CN107271394A - A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network - Google Patents
A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network Download PDFInfo
- Publication number
- CN107271394A CN107271394A CN201710342868.3A CN201710342868A CN107271394A CN 107271394 A CN107271394 A CN 107271394A CN 201710342868 A CN201710342868 A CN 201710342868A CN 107271394 A CN107271394 A CN 107271394A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- tealeaves
- sample
- infrared spectrum
- 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
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 35
- 241001122767 Theaceae Species 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 5
- 238000012937 correction Methods 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000001228 spectrum Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000001157 Fourier transform infrared spectrum Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 claims 2
- 238000000605 extraction Methods 0.000 abstract description 4
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000004611 spectroscopical analysis Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 3
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 2
- 235000017491 Bambusa tulda Nutrition 0.000 description 2
- 241001330002 Bambuseae Species 0.000 description 2
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 2
- 241000270295 Serpentes Species 0.000 description 2
- 239000011425 bamboo Substances 0.000 description 2
- 238000012850 discrimination method Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- 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
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The invention discloses the tealeaves infrared spectrum sorting technique that a kind of fuzzy Kohonen differentiates clustering network, comprise the following steps：The first step, the collection of tealeaves sample infrared spectrum；Second step, the multiplicative scatter correction (MSC) of tealeaves sample infrared spectrum.3rd step, carries out dimension-reduction treatment, then further extract feature and dimensionality reduction with linear discriminant analysis with principal component analytical method to the infrared spectrum of tealeaves；4th step, runs Fuzzy C mean cluster to obtain initial cluster center；5th step, differentiates that clustering network method carries out the classification of local tea variety using a kind of fuzzy Kohonen.The present invention solves Fuzzy Kohonen Clustering Network and data is being carried out not extract the authentication information of data during fuzzy clustering, causes the problem of cluster accuracy rate is not high.Fast with detection speed present invention can be implemented in the authentication information of Dynamic Extraction tealeaves spectroscopic data in cluster process, classification accuracy is high, and classification effectiveness is high, and the classification of different cultivars tealeaves can be achieved.
Description
Technical field
The invention belongs to field of artificial intelligence, especially a kind of fuzzy Kohonen differentiates that the tealeaves of clustering network is red
External spectrum sorting technique.
Background technology
Tealeaves is one of main crop of China, and the postharvest handling of tealeaves, quality judge and detection is always tea leaf quality
The important means of guarantee.The tea market of China is due to lacking effective tealeaves discrimination method so being pasted in tea market at present
Board phenomenon, adulterates and phenomenon ratio of mixing the spurious with the genuine is more serious, therefore the discriminating of local tea variety becomes more and more important, and studies
A kind of simple and quick local tea variety discrimination method is very important.
Infrared spectrum technology has detection speed fast, the advantages of can detecting Multiple components simultaneously.The tealeaves of different cultivars, its
Often there is difference in component and content, then the diffusing reflection spectrum obtained on the tealeaves of different cultivars is discrepant, is utilized
This principle, it is possible to achieve the Classification of Tea of different cultivars.
Fuzzy Kohonen Clustering Network is a kind of unsupervised learning method (Tsao E C, Bezdek J C, Pal N
R.Fuzzy Kohonen clustering networks.Pattern Recognition,1994,27(5):757–764.)。
Fuzzy Kohonen Clustering Network is the learning rate and more that Fuzzy C-Means Clustering (FCM) is incorporated into Kohonen clustering networks
In new strategy.As a kind of unsupervised clustering method, Fuzzy Kohonen Clustering Network can only be realized to the fuzzy poly- of data
Class, and the authentication information of data can not be extracted during fuzzy clustering, so that high cluster accuracy rate can not be obtained.
The content of the invention
The defect and deficiency existed for above-mentioned existing Fuzzy Kohonen Clustering Network, the purpose of the present invention is to propose to
A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network.This method obtains tea with infrared spectrum technology
The infrared spectrum of leaf, carries out dimension-reduction treatment to the infrared spectrum of tealeaves with principal component analysis, spectrum is carried out with linear discriminant method
The feature extraction of information, operation Fuzzy C-Means Clustering is differentiated with a kind of fuzzy Kohonen and clustered to obtain initial cluster center
Network method carries out the classification of local tea variety.
According to above-mentioned principle, the technical scheme of use comprises the following steps：
Step 1: gathering the infrared spectrum of tealeaves sample under constant-temperature constant-humidity environment：For the tealeaves sample of different cultivars,
Of infrared spectrometer these tealeaves samples are made with collection infrared spectrum to test, the infrared diffusing reflection spectrum letter of tealeaves sample is obtained
Breath, spectral information is stored in computer.
Step 2: being pre-processed to tealeaves sample infrared spectrum：With multiplicative scatter correction (MSC) to tealeaves sample infrared spectrum
Pre-processed.
Step 3: carrying out dimension-reduction treatment to tealeaves sample infrared spectrum：Using principal component analytical method (PCA) by tealeaves sample
This infrared spectrum drops to relatively low low-dimensional data from high dimensional data, and preserves these data, then is entered with linear discriminant method
Onestep extraction feature and dimensionality reduction.
Step 4: Fuzzy C-Means Clustering is to obtain initial cluster center：Mould is run to the tealeaves infrared data after dimensionality reduction
C- mean clusters are pasted, initial cluster centre is obtained.
Step 5: differentiating that clustering network method carries out the classification of local tea variety with a kind of fuzzy Kohonen：According to step 4
A kind of initial fuzzy Kohonen of cluster centre operation differentiate that clustering network method obtains fuzzy membership, be subordinate to according to fuzzy
Category degree can be classified different cultivars tealeaves.
A kind of fuzzy Kohonen in the step 5 differentiates that clustering network method is as follows：
1. initialization
Fixed tealeaves infrared spectrum sample class number c and weighted index m0Value, n is sample number, n>c>1,+∞>m0>1,
Maximum iteration t is setmaxWith the value ε of the error upper limit, initial cluster center v is seti,0(i=1,2 ... c).Set feature to
Amount number is q；
2. calculating t (t=1,2 ..., tmax) secondary iteration when learning rate αik,t
Wherein mt=m0- t Δs m, t are iterations, Δ m=(m0-1)/tmax uik,tRepresent kth during the t times iterative calculation
Individual sample is under the jurisdiction of the fuzzy membership angle value of the i-th class.uik,tIt is calculated as follows：
xkFor k-th of tealeaves infrared spectrum sample, xk∈Rp, i.e. xkDimension be p.vi,tI-th during for the t times iterative calculation
Lei Lei centers, vj,tJLei Lei centers during for the t times iterative calculation.
3. calculate the t times iteration Shi Lei center vi,t
vi,t-1I-th Lei Lei centers during for the t-1 times iterative calculation.
4. calculate collision matrix S between fuzzy classfBFuzzy totality collision matrix SfT
For the average of sample
5. according to formula：Q characteristic vector ψ before calculating1,ψ2,...,ψq(q >=1, and p>q)
6. by sample xkProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain
xk'=xk T[ψ1,ψ2,...,ψq]
By class center vi,tProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain
vi,t'=vi,t[ψ1,ψ2,...,ψq]
If 7. maxi | | v 'i,t-v′i,t-1||<ε or t>tmax, iteration terminates, otherwise, t=t+1, by xk' value be assigned to
xk, vi,t' value be assigned to vi,t, the continuation iterative calculation of return to step 2.
Beneficial effects of the present invention：
The present invention solves Fuzzy Kohonen Clustering Network and data is being carried out not extract data during fuzzy clustering
Authentication information, cause cluster accuracy rate it is not high the problem of.Present invention can be implemented in Dynamic Extraction tealeaves spectrum in cluster process
The authentication information of data, Classification of Tea accuracy rate is high.
Brief description of the drawings
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the infrared spectrogram of tealeaves sample；
Fig. 3 is that tealeaves infrared spectrum passes through the pretreated spectrograms of MSC；
Fig. 4 is the X-Y scheme obtained after linear discriminant method is handled；
Fig. 5 is that a kind of fuzzy Kohonen differentiates the fuzzy membership figure that clustering network method is produced.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment the present invention is further illustrated, but protection scope of the present invention is simultaneously
Not limited to this.
The present invention is applied to the assortment of different cultivars tealeaves, implementing procedure as shown in figure 1, specific implementation is as follows：
Embodiment：
Step 1: tealeaves sample infrared spectrum is gathered：Gather high-quality Leshan green bamboo snake, Leshan green bamboo snake inferior and Mount Emei
Mao Fengsan kind tealeaves, the sample number of every kind of tealeaves is 32, adds up to 96 samples.All tealeaves samples are ground after smashing through 40 mesh
Filter is sieved through, each sample takes 0.5g respectively with KBr by 1:Mixture 1g is taken to carry out press mold processing after 100 uniform mixing.Adopting
When collecting tealeaves infrared spectrum, laboratory temperature and relative humidity keep constant, FTIR-7600 type FTIR spectrum analyzers
Start preheating 1h.Spectroanalysis instrument scans each tealeaves sample 32 times, the wave-number range of spectral scan for 4001.569~
401.1211cm-1, sweep spacing is 1.9285cm-1, and the infrared spectrum of each tealeaves sample is the high dimensional data of 1868 dimensions.Often
Individual specimen sample 3 times, the experimental data for taking its average value to be set up as following model.It is test set that every kind of sample, which chooses 22,
Then test sample number n is 66.Remaining 10 samples are training set, then number of training nrFor 30.Test set is tea to be identified
Leaf sample, training set is known good and bad tealeaves sample.Classification number c=3 is set.The infrared spectrogram of tealeaves sample such as Fig. 2 institutes
Show.
Step 2: being pre-processed to tealeaves sample infrared spectrum：With multiplicative scatter correction (MSC) to tealeaves sample infrared spectrum
Pre-processed.Pretreated tealeaves infrared spectrogram is as shown in Figure 3.
Step 3: the dimension-reduction treatment of tealeaves sample infrared spectrum：Using principal component analytical method by tealeaves sample infrared light
Spectrum drops to 14 dimension datas from 1868 dimensions, and preserves these data.Feature is further extracted with linear discriminant method again, data are dropped
Two dimension is tieed up, the 2-D data is as shown in Figure 4.
Step 4: Fuzzy C-Means Clustering is to obtain initial cluster center：To the 2-D data of step 3 operation Fuzzy C-
Cluster centre after mean cluster (FCM), FCM iteration ends is initial as a kind of fuzzy Kohonen discriminatings clustering network method
Cluster centre vi,0。
Step 5: differentiating that clustering network method carries out the classification of local tea variety using a kind of fuzzy Kohonen：According to step
A kind of four initial fuzzy Kohonen of cluster centre operation differentiates that clustering network method obtains fuzzy membership, according to fuzzy
Degree of membership can be classified different cultivars tealeaves.
A kind of fuzzy Kohonen in the step 5 differentiates that clustering network method is as follows：
1. initialization
Fixed tealeaves infrared spectrum sample class number c=3 and weighted index m0=2, n are sample number, n>c>1,+∞>m0>
1, maximum iteration t is setmax=100 and value ε=0.00001 of the error upper limit, initial cluster center v is seti,0(i=1,
2,3) as shown in step 4.It is q=2 to set characteristic vector number；
2. calculating t (t=1,2 ..., tmax) secondary iteration when learning rate αik,t Wherein mt=m0-t
Δ m, t are iterations, Δ m=(m0-1)/tmax, uik,tRepresent that k-th of sample is under the jurisdiction of the i-th class during the t times iterative calculation
Fuzzy membership angle value.uik,tIt is calculated as follows：
xkFor k-th of tealeaves infrared spectrum sample, xk∈Rp, i.e. xkDimension be p.vi,tI-th during for the t times iterative calculation
Lei Lei centers, vj,tJLei Lei centers during for the t times iterative calculation.
3. calculate the t times iteration Shi Lei center vi,t：
Wherein, vi,t-1I-th Lei Lei centers during for the t-1 times iterative calculation.
4. calculate collision matrix S between fuzzy classfB, fuzzy overall collision matrix SfT；
Wherein, For the average of sample.
5. according to formula：Q characteristic vector ψ before calculating1,ψ2,...,ψq(q >=1, and p>q)
6. by sample xkProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain
xk'=xk T[ψ1,ψ2,...,ψq]
By class center vi,tProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain vi,t'=vi,t[ψ1,ψ2,...,ψq]。
If 7. maxi | | v 'i,t-v′i,t-1||<ε or t>tmax, iteration terminates, otherwise, t=t+1, by xk' value be assigned to
xk, vi,t' value be assigned to vi,t, the continuation iterative calculation of return to step 2.
Result of calculation：Fuzzy membership after iteration ends according to fuzzy membership as shown in figure 5, obtain local tea variety
Classification accuracy is 93.9%.
Those listed above is a series of to be described in detail only for feasibility embodiment of the invention specifically
Bright, they simultaneously are not used to limit the scope of the invention, all equivalent implementations made without departing from skill spirit of the present invention
Or change should be included in the scope of the protection.
Claims (5)
1. a kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network, it is characterised in that including following step
Suddenly：
Step 1: gathering the infrared spectrum of tealeaves sample under constant-temperature constant-humidity environment：For the tealeaves sample of different cultivars, with red
External spectrum instrument does collection infrared spectrum experiment to these tealeaves samples, obtains the infrared spectrum information that diffuses of tealeaves sample；
Step 2: being pre-processed to tealeaves sample infrared spectrum：Tealeaves sample infrared spectrum is carried out in advance with multiplicative scatter correction MSC
Processing；
Step 3: carrying out dimension-reduction treatment to tealeaves sample infrared spectrum：It is using principal component analytical method PCA that tealeaves sample is infrared
Spectrum drops to relatively low low-dimensional data from high dimensional data, and preserves these data, then is further carried with linear discriminant method
Take feature and dimensionality reduction；
Step 4: Fuzzy C-Means Clustering is to obtain initial cluster center：To after dimensionality reduction tealeaves infrared data operation Fuzzy C-
Mean cluster, obtains initial cluster centre；
Step 5: differentiating that clustering network method carries out the classification of local tea variety using fuzzy Kohonen：According to the initial of step 4
The fuzzy Kohonen of cluster centre operation differentiate that clustering network method obtains fuzzy membership, can be by not according to fuzzy membership
Classified with kind tealeaves.
2. a kind of fuzzy Kohonen according to claim 1 differentiates the tealeaves infrared spectrum sorting technique of clustering network, its
It is characterised by, the detailed process of the step one is as follows：
Several tealeaves samples are gathered, the tealeaves sample is ground after smashing and takes 0.5g through 40 mesh screens, each sample
Respectively 1 is pressed with KBr:Mixture 1g is taken to carry out press mold processing after 100 uniform mixing；When gathering tealeaves infrared spectrum, experiment
Room temperature and relative humidity keep constant, infrared spectrometric analyzer start preheating 1h；Spectroanalysis instrument scans each tealeaves sample
32 times, the wave-number range of spectral scan is 4001.569~401.1211cm-1, sweep spacing is set to 1.9285cm-1, Mei Gecha
The infrared spectrum of leaf sample is the high dimensional data of 1868 dimensions；Each specimen sample 3 times, takes its average value to be set up as following model
Experimental data；It is test set that every kind of sample, which chooses 22, and remaining 10 samples are training set, then number of training nrIt is set to
30；Test set is tealeaves sample to be identified, and training set is known good and bad tealeaves sample.
3. a kind of fuzzy Kohonen according to claim 2 differentiates the tealeaves infrared spectrum sorting technique of clustering network, its
It is characterised by, the infrared spectrometric analyzer uses FTIR-7600 type FTIR spectrum analyzers.
4. a kind of fuzzy Kohonen according to claim 1 differentiates the tealeaves infrared spectrum sorting technique of clustering network, its
It is characterised by, dimensionality reduction is specifically to drop to two dimension in the step 3.
5. a kind of fuzzy Kohonen according to claim 1 differentiates the tealeaves infrared spectrum sorting technique of clustering network, its
It is characterised by, the detailed process of the step 5 is as follows：
1) initialize
Fixed tealeaves infrared spectrum sample class number c=3 and weighted index m0=2, n are sample number, n>c>1,+∞>m0>1, if
Put maximum iteration tmax=100 and value ε=0.00001 of the error upper limit, initial cluster center v is seti,0(i=1,2,3),
It is q=2 to set characteristic vector number；
2) calculate t (t=1,2 ..., tmax) secondary iteration when learning rateWherein mt=m0- t Δ m,
T is iterations, Δ m=(m0-1)/tmax, uik,tRepresent that k-th of sample is under the jurisdiction of the fuzzy of the i-th class during the t times iterative calculation
It is subordinate to angle value；uik,tIt is calculated as follows：
<mrow>
<msub>
<mi>u</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>=</mo>
<msup>
<mrow>
<mo>&lsqb;</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>c</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>-</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>|</mo>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>-</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>j</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mfrac>
<mn>2</mn>
<mrow>
<msub>
<mi>m</mi>
<mi>t</mi>
</msub>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
</msup>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
<mo>,</mo>
<mo>&ForAll;</mo>
<mi>i</mi>
<mo>,</mo>
<mi>k</mi>
</mrow>
xkFor k-th of tealeaves infrared spectrum sample, xk∈Rp, i.e. xkDimension be p；vi,tI-th class during for the t times iterative calculation
Class center, vj,tJLei Lei centers during for the t times iterative calculation；
3) the t times iteration Shi Lei center v is calculatedi,t：
<mrow>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>=</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>+</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>&alpha;</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>-</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>&alpha;</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
</mrow>
</mfrac>
</mrow>
Wherein, vi,t-1I-th Lei Lei centers during for the t-1 times iterative calculation；
4) collision matrix S between fuzzy class is calculatedfB, fuzzy overall collision matrix SfT；
<mrow>
<msub>
<mi>S</mi>
<mrow>
<mi>f</mi>
<mi>B</mi>
</mrow>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>c</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>u</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
<mi>m</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>-</mo>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>v</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>-</mo>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
<mi>T</mi>
</msup>
<mo>;</mo>
<msub>
<mi>S</mi>
<mrow>
<mi>f</mi>
<mi>T</mi>
</mrow>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>c</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>u</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
<mi>m</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>-</mo>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>-</mo>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
<mi>T</mi>
</msup>
</mrow>
Wherein For the average of sample；
5) according to formula：Q characteristic vector ψ before calculating1,ψ2,...,ψq(q >=1, and p>q)；
6) by sample xkProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain：
xk'=xk T[ψ1,ψ2,...,ψq]
By class center vi,tProject to q characteristic vector ψ1,ψ2,...,ψqOn obtain vi,t'=vi,t[ψ1,ψ2,...,ψq]；
If 7) maxi | | v 'i,t-v′i,t-1||<ε or t>tmax, iteration terminates, otherwise, t=t+1, by xk' value be assigned to xk,
vi,t' value be assigned to vi,t, the continuation iterative calculation of return to step 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710342868.3A CN107271394A (en) | 2017-05-16 | 2017-05-16 | A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710342868.3A CN107271394A (en) | 2017-05-16 | 2017-05-16 | A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107271394A true CN107271394A (en) | 2017-10-20 |
Family
ID=60064489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710342868.3A Pending CN107271394A (en) | 2017-05-16 | 2017-05-16 | A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107271394A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107860739A (en) * | 2017-11-27 | 2018-03-30 | 江苏大学 | A kind of tealeaves mid-infrared light profile classification method of fuzzy K mediations network clustering |
| CN109685098A (en) * | 2018-11-12 | 2019-04-26 | 江苏大学 | The local tea variety classification method of cluster is separated between a kind of Fuzzy Cluster |
| CN110008989A (en) * | 2019-02-22 | 2019-07-12 | 华中光电技术研究所（中国船舶重工集团有限公司第七一七研究所） | The infrared spectroscopy recognition methods of different target under a kind of spectral signature condition of similarity |
| CN110057757A (en) * | 2018-01-18 | 2019-07-26 | 深圳市理邦精密仪器股份有限公司 | Identification, identification network establishing method and the device of hemoglobin and its derivative |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103646252A (en) * | 2013-12-05 | 2014-03-19 | 江苏大学 | Optimized fuzzy learning vector quantization apple classification method |
| CN105181650A (en) * | 2015-10-08 | 2015-12-23 | 滁州职业技术学院 | Method for quickly identifying tea varieties through near-infrared spectroscopy technology |
| CN106408012A (en) * | 2016-09-09 | 2017-02-15 | 江苏大学 | Tea infrared spectrum classification method of fuzzy discrimination clustering |
-
2017
- 2017-05-16 CN CN201710342868.3A patent/CN107271394A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103646252A (en) * | 2013-12-05 | 2014-03-19 | 江苏大学 | Optimized fuzzy learning vector quantization apple classification method |
| CN105181650A (en) * | 2015-10-08 | 2015-12-23 | 滁州职业技术学院 | Method for quickly identifying tea varieties through near-infrared spectroscopy technology |
| CN106408012A (en) * | 2016-09-09 | 2017-02-15 | 江苏大学 | Tea infrared spectrum classification method of fuzzy discrimination clustering |
Non-Patent Citations (1)
| Title |
|---|
| 刘增良 主编: "《模糊技术与应用选编（3）》", 31 December 1998, 北京航空航天大学出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107860739A (en) * | 2017-11-27 | 2018-03-30 | 江苏大学 | A kind of tealeaves mid-infrared light profile classification method of fuzzy K mediations network clustering |
| CN110057757A (en) * | 2018-01-18 | 2019-07-26 | 深圳市理邦精密仪器股份有限公司 | Identification, identification network establishing method and the device of hemoglobin and its derivative |
| CN109685098A (en) * | 2018-11-12 | 2019-04-26 | 江苏大学 | The local tea variety classification method of cluster is separated between a kind of Fuzzy Cluster |
| CN109685098B (en) * | 2018-11-12 | 2024-03-19 | 江苏大学 | Tea variety classification method for fuzzy inter-cluster separation and clustering |
| CN110008989A (en) * | 2019-02-22 | 2019-07-12 | 华中光电技术研究所（中国船舶重工集团有限公司第七一七研究所） | The infrared spectroscopy recognition methods of different target under a kind of spectral signature condition of similarity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103048273B (en) | Fruit near infrared spectrum sorting method based on fuzzy clustering | |
| CN107271394A (en) | A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network | |
| Zayas et al. | Discrimination of wheat and nonwheat components in grain samples by image analysis | |
| CN104990892B (en) | The spectrum picture Undamaged determination method for establishing model and seeds idenmtification method of seed | |
| CN108734205A (en) | A kind of simple grain for different cultivars wheat seed pinpoints identification technology | |
| CN105181650B (en) | A method of quickly differentiating local tea variety using near-infrared spectrum technique | |
| CN106408012A (en) | Tea infrared spectrum classification method of fuzzy discrimination clustering | |
| CN106680241A (en) | Novel spectrum multi-analysis classification and identification method and application thereof | |
| CN110378374A (en) | A kind of tealeaves near infrared light profile classification method that fuzzy authentication information extracts | |
| CN104374739A (en) | Identification method for authenticity of varieties of seeds on basis of near-infrared quantitative analysis | |
| CN110108644A (en) | A kind of maize variety identification method based on depth cascade forest and high spectrum image | |
| CN105548066A (en) | Method and system for distinguishing colloid types | |
| CN103822897A (en) | White spirit appraising and source-tracing method based on infrared spectroscopy | |
| CN102706813A (en) | Poa pratensis variety identification method based on hyper-spectral image | |
| CN108764288A (en) | A kind of GK differentiates the local tea variety sorting technique of cluster | |
| CN107132194A (en) | A kind of pseudo-ginseng and its adulterant discrimination method based on uv-vis spectra and Chemical Pattern Recognition | |
| CN107341807A (en) | A kind of method for extracting tobacco leaf color digital expression characteristic value | |
| CN109685098A (en) | The local tea variety classification method of cluster is separated between a kind of Fuzzy Cluster | |
| CN107192686A (en) | A kind of Possibility Fuzzy Clustering local tea variety discrimination method of fuzzy covariance matrix | |
| CN110533102A (en) | Single class classification method and classifier based on fuzzy reasoning | |
| CN107886115A (en) | A kind of tealeaves mid-infrared light profile classification method of adaptively possible C mean clusters | |
| CN104990891B (en) | A kind of seed near infrared spectrum and spectrum picture qualitative analysis model method for building up | |
| CN106290656A (en) | The method that finger printing differentiates hardwood nanmu is set up based on GC MS technology | |
| CN106570520A (en) | Infrared spectroscopy tea quality identification method mixed with GK clustering | |
| Ji et al. | Apple color automatic grading method based on machine vision |
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: 20171020 |
|
| RJ01 | Rejection of invention patent application after publication |