CN108507972A - A kind of across the time apple sugar content prediction technique of near infrared spectrum based on distance metric and semi-supervised learning - Google Patents
A kind of across the time apple sugar content prediction technique of near infrared spectrum based on distance metric and semi-supervised learning Download PDFInfo
- Publication number
- CN108507972A CN108507972A CN201810319258.6A CN201810319258A CN108507972A CN 108507972 A CN108507972 A CN 108507972A CN 201810319258 A CN201810319258 A CN 201810319258A CN 108507972 A CN108507972 A CN 108507972A
- Authority
- CN
- China
- Prior art keywords
- apple
- sample
- unmarked
- model
- lssvm
- 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
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000010238 partial least squares regression Methods 0.000 claims abstract description 8
- 239000000284 extract Substances 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 8
- 235000013399 edible fruits Nutrition 0.000 claims description 7
- 238000010521 absorption reaction Methods 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 6
- 238000001228 spectrum Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 abstract description 10
- 241000220225 Malus Species 0.000 description 77
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000008103 glucose Substances 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000050 nutritive effect Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000003971 tillage 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 provides a kind of across time apple sugar content prediction technique of the near infrared spectrum based on distance metric and semi-supervised learning, belongs to apple sugar content prediction field.This method extracts relative reflectance as characteristic parameter first from the near infrared spectrum of apple, then utilizes initial least square method supporting vector machine regression model MlssvmWith Partial Least-Squares Regression Model MplsrTo unmarked sample predictions, predicted value is obtained；Then ask unmarked sample and regression model MlssvmOr MplsrThe forecasted variances of the maximum distance of modeling sample used and two regression models to unmarked sample；Small unmarked sample is added to initial regression model M to selected distance as update collection sample with two forecast of regression model differences greatly laterlssvmOr MplsrIn modeling sample used, initial regression model M is updatedlssvmOr Mplsr；Until being demarcated to test set sample after meeting maximum iteration.The present invention is updated initial model using distance metric and semi-supervised learning, precision of prediction is high, it is easy to operate, fast and effective, there is higher robustness.
Description
Technical field
The invention belongs to apple sugar contents to predict field, and in particular to a kind of based on the close red of distance metric and semi-supervised learning
Across the time apple sugar content prediction technique of external spectrum.
Background technology
Apple have the characteristics that nutritive value it is high, can storage time it is long, it has also become global fruit.With living standard
Raising and consumption diversification, consumer is higher and higher to fruit quality and safety requirements, apple quality with safety at
The factor paid close attention to the most for consumer.Under conditions of same safety, consumer increasingly focuses on apple interior quality, such as apple
The pol of fruit.How quickly apple internal quality to be detected and is classified, is not only related to the edible quality and peace of consumer
All risk insurance hinders, and directly influences the foreign trade of apple.
Domestic and foreign scholars study the glucose prediction of apple, such as machine vision, near infrared spectrum.Wherein, closely
Infrared spectrum can reflect apple internal chemical feature, be widely used in apple sugar content prediction.
For the apple across time same breed, the near infrared light spectrum information of this method acquisition introduces new variable, considers
Tillage condition, soil environment condition and climate change, and original training set sample does not consider these new variables, drop
The low accuracy and robustness of development model.
Invention content
Purpose of the present invention is to the disadvantages more than overcoming, and provide a kind of near infrared light based on distance metric and semi-supervised learning
Across the time apple sugar content prediction technique of spectrum, improves the precision of prediction of model, easy to operate, and excellent with higher robustness etc.
Point.
Technical scheme of the present invention：
A kind of across the time apple sugar content prediction technique of near infrared spectrum based on distance metric and semi-supervised learning, step is such as
Under：
A, N number of unmarked apple sample is individually positioned near infrared spectra collection system, acquired and obtained and is unmarked
Near infrared spectrum of the apple sample under B wave band, the unmarked apple sample are current year apple sample；
B, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；It extracts under B wave band
The relative reflectance of near infrared spectrum, the characteristic parameter F as unmarked apple sample；
C, by the characteristic parameter F of the unmarked apple sample of gained in step b, input pre-establishes or newer least square
Support vector regression model MlssvmWith Partial Least-Squares Regression Model MplsrIn, obtain the initial of unmarked apple sample pol
Predicted valueWithThe recurrence mould pre-established
Type MlssvmAnd MplsrIn modeling sample be upper one year acquisition apple sample；
D, N number of unmarked apple sample is calculated to model Mlssvm(or model Mplsr) used in modeling sample maximum distanceWithWherein,It is i-th of unmarked apple sample to model
Mlssvm(or model Mplsr) used in modeling sample maximum distance,It is i-th of unmarked apple sample to model Mlssvm(or mould
Type Mplsr) used in modeling sample maximum distance,WithIt is by two calculated maximum distances of different formulas；
E, least square method supporting vector machine regression model M is calculatedlssvmWith Partial Least-Squares Regression Model MplsrIt is not marked to N number of
Remember the forecasted variances E=[e of apple sample1,...,ei,...,eN], wherein| | absolute value is sought in expression；
F, big q unmarked apple sample of maximum range value is chosen from D1 and D2 respectively, by q unmarked apple samples
This and its true pol value are respectively put intoWith
G, small q unmarked apple sample of forecasted variances value is chosen from E, by q unmarked apple samples and its sugar
Degree initial prediction is put into
H, L=[L1, L2, L3] is enabled, model M is added in LlssvmOr model MplsrIn modeling sample used, implementation model
MlssvmOr MplsrUpdate, and L is rejected from unlabelled N number of unmarked apple sample；
I, step c-h is repeated, until meeting maximum iteration T；
J, M apple samples to be predicted are placed near infrared spectra collection system, acquire and obtain the M wait for it is pre-
Near infrared spectrum of the apple sample of survey under B wave band, the apple sample to be predicted are current year apple sample；
K, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；It extracts under B wave band
The relative reflectance of near infrared spectrum, the characteristic parameter F as apple sample to be predicted；
L, the characteristic parameter F of apple sample to be predicted is brought into newer model Mslssvm(or model Mplsr) in, it treats
Prediction apple sample is predicted.
In the step d, the calculation expression of maximum distance is：
Wherein, UiIndicate the spectral signature of i-th of unmarked apple sample, SjIndicate j-th of training set that model has been established
The spectral signature of sample,Indicate the true pol value of j-th of sample of training set；Expression has been established model pair and does not mark for i-th
Remember the prediction pol value of apple sampleOr| | | | calculate mahalanobis distance.
Beneficial effects of the present invention：The purpose of the present invention is overcoming disadvantage of the existing technology, provide it is a kind of based on away from
Across the time apple sugar content prediction technique of near infrared spectrum from measurement and semi-supervised learning, can improve across time apple sugar content
Precision of prediction, it is easy to operate, quickly and effectively, and have many advantages, such as higher robustness.
Description of the drawings
The present invention is based on across the time apple sugar content predictions of the near infrared spectrum of distance metric and semi-supervised learning according to Fig. 1
Method flow schematic diagram.
Fig. 2 (a) is model MplsrThe scatter plot of test set pol actual value and predicted value before update.
Fig. 2 (b) is model MplsrThe scatter plot of updated test set pol actual value and predicted value.
Fig. 3 (a) is model MlssvmThe scatter plot of test set pol actual value and predicted value before update.
Fig. 3 (b) is model MlssvmThe scatter plot of updated test set pol actual value and predicted value.
Specific implementation mode
Below in conjunction with specific drawings and examples, the present invention will be further described.
As shown in Figure 1：A kind of across the time apple sugar content prediction of near infrared spectrum based on distance metric and semi-supervised learning
Method, steps are as follows：
Apple variety chooses Gold Delicious apple (Golden Delicious), is derived from 782 apple sample conducts in 2009
The training set of regression model, 600 unmarked apple samples for being derived from 2010 separately take 600 in 2010 as unmarked collection
A apple sample to be predicted is as collection to be predicted.
A, 600 unmarked apple samples are individually positioned near infrared spectra collection system, acquire and obtain and does not mark
Remember near infrared spectrum of the apple sample under 129 wave bands；
B, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；Extract 129 wave bands
The relative reflectance of lower near infrared spectrum, the characteristic parameter F as unmarked apple sample；
C, by the characteristic parameter F of the unmarked apple sample of gained in step b, input pre-establishes or newer least square
Support vector regression model MlssvmWith Partial Least-Squares Regression Model MplsrIn, obtain the initial of unmarked apple sample pol
Predicted valueWith
D, 600 unmarked apple samples are calculated to model Mlssvm(or model Mplsr) used in modeling sample maximum distanceWith
The calculation expression of maximum distance is：
Wherein,It is i-th of unmarked apple sample to model Mlssvm(or model Mplsr) used in modeling sample maximum
Distance,It is i-th of unmarked apple sample to model Mlssvm(or model Mplsr) used in modeling sample maximum distance,WithIt is by two calculated maximum distances of different formulas；UiIndicate the spectral signature of i-th of unmarked apple sample, SjIt indicates
The spectral signature of j-th of training set sample that model has been established,Indicate the true pol value of j-th of sample of training set；Table
Show the prediction pol value that i-th of unmarked apple sample of model pair has been establishedOr| | | | calculate mahalanobis distance；
E, least square method supporting vector machine regression model M is calculatedlssvmWith Partial Least-Squares Regression Model MplsrNot to 600
Mark the forecasted variances E=[e of apple sample1,...,ei,...,e600], wherein| | expression asks absolute
Value；
F, 5 big unmarked apple samples of maximum range value are chosen from D1 and D2 respectively, by 5 unmarked apple samples
This and its true pol value are respectively put intoWith
G, 5 small unmarked apple samples of forecasted variances value are chosen from E, by 5 unmarked apple samples and its sugar
Degree initial prediction is put into
H, L=[L1, L2, L3] is enabled, model M is added in LlssvmOr model MplsrIn modeling sample used, implementation model
MlssvmOr MplsrUpdate, and L is rejected from unlabelled 600 unmarked apple samples；
I, step c-h is repeated, until meeting maximum iteration 10；
J, 600 apple samples to be predicted are placed near infrared spectra collection system, acquire and obtain this 600
The near infrared spectrum of apple sample to be predicted under 129 wave bands；
K, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；Extract 129 wave bands
The relative reflectance of lower near infrared spectrum, the characteristic parameter F as apple sample to be predicted；
L, the characteristic parameter F of apple sample to be predicted is brought into newer model Mslssvm(or model Mplsr) in, it treats
Prediction apple sample is predicted.
Utilize model MplsrNewer experimental result, as shown in Fig. 2 (a), Fig. 2 (b) and table 1：
1 model M of tableplsrUpdated test set pol actual value and predicted value
By 1 updated model M of tableplsr7.9%, RMSET is improved to the RPT of the RPT ratios of test set not more new model
Reduce 5.5%；By Fig. 2 (a), Fig. 2 (b) it is found that with model M is not updatedplsrIt compares, updated model MplsrModel is to surveying
Try the predicted value of collection and the closer both sides for being distributed in fitted regression line of scatter plot of actual value, show based on distance metric and
The apple sugar content Partial Least Squares Regression prediction model update method of semi-supervised learning is effective.In turn, with the apple that straddles over year
Fruit sample is to model MplsrContinuous renewal, the precision of prediction of apple sample pol to be predicted will be continuously improved, also more close
Actual pol value.
Utilize model MlssvmNewer experimental result, as shown in Fig. 3 (a), Fig. 3 (b) and table 2：
2 model M of tablelssvmUpdated test set pol actual value and predicted value
By 2 updated model M of tablelssvm27.5%, RMSET is improved to the RPT of the RPT ratios of test set not more new model
Reduce 10.9%；By Fig. 3 (a), Fig. 3 (b) it is found that with model M is not updatedlssvmIt compares, updated model MlssvmModel pair
The predicted value of test set and the closer both sides for being distributed in fitted regression line of the scatter plot of actual value show to be based on distance metric
Apple sugar content least square method supporting vector machine prediction model update method with semi-supervised learning is effective.In turn, with across
Annual apple sample is to model MlssvmContinuous renewal, the precision of prediction of apple sample pol to be predicted will be continuously improved, also more
To approach actual pol value.
Claims (2)
1. a kind of across the time apple sugar content prediction technique of near infrared spectrum based on distance metric and semi-supervised learning, feature exist
In steps are as follows：
A, N number of unmarked apple sample is individually positioned near infrared spectra collection system, acquires and obtains unmarked apple
Near infrared spectrum of the sample under B wave band, the unmarked apple sample are current year apple sample；
B, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；It extracts close red under B wave band
The relative reflectance of external spectrum, the characteristic parameter F as unmarked apple sample；
C, by the characteristic parameter F of the unmarked apple sample of gained in step b, input pre-establishes or newer least square is supported
Vector machine regression model MlssvmWith Partial Least-Squares Regression Model MplsrIn, obtain the initial predicted of unmarked apple sample pol
ValueWithThe regression model pre-established
MlssvmAnd MplsrIn modeling sample be upper one year acquisition apple sample；
D, N number of unmarked apple sample is calculated to model MlssvmOr model MplsrThe maximum distance of modeling sample usedWithWherein,It is i-th of unmarked apple sample to model
MlssvmOr model MplsrThe maximum distance of modeling sample used,It is i-th of unmarked apple sample to model MlssvmOr model
MplsrThe maximum distance of modeling sample used,WithIt is by two calculated maximum distances of different formulas；
E, least square method supporting vector machine regression model M is calculatedlssvmWith Partial Least-Squares Regression Model MplsrTo N number of unmarked apple
Forecasted variances E=[the e of fruit sample1,...,ei,...,eN], wherein| | absolute value is sought in expression；
F, big q unmarked apple samples of maximum range value are chosen from D1 and D2 respectively, by q unmarked apple samples and
Its true pol value is respectively put intoWith
G, small q unmarked apple sample of forecasted variances value is chosen from E, it will be at the beginning of q unmarked apple samples and its pol
Beginning predicted value is put into
H, L=[L1, L2, L3] is enabled, model M is added in LlssvmOr model MplsrIn modeling sample used, implementation model MlssvmOr
MplsrUpdate, and L is rejected from unlabelled N number of unmarked apple sample；
I, step c-h is repeated, until meeting maximum iteration T；
J, M apple samples to be predicted are placed near infrared spectra collection system, acquire and obtain the M and is a to be predicted
Near infrared spectrum of the apple sample under B wave band, the apple sample to be predicted are current year apple sample；
K, it minimizes the needle position misalignment of near infrared spectrum using first derivative and enhances absorption peak；It extracts close red under B wave band
The relative reflectance of external spectrum, the characteristic parameter F as apple sample to be predicted；
L, the characteristic parameter F of apple sample to be predicted is brought into newer model MslssvmOr model MplsrIn, to apple to be predicted
Fruit sample is predicted.
2. across the time apple sugar of a kind of near infrared spectrum based on distance metric and semi-supervised learning according to claim 1
Spend prediction technique, which is characterized in that in the step d, the calculation expression of maximum distance is：
Wherein, UiIndicate the spectral signature of i-th of unmarked apple sample, SjIndicate j-th of training set sample that model has been established
Spectral signature,Indicate the true pol value of j-th of sample of training set；I-th of unmarked apple of model pair has been established in expression
The prediction pol value of fruit sampleOr| | | | calculate mahalanobis distance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810319258.6A CN108507972B (en) | 2018-04-11 | 2018-04-11 | Near infrared spectrum trans-year apple brix prediction method based on distance measurement and semi-supervised learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810319258.6A CN108507972B (en) | 2018-04-11 | 2018-04-11 | Near infrared spectrum trans-year apple brix prediction method based on distance measurement and semi-supervised learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108507972A true CN108507972A (en) | 2018-09-07 |
| CN108507972B CN108507972B (en) | 2020-07-24 |
Family
ID=63381440
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810319258.6A Active CN108507972B (en) | 2018-04-11 | 2018-04-11 | Near infrared spectrum trans-year apple brix prediction method based on distance measurement and semi-supervised learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108507972B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111504944A (en) * | 2020-06-02 | 2020-08-07 | 江南大学 | Statistical monitoring method of citric acid fermentation liquefied clear liquid based on near infrared spectrum |
| CN112884050A (en) * | 2021-02-26 | 2021-06-01 | 江南大学 | Quality detection method based on unsupervised active learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832926A (en) * | 2010-03-19 | 2010-09-15 | 江南大学 | Method for performing apple powder materialization non-destructive inspection by using hyper-spectral image technique |
| CN101949834A (en) * | 2010-08-02 | 2011-01-19 | 扬州福尔喜果蔬汁机械有限公司 | Method for detecting and grading internal quality of fruits |
| CN105527244A (en) * | 2015-10-26 | 2016-04-27 | 沈阳农业大学 | Near infrared spectrum-based Hanfu apple quality nondestructive test method |
| CN105675539A (en) * | 2016-01-07 | 2016-06-15 | 北京市农林科学院 | Comprehensive evaluation method of quality of agricultural products |
| CN106769970A (en) * | 2016-12-07 | 2017-05-31 | 甘肃萃英大农科技有限公司 | A kind of method of the kind, pol and acidity of Fast nondestructive evaluation apple |
| CN106841103A (en) * | 2017-03-01 | 2017-06-13 | 沈阳农业大学 | Near infrared spectrum detects fruit internal quality method and dedicated test system |
| CN107064047A (en) * | 2017-03-02 | 2017-08-18 | 兰州大学 | A kind of Fuji apple quality damage-free detection method based near infrared spectrum |
-
2018
- 2018-04-11 CN CN201810319258.6A patent/CN108507972B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832926A (en) * | 2010-03-19 | 2010-09-15 | 江南大学 | Method for performing apple powder materialization non-destructive inspection by using hyper-spectral image technique |
| CN101949834A (en) * | 2010-08-02 | 2011-01-19 | 扬州福尔喜果蔬汁机械有限公司 | Method for detecting and grading internal quality of fruits |
| CN105527244A (en) * | 2015-10-26 | 2016-04-27 | 沈阳农业大学 | Near infrared spectrum-based Hanfu apple quality nondestructive test method |
| CN105675539A (en) * | 2016-01-07 | 2016-06-15 | 北京市农林科学院 | Comprehensive evaluation method of quality of agricultural products |
| CN106769970A (en) * | 2016-12-07 | 2017-05-31 | 甘肃萃英大农科技有限公司 | A kind of method of the kind, pol and acidity of Fast nondestructive evaluation apple |
| CN106841103A (en) * | 2017-03-01 | 2017-06-13 | 沈阳农业大学 | Near infrared spectrum detects fruit internal quality method and dedicated test system |
| CN107064047A (en) * | 2017-03-02 | 2017-08-18 | 兰州大学 | A kind of Fuji apple quality damage-free detection method based near infrared spectrum |
Non-Patent Citations (4)
| Title |
|---|
| DONGSHENG GUO 等: "Model updating for apple firmness prediction based on hyperspectral scattering images with semi-supervised study and auxiliary data", 《AN ASABE MEETING PRESENTATION》 * |
| QIBING ZHU: "Wavelength Selection of Hyperspectral Scattering Image Using New Semi-supervised Affinity Propagation for Prediction of Firmness and Soluble Solid Content in Apples", 《FOOD ANAL. METHODS》 * |
| 欧阳爱国: "苹果可溶性固形物近红外光谱检测的偏最小二乘回归变量筛选研究", 《光谱学与光谱分析》 * |
| 黄文倩: "以高光谱数据有效预测苹果可溶性固形物含量", 《光谱学与光谱分析》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111504944A (en) * | 2020-06-02 | 2020-08-07 | 江南大学 | Statistical monitoring method of citric acid fermentation liquefied clear liquid based on near infrared spectrum |
| CN112884050A (en) * | 2021-02-26 | 2021-06-01 | 江南大学 | Quality detection method based on unsupervised active learning |
| CN112884050B (en) * | 2021-02-26 | 2024-04-12 | 江南大学 | Quality detection method based on unsupervised active learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108507972B (en) | 2020-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105608293B (en) | Merge Biomass retrieval method and system on the woodland of spectrum and textural characteristics | |
| Ranjan et al. | Assessment of plant nitrogen stress in wheat (Triticum aestivum L.) through hyperspectral indices | |
| CN108875913B (en) | Tricholoma matsutake rapid nondestructive testing system and method based on convolutional neural network | |
| CN102550374B (en) | Crop irrigation system combined with computer vision and multi-sensor | |
| Liu et al. | Estimating winter wheat plant water content using red edge parameters | |
| CN102788752B (en) | Non-destructive detection device and method of internal information of crops based on spectrum technology | |
| US20230024846A1 (en) | Crop yield prediction method and system | |
| CN103293111A (en) | Wheat leaf layer nitrogen content spectral monitoring mode under soil background interference and modeling method | |
| CN103940748B (en) | Based on the prediction of oranges and tangerines canopy nitrogen content and the visualization method of hyperspectral technique | |
| CN106248746A (en) | A kind of milk protein method for quickly detecting contents based on dielectric and magnetic technology | |
| CN110119717A (en) | A kind of Crop classification method based on multi-temporal NDVI and LST | |
| CN108520127A (en) | A kind of EO-1 hyperion inversion method of seeds leaf area index | |
| Dong et al. | Nondestructive egg freshness assessment from the equatorial and blunt region based on visible near infrared spectroscopy | |
| CN103714341B (en) | Hyper spectral reflectance data light spectrum signature extracting method based on overall situation sensitivity analysis | |
| CN104778349B (en) | One kind is used for rice table soil nitrogen application Classified Protection | |
| CN106485345A (en) | Cotton Gossypii time of infertility canopy SPAD value remote sensing appraising and appraising model construction method | |
| CN108507972A (en) | A kind of across the time apple sugar content prediction technique of near infrared spectrum based on distance metric and semi-supervised learning | |
| Cozzolino | The role of near-infrared sensors to measure water relationships in crops and plants | |
| CN103278467A (en) | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf | |
| Zhang et al. | Comparison of various approaches for estimating leaf water content and stomatal conductance in different plant species using hyperspectral data | |
| CN105548027A (en) | Analytical model and method for determining content of tea oil in blend oil based on near infrared spectroscopy | |
| CN105954207A (en) | Method for detecting content of chlorophyll b in canopy of jujube tree | |
| CN103913425B (en) | The Relation To Grain Protein of Winter Wheat content prediction method be coupled based on spectrum index and climatic factor and the construction method of forecast model thereof | |
| Yi et al. | Evaluating the performance of PC-ANN for the estimation of rice nitrogen concentration from canopy hyperspectral reflectance | |
| CN105527236A (en) | Method for determination of main nutritional components of agricultural product by use of spectroscopy method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |