
Wherein, Δ X (t)TFor the transposition of Δ X (t);
S302. m diagonal matrix ψ is calculated(1), ψ(2)..., ψ(m);
For i-th of diagonal matrix ψ(i), i=1,2 ..., m, calculation is as follows:
Wherein,ΛiFor Matrix C (t) preceding i eigenvalue cluster at diagonal matrix;UiFor matrix ΛiFeature
It is worth the column vector of character pair vector composition;It indicatesTransposition, diag () expression seek diagonal matrix;
S303. the estimated value of information source number n is estimated
In formula,Representing matrix seeks all set for correspond to i when the highest of track in mark operation,
Representing matrix seeks all set for correspond to i when track is minimum in mark operation;
S4. self-adaptive blind source separates:
S401. information source number is enabledIf separation matrix W (t) includes n column vector, the i-th column vector wi(t) for m dimension to
Amount, i=1,2 ..., n;
S402. by wi(t) it is multiplied with observation signal, obtains separation signal yi(t):
yi(t)=wi(t)TX(t);
S403. w is calculatedi(t) optimal cost function wi,opt(t):
Wherein, | | * | | it is Frobenius norm, | K (yi) | it is yi(t) absolute value of kurtosis,Table
Show and seeks yi(t) corresponding y when kurtosis maximum absolute valuei(t) value;
It is w by the i-th column vector in separation matrix W (t)i(t) corresponding optimal cost function w is replaced withi,opt(t);
S404. in i=1,2 ..., n, step S402~S403 is repeated, by each column vector in separation matrix W (t)
Corresponding optimal cost function is replaced with, optimal separation matrix W ' (t) is obtained;
S405. the estimation signal of source signal S (t) is found out
Wherein, S 'k(t) indicate isolated k-th of unknown liquid form product in the scattering parameter mean value signal of t moment,
Wherein: k=0,1 ..., n;
S5. dangerous material identify: the estimation signal that blind source separating is obtainedCompared with the S parameter in standard database
Compared with identifying unknown liquid product type.
The step S5 includes:
S501. for estimating signalIn include any scattering parameter mean value signal S 'k(t), k=0,1 ..., n,
By S 'k(t) it is compared respectively with the scattering parameter of each sample in standard database:
As S 'k(t) when being less than given threshold with the scattering parameter T difference of a certain sample in standard database, S 'k(t) corresponding
Unknown liquid product type it is identical as liquid form product type corresponding to scattering parameter T;At this point, if corresponding to scattering parameter T
Liquid form product be liquid dangerous material, then scattering parameter S 'k(t) corresponding unknown liquid form product is also same liquid dangerous material,
Scattering parameter T counter sample picture is shown;If liquid form product corresponding to scattering parameter T is the safe product of liquid;
S502. in the case where k=0 1 ..., n, step S501 is repeated, identifies each scattering parameter mean value signal
Corresponding unknown liquid product type.
To sum up, the present invention is based on information source number time-varying and self-adaptive blind source to separate, and realizes effective inspection to liquid dangerous material
It surveys, and multiple and different liquid form products can be detected, save the time that the public crosses safety check, improve the peace of public place
Examine efficiency;Meanwhile Sources number estimation is first carried out, blind source separating is carried out further according to estimated result, precision with higher, the later period is only
It needs the S parameter for increasing sample newly in the database that the extension of detection classification can be realized, there is good scalability.
What has been described above is only a preferred embodiment of the present invention, it should be understood that the present invention is not limited to described herein
Form, be not to be taken as excluding to outside other embodiments, and can be used for various other groups and modification and environment, and can be
In contemplated scope described herein, it is modified by the technology or knowledge of above-mentioned related fields.And what those skilled in the art were carried out
Change and transformation do not depart from the spirit and scope of the present invention, then all should be within the scope of protection of the appended claims of the present invention.
Claims (2)
1. a kind of hazardous material detection method separated based on information source number time-varying and self-adaptive blind source, it is characterised in that: including following
Step:
S1. standard database is built: using common liquid dangerous material and the safe product of liquid as sample, being acquired different samples and is existed
Scattering parameter and corresponding samples pictures in 8GHz-18GHz frequency range, build standard database;
S2. scattering parameter model construction: when being identified to several unknown liquid form products, liquid form product is carried out first scattered
Parameter acquisition is penetrated, the observation signal X (t) of t moment is obtained, and constructs scattering parameter model:
Assuming that the source signal of t moment scattering parameter is S (t)=[S1(t),S2(t),S3(t),…,Sn(t)]T, S (t) is by n
The set of unknown and independent mean value signal composition each other source signal, each unknown liquid form product is a scattering parameter
Information source, correspond to source signal S (t) in a mean value signal;Sk(t) k-th of unknown liquid form product dissipating in t moment is indicated
Penetrate mean parameter signal, k=0,1 ..., n;
Since the scattering parameter mean value signal of different unknown liquid form products can carry out mutually unknown linear hybrid, obtain final
Observation signal, thus assume source signal S (t) after unknown Linear Hybrid Systems A (t) is handled, the sight of obtained t moment
Survey the relationship of signal X (t) and source signal S (t) are as follows:
X (t)=A (t) S (t);
S3. dynamic Sources number estimation: before carrying out blind source separating, it must be determined that the number information of information source, in order to dynamically estimate
The quantity of source signal out, using information source number is estimated based on the method for cross validation, detailed process is as follows for Sources number estimation:
S301. the covariance matrix C (t) of calculating observation signal X (t):
Calculate the observation vector with zero-mean
It calculates intermediary matrix Δ X (t):
The covariance matrix C (t) of calculating observation signal:
Wherein, Δ X (t)TFor the transposition of Δ X (t);
S302. m diagonal matrix ψ is calculated(1), ψ(2)..., ψ(m);
For i-th of diagonal matrix ψ(i), i=1,2 ..., m, calculation is as follows:
Wherein,ΛiFor Matrix C (t) preceding i eigenvalue cluster at diagonal matrix;UiFor matrix ΛiCharacteristic value pair
The column vector for answering feature vector to form;It indicatesTransposition, diag () expression seek diagonal matrix;
S303. the estimated value of information source number n is estimated
In formula,Representing matrix seeks all set for correspond to i when the highest of track in mark operation,It indicates
All set for correspond to i when track is minimum in Matrix Calculating mark operation;
S4. self-adaptive blind source separates:
S401. information source number is enabledIf separation matrix W (t) includes n column vector, the i-th column vector wiIt (t) is the vector of m dimension, i
=1,2 ..., n;
S402. by wi(t) it is multiplied with observation signal, obtains separation signal yi(t):
yi(t)=wi(t)TX(t);
S403. w is calculatedi(t) optimal cost function wi,opt(t):
Wherein, | | * | | it is Frobenius norm, | K (yi) | it is yi(t) absolute value of kurtosis,Y is sought in expressioni
(t) corresponding y when kurtosis maximum absolute valuei(t) value;
It is w by the i-th column vector in separation matrix W (t)i(t) corresponding optimal cost function w is replaced withi,opt(t);
S404. in i=1,2 ..., n, step S402~S403 is repeated, each column vector in separation matrix W (t) is replaced
It is changed to corresponding optimal cost function, obtains optimal separation matrix W ' (t);
S405. the estimation signal of source signal S (t) is found out
Wherein, S 'k(t) scattering parameter mean value signal of k-th isolated of the unknown liquid form product of expression in t moment, in which: k
=0,1 ..., n;
S5. dangerous material identify: the estimation signal that blind source separating is obtainedIt is compared, knows with the S parameter in standard database
It Chu not unknown liquid product type.
2. the hazardous material detection method according to claim 1 separated based on information source number time-varying and self-adaptive blind source, special
Sign is: the step S5 includes:
S501. for estimating signalIn include any scattering parameter mean value signal S 'k(t), k=0,1 ..., n, by S 'k
(t) it is compared respectively with the scattering parameter of each sample in standard database:
As S 'k(t) when being less than given threshold with the scattering parameter T difference of a certain sample in standard database, S 'k(t) it is corresponding not
Know that liquid form product type is identical as liquid form product type corresponding to scattering parameter T;At this point, if liquid corresponding to scattering parameter T
State product is liquid dangerous material, then scattering parameter S 'k(t) corresponding unknown liquid form product is also same liquid dangerous material, will be dissipated
Parameter T counter sample picture is penetrated to be shown;If liquid form product corresponding to scattering parameter T is the safe product of liquid;
S502. in the case where k=0 1 ..., n, step S501 is repeated, identifies that each scattering parameter mean value signal is corresponding
Unknown liquid product type.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910732627.9A CN110458228A (en) | 2019-08-09 | 2019-08-09 | A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910732627.9A CN110458228A (en) | 2019-08-09 | 2019-08-09 | A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110458228A true CN110458228A (en) | 2019-11-15 |
Family
ID=68485511
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910732627.9A Pending CN110458228A (en) | 2019-08-09 | 2019-08-09 | A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110458228A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114280084A (en) * | 2021-12-23 | 2022-04-05 | 中国民航大学 | Airport security inspection auxiliary method based on radio frequency signals |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104007234A (en) * | 2014-05-16 | 2014-08-27 | 重庆大学 | Mixed gas composition identification method based on underdetermined blind source separation |
CN108665398A (en) * | 2017-03-28 | 2018-10-16 | 成都理工大学 | More article safety check algorithms based on blind source separating |
-
2019
- 2019-08-09 CN CN201910732627.9A patent/CN110458228A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104007234A (en) * | 2014-05-16 | 2014-08-27 | 重庆大学 | Mixed gas composition identification method based on underdetermined blind source separation |
CN108665398A (en) * | 2017-03-28 | 2018-10-16 | 成都理工大学 | More article safety check algorithms based on blind source separating |
Non-Patent Citations (3)
Title |
---|
周冬梅 等: "基于超宽带厘米波的液体安检感知机模型", 《电子学报》 * |
王荣杰 等: "一种适用于信源数时变的自适应盲源分离算法", 《仪器仪表学报》 * |
陈春梅: "危险品检测算法的研究与实现", 《中国优秀硕士学位论文全文数据库 社会科学Ⅰ辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114280084A (en) * | 2021-12-23 | 2022-04-05 | 中国民航大学 | Airport security inspection auxiliary method based on radio frequency signals |
CN114280084B (en) * | 2021-12-23 | 2023-09-08 | 中国民航大学 | Airport security inspection auxiliary method based on radio frequency signals |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104240455B (en) | A kind of disturbance event recognition methods in distribution type fiber-optic pipeline safety early warning system | |
Carlsten et al. | Luminosity Functions and Host-to-host Scatter of Dwarf Satellite Systems in the Local Volume | |
Kundu et al. | New Insights from Hubble Space Telescope Studies of Globular Cluster Systems. II. Analysis of 29 S0 Systems | |
US10539613B2 (en) | Analog circuit fault diagnosis method using single testable node | |
CN109784383B (en) | Rail crack identification method based on graph domain feature and DS evidence theory fusion | |
CN106096411A (en) | A kind of Android malicious code family classification method based on bytecode image clustering | |
CN104749334A (en) | Mode-recognition-based design method for biological abnormal water quality evaluation system | |
CN103954940B (en) | Radar fence interferes discrimination method with distributed compacting based on the centralization that cross bearing point clusters | |
CN110717464B (en) | Intelligent railway roadbed disease identification method based on radar data | |
CN105869146A (en) | Saliency fusion-based SAR image change detection method | |
CN105137324B (en) | A kind of more detection point failure component localization methods based on emulation disaggregated model | |
CN102360503B (en) | SAR (Specific Absorption Rate) image change detection method based on space approach degree and pixel similarity | |
CN109901111A (en) | Near-field sound source localization method based on Partial Least Squares Regression | |
CN110033160A (en) | A kind of performance appraisal system and method | |
Papoutsoglou et al. | Challenges in the multivariate analysis of mass cytometry data: the effect of randomization | |
Friedlander et al. | Dynamic agent classification and tracking using an ad hoc mobile acoustic sensor network | |
CN110458228A (en) | A kind of hazardous material detection method of information source number time-varying and self-adaptive blind source separation | |
CN115129594A (en) | Test scene processing method and device, electronic equipment and storage medium | |
CN111563532B (en) | Unknown target identification method based on attribute weight fusion | |
CN113567931A (en) | Method and system for detecting double subspace signals when training samples are insufficient | |
CN111273126B (en) | Power distribution network topology rapid sensing method | |
KR101468560B1 (en) | A method of detecting outlier pulses of radar signal based on mahalanobis distance | |
CN204989215U (en) | Wheel detector signal processing system | |
CN103743755A (en) | Method for detecting visible foreign matters in medical agents based on affinity propagation cluster | |
CN106650951A (en) | Multichannel measurement data automatic screening method based on time domain global similarity |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191115 |
|
RJ01 | Rejection of invention patent application after publication |