NL2033033B1 - Computer-based method for detection of available nutrient content in jujube orchard soil - Google Patents
Computer-based method for detection of available nutrient content in jujube orchard soil Download PDFInfo
- Publication number
- NL2033033B1 NL2033033B1 NL2033033A NL2033033A NL2033033B1 NL 2033033 B1 NL2033033 B1 NL 2033033B1 NL 2033033 A NL2033033 A NL 2033033A NL 2033033 A NL2033033 A NL 2033033A NL 2033033 B1 NL2033033 B1 NL 2033033B1
- Authority
- NL
- Netherlands
- Prior art keywords
- spectral
- data set
- nutrient content
- available nutrient
- soil
- Prior art date
Links
- 239000002689 soil Substances 0.000 title claims abstract description 115
- 239000002420 orchard Substances 0.000 title claims abstract description 63
- 235000021049 nutrient content Nutrition 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000001514 detection method Methods 0.000 title abstract description 28
- 240000008866 Ziziphus nummularia Species 0.000 title 1
- 230000003595 spectral effect Effects 0.000 claims abstract description 108
- 241001247821 Ziziphus Species 0.000 claims abstract description 62
- 235000015097 nutrients Nutrition 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 18
- 238000012216 screening Methods 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 230000002860 competitive effect Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 abstract description 9
- 230000000694 effects Effects 0.000 abstract description 5
- 238000013501 data transformation Methods 0.000 abstract description 3
- 230000007613 environmental effect Effects 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 15
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000002203 pretreatment Methods 0.000 description 3
- 235000021538 Chard Nutrition 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- ZLMJMSJWJFRBEC-UHFFFAOYSA-N Potassium Chemical compound [K] ZLMJMSJWJFRBEC-UHFFFAOYSA-N 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 244000126002 Ziziphus vulgaris Species 0.000 description 1
- 235000008529 Ziziphus vulgaris Nutrition 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004720 fertilization Effects 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910052736 halogen Inorganic materials 0.000 description 1
- 150000002367 halogens Chemical class 0.000 description 1
- 230000003301 hydrolyzing effect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 239000011591 potassium Substances 0.000 description 1
- 229910052700 potassium Inorganic materials 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/24—Earth materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/24—Earth materials
- G01N33/245—Earth materials for agricultural purposes
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01C—PLANTING; SOWING; FERTILISING
- A01C21/00—Methods of fertilising, sowing or planting
- A01C21/007—Determining fertilization requirements
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Food Science & Technology (AREA)
- Analytical Chemistry (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (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 disclosure provides a computer—based method for detection of available nutrient content in jujube orchard soil, which performs data transformation processing on multiple bands of spectral reflectance data of the jujube orchard soil using logarithmic transformation, standard normal variate and first—order differential transformation, to further screen. out, using CARS, multiple bands of spectral reflectance data corresponding to the available nutrient content of the jujube orchard soil, and build a soil available nutrient content detection model using the multiple bands of spectral reflectance data screened out; the CARS screens out feature bands of the jujube orchard soil available nutrient and the indoor spectral reflectance, reducing the redundancy between spectral bands and the interference of partial environmental noises, greatly increasing the model prediction ability for the jujube orchard soil available nutrient content, thereby achieving a higher prediction effect and an accurate prediction of the jujube orchard soil available nutrient content.
Description
P1589/NLpd
COMPUTER-BASED METHOD FOR DETECTION OF AVAILABLE NUTRIENT CONTENT
IN JUJUBE ORCHARD SOIL
The disclosure relates to the technical field of detection of available nutrient content in soil, more particularly to a comput- er-based method for detection of available nutrient content in ju- jube orchard soil.
Jujube trees have a low demand on growth environment, even can properly grow in those areas with poor soil texture and high salinization; in addition, they are simple in management and have high nutritional values and economic values. However, there are certain risks and challenges during the production management pro- cess of jujubes, the high-quality requirements of modern life for food also put forward a high demand on the production management of jujubes; therefore, scientific and accurate fertilization plans are required to develop intelligent production management so as to promote the high-quality development of the jujube industry.
At present, the existing methods for detection of soil avail- able nutrient content based on hyperspectral technologies are low in prediction accuracy and cannot accurately predict the available nutrient content in jujube orchard soil.
A computer-based method for detection of available nutrient content in jujube orchard soil, which can accurately predict the available nutrient content in jujube orchard soil, includes: acquiring a first spectral data set of a jujube orchard soil sample using a spectrometer; processing the first spectral data set of the jujube orchard soil sample with logarithmic transformation, standard normal vari- ate and first-order differential transformation respectively using a computer first processor, to obtain a second spectral data set,
a third spectral data set and a fourth spectral data set of the
Jujube orchard soil sample; according to the first spectral data set, the second spectral data set, the third spectral data set and the fourth spectral data set of the jujube orchard soil sample, screening out, using Com- petitive Adaptive Reweighted Sampling (CARS), wvarious spectral feature band data sets corresponding to each available nutrient content of the jujube orchard soil sample, and storing them into a memory; building various training data sets by a computer second pro- cessor, according to each available nutrient content of the jujube orchard soil sample and the various spectral feature band data sets corresponding to each available nutrient content of the ju- jJube orchard soil sample; training and optimizing various soil available nutrient con- tent detection models using the various training data sets, to ob- tain various optimized soil available nutrient content detection models; according to the various optimized soil available nutrient content detection models, determining a best inversion model cor- responding to each available nutrient content of the jujube or- chard soil sample with a preset model evaluation index; acquiring a spectral feature band data set to be detected corresponding to the best inversion model, the spectral feature band data set to be detected being a first spectral feature band data set, a second spectral feature band data set, a third spec- tral feature band data set or a fourth spectral feature band data set corresponding to the available nutrient content of the jujube orchard soil to be detected; and inputting the spectral feature band data set to be detected into the best inversion model, to obtain the available nutrient content of the jujube orchard soil to be detected.
Achieved following technical effects of: reducing the redundancy between spectral bands and the inter- ference of partial environmental noises, greatly increasing the model prediction ability for the jujube orchard soil available nu- trient content, thereby achieving a higher prediction effect and an accurate prediction of the jujube orchard soil available nutri- ent content.
The method for detection of available nutrient content in ju-
Jjube orchard soil includes the following.
Step 101: acquiring a first spectral data set of a jujube or- chard soil sample using a spectrometer; the first spectral data set including hyperspectral data; the hyperspectral data including multiple bands of spectral reflectance data.
Here, the hyperspectral data includes 2001 bands of spectral reflectance data within a wavelength range of 400 nm-2400 nm.
Step 102 specifically includes: processing the first spectral data set with logarithmic transformation, to obtain a second spectral data set; processing the first spectral data set with standard normal variate, to obtain a third spectral data set; and processing the first spectral data set with first-order dif- ferential transformation, to obtain a fourth spectral data set.
Step 103: according to the first spectral data set, the sec- ond spectral data set, the third spectral data set and the fourth spectral data set of the jujube orchard soil sample, screening out, using Competitive Adaptive Reweighted Sampling (CARS), a first spectral feature band data set, a second spectral feature band data set, a third spectral feature band data set and a fourth spectral feature band data set corresponding to each available nu- trient content of the jujube orchard soil sample; the spectral feature band data set including multiple bands of spectral reflec- tance data screened out.
Step 104: building a first training data set, according to each available nutrient content of the jujube orchard soil sample and the first spectral feature band data set corresponding to each available nutrient content of the jujube orchard soil sample.
Step 105: building a second training data set, according to each available nutrient content of the jujube orchard soil sample and the second spectral feature band data set corresponding to each available nutrient content of the jujube orchard soil sample.
Step 106: building a third training data set, according to each available nutrient content of the jujube orchard soil sample and the third spectral feature band data set corresponding to each available nutrient content of the jujube orchard soil sample.
Step 107: building a fourth training data set, according to each available nutrient content of the jujube orchard soil sample and the fourth spectral feature band data set corresponding to each available nutrient content of the jujube orchard soil sample.
Step 108: training and optimizing a first soil available nu- trient content detection model using the first training data set, to obtain an optimized first soil available nutrient content de- tection model.
Step 109: training and optimizing a second soil available nu- trient content detection model using the second training data set, to obtain an optimized second soil available nutrient content de- tection model.
Step 110: training and optimizing a third soil available nu- trient content detection model using the third training data set, to obtain an optimized third soil available nutrient content de- tection model.
Step 111: training and optimizing a fourth soil available nu- trient content detection model using the fourth training data set, to obtain an optimized fourth soil available nutrient content de- tection model.
Step 112: according to the optimized first soil available nu- trient content detection model, the optimized second soil availa- ble nutrient content detection model, the optimized third soil available nutrient content detection model and the optimized fourth soil available nutrient content detection model, determin- ing a best inversion model corresponding to each available nutri- ent content of the jujube orchard soil sample with a preset model evaluation index; the best inversion model being one of the opti- mized first soil available nutrient content detection model, the optimized second soil available nutrient content detection model, the optimized third soil available nutrient content detection mod- el and the optimized fourth soil available nutrient content detec- tion model.
Step 113: acquiring a spectral feature band data set to be detected corresponding to the best inversion model; the spectral feature band data set to be detected being a first spectral fea- ture band data set, a second spectral feature band data set, a 5 third spectral feature band data set or a fourth spectral feature band data set corresponding to the available nutrient content of the jujube orchard soil to be detected.
Before Step 113, the method further includes: acquiring a first spectral data set of the Jujube orchard soil to be detected; processing the first spectral data set of the jujube orchard soil to be detected with logarithmic transformation, standard nor- mal variate and first-order differential transformation respec- tively, to obtain a second spectral data set, a third spectral da- ta set and a fourth spectral data set of the jujube orchard soil to be detected; and according to the first spectral data set, the second spectral data set, the third spectral data set and the fourth spectral data set of the jujube orchard soil to be detected, screening out, us- ing CARS, a first spectral feature band data set, a second spec- tral feature band data set, a third spectral feature band data set and a fourth spectral feature band data set corresponding to the available nutrient content of the jujube orchard soil to be de- tected.
Step 114: inputting the spectral feature band data set to be detected into the best inversion model, to obtain the available nutrient content of the jujube orchard soil to be detected.
The method for detection of available nutrient content in ju-
Jjube orchard soil according to the disclosure specifically in- cludes the following steps. (1) Collection of jujube orchard soil sample
Uniformly collect mixed jujube orchard soil samples using a mesh point sampling method; for each sampling point, it is needed to collect 3 drills of soil samples with a depth of 60cm at a po- sition of 30cm inward from the edge of the crown projection, pre- treat the soil after the soil is air dried, pass the soil samples through a lmm-sieve and then mix them well and divide it into two parts, one part is used for indoor detection of soil available nu- trient content, and one part is used for detection of indoor hy- perspectral data. (2) Collection of spectral data of jujube orchard soil
Measure the indoor hyperspectral data of the soil using a
FieldSpec4-type spectrometer from the ASD company; this instrument has a wavelength range of 350-2500 nm, the spectral resolution is 3 nm in the range 350-700 nm and 10 nm in the range of 1400-2100 nm, the bandwidth is 1.4 nm (3500-1000 nm) and 1.1 nm {1001-2500 nm), and the spectral resampling interval is 1 nm. Measurement is conducted employing a handle with a halogen light source and a field angle of 25 degrees, each sample is measured 10 times re- peatedly, and white board correction is performed once every 15 minutes, and an average value of 10 times of measurement results is taken as the spectral data of this sample. (3) Pretreatment of spectral data
Remove the bands 350-399 nm and 2401-2500 nm with big noises, and only reserve the band 400~2400 nm as a raw spectral data (R) for model establishment and data analysis. Attempt to process the indoor spectral data (400~2400 nm band data, that is, the first spectral data set) of the jujube orchard soil with six data trans- formations such as reciprocal, logarithm, differential, standard normal variate, root sign, continuum removal, etc., and then screen out relevant bands of the soil available nutrient in con- junction with a data dimension reduction method.
Here, after the indoor spectral data of the jujube orchard soil is processed with the six data transformations, what is ob- tained is still 400~2400 nm (a total of 2001 bands) data, the spectral reflectance data of each band is just processed with cor- responding transformations, with the purpose of enhancing the fea- ture relationship between the soil available nutrient content and the spectral pretreated data, thereby improving the model predic- tion ability. (4) Screening of relevant wavelengths of jujube orchard soil available nutrients
The Competitive Adaptive Reweighted Sampling (CARS) extracts 67% of the total number of samples through the Monte Carlo method to build a PLS model, then calculates the weight of the absolute value of the regression coefficient, selects, employing adaptive weighting, the wavelengths in the PLS model for which the weight of the absolute value of the regression coefficient is large, and then selects a subset with a minimum root-mean-square error through N Monte Carlo runs; when a set number of iterations is reached (generally, the number of iterations is set between 50 and 1000, different numbers of iterations are set according to differ- ent soil available nutrient contents, and the number of iteration is set as 200 in the disclosure), or when the root-mean-square er- ror reaches the minimum, this algorithm stops running, thereby screening out the optimal variables relevant to the soil nutrient, that is, reserving each time, through adaptive weighting sampling, the points in the PLS model for which the weight of the absolute value of the regression coefficient is large as a new subset, re- moving the points with a small weight value, then building a PLS model based on the new subset, and after multiple times of calcu- lation, selecting the wavelengths in a subset with a minimum root- mean-square error of PLS model cross validation as feature wave- lengths, and during the course of multiple iterations, screening out spectral feature bands relevant to the soil available nutrient content. The different data pretreatment methods in Step (3) are used in combination with the CARS algorithm to screen out relevant bands of three soil available nutrients (the relevant band of each soil available nutrient includes a first spectral feature band da- ta set, a second spectral feature band data set, a third spectral feature band data set and a fourth spectral feature band data set), for building subsequent models. Here, the three soil availa- ble nutrients are soil alkali-hydrolyzable nitrogen (hydrolytic nitrogen) content (AN), soil available phosphorus content (AP) and soil available potassium content (AK). Since the CARS algorithm screens out relevant bands on the basis of the soil AN, AP and AK contents, the soil available nutrient content related wavelengths screened out according to different contents differ; therefore,
AN, AP and AK are collectively referred to as three soil available nutrients.
The subsequent relevant wavelength screening and model estab-
lishment are both conducted on the basis of the data pretreatment result of the Step (3); however, through verification by model ef- fect, finally, only four spectral data sets obtained through pre- treatments of logarithm, standard normal variate, first-order dif- ferential and raw spectral data, with an obvious improvement ef- fect, are screened out for feature wavelength screening and model establishment study.
The function of the PLS model is not just to select the wave- lengths with large weight of absolute value, but to select the wavelengths with large weight of absolute value first, then screen out, from the wavelengths with large weight of absolute value, a subset of a minimum root-mean-square error according to the re- sults of multiple times of model operations, thereby obtaining the spectral feature bands of the soil available nutrients.
Here, calculating the weight of the absolute value of the re- gression coefficient is to calculate the wavelength with large weight of absolute value, the specific calculation formula is as follows: wp = Li
Yisalail where, |q;| is the absolute value of the regression coeffi- cient of the ith variable, w; is the weight of the coefficient ab- solute value of the ith variable, n is the number of spectral wavelengths left in each sampling.
The process of screening out relevant wavelengths of jujube orchard soil available nutrients using the CARS in this step is a process of wavelength screening realized inside the CARS; when the existing CARS is employed, the algorithm itself can realize the screening of relevant wavelengths of jujube orchard soil available nutrients. (5) Building of inversion models of jujube orchard soil available nutrients
Combine the raw bands, the data transformed bands (through verification by model effect, only four spectral data sets ob- tained through pretreatments of logarithm, standard normal vari- ate, first-order differential and raw spectral data are screened out, with an obvious improvement effect), and their CARS dimen-
sion-reduced bands (soil available nutrient related optimal varia- bles of the three soil available nutrients that are screened out by CARS, wherein four optimal variable data sets are screened out for each available nutrient, and there is a total number of 12 op- timal variable data sets) with PLSR, BPNN methods respectively, to build detection models of soil available nutrient contents.
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
NL2033033A NL2033033B1 (en) | 2022-09-15 | 2022-09-15 | Computer-based method for detection of available nutrient content in jujube orchard soil |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
NL2033033A NL2033033B1 (en) | 2022-09-15 | 2022-09-15 | Computer-based method for detection of available nutrient content in jujube orchard soil |
Publications (1)
Publication Number | Publication Date |
---|---|
NL2033033B1 true NL2033033B1 (en) | 2024-03-22 |
Family
ID=84359971
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
NL2033033A NL2033033B1 (en) | 2022-09-15 | 2022-09-15 | Computer-based method for detection of available nutrient content in jujube orchard soil |
Country Status (1)
Country | Link |
---|---|
NL (1) | NL2033033B1 (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180035605A1 (en) * | 2016-08-08 | 2018-02-08 | The Climate Corporation | Estimating nitrogen content using hyperspectral and multispectral images |
CN110376139A (en) * | 2019-08-05 | 2019-10-25 | 北京绿土科技有限公司 | Soil organic matter content quantitative inversion method based on ground high-spectrum |
CN114660105A (en) * | 2022-04-08 | 2022-06-24 | 成都理工大学 | Soil heavy metal Cd content inversion method, system, medium and computer equipment |
CN115494007A (en) * | 2022-08-03 | 2022-12-20 | 北京市农林科学院信息技术研究中心 | Random forest based high-precision rapid detection method and device for soil organic matters |
-
2022
- 2022-09-15 NL NL2033033A patent/NL2033033B1/en active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180035605A1 (en) * | 2016-08-08 | 2018-02-08 | The Climate Corporation | Estimating nitrogen content using hyperspectral and multispectral images |
CN110376139A (en) * | 2019-08-05 | 2019-10-25 | 北京绿土科技有限公司 | Soil organic matter content quantitative inversion method based on ground high-spectrum |
CN114660105A (en) * | 2022-04-08 | 2022-06-24 | 成都理工大学 | Soil heavy metal Cd content inversion method, system, medium and computer equipment |
CN115494007A (en) * | 2022-08-03 | 2022-12-20 | 北京市农林科学院信息技术研究中心 | Random forest based high-precision rapid detection method and device for soil organic matters |
Non-Patent Citations (1)
Title |
---|
LIU JINBAO ET AL: "Estimation of soil organic matter content based on CARS algorithm coupled with random forest", SPECTROCHIMICA ACTA PART A: MOLECULAR AND BIOMOLECULAR SPECTROSCOPY, ELSEVIER, AMSTERDAM, NL, vol. 258, 20 April 2021 (2021-04-20), XP086575575, ISSN: 1386-1425, [retrieved on 20210420], DOI: 10.1016/J.SAA.2021.119823 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102495005B (en) | Method for diagnosing crop water deficit through hyperspectral image technology | |
CN103913432B (en) | Based on the near-infrared spectrum wavelength system of selection of particle cluster algorithm | |
US20210404952A1 (en) | Method for selection of calibration set and validation set based on spectral similarity and modeling | |
CN109324013B (en) | Near-infrared rapid analysis method for constructing crude oil property by using Gaussian process regression model | |
CN103293111A (en) | Wheat leaf layer nitrogen content spectral monitoring mode under soil background interference and modeling method | |
CN111968080A (en) | Hyperspectrum and deep learning-based method for detecting internal and external quality of Feicheng peaches | |
CN111855591A (en) | Rice overground part carbon-nitrogen ratio remote sensing inversion model and method | |
JPWO2007129648A1 (en) | Method for estimating water stress of plant leaves, apparatus for estimating water stress of plant leaves, and program for estimating water stress of plant leaves | |
CN108169165B (en) | Maltose mixture quantitative analysis method based on terahertz spectrum and image information fusion | |
CN104778349B (en) | One kind is used for rice table soil nitrogen application Classified Protection | |
CN103592255A (en) | Soft method for measuring total protein content of donkey-hide gelatin skin solution on basis of near infrared spectrum technology | |
CN111044516A (en) | Remote sensing estimation method for chlorophyll content of rice | |
CN111855593A (en) | Remote sensing inversion model and method for starch content of rice leaf | |
CN106990056A (en) | A kind of total soil nitrogen spectrum appraising model calibration samples collection construction method | |
CN107632010B (en) | Method for quantifying steel sample by combining laser-induced breakdown spectroscopy | |
CN107202784B (en) | Method for detecting process nodes in rice seed soaking and germination accelerating process | |
CN102854151B (en) | Chemometrics method for classifying sample sets in spectrum analysis | |
Newton | Development of an integrated decision-support model for density management within jack pine stand-types | |
NL2033033B1 (en) | Computer-based method for detection of available nutrient content in jujube orchard soil | |
CN112651173B (en) | Agricultural product quality nondestructive testing method based on cross-domain spectral information and generalizable system | |
AU2021102567A4 (en) | Rapid diagnosis method of soil fertility grade based on hyperspectral data | |
CN116818687A (en) | Soil organic carbon spectrum prediction method and device based on spectrum guide integrated learning | |
CN106529680A (en) | Multiscale extreme learning machine integrated modeling method based on empirical mode decomposition | |
CN113049526B (en) | Corn seed moisture content determination method based on terahertz attenuated total reflection | |
CN108007916B (en) | Confocal micro-Raman measurement model for establishing nitrogen content of rice plant by Hilbert-Huang method |