CN108760655A - A kind of apple sense of taste profile information method for visualizing - Google Patents
A kind of apple sense of taste profile information method for visualizing Download PDFInfo
- Publication number
- CN108760655A CN108760655A CN201810435827.3A CN201810435827A CN108760655A CN 108760655 A CN108760655 A CN 108760655A CN 201810435827 A CN201810435827 A CN 201810435827A CN 108760655 A CN108760655 A CN 108760655A
- Authority
- CN
- China
- Prior art keywords
- sense
- taste
- apple
- sample
- image
- 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
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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
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 invention discloses a kind of apple sense of taste profile information method for visualizing, include the following steps：S1, the acquisition based on electronic tongues apple sample sense of taste data；S2, acquisition and pretreatment based on hyperspectral technique apple sample data；The selection of S3, characteristic wave bands；The structure of S4, apple sense of taste Visualization Model；S5, after GS-SVM model foundations, pre- measured spectral values under 10 wave bands are inputted by point, the prediction of the sense of taste under each point spectral value output valve is found out and is denoted as K (i, j), it converts sense of taste value to corresponding actual color value, apple sense of taste information visualization is presented.The present invention is combined using advantage of the apple sample in EO-1 hyperion single-point difference with electronic tongues sense of taste Global Information detection technique, to realize the visualization of apple sense of taste profile information.
Description
Technical field
The present invention relates to apple sense of taste analysis fields, and in particular to a kind of apple sense of taste profile information method for visualizing.
Background technology
China is apple production big country, and yield accounts for the 65% of apple total output.Apple is rich in minerals and Wei Sheng
Element, it is soluble big, easily it is absorbed by the body, therefore have the title of " running water ", it is one of fruit that people often eat.The sense of taste of apple is believed
For breath as one of an important factor for reflecting its quality, it affects selection when whether most consumers are bought, accurate, efficient
Detection and characterize sense of taste information to breeding, plantation, storage of apple etc. have practical guided significance.Traditional Physico-chemical tests side
Method can not reflect the sense of taste sensory information of apple, and due to the factors influence such as appraise person's psychology and ambient enviroment cause it is most common
Artificial sense appraise result is not objective enough.Based on this, SA-402B types electronic tongues are as gustum intelligent bionic detecting system with it
The advantages that objective, accurate, gradually replaces application of the traditional detection method in sense of taste message context.Inspection is sampled to apple sample
It surveys, batch processed, realizes the detection of sample entirety sense of taste information, can not reflect various sense of taste information on sample space
Distribution situation.
Invention content
To solve the above problems, the present invention provides a kind of apple sense of taste profile information method for visualizing.
To achieve the above object, the technical solution that the present invention takes is：
A kind of apple sense of taste profile information method for visualizing, includes the following steps：
S1, the acquisition based on electronic tongues apple sample sense of taste data
S11, choose same batch, the fruit shape of same place picking just, the apple of uniform in size, zero defect or pollutant,
90 excellent fruits are selected by national standard GB/T10651-2008；
S12, sample number into spectrum is placed on to 20 ± 2 DEG C of room temperature, relative humidity 55 ± 5% is stored 24 hours, and humiture is kept
It is constant；
S13, each apple sample is cleaned, peels, squeeze the juice, supernatant liquor 40mL is filtered to take and is respectively placed in 2 pure surveys
In measuring cup, the detection through electronic tongues obtains three kinds of 90 × 3 dimension pols, tart flavour, saline taste sense of taste numbers at 30 seconds in order of numbers successively
According to；Before test starts, sensor first in positive and negative anodes cleaning solution cleans 90s, after 120s is cleaned in reference solution,
Continue to clean 120s, sensor balance zero 30s in another reference solution；After reaching balance, detection is proceeded by, when test
Between be 30s, every time measure after automatically into cleaning step；
S2, acquisition and pretreatment based on hyperspectral technique apple sample data
S21, high spectrum image is corrected using black and white scaling method by following formula, to eliminate the influence of noise：
In formula, RdDark image, RwThe diffusing reflection image of blank, RsThe original diffusing reflection spectrum image of apple sample, R-
The spectrogram that diffuses after correction.
S22, vectorized process is carried out to the calibrated image of black and white, obtains iamge description curve, after drawing a circle to approve target image,
Specific mask image interested is generated, sample image processing region is controlled；
S23, image cutting is carried out to the image after mask process, other light in addition to place interested of the SPECTRAL REGION after mask
It is 0 to compose numerical value.
S24, pretreated 90 apple spectrum pictures are obtained, wherein since apple is spherical fruit, therefore with height
One apple sample is divided into 4 faces when spectral selection instrument is irradiated sample, and takes number 1-5's respectively on each face
5 area-of-interests, each area-of-interest size are 300 pixels, and the spectrum for seeking 5 entire surfaces of each sample is equal
Value finally obtains the apple data of 90 × 256 dimensions；
The selection of S3, characteristic wave bands
Sweet tea, acid, the corresponding sensitive band of salty 3 kinds of basic sense of taste are chosen respectively, are determined using the coefficient of variation best sensitive
Then wave band quantity uses the method for grey relational grade (GRA) to determine between the sense of taste information of apple sample and spectrum picture
Correlation degree；
The structure of S4, apple sense of taste Visualization Model
The concentration value of each sense of taste information discrete point, wherein hyperplane function, RBF kernel functions and recurrence are predicted using SVM
Function formula (2), (3), (4) are as follows respectively：
K (x, x ')=exp (g Px-x ' P2) (2)
F (x)=w φ (x)+b (3)
In formula, the normal vector of w- hyperplane, φ (x)-nonlinear mapping functions, b- amount of bias, g- spread factors；
S5, sense of taste visualization are presented
After GS-SVM model foundations, the pre- measured spectral values under 10 wave bands are inputted by point, find out each point spectral value
Under the sense of taste prediction output valve be denoted as K (i, j), convert sense of taste value to corresponding actual color value, can by apple sense of taste information
It is presented depending on changing.
Preferably, the Gray Correlation specifically comprises the following steps：First, multiple dimension data are converted to unification
Dimensionless Form, it is a sense of taste information ordered series of numbers to define reference sequence, and it is the spectrum letter under each wave bands of 380-1038nm to compare ordered series of numbers
Ordered series of numbers is ceased to carry out dimensionless processing, and the difference of the sense of taste and spectral information dimension is eliminated with this；Then, each single sense of taste letter is sought
Cease the grey incidence coefficient of ordered series of numbers and the spectral information ordered series of numbers under all band；Finally, the grey relational grade of each wave band is sought.
Preferably, the step S4 based on genetic algorithm (GA) and grid type search for the method for (GS) come to parameter c and g into
Row optimizing, in genetic algorithm, maximum genetic algebra is 100, and initial population number is 20, and the search range of parameter c is 0 to arrive
100, g be 0 to 100；In grid type searching method, parameter optimization is carried out for interval with 0.5, the search range of parameter c and g are
2-10To 210。
Preferably, in the step S4 during establishing model, when carrying out spectroscopic data and sense of taste information MAP,
Spectral value under each wave band carry out again after mean value is sought the foundation of model；Wherein utilize Kennard-Stone methods
2/3 spectroscopic data sample is selected as training set, 1/3 spectroscopic data is selected as forecast set, is realized based on GS-SVM and GA-SVM
Spectrum-sense of taste information visuallization analysis.
The present invention utilizes the advantage and electronic tongues sense of taste Global Information detection technique in apple sample EO-1 hyperion single-point difference
It is combined, to realize the visualization of apple sense of taste profile information, to provide one kind more for the analysis of the apple sense of taste
Accurate analysis method.
Description of the drawings
Fig. 1 is that the spectrum picture in the embodiment of the present invention describes curve.
Fig. 2 is the vector mask image in the embodiment of the present invention.
Fig. 3 is the area-of-interest curve of spectrum in the embodiment of the present invention.
Fig. 4 is the coefficient of variation change curve in the embodiment of the present invention
Fig. 5 is the parameter optimization process schematic in the embodiment of the present invention；
Wherein, (a) is genetic algorithm parameter searching process；(b) it is grid type search parameter searching process.
Fig. 6 is the sense of taste visualization result figure based on GS-SVM in the embodiment of the present invention.
Specific implementation mode
In order to make objects and advantages of the present invention be more clearly understood, the present invention is carried out with reference to embodiments further
It is described in detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to limit this hair
It is bright.
Embodiment
The present embodiment choose same batch, the fruit shape of same place picking just, uniform in size, zero defect or pollutant Ah
Gram threose heart apple is research object, and excellent fruit 90 is selected by national standard GB/T10651-2008；Include the following steps：
S1, the acquisition based on electronic tongues apple sample sense of taste data
Sample number into spectrum is placed on 20 DEG C of room temperature (± 2 DEG C), relative humidity 55% (± 5%) is stored 24 hours, and temperature is kept
Humidity is basically unchanged.In order of numbers measure 90 samples successively, and each apple sample is cleaned, is peeled, is squeezed the juice, is filtered to take
Layer clear liquid 40mL is respectively placed in be measured in 2 pure measurement cups.Before test starts, sensor is first clear in positive and negative anodes cleaning solution
Wash 90s, after 120s is cleaned in reference solution, continue in another reference solution clean 120s, sensor balance returns
Zero 30s.After reaching balance, detection, testing time 30s, automatically into cleaning step after measuring every time are proceeded by.Through
The detection of electronic tongues obtains 90 × 3 three kinds of sense of taste (pol, tart flavour, saline taste) data of dimension at 30 seconds.
S2, acquisition and pretreatment based on hyperspectral technique apple sample data
In high spectrum image gatherer process, because light source under each wave band the otherness of intensity distribution and camera dark current noise
Influence, partial noise information can be mingled with.These noise informations can influence the quality of high spectrum image, and then influence high-spectrum
As the precision and stability of qualitative or quantitative analysis model.Therefore high spectrum image is corrected using black and white scaling method,
To eliminate the influence of noise, shown in formula (1).
In formula, RdDark image, RwThe diffusing reflection image of blank, RsThe original diffusing reflection spectrum image of apple sample, R-
The spectrogram that diffuses after correction.
Vectorized process is carried out to the calibrated image of black and white, obtains iamge description curve as shown in Figure 1, draws a circle to approve target figure
As after, specific mask image interested as shown in Figure 2 is generated, sample image processing region is controlled.Due to the image border of acquisition
Including more spectral noise can increase later data intractability, therefore image cutting is carried out to the image after mask process, covered
SPECTRAL REGION after film other spectra values in addition to place interested are 0.
Obtain pretreated 90 apple spectrum pictures.Wherein, since apple is spherical fruit, therefore with EO-1 hyperion
One apple sample is divided into 4 faces when sorter is irradiated sample, and takes 5 of number 1-5 respectively on each face
Area-of-interest (each area-of-interest size is about 300 pixels).The average light of 5 area-of-interests on each section
Spectrum finally obtains the apple data of 90 × 256 dimensions as shown in figure 3, seek the spectrum average of 5 entire surfaces of each sample.
The selection of S3, characteristic wave bands
Spectrum picture reacts the difference of apple internal quality, and electronic tongues are obtained by the functional group that apple sample middle reaches separate out
Sense of taste information.Therefore correlation degree between the two is determined using the method for grey relational grade (GRA), chooses sweet tea, acid, salty respectively
3 kinds of corresponding sensitive bands of the basic sense of taste.Before carrying out grey correlation analysis, best sensitive wave is determined using the coefficient of variation
Segment number.Comprehensive coefficient shows that more greatly the correlation between variable is lower, research with 2 for interval characteristic wave bands number to become
The influence of different coefficient, shown as seen from Figure 4, characteristic wave bands number achieves the highest coefficient of variation at 10.It is wherein grey
In color degree of association method, first, multiple dimension data are converted to unified Dimensionless Form, it is sense of taste information to define reference sequence
Ordered series of numbers compares ordered series of numbers and carries out dimensionless processing for spectral information ordered series of numbers under each wave bands of 380-1038nm, the sense of taste is eliminated with this
With the difference of spectral information dimension.Then, the ash of each single sense of taste information ordered series of numbers and the spectral information ordered series of numbers under all band is sought
Color incidence coefficient.Finally, the grey relational grade of each wave band is sought.The results are shown in Table 1,10 associations before each single sense of taste information
The higher characteristic wave bands of angle value.
10 characteristic wave bands before 1 each sense of taste of table
The structure of S4, apple sense of taste Visualization Model
Spectrum has discreteness with sense of taste information data, and local linear and non-linear relation are presented therebetween, and SVM
Main thought is to establish a regression hyperplane as decision surface, and multidimensional data is mapped to higher dimensional space using kernel function, is made
It is as linear as possible, to solve the local nonlinearity of initial data, finally make all data in set to hyperplane distance most
Closely.Use SVM to predict the concentration value of each sense of taste information discrete point herein based on this, wherein hyperplane function, RBF kernel functions and
Regression function formula (2), (3), (4) are as follows respectively.SVR modeling process results depend on parameter c and g, correct effective parameter
Selection has good recurrence performance to support vector machines.Therefore, it is based on genetic algorithm (GA) and grid type search (GS) herein
Method to carry out optimizing to parameter c and g, as shown in Figure 5.In genetic algorithm, maximum genetic algebra be 100, initial kind
Group's number is 20, and it is 0 to 100 that the search range of parameter c, which is 0 to 100, g,.In grid type searching method, carried out for interval with 0.5
The search range of parameter optimization, parameter c and g is 2-10To 210.During establishing model, because experiment is detected as whole sample
The sense of taste value of product, therefore when carrying out spectroscopic data and sense of taste information MAP, the spectral value under each wave band is subjected to mean value and is sought
Carry out the foundation of model again afterwards.2/3 spectroscopic data sample is wherein selected as training set using Kennard-Stone methods, 1/3
A spectroscopic data is selected as forecast set, and spectrum-sense of taste information visuallization analysis is realized based on GS-SVM and GA-SVM.As a result such as table
The parameter optimization process of 2 display GA-SVM, genetic algorithm is quickly under the conditions of it is 0.01667 to reach best fitness Mse
Filtered out modeling optimized parameter c be 6.3859, g 83.0159, the parameter optimization process of GS-SVM, grid type search up to
It is that 22.6274, g is to have filtered out modeling optimized parameter c under the conditions of being 0.0037346 to best fitness Mse
0.0019531, therefore the model for selecting GS-SVM to be sought as apple sense of taste value.
K (x, x ')=exp (g Px-x ' P2) (2)
F (x)=w φ (x)+b (3)
The normal vector of w- hyperplane in formula, φ (x)-nonlinear mapping functions, b- amount of bias, g- spread factors
The end value of table 2 c and g searching processes
S5, sense of taste visualization are presented
After GS-SVM model foundations, the pre- measured spectral values under 10 wave bands are inputted by point, find out each point spectral value
Under the sense of taste prediction output valve be denoted as K (i, j), convert sense of taste value to corresponding actual color value, can by apple sense of taste information
It is presented depending on changing, Fig. 6 is sweet taste, tart flavour, saline taste single-point sense of taste distribution map, and legend gradually thickens from navy blue to brick-red taste.
As seen from the figure, for Acker threose heart apple mainly based on sweet taste, tart flavour, saline taste is very light, and sweet taste is most dense and distribution is most wide.
Sweet taste is mainly distributed on equatorial line both sides, and central axes part is compared with sweet tea；The distributing position of tart flavour is similar to sweet taste；Saline taste be then
Equator section more uniformly and is distributed more extensive.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the principle of the present invention, it can also make several improvements and retouch, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (4)
1. a kind of apple sense of taste profile information method for visualizing, which is characterized in that include the following steps：
S1, the acquisition based on electronic tongues apple sample sense of taste data
S11, choose same batch, the fruit shape of same place picking just, the apple of uniform in size, zero defect or pollutant, by state
Mark GB/T10651-2008 selects 90 excellent fruits；
S12, sample number into spectrum is placed on to 20 ± 2 DEG C of room temperature, relative humidity 55 ± 5% is stored 24 hours, keeps humiture constant；
S13, each apple sample is cleaned, peels, squeeze the juice, supernatant liquor 40mL is filtered to take and is respectively placed in 2 pure measurement cups
In, the detection through electronic tongues obtains 90 × 3 dimension pol, tart flavours at 30 seconds, at three kinds of sense of taste data of taste in order of numbers successively；It surveys
Before runin is begun, sensor first in positive and negative anodes cleaning solution cleans 90s, after 120s is cleaned in reference solution, another
Continue to clean 120s, sensor balance zero 30s in kind reference solution；After reaching balance, detection is proceeded by, the testing time is
30s, automatically into cleaning step after measuring every time；
S2, acquisition and pretreatment based on hyperspectral technique apple sample data
S21, high spectrum image is corrected using black and white scaling method by following formula, to eliminate the influence of noise：
In formula, RdDark image, RwThe diffusing reflection image of blank, RSThe original diffusing reflection spectrum image of apple sample, R- corrections
The spectrogram that diffuses afterwards.
S22, vectorized process is carried out to the calibrated image of black and white, obtains iamge description curve, after drawing a circle to approve target image, generated
Specific mask image interested controls sample image processing region；
S23, image cutting is carried out to the image after mask process, other spectrum numbers in addition to place interested of the SPECTRAL REGION after mask
Value is 0.
S24, pretreated 90 apple spectrum pictures are obtained, wherein since apple is spherical fruit, therefore with EO-1 hyperion
One apple sample is divided into 4 faces when sorter is irradiated sample, and takes 5 of number 1-5 respectively on each face
Area-of-interest, each area-of-interest size are 300 pixels, seek the spectrum average of 5 entire surfaces of each sample, most
The apple data of 90 × 256 dimensions are obtained eventually；
The selection of S3, characteristic wave bands
Sweet tea, acid, the corresponding sensitive band of salty 3 kinds of basic sense of taste are chosen respectively, and best sensitive band is determined using the coefficient of variation
Then quantity uses the method for grey relational grade (GRA) to determine being associated between the sense of taste information of apple sample and spectrum picture
Degree；
The structure of S4, apple sense of taste Visualization Model
The concentration value of each sense of taste information discrete point, wherein hyperplane function, RBF kernel functions and regression function are predicted using SVM
Formula (2), (3), (4) are as follows respectively：
K (x, x ')=exp (g Px-x ' P2) (2)
F (x)=ω φ (x)+b (3)
In formula, the normal vector of w- hyperplane, φ (x)-nonlinear mapping functions, b- amount of bias, g- spread factors；
S5, sense of taste visualization are presented
After GS-SVM model foundations, the pre- measured spectral values under 10 wave bands are inputted by point, are found out under each point spectral value
Sense of taste prediction output valve is denoted as K (i, j), sense of taste value is converted to corresponding actual color value, by apple sense of taste information visualization
It presents.
2. a kind of apple sense of taste profile information method for visualizing as described in claim 1, which is characterized in that
The Gray Correlation specifically comprises the following steps：First, multiple dimension data are converted to unified Dimensionless Form,
To define reference sequence be sense of taste information ordered series of numbers, compares ordered series of numbers and is carried out for the spectral information ordered series of numbers under each wave bands of 380-1038nm
Dimensionless processing, the difference of the sense of taste and spectral information dimension is eliminated with this；Then, each single sense of taste information ordered series of numbers and all-wave are sought
The grey incidence coefficient of spectral information ordered series of numbers under section；Finally, the grey relational grade of each wave band is sought.
3. a kind of apple sense of taste profile information method for visualizing as described in claim 1, which is characterized in that the step S4 bases
Come to carry out optimizing to parameter c and g in the method that genetic algorithm (GA) and grid type search for (GS), it is maximum in genetic algorithm
Genetic algebra is 100, and initial population number is 20, and it is 0 to 100 that the search range of parameter c, which is 0 to 100, g,；In grid type searcher
In method, parameter optimization is carried out for interval with 0.5, the search range of parameter c and g are 2-10To 210。
4. a kind of apple sense of taste profile information method for visualizing as described in claim 1, which is characterized in that in the step S4
During establishing model, when carrying out spectroscopic data and sense of taste information MAP, the spectral value under each wave band is carried out equal
Value carries out the foundation of model again after seeking；Wherein 2/3 spectroscopic data sample is selected as training using Kennard-Stone methods
Collection, 1/3 spectroscopic data are selected as forecast set, and spectrum-sense of taste information visuallization analysis is realized based on GS-SVM and GA-SVM.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810435827.3A CN108760655B (en) | 2018-04-28 | 2018-04-28 | Apple taste map information visualization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810435827.3A CN108760655B (en) | 2018-04-28 | 2018-04-28 | Apple taste map information visualization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108760655A true CN108760655A (en) | 2018-11-06 |
| CN108760655B CN108760655B (en) | 2021-03-30 |
Family
ID=64009312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810435827.3A Active CN108760655B (en) | 2018-04-28 | 2018-04-28 | Apple taste map information visualization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108760655B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685134A (en) * | 2018-12-21 | 2019-04-26 | 龙口盛福达食品有限公司 | A kind of baby milk intelligence gustatory analysis method based on unsupervised discretization |
| CN110763698A (en) * | 2019-10-12 | 2020-02-07 | 仲恺农业工程学院 | Hyperspectral citrus leaf disease identification method based on characteristic wavelength |
| CN111289516A (en) * | 2020-03-26 | 2020-06-16 | 中国农业大学 | Method and device for detecting amino acid content of plant leaves |
| CN111289446A (en) * | 2020-03-30 | 2020-06-16 | 天津工业大学 | Method and system for detecting component concentration of complex solution |
| CN112285054A (en) * | 2020-09-08 | 2021-01-29 | 塔里木大学 | Apple identity recognition model establishing method and system and identity recognition method and system |
| EP4070076A4 (en) * | 2019-12-04 | 2024-01-17 | Profileprint Pte. Ltd. | System and method for non-destructive rapid food profiling using artificial intelligence |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060203247A1 (en) * | 2001-07-09 | 2006-09-14 | Mitsui Mining & Smelting Co., Ltd. | Apparatus and method for evaluating the interior quality of fruits and vegetables |
| CN1995987A (en) * | 2007-02-08 | 2007-07-11 | 江苏大学 | Non-destructive detection method and device for agricultural and animal products based on hyperspectral image technology |
| CN101021478A (en) * | 2007-02-13 | 2007-08-22 | 江西农业大学 | Laser inducing fluorescent high spectral image detecting method and device for fruit quality |
| CN201021922Y (en) * | 2007-02-13 | 2008-02-13 | 江西农业大学 | A no damage device for detecting fruit quality |
| 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 |
| CN101853392A (en) * | 2010-04-21 | 2010-10-06 | 河海大学 | Remote sensing hyperspectral image band selection method based on conditional mutual information |
| CN105738311A (en) * | 2015-12-01 | 2016-07-06 | 昆明睿意铂科技股份有限公司 | Apple sweetness non-damage quick detection method based on near-infrared spectrum technology |
| CN105866043A (en) * | 2016-06-06 | 2016-08-17 | 东北电力大学 | Method for detecting apple sour through hyperspectral technology |
-
2018
- 2018-04-28 CN CN201810435827.3A patent/CN108760655B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060203247A1 (en) * | 2001-07-09 | 2006-09-14 | Mitsui Mining & Smelting Co., Ltd. | Apparatus and method for evaluating the interior quality of fruits and vegetables |
| CN1995987A (en) * | 2007-02-08 | 2007-07-11 | 江苏大学 | Non-destructive detection method and device for agricultural and animal products based on hyperspectral image technology |
| CN101021478A (en) * | 2007-02-13 | 2007-08-22 | 江西农业大学 | Laser inducing fluorescent high spectral image detecting method and device for fruit quality |
| CN201021922Y (en) * | 2007-02-13 | 2008-02-13 | 江西农业大学 | A no damage device for detecting fruit quality |
| 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 |
| CN101853392A (en) * | 2010-04-21 | 2010-10-06 | 河海大学 | Remote sensing hyperspectral image band selection method based on conditional mutual information |
| CN105738311A (en) * | 2015-12-01 | 2016-07-06 | 昆明睿意铂科技股份有限公司 | Apple sweetness non-damage quick detection method based on near-infrared spectrum technology |
| CN105866043A (en) * | 2016-06-06 | 2016-08-17 | 东北电力大学 | Method for detecting apple sour through hyperspectral technology |
Non-Patent Citations (4)
| Title |
|---|
| 刘晶晶 等: "高光谱图像技术检测苹果外部损伤", 《中国食品学报》 * |
| 宫元娟 等: "寒富苹果品质无损检测光谱信息在线分析", 《沈阳农业大学学报》 * |
| 张偲敏 等: "支持向量回归机的参数择优算法", 《中国科技信息》 * |
| 马惠玲 等: "基于高光谱成像的苹果品种快速鉴别", 《农业机械学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685134A (en) * | 2018-12-21 | 2019-04-26 | 龙口盛福达食品有限公司 | A kind of baby milk intelligence gustatory analysis method based on unsupervised discretization |
| CN110763698A (en) * | 2019-10-12 | 2020-02-07 | 仲恺农业工程学院 | Hyperspectral citrus leaf disease identification method based on characteristic wavelength |
| EP4070076A4 (en) * | 2019-12-04 | 2024-01-17 | Profileprint Pte. Ltd. | System and method for non-destructive rapid food profiling using artificial intelligence |
| CN111289516A (en) * | 2020-03-26 | 2020-06-16 | 中国农业大学 | Method and device for detecting amino acid content of plant leaves |
| CN111289516B (en) * | 2020-03-26 | 2021-10-08 | 中国农业大学 | Method and device for detecting amino acid content of plant leaves |
| CN111289446A (en) * | 2020-03-30 | 2020-06-16 | 天津工业大学 | Method and system for detecting component concentration of complex solution |
| CN112285054A (en) * | 2020-09-08 | 2021-01-29 | 塔里木大学 | Apple identity recognition model establishing method and system and identity recognition method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108760655B (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108760655A (en) | A kind of apple sense of taste profile information method for visualizing | |
| CN107271382A (en) | A kind of different growing rape leaf SPAD value remote sensing estimation methods | |
| CN108872140A (en) | A kind of method and device of open air monitoring jujube quality | |
| Wang et al. | General model of multi-quality detection for apple from different origins by Vis/NIR transmittance spectroscopy | |
| CN117111092A (en) | High-spatial-resolution remote sensing water quality detection method based on machine learning | |
| CN113418874A (en) | Hyperspectral image reconstruction method and detection method for fruit quality detection | |
| Patel et al. | Rapid assessment of some physical parameters of mangoes using monochrome computer vision | |
| Mu et al. | Non‐destructive detection of blueberry skin pigments and intrinsic fruit qualities based on deep learning | |
| CN110263969B (en) | Dynamic prediction system and prediction method for quality of apples with shelf life | |
| CN104198397B (en) | The method that chamber crop nutrient content is detected under N P and K reciprocation | |
| Guo et al. | Dynamic Nondestructive Detection Models of Apple Quality in Critical Harvest Period Based on Near-Infrared Spectroscopy and Intelligent Algorithms | |
| Qian et al. | Development and application of crop monitoring system for detecting chlorophyll content of tomato seedlings | |
| Wang et al. | Determination of watermelon soluble solids content based on visible/near infrared spectroscopy with convolutional neural network | |
| CN110118735A (en) | A kind of high light spectrum image-forming detection method and device detecting bergamot pear male and female | |
| Li et al. | Multi-sensor multispectral reconstruction framework based on projection and reconstruction | |
| CN106124447A (en) | A kind of based on the method for soluble solid content in near-infrared spectral analysis technology detection Fructus Fragariae Ananssae | |
| CN113807320B (en) | Human skin color recognition method and system based on multispectral imaging technology | |
| CN107389620A (en) | A kind of method of accurate quick measure Cherry Tomato Fruit surface gloss | |
| NL2027491B1 (en) | Method for rapidly identifying identities of pumpkin seeds | |
| CN104574341B (en) | A kind of determination method and apparatus of sugar degree | |
| CN115015180B (en) | Fruit sugar degree nondestructive testing method based on spectral reflectance reconstruction technology | |
| Ueno et al. | Data driven calibration of color-sensitive optical sensor by supervised learning for botanical application | |
| Abe et al. | Quantitative Determination of Optimal Harvest Time by Surface Color Analysis of Tree Fruits | |
| Xu et al. | Nondestructive detection of yellow peach quality parameters based on 3D-CNN and hyperspectral images | |
| FR2949157A1 (en) | PROCESS FOR APPROXIMATING WEATHER VARIABLES |
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 |