CN105445218A - Establishing method of self-adaptive model for detection of content of protein of rapeseeds on basis of mid-infrared spectrum - Google Patents
Establishing method of self-adaptive model for detection of content of protein of rapeseeds on basis of mid-infrared spectrum Download PDFInfo
- Publication number
- CN105445218A CN105445218A CN201510828543.7A CN201510828543A CN105445218A CN 105445218 A CN105445218 A CN 105445218A CN 201510828543 A CN201510828543 A CN 201510828543A CN 105445218 A CN105445218 A CN 105445218A
- Authority
- CN
- China
- Prior art keywords
- model
- spectrum
- sample
- rapeseed
- protein content
- 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
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 18
- 238000001514 detection method Methods 0.000 title claims abstract description 12
- 240000002791 Brassica napus Species 0.000 title abstract description 53
- 102000004169 proteins and genes Human genes 0.000 title abstract description 6
- 108090000623 proteins and genes Proteins 0.000 title abstract description 6
- 235000011293 Brassica napus Nutrition 0.000 title abstract 3
- 235000004977 Brassica sinapistrum Nutrition 0.000 claims abstract description 65
- 238000001228 spectrum Methods 0.000 claims abstract description 44
- 239000000126 substance Substances 0.000 claims abstract description 23
- 244000188595 Brassica sinapistrum Species 0.000 claims abstract 15
- 230000003044 adaptive effect Effects 0.000 claims description 26
- 238000004611 spectroscopical analysis Methods 0.000 claims description 25
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 16
- 230000003595 spectral effect Effects 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 230000015572 biosynthetic process Effects 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 8
- 229910052757 nitrogen Inorganic materials 0.000 claims description 8
- 238000003786 synthesis reaction Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 239000000835 fiber Substances 0.000 claims description 4
- 238000004476 mid-IR spectroscopy Methods 0.000 claims description 4
- 239000011521 glass Substances 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000001834 photoacoustic spectrum Methods 0.000 abstract description 8
- 239000003153 chemical reaction reagent Substances 0.000 abstract description 3
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 230000008030 elimination Effects 0.000 abstract 1
- 238000003379 elimination reaction Methods 0.000 abstract 1
- 238000009499 grossing Methods 0.000 abstract 1
- 238000012821 model calculation Methods 0.000 abstract 1
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000012163 sequencing technique Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 8
- MHAJPDPJQMAIIY-UHFFFAOYSA-N Hydrogen peroxide Chemical compound OO MHAJPDPJQMAIIY-UHFFFAOYSA-N 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000004867 photoacoustic spectroscopy Methods 0.000 description 5
- QAOWNCQODCNURD-UHFFFAOYSA-N Sulfuric acid Chemical compound OS(O)(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-N 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000003921 oil Substances 0.000 description 4
- 235000019198 oils Nutrition 0.000 description 4
- 239000002689 soil Substances 0.000 description 4
- HEMHJVSKTPXQMS-UHFFFAOYSA-M Sodium hydroxide Chemical compound [OH-].[Na+] HEMHJVSKTPXQMS-UHFFFAOYSA-M 0.000 description 3
- KGBXLFKZBHKPEV-UHFFFAOYSA-N boric acid Chemical compound OB(O)O KGBXLFKZBHKPEV-UHFFFAOYSA-N 0.000 description 3
- 239000004327 boric acid Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 235000013311 vegetables Nutrition 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 238000009835 boiling Methods 0.000 description 2
- 239000008367 deionised water Substances 0.000 description 2
- 229910021641 deionized water Inorganic materials 0.000 description 2
- 239000001307 helium Substances 0.000 description 2
- 229910052734 helium Inorganic materials 0.000 description 2
- SWQJXJOGLNCZEY-UHFFFAOYSA-N helium atom Chemical compound [He] SWQJXJOGLNCZEY-UHFFFAOYSA-N 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 239000001117 sulphuric acid Substances 0.000 description 2
- 235000011149 sulphuric acid Nutrition 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 240000008042 Zea mays Species 0.000 description 1
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 1
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 150000001413 amino acids Chemical class 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 235000014121 butter Nutrition 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 238000009614 chemical analysis method Methods 0.000 description 1
- 238000009841 combustion method Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 235000005822 corn Nutrition 0.000 description 1
- 230000029087 digestion Effects 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 239000008157 edible vegetable oil Substances 0.000 description 1
- 235000020776 essential amino acid Nutrition 0.000 description 1
- 239000003797 essential amino acid Substances 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 238000009396 hybridization Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004570 mortar (masonry) Substances 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000010926 purge Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 230000002522 swelling effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000004448 titration Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000005406 washing 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
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
An establishing method of a self-adaptive model for detection of content of protein of rapeseeds on the basis of a mid-infrared spectrum comprises steps as follows: I, mid-infrared photoacoustic spectrum information is sampled and acquired, continuous scanning is performed for multiple times, and an average spectrum is obtained; II, the content of the protein of the rapeseeds is chemically analyzed, and a chemical reference value is obtained; III, noise elimination, smoothing and standardized preprocessing are performed on the spectrum information; IV, spectrum data are calculated with an Euclidean distance method, and sequencing from small to large is performed; V, a prediction model based on the spectrum information is established; VI, a mid-infrared photoacoustic spectrum of to-be-tested samples is taken into the model, the content of the protein of the to-be-tested rapeseed samples is calculated and predicted, and a predicted value is obtained. The method is high in sensitivity, simple to operate and nondestructive to samples, the content of required samples is small, no chemical reagent is used, the method has the universality, the prediction result is accurate, invalid information interference of the samples can be effectively reduced, a sample modeling set can be optimized and simplified, and model calculation is quick and accurate.
Description
Technical field
The invention belongs to agriculture test method field, be specifically related to a kind of based on the method for building up of middle infrared spectrum to quick, accurate, the Non-Destructive Testing adaptive model of rapeseed protein content.
Background technology
Rape is one of large oil crops in the world four, and China is rape seed area 7,000,000 hectares throughout the year, and total product 1,200 ten thousand tons, accounts for other major oil vegetable seed producing countries of more than 60%, the Yuan Chao world of national edible vegetable oil total production.Rape-seed production occupies extremely important status in Chinese national economy, and the examination and controlling of rapeseed nutritional quality is that the important technology that china rape is produced supports.Nitrogen content is the important quality index of rapeseed, in dregs of rapeseed cake after oil expression, protein content reaches 35%-39%, compare corn, wheat exceeds 3-4 doubly, and Amino Acids in Proteins ratio of components is more reasonable, essential amino acids content is higher, is the important sources of feed protein.Thus, carry out quick, accurate and harmless mensuration to protein content in rapeseed, quality and the rape cake feedstuff quality of assessment rapeseed are significant
The detection method of traditional protein content has Kjeldahl's method, Thomas's combustion method etc., and these methods not only need to carry out pre-service to rapeseed sample, damage sample, and complicated operation, and cost is higher.Infrared spectrum technology is for the qualitative and quantitative analysis of agricultural product in recent years, but how to be undertaken by near-infrared spectral reflectance technology.When near-infrared spectrum technique characterizes rape variety, required sample size is generally 3-5g, can produce the small sample kind that selfing and hybridization etc. are important when breeding work, and these small sample seed amounts are low, can not meet sample size needs; Some rapeseed research needs special spectral device and system, comparatively complicated.Therefore, how detecting rapeseed content fast, improving the precision of prediction of infrared spectrum, is also a current important subject.
Middle infrared spectrum is that the fundamental frequency of material absorbs, and absorbance is large, and characteristic is strong.Optoacoustic spectroscopy is based on modern optoacoustic switch technology, a branch of infrared light incides optoacoustic annex, sample produces thermal effect after receiving Infrared irradiation, and by heat transfer to the gaseous state in sample cell, understand Swelling and contraction after gas is heated thus produce heat wave, heat wave is detected by the microphone of sensitivity, is optoacoustic spectroscopy, and optoacoustic spectroscopy can be exchanged into electric signal and obtains infrared photoacoustic spectra.Highly sensitive, the required sample size of optoacoustic spectroscopy is few, do not affect by sample shape, without the need to pre-treatment, to sample nondestructive, can realize in-site detecting, demonstrates good application potential in agricultural.
Mid-infrared spectral wavelength coverage is at 4000-400cm
-1, spectral information enriches, and peak overlap does not have near-infrared region serious, and detectability is more sensitive.But the demand of existing mid-infrared light spectral technology to test sample still there are certain requirements, the sample that Few sample of rapeseed seed amount is low can not be met, have some limitations, cannot be directly used in simply rapeseed protein content is carried out fast, accurately, harmless mensuration.
Summary of the invention
The object of this invention is to provide a kind of method for building up of the adaptive model quick to rapeseed protein content, accurate, harmless based on mid-infrared light acousto-optic spectrum spectrum.By detecting the infrared photoacoustic spectra information of rapeseed sample, utilize the spectroscopic data of the sample spectra information gathered and the rapeseed protein content measured, calculated by Euclidean distance, the training sample spectra the most similar to spectrum to be predicted is sorted from small to large by Euclidean distance, the spectroscopic data of new sort is built spectroscopic data model by partial least square method, further model is optimized by index of correlation value, can realizes carrying out fast the protein content of testing sample, accurately, Nondestructive Detection.
Technical scheme of the present invention is as follows:
A method for building up for middle infrared spectrum rapeseed protein content detection adaptive model, comprises following steps:
Step one: rapeseed samples every part of 100-200mg, is placed in FT-mid-IR fiber optics spectroscopy instrument-optoacoustic annex sample cell, and gather mid-infrared light acousto-optic spectrum information, continuous sweep repeatedly, is averaged spectrum;
Step 2: in step one light-metering spectrum rapeseed 5g, chemical analysis is carried out to its protein content, is multiplied by 6.25 in the result of nitrogen content, do the protein content of rapeseed, obtain chemical reference value;
Step 3: rapeseed spectral information step one gathered carries out de-noising, level and smooth, standardization pre-service;
Step 4: spectroscopic data step 3 handled well is calculated by Euclidean distance method; And the Euclidean distance calculated is sorted from small to large according to the distance the most close with spectrum to be predicted;
Step 5: the spectroscopic data of new sort, by partial least square method, sets up the rapeseed protein content prediction model based on spectral information.
The mid-infrared light acousto-optic bands of a spectrum of testing sample rapeseed are entered model by step 6: utilize the adaptive model set up in step 5, carry out calculating and predict, draw predicted value to the protein content of rapeseed sample to be measured;
Then step 2 chemical method gained chemistry reference value and adaptive model is utilized to predict that the predicted value of gained compares, the constructed mid-infrared light acousto-optic spectrum model chemical reference value that pre-and Kjeldahl's method detects to the predicted value of rapeseed protein content is basically identical, predicts the outcome reliably.
More optimize and more particularly, the method for operating of each step of the present invention is:
Step one: rapeseed samples every part of 100-200mg, be placed in FT-mid-IR fiber optics spectroscopy instrument (Nicolet8700, ThermoFisherScientific, USA)-optoacoustic annex (PA300, MTEC, USA), in sample cell, gather mid-infrared light acousto-optic spectrum information, collection wavelength coverage is 4000-800cm
-1, scanning resolution is 8cm
-1; Index glass speed is 0.4747cm
-1s, 64 continuous sweep is averaged spectrum.
Step 2: to its protein content, chemical analysis is carried out to the rapeseed 5g that in step one, light-metering has been composed, utilize full-automatic Kjeldahl determination device (Buchi339, CH), adopt Kjeldahl nitrogen determination, be multiplied by the result of nitrogen content the protein content that 6.25 do rapeseed, obtain chemical reference value.
Step 3: rapeseed spectral information step one gathered carries out de-noising, level and smooth, standardization pre-service (Du Changwen, " soil infrared photoacoustic spectra Principle and application ", Science Press, Beijing, 2012 editions).
Step 4: sample to be tested spectrum step 3 handled well and training sample spectroscopic data are calculated by Euclidean distance method.And the Euclidean distance calculated is sorted from small to large according to the distance the most close with spectrum to be predicted.
Step 5: the spectroscopic data of new sort, by partial least square method, sets up the rapeseed protein content prediction model based on spectral information.By index of correlation to model parameter as coefficient R
2, root-mean-square error RMSE and model prediction performance synthesis evaluation index RPD, makes the coefficient R of model
2close to 1, root-mean-square error RMSE is close to 0, and model prediction performance synthesis evaluation index RPD is greater than 2, can obtain the optimal sample number setting up forecast model.By to the comparison of model and optimization, finally obtain optimum adaptive model.Model is set up and is carried out (Mathwork, USA) with computation process by Matlab2013a software.
The mid-infrared light acousto-optic bands of a spectrum of testing sample rapeseed are entered model by step 6: utilize the adaptive model set up in step 5, carry out calculating and predict, draw predicted value to the protein content of rapeseed sample to be measured; Then step 7 chemical method gained chemistry reference value and adaptive model is utilized to predict that the predicted value of gained compares, the constructed mid-infrared light acousto-optic spectrum model chemical reference value that pre-and Kjeldahl's method detects to the predicted value of rapeseed protein content is basically identical, predicts the outcome reliably.
The advantages such as the method that the present invention utilizes Chemical Measurement to combine with infrared spectrum conveniently has fast in agricultural production, harmless, accurate, set up a kind of based on mid-infrared spectral adaptive model, can carry out fast rapeseed protein content, accurately, the mensuration that can't harm.Overcome rapeseed kind seed amount low, to implement in actual production comprehensive rapeseed quality control, maximum oil vegetable seed production economy effect, reduce testing cost have important practical significance.
Beneficial effect of the present invention is:
One, the present invention utilizes mid-infrared light acousto-optic to compose rapeseed protein content, and highly sensitive, required sample size is few; Simple to operate, to sample nondestructive, do not use chemical reagent.
Two, by detecting the infrared photoacoustic spectra information architecture adaptive model of rapeseed sample, be characterized in setting up to each sample spectra the adaptive model system meeting its spectral signature, this adaptive model has universality, model is sane, estimated performance is good, result accuracy is high, model prediction accuracy and background technology actual value close, predict the outcome accurately.
Three, by the structure of adaptive model, effectively can reduce the interference of sample invalid information, optimization, reduced sample modeling collection, model calculates quick, accurate.
Accompanying drawing explanation
Fig. 1 is rapeseed protein content prediction model scatter diagram.
Embodiment
Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.
If do not specialize, the conventional means that technological means all in embodiment is well known to those skilled in the art.
Embodiment 1, the method for building up of quick to rapeseed protein content, accurate, the harmless adaptive model of spectrum is composed based on mid-infrared light acousto-optic:
For 180 parts, formation testing vegetable seed sample, sample is carried out random assortment, and wherein 70 percent (126 parts) are as training sample set, and 30 percent (54 parts) try sample set for waiting.
One, training sample set and the every increment product of sample to be tested collection are got 100-200mg, be placed in FT-mid-IR fiber optics spectroscopy instrument (Nicolet8700, ThermoFisherScientific, USA)-optoacoustic annex (PA300, MTEC, USA) in sample cell, sample is put into photoacoustic cell and is no more than 2/3, closes sample cell, open high-pure helium air valve after being placed by sample, purge with high-purity helium, remove steam and carbon dioxide in photoacoustic cell.Gather mid-infrared light acousto-optic spectrum information, collection wavelength coverage is 4000-800cm
-1, scanning resolution is 8cm
-1; Index glass speed is 0.4747cm
-1s, 64 continuous sweep is averaged spectrum.
Two, for sample rapeseed every part 5g originally, chemical analysis is carried out to its protein content by all, utilize full-automatic Kjeldahl determination device (Buchi339, CH), adopt Kjeldahl nitrogen determination.Concrete steps are as follows:
Every part of rapeseed 5g dries 6h at 100 DEG C, uses agate mortar grinds sample after drying after sample;
The rapeseed powder taking the levigate oven dry of 0.5g is placed in 50mL and disappears and boil pipe, instills a small amount of deionized water and soaks sample, add the 10mL concentrated sulphuric acid, shake up, hold over night.Disappearing after spending the night is boiled pipe and is placed in constant temperature digestion instrument (LabTechEHD36, USA) and disappears and boil.Disappear when boiling and prevent curved neck funnel at the bottleneck boiling pipe that disappears, first disappear with slow fire and boil, temperature is set to 150 DEG C, and decompose until the concentrated sulphuric acid raised temperature again of emerging in a large number after the supercilious look, temperature is set to 250 DEG C, takes off when solution is uniform brownish black.Add 10 hydrogen peroxide after slightly cold to continue to disappear and boil, so in triplicate, the hydrogen peroxide of interpolation be successively reduced to disappear boil liquid become colorless or limpid after, then heat about 8min, eliminate hydrogen peroxide, taking off disappears boils pipe cooling.With the curved neck funnel of a small amount of deionized water rinsing, washing fluid flows into and disappears and boil pipe, after will disappear and boil liquid and be settled in 100mL volumetric flask and shake up (Lu Rukun " soil agrochemistry chemical analysis method ", Chinese agriculture Science Press, Beijing, 2000 editions).
In Kjeldahl's method, reagent comprises 10molL
-1naOH, 2% boric acid, 0.02molL
-1dilution heat of sulfuric acid, boric acid the indicator (-bromcresol green that methylates of 0.099:0.066).The running parameter of azotometer is set to: disappear and boil liquid consumption 20mL, and sodium hydroxide concentration is 30mL, and boric acid consumption is 15mL, and the pH of titration end-point is 4.65.During mensuration, each sample arranges a blank assay, supposes a Quality control samples every 20 samples.
Adopt the result of the nitrogen content of Kjeldahl nitrogen determination to be multiplied by the protein content that 6.25 do rapeseed, obtain chemical reference value;
Three, sample spectra information unification is carried out de-noising, level and smooth, standardization pre-service (Du Changwen, " soil infrared photoacoustic spectra Principle and application ", Science Press, Beijing, 2012 editions), concrete grammar is as follows:
Denoising smooth method adopts wavelet filtering to carry out, and provides a lot of digital filtering function in Matlab software, and adopt filtfilt function to carry out, its syntactic structure is:
[b,a]=butter(n,wn,’low’)
SpectrumF=filtfilt(b,a,spectrum)
Wherein variable a, b have been footage application definition a n rank cutoff frequency is the butterworth filter of wn, and low represents that cutoff frequency is the low-pass filtering of wn, and after spectrum Spectrum filtering, rreturn value is SpectrumF.
Standardization adopts the normalization function provided in Matlab software:
[pn,ps]=mapminmas(spectrum)
Wherein pn is the data after standard, and ps is the data containing former data average and standard deviation information.
Four, sample to be tested spectrum good for pre-service and training sample spectroscopic data are calculated by Euclidean distance method.Euclidean distance D computing formula between two samples is as follows:
Wherein X
isfor the spectroscopic data of training sample, X
imfor the spectroscopic data of sample to be tested; Often group spectroscopic data in training sample is taken out one by one, calculates with the spectroscopic data of sample to be tested, draw sample to be tested and its distance, and list with a matrix type.And by the D value of Euclidean distance according to resequencing from small to large, show that new spectrum that the distance the most close with spectrum to be predicted carries out sorting from small to large sorts training sample set.
Five, the spectroscopic data of new sort is by partial least square method, sets up the rapeseed protein content prediction model based on spectral information.Pass through coefficient R
2, root-mean-square error RMSE and model prediction performance synthesis evaluation index RPD calculates, and makes the coefficient R of model
2close to 1, root-mean-square error RMSE is close to 0, and model prediction performance synthesis evaluation index RPD is greater than 2, can draw the optimal sample number setting up forecast model.By to the comparison of model and optimization, finally obtain optimum adaptive model.Model is set up and is carried out (Mathwork, USA) with computation process by Matlab2013a software.
Wherein y and y' is chemical reference value and the infrared spectrum predicted value of a sample,
for the average of sample chemical reference value, n is sample capacity, and SD is the standard deviation of simple chemical reference value.Wherein R
2more less close to 1, RMSE, illustrate that the estimated performance of model is better.As RPD>2, think that model quality is excellent; As 1.5<PRD<2, think that model can accept; As RPD<1.5, then think model poor unacceptable (Du Changwen, " soil infrared photoacoustic spectra Principle and application ", Science Press, Beijing, 2012 editions).
The R of model is drawn by parameter optimization
2be 0.9205, RMSE be 0.594%, RPD value be 2.60, think that this model quality is excellent, quantitative effect is better.Fig. 1 mid point shape scatter diagram represents the scatter diagram of the pre-actual value of the training sample of this model and predicted value, fit line when figure bend is perfect forecast, and the training sample of rapeseed is roughly distributed in fit line both sides, does not have larger prediction deviation.
Six, the rapeseed optoacoustic spectroscopy data of testing sample are brought in the adaptive model that above-mentioned steps five sets up and the chemical score corresponding to the spectral value of sample to be tested is predicted, draw predicted value, then utilize chemical method gained chemistry reference value to compare.In Fig. 1, square scatter diagram represents the sample to be tested actual value of this model and the scatter diagram of predicted value, fit line when figure bend is perfect forecast, the R of sample to be tested
2be 0.9013, RMSE be 0.633%, RPD value be 2.51.The sample to be tested of rapeseed is roughly distributed in fit line both sides, does not have larger prediction deviation.
Visible, the chemical reference value that constructed mid-infrared light acousto-optic spectrum model detects the predicted value of rapeseed protein content and Kjeldahl's method is basically identical, predicts the outcome reliable.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the prerequisite not departing from the technology of the present invention principle; can also make some improvement and connect retouching, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (6)
1. a method for building up for middle infrared spectrum rapeseed protein content detection adaptive model, comprises following steps:
Step one: rapeseed samples every part of 100-200mg, is placed in FT-mid-IR fiber optics spectroscopy instrument-optoacoustic annex sample cell, and gather mid-infrared light acousto-optic spectrum information, continuous sweep repeatedly, is averaged spectrum;
Step 2: in step one light-metering spectrum rapeseed 5g, chemical analysis is carried out to its protein content, is multiplied by 6.25 in the result of nitrogen content, do the protein content of rapeseed, obtain chemical reference value;
Step 3: rapeseed spectral information step one gathered carries out de-noising, level and smooth, standardization pre-service;
Step 4: spectroscopic data step 3 handled well is calculated by Euclidean distance method; And the Euclidean distance calculated is sorted from small to large according to the distance the most close with spectrum to be predicted;
Step 5: the spectroscopic data of new sort, by partial least square method, sets up the rapeseed protein content prediction model based on spectral information;
The mid-infrared light acousto-optic bands of a spectrum of testing sample rapeseed are entered model by step 6: utilize the adaptive model set up in step 5, carry out calculating and predict, draw predicted value to the protein content of rapeseed sample to be measured.
2. the method for building up of middle infrared spectrum rapeseed protein content detection adaptive model according to claim 1, is characterized in that, in step one, gathering mid-infrared light acousto-optic spectrum information is that collection wavelength coverage is 4000-800cm
-1, scanning resolution is 8cm
-1; Index glass speed is 0.4747cm
-1s, 64 continuous sweep is averaged spectrum.
3. the method for building up of middle infrared spectrum rapeseed protein content detection adaptive model according to claim 1, it is characterized in that, in step 4, sample to be tested spectrum good for pre-service and training sample spectroscopic data are calculated by Euclidean distance method, the Euclidean distance D computing formula between two samples is as follows:
Wherein X
isfor the spectroscopic data of training sample, X
imfor the spectroscopic data of sample to be tested; Often group spectroscopic data in training sample is taken out one by one, calculates with the spectroscopic data of sample to be tested, draw sample to be tested and its distance, and list with a matrix type; And by the D value of Euclidean distance according to resequencing from small to large, show that new spectrum that the distance the most close with spectrum to be predicted carries out sorting from small to large sorts training sample set.
4. according to the method for building up of the middle infrared spectrum rapeseed protein content detection adaptive model one of claim 1-3 Suo Shu, it is characterized in that, the operation of step 5 is: the spectroscopic data of new sort, by partial least square method, sets up the rapeseed protein content prediction model based on spectral information; By index of correlation to model parameter as coefficient R
2, root-mean-square error RMSE and model prediction performance synthesis evaluation index RPD, makes the coefficient R of model
2close to 1, root-mean-square error RMSE is close to 0, and model prediction performance synthesis evaluation index RPD is greater than 2, obtains the optimal sample number setting up forecast model; By to the comparison of model and optimization, finally obtain optimum adaptive model.
5. the method for building up of middle infrared spectrum rapeseed protein content detection adaptive model according to claim 4, is characterized in that, the model in step 5 is set up and computation process, is undertaken by Matlab2013a software.
6. the method for building up of middle infrared spectrum rapeseed protein content detection adaptive model according to claim 5, is characterized in that, the model in step 5 is set up and computation process, is pass through coefficient R
2, root-mean-square error RMSE and model prediction performance synthesis evaluation index RPD calculates, and makes the coefficient R of model
2close to 1, root-mean-square error RMSE is close to 0, and model prediction performance synthesis evaluation index RPD is greater than 2, can draw the optimal sample number setting up forecast model; By to the comparison of model and optimization, finally obtain optimum adaptive model; Model is set up and with the formula of computation process is:
Wherein y and y' is chemical reference value and the infrared spectrum predicted value of a sample,
for the average of sample chemical reference value, n is sample capacity, and SD is the standard deviation of simple chemical reference value; Wherein R
2more less close to 1, RMSE, illustrate that the estimated performance of model is better; As RPD>2, think that model quality is excellent; As 1.5<PRD<2, think that model can accept; As RPD<1.5, then think that model is poor unacceptable.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510828543.7A CN105445218B (en) | 2015-11-25 | 2015-11-25 | The method for building up of middle infrared spectrum rapeseed protein content detection adaptive model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510828543.7A CN105445218B (en) | 2015-11-25 | 2015-11-25 | The method for building up of middle infrared spectrum rapeseed protein content detection adaptive model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105445218A true CN105445218A (en) | 2016-03-30 |
CN105445218B CN105445218B (en) | 2018-12-18 |
Family
ID=55555673
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510828543.7A Active CN105445218B (en) | 2015-11-25 | 2015-11-25 | The method for building up of middle infrared spectrum rapeseed protein content detection adaptive model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105445218B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107024445A (en) * | 2017-04-17 | 2017-08-08 | 中国科学院南京土壤研究所 | The modeling method and detection method of the quick detection of Nitrate in Vegetable |
CN107067034A (en) * | 2017-04-15 | 2017-08-18 | 温州大学 | A kind of method and system of quick identification ir data classification |
CN107576630A (en) * | 2017-09-28 | 2018-01-12 | 中国科学院昆明植物研究所 | The middle infrared spectrum portable detection system of plant various heavy content is determined simultaneously |
CN108254333A (en) * | 2017-12-14 | 2018-07-06 | 中国科学院南京土壤研究所 | Nitrogenous pesticide residue rapid detection method based on infrared photoacoustic spectra |
CN108507964A (en) * | 2018-03-07 | 2018-09-07 | 中国科学院西北高原生物研究所 | A kind of method that spectral model method quickly measures protein content in cynomorium songaricum |
CN110057756A (en) * | 2019-05-20 | 2019-07-26 | 南京农业大学 | A kind of infrared photoacoustic spectra detection method of wheat flour quality |
CN113792082A (en) * | 2021-09-02 | 2021-12-14 | 深圳创景数科信息技术有限公司 | Fabric component retrieval method based on database |
CN115201144A (en) * | 2022-09-14 | 2022-10-18 | 武汉工程大学 | Quantitative detection method, system and medium for amino acid and protein of rapeseed |
CN115931828A (en) * | 2023-02-17 | 2023-04-07 | 华谱智能科技(天津)有限公司 | Component analysis and prediction method, unit and system suitable for complex soil matrix |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101936893A (en) * | 2010-07-30 | 2011-01-05 | 华中农业大学 | Method for detecting protein and amino acid in rapeseeds |
US20120237949A1 (en) * | 2008-07-22 | 2012-09-20 | Board Of Trustees Of Michigan State University | Specific detection and quantification of phosphatidic acid using an arabidopsis trigalactosyldiacylglycerol-4 (tgd4) protein |
US20130055453A1 (en) * | 2011-07-13 | 2013-02-28 | Ms Technologies, Llc | Stacked herbicide tolerance event 8264.42.32.1, related transgenic soybean lines, and detection thereof |
CN103592225A (en) * | 2013-10-31 | 2014-02-19 | 中国科学院南京土壤研究所 | Soil infrared photoacoustic spectrum rapid determinator and soil physical and chemical property determination method |
-
2015
- 2015-11-25 CN CN201510828543.7A patent/CN105445218B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120237949A1 (en) * | 2008-07-22 | 2012-09-20 | Board Of Trustees Of Michigan State University | Specific detection and quantification of phosphatidic acid using an arabidopsis trigalactosyldiacylglycerol-4 (tgd4) protein |
CN101936893A (en) * | 2010-07-30 | 2011-01-05 | 华中农业大学 | Method for detecting protein and amino acid in rapeseeds |
US20130055453A1 (en) * | 2011-07-13 | 2013-02-28 | Ms Technologies, Llc | Stacked herbicide tolerance event 8264.42.32.1, related transgenic soybean lines, and detection thereof |
CN103592225A (en) * | 2013-10-31 | 2014-02-19 | 中国科学院南京土壤研究所 | Soil infrared photoacoustic spectrum rapid determinator and soil physical and chemical property determination method |
Non-Patent Citations (3)
Title |
---|
ROBERTO KAWAKAMI HARROP GALVAO ET AL: "A method for calibration and validation subset partitioning", 《TALANTA》 * |
陆宇振 等: "红外光声光谱法测定油菜籽品质参数", 《分析化学研究报告》 * |
陆振宇 等: "红外光声光谱技术结合支持向量机鉴别油菜籽品种", 《计算机与应用化学》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107067034A (en) * | 2017-04-15 | 2017-08-18 | 温州大学 | A kind of method and system of quick identification ir data classification |
CN107067034B (en) * | 2017-04-15 | 2020-06-23 | 温州大学 | Method and system for rapidly identifying infrared spectrum data classification |
CN107024445A (en) * | 2017-04-17 | 2017-08-08 | 中国科学院南京土壤研究所 | The modeling method and detection method of the quick detection of Nitrate in Vegetable |
CN107576630A (en) * | 2017-09-28 | 2018-01-12 | 中国科学院昆明植物研究所 | The middle infrared spectrum portable detection system of plant various heavy content is determined simultaneously |
CN108254333A (en) * | 2017-12-14 | 2018-07-06 | 中国科学院南京土壤研究所 | Nitrogenous pesticide residue rapid detection method based on infrared photoacoustic spectra |
CN108507964A (en) * | 2018-03-07 | 2018-09-07 | 中国科学院西北高原生物研究所 | A kind of method that spectral model method quickly measures protein content in cynomorium songaricum |
CN110057756A (en) * | 2019-05-20 | 2019-07-26 | 南京农业大学 | A kind of infrared photoacoustic spectra detection method of wheat flour quality |
CN113792082A (en) * | 2021-09-02 | 2021-12-14 | 深圳创景数科信息技术有限公司 | Fabric component retrieval method based on database |
CN115201144A (en) * | 2022-09-14 | 2022-10-18 | 武汉工程大学 | Quantitative detection method, system and medium for amino acid and protein of rapeseed |
CN115201144B (en) * | 2022-09-14 | 2022-12-09 | 武汉工程大学 | Quantitative detection method, system and medium for amino acid and protein of rapeseed |
CN115931828A (en) * | 2023-02-17 | 2023-04-07 | 华谱智能科技(天津)有限公司 | Component analysis and prediction method, unit and system suitable for complex soil matrix |
CN115931828B (en) * | 2023-02-17 | 2023-06-16 | 华谱智能科技(天津)有限公司 | Component analysis and prediction method, unit and system suitable for complex soil matrix |
Also Published As
Publication number | Publication date |
---|---|
CN105445218B (en) | 2018-12-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105445218A (en) | Establishing method of self-adaptive model for detection of content of protein of rapeseeds on basis of mid-infrared spectrum | |
CN103134767B (en) | Method for liquor quality identification through infrared spectrum revision | |
CN101609042A (en) | Hand-held soil nutrient nondestructive measurement system based near infrared spectrum | |
CN104764699B (en) | A kind of method for measuring edible oil acid value | |
CN109669023A (en) | A kind of soil attribute prediction technique based on Multi-sensor Fusion | |
CN105486655A (en) | Rapid detection method for organic matters in soil based on infrared spectroscopic intelligent identification model | |
CN102928379A (en) | Method for rapidly detecting polysaccharide content of health enhancement oral liquid by adopting near infrared spectrum technology | |
CN105628644A (en) | Device and method for on-line monitoring of protein enzymolysis process based on in-situ real-time spectrum | |
CN104778349B (en) | One kind is used for rice table soil nitrogen application Classified Protection | |
CN201503392U (en) | Handheld soil nutrient nondestructive measurement device based on near infrared spectrum | |
CN103558180A (en) | Method for quickly detecting physicochemical indexes in saccharification leavening agents through near infrared | |
CN110144280A (en) | A kind of white liquor method and its online simulation based near infrared spectrum on-line checking plucks wine device | |
CN108072627A (en) | It is a kind of that the method for amino-acid nitrogen and total acid content in soy sauce is quickly detected with middle infrared spectrum | |
CN103487398B (en) | A kind of analytical method of lysine fermentation liquor | |
CN101556245A (en) | Chlorophyll measurement method based on RGB digital signal | |
CN102768191B (en) | Method for easily detecting trace thallium in water | |
CN106706554A (en) | Method for rapidly and nondestructively determining content of straight-chain starch of corn single-ear grains | |
WO2024061126A1 (en) | Method, apparatus, and system for evaluating polyhydroxyalkanoate content, and device | |
CN109406447A (en) | A kind of near infrared detection method of tannin in sorghum | |
CN103134770A (en) | Method for eliminating influence on infrared spectrum detection of soil total nitrogen content from moisture | |
CN105386263A (en) | Dyeing process monitoring method | |
CN103728269A (en) | Method for near infrared rapid detection of physical and chemical indicators in wine stock | |
CN104374731A (en) | MIR spectral analysis model and method for identifying excessive cadmium content of rice. | |
CN112362812A (en) | Remote sensing inversion model and method for rice leaf chlorophyll carotenoid content ratio based on Lars algorithm | |
CN101738339B (en) | Near-infrared complete pretreatment device for blended gasoline |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |