CN106092957A - The near infrared spectrum recognition methods of mahogany furniture - Google Patents
The near infrared spectrum recognition methods of mahogany furniture Download PDFInfo
- Publication number
- CN106092957A CN106092957A CN201610384194.9A CN201610384194A CN106092957A CN 106092957 A CN106092957 A CN 106092957A CN 201610384194 A CN201610384194 A CN 201610384194A CN 106092957 A CN106092957 A CN 106092957A
- Authority
- CN
- China
- Prior art keywords
- spectrum
- near infrared
- furniture
- sample
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 27
- 241000158728 Meliaceae Species 0.000 title claims abstract description 25
- 238000001228 spectrum Methods 0.000 claims abstract description 44
- 240000005020 Acaciella glauca Species 0.000 claims description 17
- 235000003499 redwood Nutrition 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 13
- 238000009499 grossing Methods 0.000 claims description 11
- 239000011121 hardwood Substances 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 9
- 238000009795 derivation Methods 0.000 claims description 8
- 230000008030 elimination Effects 0.000 claims description 6
- 238000003379 elimination reaction Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 241001062570 Pterocarpus angolensis Species 0.000 claims description 3
- 241001544576 Pterocarpus macrocarpus Species 0.000 claims description 3
- 239000002253 acid Substances 0.000 claims description 3
- 244000138993 panchioli Species 0.000 claims description 2
- 238000002360 preparation method Methods 0.000 claims description 2
- 230000002349 favourable effect Effects 0.000 abstract description 3
- 230000003595 spectral effect Effects 0.000 description 13
- 239000010410 layer Substances 0.000 description 11
- 230000008859 change Effects 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 7
- 239000000126 substance Substances 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 239000000203 mixture Substances 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000002023 wood Substances 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 239000011229 interlayer Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000002512 chemotherapy Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012850 discrimination method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 229920005610 lignin Polymers 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 240000001606 Adenanthera pavonina Species 0.000 description 1
- 235000011470 Adenanthera pavonina Nutrition 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 229920002488 Hemicellulose Polymers 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 229920002678 cellulose Polymers 0.000 description 1
- 239000001913 cellulose Substances 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000491 multivariate analysis Methods 0.000 description 1
- 238000010298 pulverizing process Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000001845 vibrational spectrum 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
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention proposes the near infrared spectrum recognition methods of a kind of mahogany furniture, including furniture sample prepare near infrared spectrum gather selecting and the step such as the checking setting up discrimination model extracting discrimination model of Pretreated spectra characteristic wavelength.The present invention uses the near infrared light spectrum signal of miniature near infrared spectra collection sample, differentiate with chemometrics method, by setting up the discrimination model of true and false mahogany furniture, the quick nondestructive utilizing the true and false mahogany furniture of model realization built up differentiates, has the various features such as lossless, efficient, quick, portable, low cost, favorable reproducibility and accuracy rate height.
Description
Technical field
The present invention relates to spectral discrimination method, be specifically related to the near infrared spectrum recognition methods of a kind of mahogany furniture.
Background technology
Rare timber kind is various, and different types of timber price is widely different, causes the furniture price after making finished product
Also there is huge difference, wherein the most expensive with mahogany furniture.Mahogany furniture has the strongest Chinese culture intension, and China is " red
Wood " national standard (GB/T 18107-2000), 5 genus 33 kinds of rare timber of 8 class are classified as redwood.Mahogany furniture is as high-grade family
The synonym of tool, integrates collectibles, consumer goods, investment goods, adds that raw material is the most rare, and its price soars all the way.Redwood
When the river rises the boat goes up for furniture market, and many bad businessmans, in order to earn profit, use and have the low and middle-grade rare of similar features with redwood
Timber pretends to be high-grade redwood, or other redwood inferior that adulterate in mahogany furniture, severe jamming market order, compromises wide
The interests of big consumer.
Common redwood discrimination method mainly relies on special instrument or has the expert of abundant industry experience to put into extensive work
Just can complete.These methods have the shortcomings such as time-consuming, cost is high, destructive, chemical reagent pollutes, operation is complicated, are only suitable for reality
Test room to use, and the Study on Identification of the most a lot of timber is all built upon sample is carried out the processed such as pulverizing, stripping and slicing
On the basis of, test format mostly is censorship or sampling observation, and coverage rate is greatly limited, and is difficult to ensure that whole finished product in actual applications
Furniture true and false.The near-infrared spectrum technique being combined with Chemical Measurement in recent years develops in terms of log detection with timber identification
Rapidly, it is lossless, efficient, quick, low cost, favorable reproducibility, be easy to the unique advantages such as on-line checking, differentiates to provide for timber
A kind of new detection technique.At present, have numerous studies both at home and abroad and show that near-infrared spectrum technique differentiates the feasibility of timber, but
Research is not the most deep into and quickly differentiates the near-infrared of some confusing rare wood furnitures in Chinese market.
Near infrared region refers to wavelength electromagnetic wave in the range of 780～2526 nm by ASTM definition.Near infrared spectrum belongs to
Molecular vibration spectrum, results from the vibration of covalent chemical bond anharmonic energy level, is frequency multiplication and the combination frequency of anharmonic vibration, is positioned at visible ray
And between mid-infrared light district, it is adaptable to measuring the material containing groups such as C-H, N-H, O-H, the spectrum produced due to different groups is inhaled
Receiving peak position and intensity is the most different, according to Lambert-Beer absorption law, absorption spectrum can be along with sample composition composition or knot
Fruit change and produce change.
Near-infrared spectrum technique mainly includes near infrared spectrometer, chemo metric software and multivariate calibration model etc..Closely
Infrared spectrometer is the equipment for gathering sample near infrared spectrum, and chemo metric software is for associating spectrum and sample
The instrument of moral character matter, and calibration model is for reflecting that between sample spectra and character, the quantitative or qualitative work of corresponding relation is bent
Line.The chemometrics method of near-infrared spectrum technique relates generally to three aspect contents: one is preprocessing procedures research, right
Sample spectrum carries out pretreatment, reduces to such an extent as to eliminates the impact that spectrum is caused by various Aimless factors；Two is spectral signature
The selection of wavelength and extraction, extract the information relevant with class object selectively and suppress the shadow of irrelevant feature and noise
Ring；Three is the research of near infrared light spectrum correction method, to setting up sane, reliable, highly sensitive calibration model.
In the initial data of near infrared spectrometer collection in addition to comprising the information relevant to simple chemical structure, also wrap simultaneously
Containing spectrogram information being produced the noise signal disturbed, thus affect the foundation of model and the prediction to unknown sample.Therefore, adopt
Carry out noise elimination by smoothing processing, improve signal to noise ratio；Method of Seeking Derivative is used to carry out baseline correction process；Use orthogonal signalling school
Just carry out spectral filtering, delete system change undesirable in data, to improve the performance of model.
During near-infrared spectrum analysis, when using complete wavelength range to set up model, computationally intensive, calculate speed
Slowly.And owing to the spectral information at some spectral region sample is the most weak or uncorrelated with the composition of sample or character, introduce this
The variable of sample can cause the precision reduction of institute's established model even to make a mistake.The selection of characteristic wavelength can simplify mould with extracting
Type, more importantly due to the rejecting of uncorrelated or non-linear variable, prediction rate and the stability of calibration model improve.Without letter
It is a kind of wavelength extraction algorithm set up based on PLS coefficient that breath variable eliminates, for eliminating without information variable
Method.The method carrys out the reliability of each variable in evaluation model by introducing stability value, thus chooses.The method is
Being widely used in the selection of spectral variables, wherein Monte Carlo is best without information variable removing method effect.Successive projection is calculated
Method is to find the set of variables containing MIN redundancy in light spectrum matrix so that the synteny between variable reaches
Little, farthest decrease information overlap, simplify mathematical model, be widely used in the extraction of sample wavelength.
In the actual application of near-infrared spectral analysis technology, the most only need to know kind or the place of production etc. of sample, not
It is to be appreciated that the constituent content in sample, i.e. qualitative analysis problem, at this moment need the pattern-recongnition method using in Chemical Measurement.
Extreme learning machine randomly generates input layer and the connection weights of implicit interlayer and the threshold value of hidden layer neuron, in the training process
Only the number of neuron in hidden layer need to be set, just can obtain unique optimal solution.Compared with traditional training method, the party
Calligraphy learning speed is fast, Generalization Capability good.
To sum up, it is achieved mahogany furniture main on domestic market and the quick discriminating of the famous and precious hardwood furniture of non-redwood, to guarantor
Protect consumer's interests, promote that the sound development in mahogany furniture market has important function.
Summary of the invention
The technical problem to be solved is to provide the near infrared spectrum recognition methods of a kind of mahogany furniture, and utilizing should
Method can identify the true and false of mahogany furniture exactly.
For solving above-mentioned technical problem, the technical solution used in the present invention is: the near infrared spectrum of a kind of mahogany furniture
Recognition methods, including following methods step:
(1). the preparation of furniture sample: selecting mahogany furniture and the non-redwood famous and precious hardwood furniture sample of standard, sample includes Lignum Santali Albi
Lignum pterocarpi indici, Barry yellow wingceltis, toe yellow wingceltis, Burma padauk, Dalbergia louvelii, nick yellow wingceltis, Sulawesi Scobis Diospyroris Ebeni, kiaat,
Africa acid branch, these 10 kinds of furniture samples of hardwood nanmu；
(2). the collection of near infrared spectrum: utilize miniature near infrared spectrometer to gather near infrared light at the diverse location of sample surface
Spectrum, including the diverse location of every furniture, spectrum is converted into a spectrum after average and represents a sample；Gather spectrum model
It is trapped among 1000nm 1650nm；
(3). Pretreated spectra: use Savitzky-Golay convolution smoothing techniques smooth high frequencies noise, improves signal to noise ratio, uses
The spectrogram that the Savitzky-Golay convolution method of derivation near infrared spectrum to gathering occurs offsets or drift phenomenon, and spectrum spectrum
Line overlap phenomenon carries out baseline correction, uses Orthogonal Signal Correction Analyze to eliminate part unrelated with predictive value in spectrum；
(4). the selection of characteristic wavelength and extraction: use Monte Carlo to carry out feature without information variable method of elimination and successive projection method
The selection of wavelength and extraction；
(5). the foundation of discrimination model: select 2/3rds quantity respectively from mahogany furniture and the famous and precious hardwood furniture of non-redwood
Sample constitutes the training set sample modeled, preprocessing procedures when modeling differentiates with reality and characteristic wavelength point necessary
Cause, set up extreme learning machine model respectively；
(6). the checking of discrimination model: residue 1/3rd number of samples are verified by the model of call establishment, it determines accuracy
Close to 100%.
The Savitzky-Golay convolution smoothing techniques of the present invention, smoothing processing is to eliminate a kind of method that noise is the most frequently used,
Its basic assumption be the noise that spectrum contains be zero equal random white noise, if repetitive measurement is averaged just can effectively smooth height
Frequently noise, improves signal to noise ratio.Conventional signal smoothing method has rolling average smoothing techniques and Savitzky-Golay convolution to smooth
Method.The present invention uses Savitzky-Golay convolution smoothing techniques, compares rolling average smoothing techniques, Savitzky-Golay convolution
Smoothing techniques is, by multinomial, the data in moving window are carried out polynomial least mean square fitting, is substantially a kind of weighting
Averaging method, more emphasizes central role of central point.
The Savitzky-Golay convolution method of derivation of the present invention, owing to instrument, sample background and other factors affect, gathers
Near infrared spectrum spectrogram skew or drift phenomenon often occur, and light is caused for interfering between sample different component
The phenomenon of spectrum overlap of spectral lines, can use the method for derivation to carry out baseline correction process.Conventional spectrum Method of Seeking Derivative typically has two
Kind: direct differential method and Savitzky-Golay convolution method of derivation.The spectrum high for resolution, wavelength sampled point is many, uses
Spectrum after direct differential method derivation is more or less the same with actual, but for the few spectrum of wavelength sampled point, required by the method
Derivative Error is relatively big, and Savitzky-Golay convolution method of derivation therefore can be used to calculate.
The Orthogonal Signal Correction Analyze (Orthogonal Signal Correction, OSC) of the present invention is Wold in 1998
Deng a kind of preprocessing procedures of proposition, its basic thought is to utilize orthogonalization method, eliminates in spectrum unrelated with predictive value
Part, thus obtain the spectrum of " pure ".OSC is mainly used in spectral filtering, deletes system change undesirable in data such as
Baseline drifts etc., to improve the performance of model.
The Monte Carlo of the present invention is without information variable method of elimination, and it is a kind of based on offset minimum binary for eliminating without information variable
The wavelength extraction algorithm that (Partial Least Squares, PLS) regression coefficient is set up, for eliminating the side without information variable
Method.The method carrys out the reliability of each variable in evaluation model by introducing stability value, thus chooses.The method by
Being widely used in the selection of spectral variables, wherein Monte Carlo is best without information variable elimination (MC-UVE) method effect.
The successive projection method (Successive Projections Algorithm, SPA) of the present invention is at light spectrum matrix
Middle searching contains the set of variables of MIN redundancy so that the synteny between variable minimizes, farthest
Decrease information overlap, simplify mathematical model, be widely used in the extraction of sample wavelength.
Extreme learning machine model of the present invention, extreme learning machine (Extreme Learning Machine, ELM) produces at random
Raw input layer connects weights and the threshold value of hidden layer neuron with implicit interlayer, the most only need to arrange in hidden layer
The number of neuron, just can obtain unique optimal solution.Compared with traditional training method, the method pace of learning is fast, general
Change performance good.
The present invention uses the near infrared spectrum recognition methods of the mahogany furniture designed by technique scheme, red by gathering
Wooden furniture and non-redwood famous and precious hardwood furniture sample, utilize miniature near infrared spectrometer at the not coordination of every furniture sample surface
Put collection near infrared spectrum, after Pretreated spectra, extract characteristic wavelength, by multivariate data analysis method, set up respectively
The discrimination model of true and false mahogany furniture, utilizes the quick nondestructive of the true and false mahogany furniture of model realization built up to differentiate.The present invention carries
The method of confession has lossless, efficient, quick, portable, low cost, favorable reproducibility and accuracy rate high, it is adaptable to redwood man
Tool kind differentiates.
Accompanying drawing explanation
Fig. 1 represents 10 kinds of mahogany furnitures and the near infrared light spectrogram of the famous and precious hardwood furniture of non-redwood；
Fig. 2 represents present invention characteristic wavelength scattergram after characteristic wavelength selects and extracts.
Detailed description of the invention
The near infrared spectrum recognition methods of mahogany furniture of the present invention, including following methods:
(1). use miniature near infrared spectrometer to gather red sandalwood, Barry yellow wingceltis, toe yellow wingceltis, Burma padauk, the black Huang of Lushi
Wingceltis, nick yellow wingceltis, Sulawesi Scobis Diospyroris Ebeni, kiaat, Africa acid branch, the spectrogram of 10 kinds of rare wood furnitures of hardwood nanmu are also protected
Deposit.Diverse location in sample surface gathers near infrared spectrum, and including the diverse location of every furniture, spectrum turns after average
Turn to a spectrum and represent a sample；Collection spectral region is at 1000nm 1650nm, and the spectral information in this region is main
Relevant with redwood and the structure of non-redwood hardwood and chemical composition, including cellulose, hemicellulose, lignin (lignin) and extracting
Thing etc.；As shown in Figure 1.
(2). by the relatively different preprocessing procedures impact on modeling accuracy, the optimum preprocess method of final selection
For Savitzky-Golay smooth (SG smooths), Savitzky-Golay1 order derivative (SG1 leads on rank) and Orthogonal Signal Correction Analyze.Former
Beginning spectrum will not have bigger change after SG smooths pretreatment to spectrum, but noise substantially weakens.Lead pre-through the smooth+SG1 rank of SG
After process, overlapping absorbance peak spectrally is exaggerated, after+OSC pretreatment is led on the smooth+SG1 rank of SG, between one species sample
Obvious difference reduce.
In the initial data of near infrared spectrometer collection in addition to comprising the information relevant to simple chemical structure, also wrap simultaneously
Containing spectrogram information being produced the noise signal disturbed, thus affect the foundation of model and the prediction to unknown sample.Therefore, light
Modal data pretreatment is mainly used in garbled data, eliminates noise and the impact on data message of other factors, for calibration model
Set up and the Accurate Prediction of unknown sample lays the first stone.Use the Savitzky-Golay convolution smoothing techniques can smooth high frequencies effectively
Noise, improves signal to noise ratio.Owing to instrument, sample background and other factors affect, often there is spectrogram in the near infrared spectrum of collection
Skew or drift phenomenon, and for the phenomenon interfering derivative spectomstry overlap of spectral lines between sample different component, can use
Savitzky-Golay convolution method of derivation is carried out at baseline correction.Orthogonal Signal Correction Analyze (Orthogonal Signal
Correction, OSC) it is exactly to utilize orthogonalization method, eliminate part unrelated with predictive value in spectrum, thus obtain " pure
Only " spectrum.OSC is mainly used in spectral filtering, deletes system change such as baseline drift etc. undesirable in data, to improve
The performance of model.
(3). after characteristic wavelength is extracted without information variable method of elimination and successive projection method in Monte Carlo, number of features is nearly
Reduce 97%, significantly have compressed wave band quantity, reduce the amount of storage of spectroscopic data, improve model and calculate speed, such as Fig. 2 institute
Show.
(4). every kind of furniture is set up extreme learning machine model, and extreme learning machine model hidden layer excitation function selects
" Sigmoidal " function, hidden layer neuron number is 70, and calibration set accuracy is 100%, and checking collection accuracy is 100%.Its
In, extreme learning machine (Extreme Learning Machine, ELM) randomly generates the connection weight of input layer and implicit interlayer
Value and the threshold value of hidden layer neuron, the most only need to arrange the number of neuron in hidden layer, just can obtain only
The optimal solution of one.Compared with traditional training method, the method pace of learning is fast, Generalization Capability good." Sigmoidal " function
(sigmoidal founction) if be f (x) with initial point as point symmetry, then have f (-x)=-f (x), when select " Sigmoidal "
During function, Model checking performance is more stable, it determines precision is the highest.
The present invention uses the near infrared light spectrum signal of miniature near infrared spectra collection sample, carries out with chemometrics method
Differentiate.First gather a number of sample spectrum for modeling, spectrum is carried out pretreatment the most again, extracts characteristic wavelength,
The independent soft type method of rear employing offset minimum binary diagnostic method, bunch class, support vector machine and extreme learning machine are modeled point respectively
By experiment, analysis, differentiates that result determines optimum modeling method.The present invention is not limited to the rare wood furniture of described above ten kind
Sample, the rare wood furniture sample for other kind is equally applicable.
Claims (1)
1. the near infrared spectrum recognition methods of a mahogany furniture, it is characterised in that: include following methods step:
(1). the preparation of furniture sample: selecting mahogany furniture and the non-redwood famous and precious hardwood furniture sample of standard, sample includes Lignum Santali Albi
Lignum pterocarpi indici, Barry yellow wingceltis, toe yellow wingceltis, Burma padauk, Dalbergia louvelii, nick yellow wingceltis, Sulawesi Scobis Diospyroris Ebeni, kiaat,
Africa acid branch, these 10 kinds of furniture samples of hardwood nanmu；
(2). the collection of near infrared spectrum: utilize miniature near infrared spectrometer to gather near infrared light at the diverse location of sample surface
Spectrum, including the diverse location of every furniture, spectrum is converted into a spectrum after average and represents a sample；Gather spectrum model
It is trapped among 1000nm 1650nm；
(3). Pretreated spectra: use Savitzky-Golay convolution smoothing techniques smooth high frequencies noise, improves signal to noise ratio, uses
The spectrogram that the Savitzky-Golay convolution method of derivation near infrared spectrum to gathering occurs offsets or drift phenomenon, and spectrum spectrum
Line overlap phenomenon carries out baseline correction, uses Orthogonal Signal Correction Analyze to eliminate part unrelated with predictive value in spectrum；
(4). the selection of characteristic wavelength and extraction: use Monte Carlo to carry out feature without information variable method of elimination and successive projection method
The selection of wavelength and extraction；
(5). the foundation of discrimination model: select 2/3rds quantity respectively from mahogany furniture and the famous and precious hardwood furniture of non-redwood
Sample constitutes the training set sample modeled, preprocessing procedures when modeling differentiates with reality and characteristic wavelength point necessary
Cause, set up extreme learning machine model respectively；
(6). the checking of discrimination model: residue 1/3rd number of samples are verified by the model of call establishment, it determines accuracy
Close to 100%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610384194.9A CN106092957A (en) | 2016-06-02 | 2016-06-02 | The near infrared spectrum recognition methods of mahogany furniture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610384194.9A CN106092957A (en) | 2016-06-02 | 2016-06-02 | The near infrared spectrum recognition methods of mahogany furniture |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106092957A true CN106092957A (en) | 2016-11-09 |
Family
ID=57448094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610384194.9A Pending CN106092957A (en) | 2016-06-02 | 2016-06-02 | The near infrared spectrum recognition methods of mahogany furniture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106092957A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106525756A (en) * | 2016-12-02 | 2017-03-22 | 赣州市检验检疫科学技术研究院 | Apparatus and method for rapid identification of timber varieties |
| CN106770003A (en) * | 2016-11-21 | 2017-05-31 | 无锡迅杰光远科技有限公司 | Wood Identification Method and system based on near-infrared spectrum technique |
| CN107091815A (en) * | 2017-05-05 | 2017-08-25 | 张方达 | A kind of method for identifying rosewood |
| CN107748144A (en) * | 2017-11-13 | 2018-03-02 | 中国科学院昆明植物研究所 | The middle infrared spectrum detecting system of quick measure SOIL CARBON AND NITROGEN and its stable isotope |
| CN108645809A (en) * | 2018-06-27 | 2018-10-12 | 广西民族大学 | A kind of method that near-infrared spectrum technique quickly identifies rosin original tree species |
| CN108956528A (en) * | 2018-08-01 | 2018-12-07 | 浙江农林大学 | The near-infrared Undamaged determination method of mahogany furniture |
| CN113392586A (en) * | 2021-06-10 | 2021-09-14 | 闽江学院 | Vegetable oil identification method based on R language and orthogonal partial least square discriminant analysis |
| CN113720797A (en) * | 2021-08-30 | 2021-11-30 | 四川轻化工大学 | Online rapid quality-measuring liquor taking method for liquor distillation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1936552A (en) * | 2006-10-12 | 2007-03-28 | 中国林业科学研究院木材工业研究所 | Method for identifying red sandalwood by near-infrared ray |
| CN101299041A (en) * | 2008-06-10 | 2008-11-05 | 中国林业科学研究院木材工业研究所 | Method for quickly predicting rattan cane anatomical property |
| CN103900971A (en) * | 2014-04-02 | 2014-07-02 | 云南大学 | Method for distinguishing interrelation between pinus yunnanensis and sibling species by using FT-NIR (Fourier Transform Near-Infrared Spectroscopy) |
| CN105136738A (en) * | 2015-09-29 | 2015-12-09 | 中国林业科学研究院林产化学工业研究所 | Near-infrared-based method for identifying tree varieties ranging from eucalyptus-category tree varieties to acacia-mangium-category tree varieties |
-
2016
- 2016-06-02 CN CN201610384194.9A patent/CN106092957A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1936552A (en) * | 2006-10-12 | 2007-03-28 | 中国林业科学研究院木材工业研究所 | Method for identifying red sandalwood by near-infrared ray |
| CN101299041A (en) * | 2008-06-10 | 2008-11-05 | 中国林业科学研究院木材工业研究所 | Method for quickly predicting rattan cane anatomical property |
| CN103900971A (en) * | 2014-04-02 | 2014-07-02 | 云南大学 | Method for distinguishing interrelation between pinus yunnanensis and sibling species by using FT-NIR (Fourier Transform Near-Infrared Spectroscopy) |
| CN103900971B (en) * | 2014-04-02 | 2015-06-03 | 云南大学 | Method for distinguishing interrelation between pinus yunnanensis and sibling species by using FT-NIR (Fourier Transform Near-Infrared Spectroscopy) |
| CN105136738A (en) * | 2015-09-29 | 2015-12-09 | 中国林业科学研究院林产化学工业研究所 | Near-infrared-based method for identifying tree varieties ranging from eucalyptus-category tree varieties to acacia-mangium-category tree varieties |
Non-Patent Citations (1)
| Title |
|---|
| 寿国忠: "《现代农林业精细化管理》", 31 January 2016 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106770003A (en) * | 2016-11-21 | 2017-05-31 | 无锡迅杰光远科技有限公司 | Wood Identification Method and system based on near-infrared spectrum technique |
| CN106525756A (en) * | 2016-12-02 | 2017-03-22 | 赣州市检验检疫科学技术研究院 | Apparatus and method for rapid identification of timber varieties |
| CN107091815A (en) * | 2017-05-05 | 2017-08-25 | 张方达 | A kind of method for identifying rosewood |
| CN107748144A (en) * | 2017-11-13 | 2018-03-02 | 中国科学院昆明植物研究所 | The middle infrared spectrum detecting system of quick measure SOIL CARBON AND NITROGEN and its stable isotope |
| CN108645809A (en) * | 2018-06-27 | 2018-10-12 | 广西民族大学 | A kind of method that near-infrared spectrum technique quickly identifies rosin original tree species |
| CN108956528A (en) * | 2018-08-01 | 2018-12-07 | 浙江农林大学 | The near-infrared Undamaged determination method of mahogany furniture |
| CN113392586A (en) * | 2021-06-10 | 2021-09-14 | 闽江学院 | Vegetable oil identification method based on R language and orthogonal partial least square discriminant analysis |
| CN113392586B (en) * | 2021-06-10 | 2022-06-14 | 闽江学院 | Vegetable oil identification method based on R language and orthogonal partial least square discriminant analysis |
| CN113720797A (en) * | 2021-08-30 | 2021-11-30 | 四川轻化工大学 | Online rapid quality-measuring liquor taking method for liquor distillation |
| CN113720797B (en) * | 2021-08-30 | 2024-02-09 | 四川轻化工大学 | Online rapid quality measuring and liquor picking method for distilled liquor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106092957A (en) | The near infrared spectrum recognition methods of mahogany furniture | |
| Vitale et al. | A rapid and non-invasive method for authenticating the origin of pistachio samples by NIR spectroscopy and chemometrics | |
| Barnaba et al. | Portable NIR‐AOTF spectroscopy combined with winery FTIR spectroscopy for an easy, rapid, in‐field monitoring of Sangiovese grape quality | |
| CN104990894B (en) | A kind of gasoline property detection method based on weighting absorbance and similar sample | |
| CN107748146A (en) | A kind of crude oil attribute method for quick predicting based near infrared spectrum detection | |
| CN103278473B (en) | The mensuration of pipering and moisture and method for evaluating quality in white pepper | |
| CN105806824A (en) | Quick edible oil variety identification method based on multisource spectral feature fusion | |
| CN109520962A (en) | A kind of grape wine near infrared spectrum detection method | |
| CN101995389A (en) | Method for fast recognition of crude oil variety through near infrared spectrum | |
| CN105424641A (en) | Crude oil type near infrared spectrum identification method | |
| CN104132896A (en) | Method for rapidly identifying adulterated meat | |
| CN104251839A (en) | Spectrum separation detection method of compositions of south-Xinjiang red date sample for south-Xinjiang red date modeling | |
| CN106770003A (en) | Wood Identification Method and system based on near-infrared spectrum technique | |
| Ma et al. | The rapid determination of total polyphenols content and antioxidant activity in Dendrobium officinale using near-infrared spectroscopy | |
| CN102937575B (en) | Watermelon sugar degree rapid modeling method based on secondary spectrum recombination | |
| CN102841069B (en) | Method for rapidly identifying types of crude oil by using mid-infrared spectrum | |
| CN105548027A (en) | Analytical model and method for determining content of tea oil in blend oil based on near infrared spectroscopy | |
| CN103645144A (en) | Method for quantitatively analyzing components in white spirit | |
| CN105528580A (en) | Hyperspectral curve matching method based on absorption peak characteristic | |
| Lei et al. | Geographic origin identification of coal using near-infrared spectroscopy combined with improved random forest method | |
| CN107402192A (en) | A kind of method of quick analysis essence and flavoring agent quality stability | |
| Li et al. | Multi-information based on ATR-FTIR and FT-NIR for identification and evaluation for different parts and harvest time of Dendrobium officinale with chemometrics | |
| CN106226267B (en) | A kind of near-infrared assay method of dry chili color value | |
| CN111523587A (en) | Woody plant species spectrum identification method based on machine learning | |
| CN103308637A (en) | Gas chromatography-mass spectrometry method for identifying dalbergia odorifera and dalbergia tonkinensi |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170111 Address after: 311300 Ling'an, Hangzhou, North Ring Road, No. 88, No. Applicant after: Zhejiang A&F University Applicant after: Near Hangzhou, photoelectric technology Co. Ltd. Address before: 311300 Ling''an, Hangzhou, North Ring Road, No. 88, No. Applicant before: Zhejiang A&F University |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161109 |