CN105044022B - A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness - Google Patents
A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness Download PDFInfo
- Publication number
- CN105044022B CN105044022B CN201510477871.7A CN201510477871A CN105044022B CN 105044022 B CN105044022 B CN 105044022B CN 201510477871 A CN201510477871 A CN 201510477871A CN 105044022 B CN105044022 B CN 105044022B
- Authority
- CN
- China
- Prior art keywords
- hardness
- wheat
- data
- near infrared
- sample
- 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.)
- Expired - Fee Related
Links
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of method based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness and applications, belong to grain quality detection technique field.Method provided by the present invention is the near-infrared monitoring model by establishing wheat hardness, wheat samples to be checked are scanned using near infrared spectrometer, obtain sample near infrared light spectral curve, after carrying out data processing near infrared light spectral curve, the hardness of wheat samples to be checked is determined using detection model further according to the data obtained.Method provided by the present invention has that at low cost, pollution-free, detection speed is fast, objectivity is high, and may be implemented in line analysis, without geographical restrictions, can be measured in real time.
Description
Technical field
The present invention relates to a kind of method based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness and applications, belong to
Grain quality detection technique field.
Background technology
Wheat kernel texture it is soft, actually evaluate wheat quality and edible quality an important indicator, and with it is small
Wheat breeding and trade price etc. are many-sided closely related, hardness be domestic and international wheat market classification and price important evidence it
One and various countries important one of the breeding objective of breeder.Wheat hardness is defined as resistance suffered when broken seed,
It is crushed required power when seed.Hardness is determined by the bond strength in albuminous cell between protein matrix and starch
, this bond strength is controlled by heredity.The powder quality processed and grain hardness of wheat are closely related, and the variation of wheat hardness can make
The very big change of generations such as each system goods in process inventory and quality, each equipment work efficiency, processing dynamics consumption in wheat flour milling flow
Change.The hardness of advance measurement materials of wheat determines for adjusting flouring technology flow and corresponding technical parameter in time and matches wheat
Scheme, the material balance for keeping flow and stable, the raising production efficiency of production etc., all have important technological guidance's meaning.It presses
According to the difference of hardness, wheat is divided into three apparent hardness leveles by foreign countries：Soft wheat, hard wheat and durum wheat.
The hardness that wheat is measured using conventional method, take a lot of work, time-consuming or formality it is complicated, and be unable to on-line checking.It is close several
Year, with industry, agricultural and the industries such as pharmacy to more rapidly, the pursuit of low consumption and non-destructive testing, spectral technique becomes
The emphasis of people's research, is more and more paid attention to, its many applications have been put into American Association of Cereal Chemists's mark
Quasi- method (AACC).Near-infrared (NIR) detection because have detection speed is fast, can be achieved to detect entry simultaneously, accuracy height,
The features such as sample representativeness is good and at low cost so that near infrared technology becomes a kind of huge environmentally protective analysis skill of application potential
Art.Using NIR methods measurement wheat hardness foreign countries it has been reported that AACC recommends method to use 1680 and 2230nm, two wavelength points,
And provide three regression coefficients, but it is standardized instrument and sample comminution object, otherwise evaluated error compared with
Greatly.So actual use NIR methods measure the report of wheat hardness and few at present, people, which are still dedicated to studying other methods, to be measured
Wheat hardness, however NIR methods is applied to measure the other qualities of wheat and the technology classified by certain standard to wheat and side
Method is very ripe, and can reach higher precision.
Invention content
To solve the problems, such as that prior art detection cycle is long, testing cost high efficiency is low, cumbersome, the present invention provides
A method of based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness, the technical solution taken is as follows：
The purpose of the present invention is to provide a kind of methods based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness.
This method is the near infrared detection model by establishing wheat hardness, scans wheat samples to be checked using near infrared spectrometer, obtains
Sample near infrared spectrum data is taken, after carrying out data processing near infrared spectrum data, detection is utilized further according to the data obtained
Model determines the hardness of wheat samples to be checked.
The step of the method, is as follows：
1) it utilizes near infrared spectrometer to scan wheat samples, obtains the near infrared spectrum data of sample；
2) stiffness method determination step 1 is utilized) hardness of the wheat samples, and establish wheat hardness database；
3) near infrared spectrum data obtained by step 1) is divided into calibration set and prediction using the SPXY algorithms after optimization
Collection, pre-processes calibration set and forecast set spectroscopic data, obtains preprocessed data；
4) it is corresponded to using pretreated calibration set spectrum data matrix obtained by step 3) as the input of model with calibration set
Desired output of the wheat hardness data as model, establish preliminary radial basis function neural network model, recycle step 3)
The data of forecast set are verified and are corrected to preliminary radial basis function neural network model after gained pretreatment, are finally examined
Survey model；
5) it utilizes near infrared spectrometer to scan wheat samples to be checked, obtains sample near infrared spectrum data, it is closely red to gained
After external spectrum data are pre-processed, the hardness of wheat samples to be measured is measured using the detection model obtained by step 4).
Preferably, step 1) is described scans wheat samples using near infrared spectrometer, and scanning range 400-2498nm divides
Resolution is 8nm, and scanning times are 2 times.
Preferably, step 2) the stiffness method is to take 24.99-25.01g samples, is crushed simultaneously using hardness analyzer
Determination sample hardness, grinding time 50s weigh and are calculated finally by automatic weighing system final result or weigh by hand simultaneously
According to following formula result of calculation：
In formula, HI (%) indicates correction to moisture 12%, stiffness when 25 DEG C of environment temperature, m1(g) it indicates to crush
Afterwards by the sample quality of sieve, w (%) indicates the moisture of sample, k1Indicate moisture correction coefficient, k2Indicate temperature correction
Coefficient.
Preferably, the pretreatment of the step 3) near infrared spectrum data is to reject spectroscopic data using mahalanobis distance method
In exceptional value, recycle optimization after SPXY algorithms spectroscopic data is divided.
Preferably, the SPXY algorithms after the step 3) optimization are first to select two farthest point datas of Euclidean distance into school
Positive collection, recycles SPXY methods to choose remaining calibration set data from remaining sample.
Preferably, the pretreatment described in step 3), be standard normal variable transformation, first derivative, second dervative and continuously
Projection algorithm handles one or more of spectrum.
Preferably, the step 4) detection model is an accurate Radial Basis Function neural using newrbe function creations
Network (RBF) model.RBF networks include input layer, hidden layer and output layer, pretreated calibration set spectrum data matrix conduct
The input of model is defaulted as 0 using the corresponding wheat hardness data of calibration set as the desired output of model, mean square error,
SPREAD values take 1200, and hidden layer basic function uses Gaussian function.Radial basis function network is a kind of partial approximation network, and one is most
Big advantage is to obtain weights W by linear least square, therefore have faster processing speed.
The method is as follows：
1) near infrared spectrometer is utilized to scan wheat samples, scanning range 400-2498nm, resolution ratio 8nm, scanning
Number is 2 times, obtains the near infrared light spectral curve of sample；
2) the 24.99-25.01g steps 1) wheat samples are taken, simultaneously determination sample hardness is crushed using hardness analyzer,
Grinding time is 50s, and final hardness is weighed and calculated finally by automatic weighing system, and establishes wheat hardness database；
3) it utilizes mahalanobis distance method to reject the exceptional value in the curve of spectrum, recycles the SPXY algorithms after optimization by gained
Near infrared spectrum data is divided into calibration set and forecast set, and optimization calibration set and forecast set spectrum are converted using standard normal variable
Data, while compression processing is carried out to calibration set and forecast set spectroscopic data using successive projection algorithm, obtain preprocessed data；
SPXY algorithms after the optimization are first to select two farthest point datas of Euclidean distance into calibration set, are being selected from remaining sample
Take remaining calibration set data；
4) using the pretreated calibration set spectrum data matrix of gained as the input of model, with the corresponding wheat of calibration set
Desired output of the hardness data as model, mean square error are defaulted as 0, establish preliminary radial basis function neural network model, then
Preliminary radial basis function neural network model is verified and is corrected using the data of forecast set after pretreatment obtained by step 3),
Finally obtain detection model；
5) wheat samples to be checked are scanned using near infrared spectrometer according to step 1) the method, obtains sample near infrared light
Modal data after being pre-processed near infrared spectrum data further according to step 3) the method, utilizes the detection obtained by step 4)
The hardness of model determination wheat samples to be measured.
The either method is used equally for the hardness of detection wheat.
What the present invention obtained has the beneficial effect that：
1. the method for the present invention has the characteristics that at low cost, free of contamination.Any examination is not needed using near-infrared spectrum technique
Agent does not consume sample, without discharging pollutants.Compared with conventional method, cost can be reduced and it is one with environmental protection
Kind " green analysis " technology.
2. the detection speed of the method for the present invention is fast, detection wheat type is more, and use is small, and general measure takes time and effort.Often
Sample data volume is excessive when advising Method Modeling, and there are more disturbing factors, influence the stability of model, and the present invention passes through to light
Modal data is pre-processed, and compressed spectrum data carry out the further stability for enhancing model.After model foundation, sample is measured
Time only needs tens seconds, increases substantially detection efficiency.
3. the method objectivity of the present invention is higher.The needs such as conventional wheat hardness measurement method such as vitreousness method artificially measure,
Error is larger, and near infrared detection method is essentially all machine operation, substantially eliminates human error.
4. on-line analysis may be implemented.Near infrared detection method without geographical restrictions, can be measured in real time.
Description of the drawings
Fig. 1 is the flow chart that the present invention detects.
Fig. 2 is wheat samples atlas of near infrared spectra；
(wherein abscissa represents the wavelength of spectrum, and ordinate represents the absorbance of spectrum, and each spectrum corresponds to each sample
The hardness of product).
Fig. 3 is the spectrum point of successive projection algorithms selection；
(wherein abscissa represents wave point, and ordinate is absorbance value, and the square in figure represents the wave point position that screening obtains
It sets).
Fig. 4 is RBF schematic network structures.
Specific implementation mode
With reference to specific embodiment, the present invention will be further described, but the present invention should not be limited by the examples.
Following embodiment material therefor, reagent, method and instrument are this field conventional material, examination without specified otherwise
Agent, method and instrument, those skilled in the art can be obtained by commercial channel.
Embodiment 1
1, the sample selected by the present embodiment is each from the whole nation by agricultural product quality and safety research institute of Heilongjiang Academy of Agricultural Sciences
The wheat samples in 2013 that ground wheat main producing region is collected collect 111, sample altogether.
Spectroscopic data is obtained using near-infrared spectrometers, detailed process is：Temperature control is in 25 DEG C of room temperature when scanning
Left and right；Sample needs to strike off rim of a cup after being packed into specimen cup when filling sample, makes sample surfaces and specimen cup edge to flat as possible；
Sample is put into the prototype sulculus of diameter 35mm, 10mm a depth and is scanned, sweep spacing is 8nm, scanning 2
It is secondary, 262 spectrum points are generated, such as the atlas of near infrared spectra that Fig. 2 is 4 wheats.
2, routine experiment is carried out to the wheat samples after near-infrared spectrometers scan, accurately weighs the sample prepared
Product (25.00 ± 0.01) g；Hardness analyzer end cap is opened, by a cavity (recess portion between two knives) for crushing system rotor
Alignment feed inlet upwards, closes and locks end cap；Hooper door is opened, the sample weighed is all poured into feed hopper, is closed
Close hooper door；Open analyzer, after sample comminution 50s, autostop；End cap is opened after instrument comes to a complete stop, carefully by splicing
Bucket, screen system take out together, and according to the regulation of instrument specification, the carry-over on sieve is cleaned up.To prevent in cleaning
Only screen system is detached with receiving hopper, in case the carry-over on sieve falls into the substance in receiving hopper and (or) in receiving hopper and spreads
Go out；It is weighed together with receiving hopper, screen system and shines lower object, screenings quality is obtained after deducting receiving hopper, screen system quality
M1 is accurate to 0.01g；Instrument crushing system, receiving hopper, screen system etc. are cleaned up, used in case next time measures.It is equipped with and claims
The instrument for measuring computing system, calculates after weighing and prints result automatically.It is not equipped with and weighs computing system, based on formula (1)
It calculates：
In formula, HI (%) indicates correction to moisture 12%, and stiffness when 25 DEG C of environment temperature, m1 (g) indicates to crush
Afterwards by the sample quality of sieve, w (%) indicates that the moisture of sample, k1 indicate that moisture correction coefficient, k2 indicate temperature school
Positive coefficient.All experimental procedures are executed in strict accordance with standard GB/T/T 21304, complete by academy of agricultural sciences professional researcher
At.
3, the abnormal sample in sample is selected using mahalanobis distance method, and is rejected, 111 samples are remaining after processing
Then 108 samples divide entire sample set using the SPXY methods after optimization.Final choice 84 is representative
Sample as calibration set, for establishing model, calibration set has 24 samples.
4, calibration set spectroscopic data is optimized using standard normal variable transformation SNV, to eliminate the shadow of correction scattering
It rings.
5, compression processing is carried out to spectroscopic data using successive projection algorithm, obtained wavelength points 47 are as shown in Figure 3.
6, radial basis function (RBF) neural network model (Fig. 4), pretreated spectrum number are established using calibration set sample
Input according to matrix as model, the output of the hardness of the sample corresponding to these spectrum as model, mean square error are defaulted as
0, by repeatedly testing, when SPREAD values take 1200, the effect of model is relatively good, and hidden layer basic function uses Gaussian function, mould
The discriminant coefficient R of type2It is 0.90, prediction standard difference SEP is 3.02, and relation analysis error RPD is 3.11, the R of model2More approach
In 1, the stability of model is better.
According to the discriminant coefficient of model, prediction standard is poor, relation analysis error constantly adjusts the parameter of model.
Embodiment 2
1, step 1-3 is identical with the step 1-3 in embodiment 1.
2, radial basis function (RBF) neural network model is established using the original spectrum of untreated calibration set sample,
Using forecast set sample come checking R BF models, the R of model2For 0.79, RPD 2.19, SEP 4.30.
Embodiment 3
1, step 1-3 is identical with the step 1-3 in embodiment 1.
2, the spectroscopic data of all samples is optimized using standard normal variable transformation SNV, to eliminate correction scattering
Influence, then second derivative spectra is obtained by Mathematic calculation method.
3, offset minimum binary (PLS) model is established using calibration set sample, is verified using forecast set sample and corrects PLS
Model, the R of model2For 0.85, RPD 2.57, SEP 3.66.
Embodiment 4
1, step 1-3 is identical with the step 1-3 in embodiment 1.
2, the spectroscopic data of all samples is optimized using standard normal variable transformation SNV, to eliminate correction scattering
Influence, then first derivative spectrum is obtained by Mathematic calculation method.
3, compression processing is carried out to initial data using successive projection algorithm.
4, BP neural network model is established using calibration set sample, BP models is tested just and corrected using forecast set sample,
The R of model2For 0.60, RPD 1.59, SEP 5.92.
Embodiment 5
After establishing detection model, inventor arrives again purchases and extracts 4 kinds of wheats in the market, each 6 samples, amounts to
24 samples.Determination of Hardness is carried out using the detection model constructed by embodiment 1-4.Infrared diaphanoscopy and stiffness method measure
Etc. processes it is identical as listed method in embodiment 1, final detection result is as shown in table 1.As can be seen from the table, implement
The estimated performance for the RBF neural network model established in example 1 is best, and error is minimum.
The testing result of the different detection models of table 1
The different Model checking coefficients of table 2, relation analysis error, prediction standard are poor
| R2 | RPD | SEP | |
| Embodiment 1 | 0.90 | 3.02 | 3.11 |
| Embodiment 2 | 0.79 | 2.19 | 4.30 |
| Embodiment 3 | 0.85 | 2.57 | 3.66 |
| Embodiment 4 | 0.60 | 1.59 | 5.92 |
From table 2 it can be seen that 1 discriminant coefficient R of embodiment2Closest to 1, relation analysis error amount RPD is maximum, prediction standard
Poor SEP is minimum, it can be seen that the forecast result of model of embodiment 1 is fine, can be satisfied with practical application.
Although the present invention is disclosed as above with preferred embodiment, it is not limited to the present invention, any to be familiar with this
The people of technology can do various changes and modification, therefore the protection of the present invention without departing from the spirit and scope of the present invention
Range should be subject to what claims were defined.
Claims (6)
1. a kind of method based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness, which is characterized in that steps are as follows：
1) it utilizes near infrared spectrometer to scan wheat samples, obtains the near infrared spectrum data of sample；
2) stiffness method determination step 1 is utilized) hardness of the wheat samples, and establish wheat hardness database；
3) mahalanobis distance method is utilized to reject the exceptional value in spectroscopic data, it will be obtained by step 1) using the SPXY algorithms after optimization
Near infrared spectrum data is divided into calibration set and forecast set, is pre-processed to calibration set and forecast set spectroscopic data, obtains pre-
Handle data；SPXY algorithms after the optimization are first to select two farthest point datas of Euclidean distance into calibration set, recycle
SPXY methods choose remaining calibration set data from remaining sample；
4) corresponding small with calibration set using pretreated calibration set spectrum data matrix obtained by step 3) as the input of model
Desired output of the wheat hardness data as model, establishes preliminary radial basis function neural network model, recycle step 3) gained
The data of forecast set are verified and are corrected to preliminary radial basis function neural network model after pretreatment, finally obtain detection mould
Type；
5) it utilizes near infrared spectrometer to scan wheat samples to be checked, sample near infrared spectrum data is obtained, to gained near infrared light
After modal data is pre-processed, the hardness of wheat samples to be measured is measured using the detection model obtained by step 4).
2. claim 1 the method, which is characterized in that step 1) is described to scan wheat samples using near infrared spectrometer, sweeps
Ranging from 400-2498nm, resolution ratio 8nm are retouched, scanning times are 2 times.
3. claim 1 the method, which is characterized in that step 2) the stiffness method, is to take 24.99-25.01g samples,
It is crushed using hardness analyzer and determination sample hardness, grinding time 50s is weighed and calculated finally by automatic weighing system
Final result or craft weigh and according to following formula result of calculation：
In formula, HI (%) indicates correction to moisture 12%, and stiffness when 25 DEG C of environment temperature, m1 (g) indicates to lead to after crushing
The sample quality of sieve is crossed, w (%) indicates that the moisture of sample, k1 indicate that moisture correction coefficient, k2 indicate temperature correction system
Number.
4. claim 1 the method, which is characterized in that the pretreatment described in step 3), for standard normal variable transformation, single order
One or more of derivative, second dervative and successive projection algorithm process spectrum.
5. claim 1 the method, which is characterized in that the step 4) detection model is radial basis function neural network mould
Type, model are using pretreated calibration set spectroscopic data as mode input, in the corresponding wheat hardness database of calibration set
Desired output of the data as model, mean square error 0, SPREAD values are 1200, using Gaussian function as hidden layer basic function into
Row structure.
6. claim 1 the method, which is characterized in that be as follows：
1) near infrared spectrometer is utilized to scan wheat samples, scanning range 400-2498nm, resolution ratio 8nm, scanning times
It is 2 times, obtains the near infrared light spectral curve of sample；
2) the 24.99-25.01g steps 1) wheat samples are taken, simultaneously determination sample hardness is crushed using hardness analyzer, crushes
Time is 50s, and final hardness is weighed and calculated finally by automatic weighing system, and establishes wheat hardness database；
3) it utilizes mahalanobis distance method to reject the exceptional value in the curve of spectrum, recycles the SPXY algorithms after optimization by the close red of gained
External spectrum data are divided into calibration set and forecast set, and optimization calibration set and forecast set spectrum number are converted using standard normal variable
According to, while compression processing is carried out to calibration set and forecast set spectroscopic data using successive projection algorithm, obtain preprocessed data；Institute
The SPXY algorithms after optimization are stated, are first to select two farthest point datas of Euclidean distance into calibration set, recycle SPXY methods from surplus
Remaining calibration set data is chosen in remaining sample；
4) using the pretreated calibration set spectrum data matrix of gained as the input of model, with the corresponding wheat hardness of calibration set
Desired output of the data as model, it is 1200 that mean square error, which is defaulted as 0, SPREAD values, using Gaussian function as hidden layer basic function
Establish preliminary radial basis function neural network model, recycle step 3) after gained pretreatment forecast set data to preliminary radial
Basis function neural network model is verified and is corrected, and detection model is finally obtained；
5) wheat samples to be checked are scanned using near infrared spectrometer according to step 1) the method, obtains sample near infrared spectrum number
According to after being pre-processed near infrared spectrum data further according to step 3) the method, utilizing the detection model obtained by step 4)
Measure the hardness of wheat samples to be measured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510477871.7A CN105044022B (en) | 2015-08-06 | 2015-08-06 | A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510477871.7A CN105044022B (en) | 2015-08-06 | 2015-08-06 | A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105044022A CN105044022A (en) | 2015-11-11 |
| CN105044022B true CN105044022B (en) | 2018-09-21 |
Family
ID=54450740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510477871.7A Expired - Fee Related CN105044022B (en) | 2015-08-06 | 2015-08-06 | A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105044022B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803009A (en) * | 2015-11-26 | 2017-06-06 | 北京光巨力信息技术有限公司 | A kind of near infrared no-wound blood glucose meter correcting algorithm based on manifold learning and RBF neural |
| CN106485049B (en) * | 2016-09-09 | 2019-03-26 | 黑龙江大学 | A kind of detection method of the NIRS exceptional sample based on Monte Carlo cross validation |
| CN108318445A (en) * | 2018-04-10 | 2018-07-24 | 江苏大学 | A kind of near infrared technology qualitative discrimination wheat whether the detection method of heated denaturalization |
| CN208420696U (en) * | 2018-05-22 | 2019-01-22 | 南京农业大学 | Wheat based on near-infrared spectrum technique infects head blight grade on-line detecting system |
| CN108875913B (en) * | 2018-05-30 | 2021-09-17 | 江苏大学 | Tricholoma matsutake rapid nondestructive testing system and method based on convolutional neural network |
| CN109580525A (en) * | 2018-12-03 | 2019-04-05 | 益海嘉里（兖州）粮油工业有限公司 | A kind of detection method of quick predict wheat baking quality |
| CN109636074A (en) * | 2019-02-01 | 2019-04-16 | 中国农业大学 | The near-infrared method for quick predicting of pig digestible energy and metabolic energy in a kind of sorghum |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6100526A (en) * | 1996-12-30 | 2000-08-08 | Dsquared Development, Inc. | Grain quality monitor |
| US6596930B1 (en) * | 1998-05-22 | 2003-07-22 | The United States Of America As Represented By The Secretary Of Agriculture | Modification of cereal grain hardness via expression of puroindoline protein |
| CN101713731B (en) * | 2009-11-09 | 2012-02-29 | 北京中医药大学 | Method for distinguishing coating quality of medicine preparation |
| CN103149174A (en) * | 2013-03-19 | 2013-06-12 | 周口市农业科学院 | Method for rapidly detecting yellow pigment content in wheat grain |
| CN104020135B (en) * | 2014-06-18 | 2016-09-21 | 中国科学院重庆绿色智能技术研究院 | Calibration model modeling method based near infrared spectrum |
| CN104749126A (en) * | 2015-02-28 | 2015-07-01 | 华北水利水电大学 | Wheat hardness prediction method based on near infrared hyperspectral image analysis |
-
2015
- 2015-08-06 CN CN201510477871.7A patent/CN105044022B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN105044022A (en) | 2015-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105044022B (en) | A kind of method and application based on near-infrared spectrum technique Fast nondestructive evaluation wheat hardness | |
| Zareiforoush et al. | Design, development and performance evaluation of an automatic control system for rice whitening machine based on computer vision and fuzzy logic | |
| CN108680515B (en) | Single-grain rice amylose quantitative analysis model construction and detection method thereof | |
| CN104142311B (en) | A kind of method of near-infrared spectrum technique prediction torch pine pine resin yield | |
| MX2008012927A (en) | Method for multivariate analysis in predicting a trait of interest. | |
| Nadimi et al. | Examination of wheat kernels for the presence of Fusarium damage and mycotoxins using near-infrared hyperspectral imaging | |
| CN107515203A (en) | The research of near infrared technology quantitative analysis rice single grain amylose content | |
| CN108169165B (en) | Maltose mixture quantitative analysis method based on terahertz spectrum and image information fusion | |
| Shahin et al. | Quantification of mildew damage in soft red winter wheat based on spectral characteristics of bulk samples: a comparison of visible-near-infrared imaging and near-infrared spectroscopy | |
| Mabille et al. | Parametric modelling of wheat grain morphology: a new perspective | |
| CN105021564A (en) | Method for determining content of ergosterol in tobacco based on near infrared spectroscopic analysis technology | |
| CN108613943B (en) | Near-infrared single-grain crop component detection method based on spectrum morphology transfer | |
| Chen et al. | Prediction of milled rice grades using Fourier transform near-infrared spectroscopy and artificial neural networks | |
| CN109540837A (en) | The method that near-infrared quickly detects Boehmeria nivea leaves wood fibre cellulose content | |
| WO2020248961A1 (en) | Method for selecting spectral wavenumber without reference value | |
| Jiang et al. | Rapid nondestructive detecting of wheat varieties and mixing ratio by combining hyperspectral imaging and ensemble learning | |
| CN106226267B (en) | A kind of near-infrared assay method of dry chili color value | |
| CN109932336A (en) | A kind of method for quick identification of wholemeal | |
| CN115201148A (en) | Method for rapidly detecting content of fatty acid in rice based on near infrared spectrum | |
| CN104316492A (en) | Method for near-infrared spectrum measurement of protein content in potato tuber | |
| Symons et al. | Measurement of hard vitreous kernels in durum wheat by machine vision | |
| Tian et al. | Nondestructive testing for wheat quality with sensor technology based on Big Data | |
| CN117723505A (en) | Near infrared spectrum-based rapid detection method for rice seed grain content grade | |
| Sinelli et al. | Prediction of semolina technological quality by FT-NIR spectroscopy | |
| Yang | Development of an integrated variety and appearance quality measurement system for milled rice |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Ye Dandan Inventor after: Sun Laijun Inventor after: Liu Yangyang Inventor after: Hui Guangyan Inventor after: Wang Jianan Inventor before: Sun Laijun Inventor before: Hui Guangyan Inventor before: Wang Jianan |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180921 Termination date: 20210806 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |