CN108446599A - A kind of high spectrum image wave band fast selecting method of p value statistic modeling independence - Google Patents
A kind of high spectrum image wave band fast selecting method of p value statistic modeling independence Download PDFInfo
- Publication number
- CN108446599A CN108446599A CN201810161250.1A CN201810161250A CN108446599A CN 108446599 A CN108446599 A CN 108446599A CN 201810161250 A CN201810161250 A CN 201810161250A CN 108446599 A CN108446599 A CN 108446599A
- Authority
- CN
- China
- Prior art keywords
- wave band
- value
- band
- matrix
- wave
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Complex Calculations (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to the high spectrum image wave band fast selecting method that a kind of p value statistic models independence, this method includes mainly image importing, p value normalized set, calculating target function and selects the links such as wave band.Image imports：Import the high spectrum image of wave band to be selected.P value normalized set：Obtain the p value statistic of the Pearson linear correlations between the wave band of high spectrum image.Calculating target function：Count the sum of the p value between each wave band of high spectrum image and its all band, the i.e. independence of wave band.Choose wave band：The independence of each wave band is ranked up, wave band is chosen.The present invention carries out waveband selection to high spectrum image, and algorithm complexity is low, efficient, can reach instant or even real-time effect.
Description
Technical field
The present invention relates to remote sensing fields, specifically provide a kind of high spectrum image wave band of p value statistic modeling independence
Fast selecting method for EO-1 hyperion dimensionality reduction and carries out image classification.
Background technology
The characteristics of being limited to " big data quantity, High redundancy " so that high spectrum image is not easy to by high efficiency and accurately
Realize the typical cases such as mixed solution, classification, target detection and physical quantity inverting, and dimensionality reduction is effectively to solve the problems, such as this main means
One of.One of two kinds of main realization methods as dimensionality reduction, waveband selection is to seek the characteristic wave of " contain much information, independence strong "
Section realizes the simplification of feature space.The wave of large information capacity and strong independence can be effectively selected in spite of certain methods and software
Section, but generally require to take considerable time, need stronger hardware system as support.
Invention content
The technical problem to be solved in the present invention：It overcomes the deficiencies of the prior art and provide a kind of with efficient EO-1 hyperion
Image band fast selecting method.This method includes mainly that image imports, Pearson linearly dependent coefficients calculate, p value statistic
Calculate, the independence index of wave band calculates and selects wave band, calculate it is simple, it is real-time.
The technical solution adopted by the present invention：A kind of high spectrum image wave band of p value statistic modeling independence quickly selects
Method, its step are as follows：
Step (1), the high spectrum image for importing wave band to be selected, and identical standardization is carried out to each wave band；
Step (2), according to step (1) obtain as a result, carrying out wave band Pearson linearly dependent coefficients r between any two
Calculating, finally obtain full wave r value matrixs；
Step (3), according to step (2) obtain as a result, carrying out the meter of wave band correlation analysis p value statistic between any two
It calculates, finally obtains full wave p value matrix；
Step (4), according to step (3) obtain as a result, each column to p value matrix carries out descending arrangement respectively, dropped
Sequence p value matrix and corresponding band number matrix；
Step (5), according to step (4) obtain as a result, take the preceding k rows of descending p value matrix and corresponding band number matrix,
And the sum of the p value in the corresponding descending p value matrix of each wave band number in band number matrix is counted, as each wave band is selecting
Independence index when k wave band；
Step (6), according to step (5) obtain as a result, the independence index shut out to each wave carries out descending arrangement, the row of taking
K forward wave band of name, the as result of waveband selection.
Further, step (2) medium wave band Pearson linearly dependent coefficient r values between any two and all band r values
The calculating process of matrix is as follows：
Step (21) decentralization：Each wave band subtracts wave band mean value；
Step (22) calculates the covariance between wave band；
Step (23) is standardized covariance：Covariance between wave band divided by their standard deviation, you can obtain wave
The Pearson linearly dependent coefficient r values of section between any two；
Step (24) carries out full wave arbitrary wave band two-by-two the operation of step (21)-(23), you can obtains all band r
Value matrix.
Further, the linearly related p value statistic between any two of the wave band in the step (3) and all band p value square
The calculating process of battle array is as follows：
Step (31) constructs the t statistics that one degree of freedom is υ；Preferably, ν=N-2 is set, and wherein N is correlation analysis
In total sample number.
Step (32) calculating Alpha's function A (t | ν)；
Step (33) calculates p value statistic, the as linearly related p value statistic of wave band between any two；
Step (34) carries out full wave arbitrary wave band two-by-two the operation of step (31)-(33), you can obtains all band p
Value matrix.
Further, the acquisition process of the descending p value matrix in the step (4) and corresponding band number matrix is such as
Under：
Step (41) carries out descending arrangement respectively to each row of p value matrix, obtains descending p value matrix；
Step (42) obtains the corresponding band number of row p value, obtains corresponding wave while obtaining descending p value row
Section serial number matrix.
The advantages of the present invention over the prior art are that：
(1) the invention belongs to unsupervised dimension reduction methods, adaptively automatic to result end by data terminal without setting threshold value
It completes to resolve；
(2) algorithm complexity of method proposed by the invention is relatively low, while lower space complexity is to system hardware
Level requirement is relatively low；
(3) present invention can carry out high spectrum image the instant or even real-time selection of wave band, in order to be effectively performed
The second selecting etc. of image classification, target signature wave band.
Description of the drawings
Fig. 1 is flow chart of the present invention.
Specific implementation mode
Invention is further described in detail with reference to the accompanying drawings and detailed description.P value system provided by the invention
The high spectrum image wave band fast selecting method of metering modeling independence, as shown in Figure 1, the key step for including is as follows：
Step (1) imports data and initializes：The high spectrum image of wave band to be selected is imported, and phase is carried out to each wave band
Same standardization.
The initiating terminal of hyperspectral image band selection method of the present invention is the importing of hyperspectral image data, is then needed
Data are standardized etc. with pretreatments.
A wave band B standard in high spectrum image turns to Bnorm, standardisation process is represented by formula (1)：
Bnorm=(B-Bmin)/(Bmax-Bmin) (1)
Wherein, Bmax、BminThe respectively maximum value of B and minimum value.
By the high spectrum image M after standardizationi,j,nEach band image be converted to column vector, obtain matrix Mij,n：
Wherein, it is M that i, j, n, which distinguish n,i,j,nRow, column, wave band sum, ij=i × j be Mij,nLine number.
Step (2) calculates wave band correlation coefficient r and all band r value matrixs：According to step (1) obtain as a result, into traveling wave
The calculating of the Pearson linearly dependent coefficients r of section between any two.
High spectrum image M based on initializationij,n, formula (2) calculates separately the Pearson between arbitrary two wave band
Related coefficient, the formula include the decentralization of data, covariance calculating etc.：
Wherein, N is total sample number, i.e. N=ij=i × j；Binary sequenceIndicate Mij,nIn any two row
Vector,WithRespectivelyWithMean value.
The calculating that correlation coefficient r is carried out between repetition step (2) two wave band arbitrary to all band, finally obtains full wave
R value matrixs.
Step (3), the p value statistic and all band p value matrix for calculating wave band correlation coefficient r：It is obtained according to step (2)
As a result, carrying out the calculating of wave band Linear correlative analysis p value statistic between any two and all band p value matrix.
P value is proposed by statistician Sir Ronald Aylmer Fisher, and in null hypothesis, (or zero is false for expression
If null hypothesis) when being true, there is the probability of or more extreme case identical with current observations appearance.In correlation
In the hypothesis testing of analysis, null hypothesis be without related (no correlation), at this point, p value is a kind of statistic of sample,
Size is solved by constructing the t statistics that one degree of freedom is υ：
P=1-A (t | ν) (3)
ν=N-2 (6)
Wherein, beta functionAlternatively, it is also possible to by gamma function Γ come indirectly
Solve Β functions.The p value of wave band between any two is just calculated respectively by correlation coefficient r matrix, obtains related coefficient p value matrix
Pnxn。
The calculating of p value, finally obtains full wave p value matrix between repetition step (3) two wave band arbitrary to all band.
Step 4: related coefficient p value matrix between descending processing wave band：According to step (3) obtain as a result, to p value matrix
It is each row respectively carry out descending arrangement, obtain descending p value matrix and corresponding band number matrix.
Remove p value matrix Pn×nP is obtained after medium wave band and the p value of its own(n-1)×n；To P(n-1)×nEach row descending arrangement, choosing
K rows composition matrix PS before selectingk×n, and its corresponding wave band number is formed into matrix Bk×n；
Step 5: calculating target function, object function is for indicating that certain wave band is selecting independence when k wave band to refer to
Mark：According to step (4) obtain as a result, take the preceding k rows of descending p value matrix and corresponding band number matrix, and count wave band sequence
The sum of p value in number matrix in the corresponding descending p value matrix of each wave band number, as each wave band is when selecting k wave band
Independence index；
Construct the reference value object function f of waveband selectionk(i)：
Wherein, fk(i) it is p values the sum of of the element i on PS corresponding positions in B, indicates when selecting k wave band, i-th
The selection reference value of wave band.
Step 6: selecting wave band：According to step (5) obtain as a result, independence to each wave band when selecting k wave band
Property index carry out descending arrangement, take k wave band in the top, the as result of waveband selection.
To fk(i) (i=1,2 ..., n) carries out descending arrangement, the corresponding wave band of k value before selection.
Recited above is only to embody a kind of embodiment of high spectrum image wave band fast selecting method of the present invention.The present invention
It is not limited to above-described embodiment.The specification of the present invention is not limit the scope of the claims for illustrating.For ability
The technical staff in domain, it is clear that can have many replacements, improvement and variation.It is all using equivalent substitution or equivalent transformation formed
Technical solution is all fallen within the protection domain of application claims.
Claims (4)
1. a kind of high spectrum image wave band fast selecting method of p value statistic modeling independence, it is characterised in that：Including as follows
Step：
Step (1), the high spectrum image for importing wave band to be selected, and identical standardization is carried out to each wave band；
Step (2), according to step (1) obtain as a result, carrying out the meter of wave band Pearson linearly dependent coefficients r between any two
It calculates, finally obtains full wave r value matrixs；
Step (3), according to step (2) obtain as a result, carrying out the meter of wave band Linear correlative analysis p value statistic between any two
It calculates, finally obtains full wave p value matrix；
Step (4), according to step (3) obtain as a result, to p value matrix it is each row respectively carry out descending arrangement, obtain descending p value
Matrix and corresponding band number matrix；
Step (5), according to step (4) obtain as a result, take the preceding k rows of descending p value matrix and corresponding band number matrix, and unite
The sum of the p value in the corresponding descending p value matrix of each wave band number in band number matrix is counted, as each wave band is in selection k
Independence index when wave band；
Step (6), according to step (5) obtain as a result, selecting independence index when k wave band to drop each wave band
Sequence arranges, and takes k wave band in the top, the as result of waveband selection.
2. a kind of high spectrum image wave band fast selecting method of p value statistic modeling independence according to claim 1,
It is characterized in that：Step (2) medium wave band Pearson linearly dependent coefficient r values between any two and all band r value matrixs
Calculating process is as follows：
Step (21) decentralization：Each wave band subtracts wave band mean value；
Step (22) calculates the covariance between wave band；
Step (23) is standardized covariance：Covariance between wave band divided by their standard deviation, you can obtain wave band two
Pearson linearly dependent coefficient r values between two；
Step (24) carries out full wave arbitrary wave band two-by-two the operation of step (21)-(23), you can obtains all band r value squares
Battle array.
3. a kind of high spectrum image wave band fast selecting method of p value statistic modeling independence according to claim 1,
It is characterized in that：The meter of wave band in the step (3) linearly related p value statistic between any two and all band p value matrix
Calculation process is as follows：
Step (31) constructs the t statistics that one degree of freedom is υ；
Step (32) calculating Alpha's function A (t | ν)；
Step (33) calculates p value statistic, the as linearly related p value statistic of wave band between any two；
Step (34) carries out full wave arbitrary wave band two-by-two the operation of step (31)-(33), you can obtains all band p value square
Battle array.
4. a kind of high spectrum image wave band fast selecting method of p value statistic modeling independence according to claim 1,
It is characterized in that：The acquisition process of descending p value matrix and corresponding band number matrix in the step (4) is as follows：
Step (41) carries out descending arrangement respectively to each row of p value matrix, obtains descending p value matrix；
Step (42) obtains the corresponding band number of row p value, obtains corresponding wave band sequence while obtaining descending p value row
Number matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810161250.1A CN108446599B (en) | 2018-02-27 | 2018-02-27 | Hyperspectral image band rapid selection method of p-value statistical modeling independence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810161250.1A CN108446599B (en) | 2018-02-27 | 2018-02-27 | Hyperspectral image band rapid selection method of p-value statistical modeling independence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108446599A true CN108446599A (en) | 2018-08-24 |
| CN108446599B CN108446599B (en) | 2021-11-05 |
Family
ID=63193086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810161250.1A Active CN108446599B (en) | 2018-02-27 | 2018-02-27 | Hyperspectral image band rapid selection method of p-value statistical modeling independence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108446599B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112884035A (en) * | 2021-02-07 | 2021-06-01 | 山东科技大学 | Noise robust hyperspectral image band selection method |
| CN113310930A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Spectral identification method of high-temperature sterilized milk, pasteurized milk and pasteurized milk mixed with high-temperature sterilized milk |
| CN113310937A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for rapidly identifying high-temperature sterilized milk, pasteurized fresh milk of dairy cow and reconstituted milk of milk powder |
| CN113310932A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Rapid identification method for adding high-temperature sterilized milk into pasteurized buffalo fresh milk |
| CN113310936A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Rapid identification method for four high-temperature sterilized commercial milks |
| CN113310934A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for quickly identifying milk cow milk mixed in camel milk and mixing proportion thereof |
| CN113310928A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for rapidly identifying high-temperature sterilized milk with shelf life within and out of date |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807301A (en) * | 2010-03-17 | 2010-08-18 | 北京航空航天大学 | High spectral image target detection method based on high order statistic |
| CN102135981A (en) * | 2010-01-25 | 2011-07-27 | 安克生医股份有限公司 | Method for multi-layer classifier |
| US20120076406A1 (en) * | 2009-07-20 | 2012-03-29 | Kevin Fisher | System and method for progressive band selection for hyperspectral images |
| CN102609703A (en) * | 2012-03-05 | 2012-07-25 | 中国科学院对地观测与数字地球科学中心 | Method and device for detecting target ground object in hyperspectral image |
| CN102903006A (en) * | 2012-09-14 | 2013-01-30 | 哈尔滨工程大学 | Hyperspectral image waveband selecting method |
| US8464095B1 (en) * | 2012-11-15 | 2013-06-11 | DSSD, Inc. | Method and system for multi-dimensional raid reconstruction and defect avoidance |
| EP2704100A1 (en) * | 2012-08-29 | 2014-03-05 | GE Aviation Systems LLC | Method for simulating hyperspectral imagery |
| KR101463354B1 (en) * | 2013-10-22 | 2014-11-18 | 영남대학교 산학협력단 | Apparatus for target detection of hyperspectral image and method thereof |
| CN104268582A (en) * | 2014-08-26 | 2015-01-07 | 中国科学院遥感与数字地球研究所 | Band selection method and device of hyperspectral images |
| CN104915636A (en) * | 2015-04-15 | 2015-09-16 | 北京工业大学 | Remote sensing image road identification method based on multistage frame significant characteristics |
| CN105426543A (en) * | 2015-12-28 | 2016-03-23 | 辽宁师范大学 | Image retrieval method based on Bessel statistic model |
| CN105931223A (en) * | 2016-04-13 | 2016-09-07 | 浙江大学 | Band ratio method based maize embryo segmentation method in high-spectral reflection image |
| CN106915462A (en) * | 2017-02-14 | 2017-07-04 | 福建兴宇信息科技有限公司 | Forestry pests ＆ diseases intelligent identifying system based on multi-source image information |
| CN107480699A (en) * | 2017-07-13 | 2017-12-15 | 电子科技大学 | A kind of intrusion detection method based on channel condition information and SVMs |
| CN107679676A (en) * | 2017-10-27 | 2018-02-09 | 河海大学 | A kind of city based on numerical simulation is low to influence exploitation Optimal Configuration Method |
-
2018
- 2018-02-27 CN CN201810161250.1A patent/CN108446599B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120076406A1 (en) * | 2009-07-20 | 2012-03-29 | Kevin Fisher | System and method for progressive band selection for hyperspectral images |
| CN102135981A (en) * | 2010-01-25 | 2011-07-27 | 安克生医股份有限公司 | Method for multi-layer classifier |
| CN101807301A (en) * | 2010-03-17 | 2010-08-18 | 北京航空航天大学 | High spectral image target detection method based on high order statistic |
| CN102609703A (en) * | 2012-03-05 | 2012-07-25 | 中国科学院对地观测与数字地球科学中心 | Method and device for detecting target ground object in hyperspectral image |
| EP2704100A1 (en) * | 2012-08-29 | 2014-03-05 | GE Aviation Systems LLC | Method for simulating hyperspectral imagery |
| CN102903006A (en) * | 2012-09-14 | 2013-01-30 | 哈尔滨工程大学 | Hyperspectral image waveband selecting method |
| US8464095B1 (en) * | 2012-11-15 | 2013-06-11 | DSSD, Inc. | Method and system for multi-dimensional raid reconstruction and defect avoidance |
| KR101463354B1 (en) * | 2013-10-22 | 2014-11-18 | 영남대학교 산학협력단 | Apparatus for target detection of hyperspectral image and method thereof |
| CN104268582A (en) * | 2014-08-26 | 2015-01-07 | 中国科学院遥感与数字地球研究所 | Band selection method and device of hyperspectral images |
| CN104915636A (en) * | 2015-04-15 | 2015-09-16 | 北京工业大学 | Remote sensing image road identification method based on multistage frame significant characteristics |
| CN105426543A (en) * | 2015-12-28 | 2016-03-23 | 辽宁师范大学 | Image retrieval method based on Bessel statistic model |
| CN105931223A (en) * | 2016-04-13 | 2016-09-07 | 浙江大学 | Band ratio method based maize embryo segmentation method in high-spectral reflection image |
| CN106915462A (en) * | 2017-02-14 | 2017-07-04 | 福建兴宇信息科技有限公司 | Forestry pests ＆ diseases intelligent identifying system based on multi-source image information |
| CN107480699A (en) * | 2017-07-13 | 2017-12-15 | 电子科技大学 | A kind of intrusion detection method based on channel condition information and SVMs |
| CN107679676A (en) * | 2017-10-27 | 2018-02-09 | 河海大学 | A kind of city based on numerical simulation is low to influence exploitation Optimal Configuration Method |
Non-Patent Citations (5)
| Title |
|---|
| CHENIN-I CHANG 等: "Progressive Band Selection of Spectral Unmixing for Htperspectral Imagery", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 * |
| R. RESHMA 等: "Dimensionality Reduction Using Band Selection Technique for Kernel Based Hyperspectral Image Classification", 《PROCEDIA COMPUTER SCIENCE》 * |
| 刘春红 等: "一种新的高光谱遥感图像降维方法", 《中国图象图形学报》 * |
| 张爱武 等: "非线性变换和信息相邻相关的高光谱自适应波段选择", 《红外与激光工程》 * |
| 杜华强 等: "分形维数作为高光谱遥感数据波段选择的一个指标", 《遥感技术与应用》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112884035A (en) * | 2021-02-07 | 2021-06-01 | 山东科技大学 | Noise robust hyperspectral image band selection method |
| CN113310930A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Spectral identification method of high-temperature sterilized milk, pasteurized milk and pasteurized milk mixed with high-temperature sterilized milk |
| CN113310937A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for rapidly identifying high-temperature sterilized milk, pasteurized fresh milk of dairy cow and reconstituted milk of milk powder |
| CN113310932A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Rapid identification method for adding high-temperature sterilized milk into pasteurized buffalo fresh milk |
| CN113310936A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Rapid identification method for four high-temperature sterilized commercial milks |
| CN113310934A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for quickly identifying milk cow milk mixed in camel milk and mixing proportion thereof |
| CN113310928A (en) * | 2021-05-10 | 2021-08-27 | 华中农业大学 | Method for rapidly identifying high-temperature sterilized milk with shelf life within and out of date |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108446599B (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108446599A (en) | A kind of high spectrum image wave band fast selecting method of p value statistic modeling independence | |
| CN110321963B (en) | Hyperspectral image classification method based on fusion of multi-scale and multi-dimensional space spectrum features | |
| CN112364730B (en) | Hyperspectral ground object automatic classification method and system based on sparse subspace clustering | |
| CN105022754B (en) | Object classification method and device based on social network | |
| CN106326288A (en) | Image search method and apparatus | |
| CN108509911A (en) | Interference signal recognition methods based on convolutional neural networks | |
| CN106575379A (en) | Improved fixed point integer implementations for neural networks | |
| CN110363229A (en) | A kind of characteristics of human body's parameter selection method combined based on improvement RReliefF and mRMR | |
| CN109101938A (en) | A kind of multi-tag age estimation method based on convolutional neural networks | |
| CN109116312A (en) | Radar signal classification method based on QMFB and convolutional neural networks | |
| CN115953665A (en) | Target detection method, device, equipment and storage medium | |
| CN109787929A (en) | Signal modulate method, electronic device and computer readable storage medium | |
| CN110070082A (en) | Licence plate recognition method, device, equipment and storage medium | |
| CN110188825A (en) | Image clustering method, system, equipment and medium based on discrete multiple view cluster | |
| CN109033833A (en) | A kind of malicious code classification method based on multiple features and feature selecting | |
| He et al. | Recognition of soybean pods and yield prediction based on improved deep learning model | |
| CN115049814B (en) | Intelligent eye protection lamp adjusting method adopting neural network model | |
| CN109409407A (en) | A kind of industry monitoring data clustering method based on LE algorithm | |
| Allam Jr et al. | Paying attention to astronomical transients: introducing the time-series transformer for photometric classification | |
| CN105760857B (en) | A kind of target in hyperspectral remotely sensed image object detection method | |
| Vaddi et al. | Hyperspectral remote sensing image classification using combinatorial optimisation based un‐supervised band selection and CNN | |
| Zhou et al. | Hyperspectral image change detection by self-supervised tensor network | |
| CN113343801A (en) | Automatic wireless signal modulation and identification method based on lightweight convolutional neural network | |
| Zhan et al. | Computational colour constancy based on convolutional neural networks with a cross‐level architecture | |
| CN111783688B (en) | Remote sensing image scene classification method based on convolutional neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |