CN109324000A - Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR - Google Patents
Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR Download PDFInfo
- Publication number
- CN109324000A CN109324000A CN201811325899.9A CN201811325899A CN109324000A CN 109324000 A CN109324000 A CN 109324000A CN 201811325899 A CN201811325899 A CN 201811325899A CN 109324000 A CN109324000 A CN 109324000A
- Authority
- CN
- China
- Prior art keywords
- cars
- miv
- solids content
- soluble solids
- svr
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Biochemistry (AREA)
- Game Theory and Decision Science (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Multimedia (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR that the invention discloses a kind of.A kind of Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR of the present invention, comprising: obtain the original high spectrum image of each bergamot pear sample, and black and white correction is carried out to high spectrum image；Area-of-interest is extracted, high-spectral data is obtained；It measures the soluble solids content of each bergamot pear sample and divides sample set；Data are pre-processed using standard normal variable method, then extract characteristic wavelength；Using the spectral information that full spectrum of wavelengths information and characteristic wavelength selection method obtain as input vector, support vector regression prediction model is established；According to model prediction as a result, assessment models performance, selects the best characteristic wavelength selection method of prediction effect.Beneficial effects of the present invention: using the combinational algorithm screening characteristic wavelength of competitive adaptive weight weighting algorithm and average influence value-based algorithm to modeling analysis.
Description
Technical field
The present invention relates to Kuerle delicious pear soluble solids contents to predict field, and in particular to one kind is based on CARS-MIV-
The Kuerle delicious pear soluble solids content prediction technique of SVR.
Background technique
High light spectrum image-forming technology has merged spectral information and image information, the chemical component of energy reflected sample and microcosmic knot
Structure.Spectral reflectance factor of many scholars at home and abroad according to sample, combined data transmit interactive device and multivariate statistics tool
The non-destructive testing of fruit is studied, research shows that can effectively be detected using high light spectrum image-forming technology, fruit is common to be lacked
The sunken, index of quality (such as hardness, pol, moisture) and maturity etc..
There are following technical problems for traditional technology:
The detection research of existing Kuerle delicious pear soluble solids content is made a general survey of, some scholars are using competitive adaptive weight
Weighting algorithm (CARS) extracts characteristic wavelength, and in CARS method, variable regression coefficient can be due to the random selection of modeling sample
It changes, the absolute value of regression coefficient cannot reflect variable importance comprehensively, thus influence model inspection precision.
Summary of the invention
The technical problem to be solved in the present invention is to provide one kind precisely, quickly based on the Kuerle of CARS-MIV-SVR
Bergamot pear soluble solids content prediction technique introduces influence size of the Mean Impact Value reflection independent variable to output neuron, into
One step filters out the biggish variable of correlation to modeling analysis, improves model prediction accuracy.
In order to solve the above-mentioned technical problems, the present invention provides a kind of Kuerle delicious pear based on CARS-MIV-SVR is solvable
Property solid content prediction technique, comprising:
Obtain the original high spectrum image of each bergamot pear sample；
Area-of-interest is extracted, high-spectral data is obtained；
It measures the soluble solids content of each bergamot pear sample and divides sample set；
Data are pre-processed using standard normal variable method, then extract characteristic wavelength；
Using the spectral information that full spectrum of wavelengths information and characteristic wavelength selection method obtain as input vector, establishes and support
Vector regression prediction model；
According to model prediction as a result, assessment models performance, selects the best characteristic wavelength selection method of prediction effect.
In one of the embodiments, " obtain the original high spectrum image of each bergamot pear sample, and to high spectrum image into
The correction of row black and white；" in Image Acquisition operation carried out in camera bellows.
In one of the embodiments, after " the original high spectrum image for obtaining each bergamot pear sample ", to high spectrum image
Black and white correction is carried out, then carries out " extracting area-of-interest, obtaining high-spectral data；".
" black and white correction is carried out to high spectrum image in one of the embodiments,；" in black and white updating formula it is as follows:
Wherein, RwFor white uncalibrated image, RbImage, R are corrected for blackboard0For raw noise image, R is the bloom after correction
Spectrogram picture.
It " measures the soluble solids content of each bergamot pear sample in one of the embodiments, and divides sample set；" in,
Using Kennard-Stone method according to the ratio cut partition calibration set and forecast set of 2:1.
In one of the embodiments, " data is pre-processed using standard normal variable method, then extract feature
Wavelength；" in use the combinational algorithm (CARS-MIV) of competitive adaptive weight weighting algorithm and average influence value-based algorithm, competitive
Adaptive weight weighting algorithm (CARS), successive projection algorithm (SPA) or Monte Carlo are selected without information variable elimination algorithm (MCUVE)
Select characteristic wavelength.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
The step of computer program, the processor realizes any one the method when executing described program.
A kind of computer readable storage medium, is stored thereon with computer program, realization when which is executed by processor
The step of any one the method.
A kind of processor, the processor is for running program, wherein described program executes described in any item when running
Method.
Beneficial effects of the present invention:
Using the combinational algorithm screening characteristic wavelength of competitive adaptive weight weighting algorithm and average influence value-based algorithm to
Modeling analysis reduces data redundancy under the premise of guaranteeing model prediction accuracy, realizes Kuerle delicious pear soluble solids and contains
Amount is effectively predicted.
Detailed description of the invention
Fig. 1 is the bloom in the Kuerle delicious pear soluble solids content prediction technique the present invention is based on CARS-MIV-SVR
Compose image capturing system schematic diagram.
Fig. 2 is invention based in the Kuerle delicious pear soluble solids content prediction technique of CARS-MIV-SVR
Pears sample original spectrum reflectance curve.
Fig. 3 is the corresponding MIV value column diagram of 42 characteristic wavelengths that CARS method filters out.
Specific embodiment
The present invention will be further explained below with reference to the attached drawings and specific examples, so that those skilled in the art can be with
It more fully understands the present invention and can be practiced, but illustrated embodiment is not as a limitation of the invention.
1. high spectrum image acquisition and correction
1.1. high spectrum image acquires
Hand picking is of moderate size, the not damaged bergamot pear in surface totally 157 and number consecutively.We are made by manual control
It obtains room temperature and is maintained at 21 DEG C of constant temperature, bergamot pear sample is placed in thermostatic chamber 24 hours, high spectrum image acquisition is then carried out,
For eliminating influence of the temperature to final result.
Present invention application high spectrum image acquisition system obtains high spectrum image, and Fig. 1 is that high spectrum image acquisition system is shown
It is intended to.High light spectrum image-forming acquisition system is acquired soft by hyperspectral imager, double CCD cameras, 4 halogen lamp and a sets of data
Part and computer composition.To reduce influence of the extraneous light to experimental data, Image Acquisition operation carries out in camera bellows, only by
Optical fiber halogen lamp provides light source.The SOC710VP hyperspectral imager that the service system is produced using the U.S., SOC710VP spectrum
Resolution ratio is 4.6875nm, can obtain 400~1000nm wavelength, amounts to the quality data of 128 wavelength.
The bergamot pear for being used to test is placed horizontally at lifting platform center and tested region aligned with camera camera lens, adjustment are taken the photograph
As head focal length makes imaging clearly, setting lifting platform height is 45cm, time of integration 25ms.
1.2. high spectrum image corrects
Since, there are many dark current and the inhomogeneities of illumination, the high spectrum image of acquisition is not in EO-1 hyperion camera
It can be directly used in data analysis, need first to carry out it influence that dark current in camera is eliminated in black and white correction, black and white correction is public
Formula is as follows:
After sample collection, keep every system parameter setting constant, scanning U.S. NIST, which traces to the source, to be calibrated reference plate and obtain
To white uncalibrated image Rw, then cover camera lens lid acquisition blackboard correction image Rb。R0For raw noise image, foundation public affairs
Formula (1) calculates, the high spectrum image R after being corrected.
2. obtaining high-spectral data to area-of-interest is extracted using ENVI software
The spectral information of entire bergamot pear sample surface is contained in high spectrum image after correction, uses ENVI5.3 software
Manually choose 10 × 10 pixel of bergamot pear equatorial positions spectral information as area-of-interest (Region of Interest,
ROI), the region averaged spectrum reflectivity is calculated.
3. measuring the soluble solids content (SSC) of each bergamot pear sample and dividing sample set
3.1. soluble solids content measures
Sample is measured using handheld digital refractometer (Digital Hand-held " Pocket " Refractometer)
SSC value, zero calibration is carried out to refractometer with distilled water before first measurement.At each bergamot pear sample equator position, belt leather is cut
Then manual compression juice drop records the SSC value of bergamot pear sample in refractometer solution holding region to marked region pulp.
3.2. sample set divides
Kennard-Stone (KS) sample set method can guarantee that sample is uniformly distributed according to space length in training set,
Improve prediction model stability and accuracy.Using KS method according to the ratio cut partition calibration set and forecast set of 2:1.
Table 1 is the sample in the Kuerle delicious pear soluble solids content prediction technique the present invention is based on CARS-MIV-SVR
Collect division result
The SSC range of calibration set is 8.9~13.2 ° of Brix, and the SSC range of forecast set is 9~13.2 ° of Brix, calibration set
SSC range preferably cover forecast set, the SSC average and standard deviation of two datasets is close.
4. Pretreated spectra and characteristic wavelength extract
4.1. Pretreated spectra
After black and white correction in obtained spectroscopic data in addition to including bergamot pear essential information, there is also noise of instrument,
Environment stray light etc. predicts unrelated information with bergamot pear SSC value, thus needs to pre-process original spectral data, maximum journey
Degree purification data.It is to carry out the spectral reflectance after data prediction to original spectrum using standard normal variable method shown in Fig. 2
Rate curve.The curve of spectrum trend of 157 samples is similar, without obvious exceptional sample.Spectral absorption peak is Ku Er at 680nm
Caused by strangling the absorption of bergamot pear epidermis Determination of Chlorophyll, absorption peak is related with Kuerle delicious pear sugar content at 820nm, and peak value becomes at 920nm
Change is caused by moisture absorption.
4.2. characteristic wavelength is extracted
The each pixel spectral wavelength of high spectrum image that high spectrum image acquisition system obtains is more continuous, adjacent wavelength
Between similitude it is very high, there are mass data redundancies, will affect the timeliness and accuracy of multi-variables analysis.Therefore, selection can
The wavelength variable subset for sufficiently characterizing whole wavelength informations is particularly important.
CARS-MIV wavelength selection algorithm basic ideas are: being established first using Monte Carlo sampling selection calibration set sample
Then offset minimum binary (PLS) model is chosen by adaptively weighing weight sampling technology (ARS) and decaying exponential function (EDF)
The biggish wavelength of regression coefficient is corresponded in PLS model, obtains the minimum subset of RMSECV value using cross validation.It will in this research
Monte Carlo sampling number is set as 50 times, and cross validation uses ten folding cross validations.It finally found that, when sampling number is 14,
RMSECV value reaches minimum value 0.4894, at this time includes 42 variables in variable subset, as the feature of CARS method choice out
Variable.
In CARS method, modeling sample is randomly choosed by Monte Carlo sampling and is generated, and variable regression coefficient can be because of sample
Random selection and change, the absolute value of regression coefficient can not reflect the importance of variable comprehensively, thus influence model essence
Degree.To reduce this influence, present invention introduces Mean Impact Value (MIV) algorithms to carry out postsearch screening to independent variable, is guaranteeing mould
Data scale is further simplified under the premise of type precision of prediction, improves model robustness.
Mean Impact Value (MIV) algorithm is that influence importance of the neuron to output neuron is inputted in neural network most
Good evaluation index.Independent variable number, IV are indicated with NiIndicate that the influence value of i-th of independent variable, MIV indicate i-th of independent variable
Mean Impact Value, steps are as follows: independent variable x in training set Xi(1≤i≤N) is increased separately or is reduced on the basis of initial value
10%, constitute two new independent variable xi1And xi2, by xi1And xi2Original training set X is added and replaces independent variable xi, newly trained
Collect Pi1And Pi2.Calculate each independent variable xiThe influence value and IV of (1≤i≤N)iMean Impact Value MIVi, such as formula (2) and public affairs
Shown in formula (3).MIV value sign symbol represents relevant direction, and absolute value, which is represented, influences importance to model.To all from change
The MIV absolute value of amount carries out descending sort, obtains independent variable and exports the precedence table for influencing relative importance on network, to judge
Influence degree of the input feature vector to model result.
IVi=yi1-yi2=svr_train (Pi1)-svr_train(Pi2) (2)
Fig. 3 is the corresponding MIV value column diagram of 42 characteristic wavelengths that CARS method filters out.By 42 characteristic wavelengths according to
Its MIV value arranges sequence, it is known that preceding 33 variable MIV accumulation contribution rate is greater than 95%, so characteristic wave long number after postsearch screening
Amount is reduced to 33.
5. establishing branch using the spectral information that full spectrum of wavelengths information and characteristic wavelength selection method obtain as input vector
Hold vector regression (SVR) prediction model.
Model is established to Kuerle using support vector regression (Support Vector Regression, SVR) in experiment
Bergamot pear SSC is predicted.SVR introduces kernel function, and lower dimensional space is mapped to higher dimensional space, converts line for nonlinear problem
Property separable problem solve, solving the higher dimensional space modeling problem of small sample, avoid dimension disaster, the present invention selects radial
Basic function (Radial Basis Function, RBF) is used as kernel function.
Respectively using full spectrum (FS) data, the characteristic wavelength data selected as input, bergamot pear soluble solids content is surveyed
Magnitude establishes SVR prediction model and is compared analysis to result as output.
6. according to model prediction as a result, assessment models performance, selects the best characteristic wavelength selection method of prediction effect
It is as shown in table 2 using different characteristic Wavelength selecting method modeling result.As shown in Table 2, it is established using CARS method
SVR prediction model Rc value is that 0.98241, Rp value is 0.90841, compared to FS-SVR, SPA-SVR, MCUVE-SVR model,
Rc, Rp value are more excellent, while RMSEC and RMSEP value is all smaller.Although characteristic wavelength quantity is more after the screening of CARS method
(being 2.1 times of the characteristic wavelength quantity of SPA selection), but it is smaller in view of sample data volume, and calculating time difference is little, so
In this 3 kinds basic characteristic wavelength selection methods of CRAS, SPA, MCUVE, the SVR prediction model based on CARS algorithm has phase
To preferable estimated performance.
CARS-SVR and CARS-MIV-SVR modeling result are compared, compared to CARS-SVR, CARS-MIV-SVR method is special
It levies number of wavelengths and reduces 21.43%, the two RcValue is not much different, RpIt is 0.94631 that value is increased by 0.90841, while RMSEP value has
It is reduced.This prove CARS-MIV characteristic wavelength extracting method can Effective selection go out characterize spectrum most information wavelength,
The redundancy in initial data is removed, reduces and calculates time, lift scheme precision of prediction.
The present invention uses calibration set related coefficient (Rc), calibration set root-mean-square error (RMSEC), forecast set related coefficient
(Rp) and four parameters of forecast set root-mean-square error (RMSEP) carry out the precision of prediction of assessment models, parameters calculation formula is such as
Under:
In formula, Nc is calibration set number of samples, and Np is forecast set number of samples,Indicate the SSC prediction of i-th of sample
Value, yiRepresent the SSC measured value of i-th of sample, ycAnd ypThe respectively SSC mean value of calibration set sample, forecast set sample.
Table 2 based in the Kuerle delicious pear soluble solids content prediction technique of CARS-MIV-SVR based on different characteristic
Wavelength selecting method modeling result compares
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
The step of computer program, the processor realizes any one the method when executing described program.
A kind of computer readable storage medium, is stored thereon with computer program, realization when which is executed by processor
The step of any one the method.
A kind of processor, the processor is for running program, wherein described program executes described in any item when running
Method.
Embodiment described above is only to absolutely prove preferred embodiment that is of the invention and being lifted, protection model of the invention
It encloses without being limited thereto.Those skilled in the art's made equivalent substitute or transformation on the basis of the present invention, in the present invention
Protection scope within.Protection scope of the present invention is subject to claims.
Claims (9)
1. a kind of Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR characterized by comprising
Obtain the original high spectrum image of each bergamot pear sample；
Area-of-interest is extracted, high-spectral data is obtained；
It measures the soluble solids content of each bergamot pear sample and divides sample set；
Data are pre-processed using standard normal variable method, then extract characteristic wavelength；
The spectral information obtained using full spectrum of wavelengths information and characteristic wavelength selection method establishes supporting vector as input vector
Regressive prediction model；
According to model prediction as a result, assessment models performance, selects the best characteristic wavelength selection method of prediction effect.
2. the Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR as described in claim 1,
It is characterized in that, " obtains the original high spectrum image of each bergamot pear sample, and black and white correction is carried out to high spectrum image；" in image
Acquisition operation carries out in camera bellows.
3. the Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR as described in claim 1,
Be characterized in that, after " the original high spectrum image for obtaining each bergamot pear sample ", to high spectrum image carry out black and white correction, then into
Row " extracts area-of-interest, obtains high-spectral data；".
4. the Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR as claimed in claim 3,
It is characterized in that, " black and white correction is carried out to high spectrum image；" in black and white updating formula it is as follows:
Wherein, RwFor white uncalibrated image, RbImage, R are corrected for blackboard0For raw noise image, R is the high-spectrum after correction
Picture.
5. the Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR as described in claim 1,
It is characterized in that, " measures the soluble solids content of each bergamot pear sample and divide sample set；" in,
Using Kennard-Stone method according to the ratio cut partition calibration set and forecast set of 2:1.
6. the Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR as described in claim 1,
It is characterized in that, " data is pre-processed using standard normal variable method, then extract characteristic wavelength；" middle using competitive
The combinational algorithm (CARS-MIV) of adaptive weight weighting algorithm and average influence value-based algorithm, competitive adaptive weight weighting algorithm
(CARS), successive projection algorithm (SPA) or Monte Carlo are without information variable elimination algorithm (MCUVE) selection characteristic wavelength.
7. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes any one of claims 1 to 6 the method when executing described program
Step.
8. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of any one of claims 1 to 6 the method is realized when row.
9. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run
Benefit requires 1 to 6 described in any item methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325899.9A CN109324000A (en) | 2018-11-08 | 2018-11-08 | Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325899.9A CN109324000A (en) | 2018-11-08 | 2018-11-08 | Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109324000A true CN109324000A (en) | 2019-02-12 |
Family
ID=65261223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811325899.9A Pending CN109324000A (en) | 2018-11-08 | 2018-11-08 | Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109324000A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109765190A (en) * | 2019-02-20 | 2019-05-17 | 中国水稻研究所 | A method of barnyard grass in paddy field is identified using high light spectrum image-forming technology |
| CN109839358A (en) * | 2019-01-22 | 2019-06-04 | 北京农业质量标准与检测技术研究中心 | Analyzing The Quality of Agricultural Products method and device |
| CN109975217A (en) * | 2019-03-26 | 2019-07-05 | 贵阳学院 | Plum soluble solid content value detection method based on Hyperspectral imager |
| CN110020679A (en) * | 2019-03-25 | 2019-07-16 | 中国科学院半导体研究所 | Classification method and device based on one-way analysis of variance selection bloom spectrum wavelength |
| CN110095436A (en) * | 2019-05-30 | 2019-08-06 | 江南大学 | Apple slight damage classification method |
| CN110455722A (en) * | 2019-08-20 | 2019-11-15 | 中国热带农业科学院橡胶研究所 | Rubber tree blade phosphorus content EO-1 hyperion inversion method and system |
| CN110553983A (en) * | 2019-09-05 | 2019-12-10 | 东北农业大学 | Method for rapidly detecting potassium sorbate in milk |
| CN113008815A (en) * | 2021-02-24 | 2021-06-22 | 浙江工业大学 | Hyperspectral image information-based method for nondestructive detection of total flavonoids in spina date seeds |
| CN117668766A (en) * | 2023-11-23 | 2024-03-08 | 沈阳农业大学 | Blueberry soluble solid detection method, device, equipment and medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107300536A (en) * | 2017-08-25 | 2017-10-27 | 天津商业大学 | Soluble solid content Forecasting Methodology after mango impact injury based on EO-1 hyperion |
-
2018
- 2018-11-08 CN CN201811325899.9A patent/CN109324000A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107300536A (en) * | 2017-08-25 | 2017-10-27 | 天津商业大学 | Soluble solid content Forecasting Methodology after mango impact injury based on EO-1 hyperion |
Non-Patent Citations (5)
| Title |
|---|
| JIAN-LAN JIANG ET AL.: "A Novel Approach to Active Compounds Identification Based on Support Vector Regression Model and Mean Impact Value", 《CHEM BIOL DRUG DES》 * |
| MATLAB中文论坛编著: "《MATLAB神经网络30个案例分析》", 30 April 2010, 北京航空航天大学出版社 * |
| 严衍禄等编著: "《近红外光谱分析的原理、技术与应用》", 31 January 2013, 中国轻工业出版社 * |
| 梁琨 等: "基于高光谱和CARS-IRIV算法的‘库尔勒香梨’可溶性固形物含量检测", 《南京农业大学学报》 * |
| 詹白勺 等: "高光谱技术结合CARS算法的库尔勒香梨可溶性固形物定量测定", 《光谱学与光谱分析》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109839358A (en) * | 2019-01-22 | 2019-06-04 | 北京农业质量标准与检测技术研究中心 | Analyzing The Quality of Agricultural Products method and device |
| CN109839358B (en) * | 2019-01-22 | 2021-08-10 | 北京农业质量标准与检测技术研究中心 | Agricultural product quality analysis method and device |
| CN109765190A (en) * | 2019-02-20 | 2019-05-17 | 中国水稻研究所 | A method of barnyard grass in paddy field is identified using high light spectrum image-forming technology |
| CN110020679A (en) * | 2019-03-25 | 2019-07-16 | 中国科学院半导体研究所 | Classification method and device based on one-way analysis of variance selection bloom spectrum wavelength |
| CN109975217A (en) * | 2019-03-26 | 2019-07-05 | 贵阳学院 | Plum soluble solid content value detection method based on Hyperspectral imager |
| CN110095436A (en) * | 2019-05-30 | 2019-08-06 | 江南大学 | Apple slight damage classification method |
| CN110455722A (en) * | 2019-08-20 | 2019-11-15 | 中国热带农业科学院橡胶研究所 | Rubber tree blade phosphorus content EO-1 hyperion inversion method and system |
| CN110553983A (en) * | 2019-09-05 | 2019-12-10 | 东北农业大学 | Method for rapidly detecting potassium sorbate in milk |
| CN110553983B (en) * | 2019-09-05 | 2022-03-04 | 东北农业大学 | Method for rapidly detecting potassium sorbate in milk |
| CN113008815A (en) * | 2021-02-24 | 2021-06-22 | 浙江工业大学 | Hyperspectral image information-based method for nondestructive detection of total flavonoids in spina date seeds |
| CN117668766A (en) * | 2023-11-23 | 2024-03-08 | 沈阳农业大学 | Blueberry soluble solid detection method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109324000A (en) | Kuerle delicious pear soluble solids content prediction technique based on CARS-MIV-SVR | |
| CN111968080B (en) | Method for detecting quality of inside and outside of Feicheng peaches based on hyperspectral and deep learning | |
| CN103411973B (en) | A kind of based on anthocyanin content method for measuring in the vinifera pericarp of EO-1 hyperion | |
| CN108956505B (en) | Method and device for detecting concentration of chlorophyll a in small water body based on Sentinel-2 image | |
| Qin et al. | Prediction of apple internal quality using spectral absorption and scattering properties | |
| CN108956604A (en) | A method of Eriocheir sinensis quality is identified based on hyper-spectral image technique | |
| CN108760655B (en) | Apple taste map information visualization method | |
| Yuan et al. | Non-invasive measurements of ‘Yunhe’pears by vis-NIRS technology coupled with deviation fusion modeling approach | |
| CN109540836A (en) | Near infrared spectrum pol detection method and system based on BP artificial neural network | |
| CN111044483A (en) | Method, system and medium for determining pigment in cream based on near infrared spectrum | |
| Haijun et al. | Predicting sandy soil moisture content with hyperspectral imaging | |
| CN116482317B (en) | Lake water nutrition state real-time monitoring method, system, equipment and medium | |
| CN105021535A (en) | Nondestructive testing method and system of fatty acid content of rice | |
| CN109975217A (en) | Plum soluble solid content value detection method based on Hyperspectral imager | |
| CN113607656A (en) | Leaf chlorophyll content monitoring method and system based on hyperspectral imaging | |
| Zhaoyong et al. | Detection of moldy core in apples and its symptom types using transmittance spectroscopy | |
| Ekramirad et al. | Application of hyperspectral imaging and acoustic emission techniques for apple quality prediction | |
| Xin et al. | Construction of spectral detection models to evaluate soluble solids content and acidity in Dangshan pear using two different sensors | |
| CN105866043A (en) | Method for detecting apple sour through hyperspectral technology | |
| CN104198397A (en) | Method for detecting nutrient content of greenhouse crops under nitrogen, phosphorus and potassium interaction | |
| CN116310788A (en) | Plant leaf nitrogen content hyperspectral modeling method based on map fusion | |
| CN116148210A (en) | Fruit quality detection method and device | |
| CN114813593A (en) | Method for detecting total acid content of fermented grains based on hyperspectral imaging technology | |
| CN115931727A (en) | Chlorophyll a remote sensing inversion method, device, system and storage medium | |
| Ali et al. | Prospect inversion for indirect estimation of leaf dry matter content and specific leaf area |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190212 |