CN108760655B - Apple taste map information visualization method - Google Patents
Apple taste map information visualization method Download PDFInfo
- Publication number
- CN108760655B CN108760655B CN201810435827.3A CN201810435827A CN108760655B CN 108760655 B CN108760655 B CN 108760655B CN 201810435827 A CN201810435827 A CN 201810435827A CN 108760655 B CN108760655 B CN 108760655B
- Authority
- CN
- China
- Prior art keywords
- taste
- apple
- spectral
- image
- information
- 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.)
- Active
Links
- 235000019640 taste Nutrition 0.000 title claims abstract description 72
- 238000007794 visualization technique Methods 0.000 title claims description 6
- 230000003595 spectral effect Effects 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000001228 spectrum Methods 0.000 claims abstract description 11
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 238000010276 construction Methods 0.000 claims abstract description 4
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 241000220225 Malus Species 0.000 claims description 58
- 230000008569 process Effects 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 9
- 235000013399 edible fruits Nutrition 0.000 claims description 9
- 230000002068 genetic effect Effects 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 9
- 235000021016 apples Nutrition 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 239000012088 reference solution Substances 0.000 claims description 6
- 239000006228 supernatant Substances 0.000 claims description 6
- 235000009508 confectionery Nutrition 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 235000019614 sour taste Nutrition 0.000 claims description 5
- 238000005520 cutting process Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 claims description 3
- 239000003344 environmental pollutant Substances 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 231100000719 pollutant Toxicity 0.000 claims description 3
- 239000000243 solution Substances 0.000 claims description 3
- 239000004575 stone Substances 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 9
- 238000012800 visualization Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 9
- 238000009826 distribution Methods 0.000 description 5
- 235000019643 salty taste Nutrition 0.000 description 5
- 235000019605 sweet taste sensations Nutrition 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 239000011449 brick Substances 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 125000000524 functional group Chemical group 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 239000011664 nicotinic acid Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 235000019613 sensory perceptions of taste Nutrition 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000035923 taste sensation Effects 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
本发明公开了一种苹果味觉图谱信息可视化方法，包括如下步骤：S1、基于电子舌苹果样本味觉数据的获取；S2、基于高光谱技术苹果样本数据的获取与预处理；S3、特征波段的选取；S4、苹果味觉可视化模型的构建；S5、在GS‑SVM模型建立之后，将10个波段下的预测光谱值按点输入，求出各点光谱值下的味觉预测输出值记为K(i，j)，将味觉值转化为对应的实际色彩值，将苹果味觉信息可视化呈现。本发明利用苹果样本在高光谱单点差异上的优点与电子舌味觉整体信息检测技术相结合，从而实现了苹果味觉图谱信息的可视化。
The invention discloses a method for visualizing apple taste spectrum information, comprising the following steps: S1, acquisition of taste data of apple samples based on electronic tongue; S2, acquisition and preprocessing of apple sample data based on hyperspectral technology; S3, selection of characteristic bands S4, the construction of apple taste visualization model; S5, after the GS-SVM model is established, the predicted spectral values under 10 bands are input by point, and the predicted output value of taste under the spectral values of each point is obtained and recorded as K(i , j), convert the taste value into the corresponding actual color value, and visualize the apple taste information. The invention utilizes the advantages of the apple sample in the hyperspectral single point difference and combines the electronic tongue taste overall information detection technology, thereby realizing the visualization of the apple taste spectrum information.
Description
Technical Field
The invention relates to the field of apple taste analysis, in particular to an apple taste map information visualization method.
Background
China is a big apple producing country, and the yield of China accounts for 65% of the total apple yield. The apple fruit is rich in minerals and vitamins, has high solubility and is easy to be absorbed by human bodies, so that the apple fruit is called as 'running water', and is one of the fruits which are often eaten by people. As one of the important factors for reflecting the quality of the apples, the taste information of the apples influences the selection of most consumers whether to buy, and has practical guiding significance for breeding, planting, storing and the like of the apples by accurately and efficiently detecting and representing the taste information. The traditional physical and chemical detection method cannot reflect the taste and sense information of the apples, and the most common artificial sense evaluation result is not objective due to the influences of factors such as the psychology of an evaluator, the surrounding environment and the like. Based on this, the SA-402B type electronic tongue as a single-taste intelligent bionic detection system gradually replaces the application of the traditional detection method in the aspect of taste information by the advantages of objectivity, accuracy and the like. Sampling detection and batch processing are carried out on the apple samples, so that the detection of the whole taste information of the samples is realized, and the distribution condition of various taste information on the sample space cannot be reflected.
Disclosure of Invention
In order to solve the problems, the invention provides an apple taste sensation map information visualization method.
In order to achieve the purpose, the invention adopts the technical scheme that:
an apple taste map information visualization method comprises the following steps:
s1 obtaining taste data based on electronic tongue apple sample
S11, selecting apples with regular shape, uniform size and no defects or pollutants, which are picked in the same batch and same place, and selecting 90 superior fruits according to the national standard GB/T10651-2008;
s12, numbering the samples, placing the samples at the normal temperature of 20 +/-2 ℃, storing the samples for 24 hours at the relative humidity of 55 +/-5%, and keeping the temperature and the humidity unchanged;
s13, cleaning, peeling, juicing and filtering each apple sample to obtain 40mL of supernatant, respectively placing the supernatant into 2 pure measuring cups, and sequentially detecting by an electronic tongue according to the numbering sequence to obtain three taste data of 90 multiplied by 3 sugar degree, sour taste and salty taste at 30 seconds; before the test starts, the sensor is firstly cleaned in positive and negative cleaning solution for 90s, cleaned in reference solution for 120s after the test is finished, cleaned in another reference solution for 120s continuously, and the balance of the sensor returns to zero for 30 s; after the balance is achieved, starting to detect, wherein the test time is 30s, and automatically entering a cleaning step after each measurement is finished;
s2, obtaining and preprocessing apple sample data based on hyperspectral technology
S21, correcting the hyperspectral image by adopting a black and white calibration method according to the following formula so as to eliminate the influence of noise:
in the formula, RdDark image, RwDiffuse reflection image of whiteboard, Rs-original diffuse reflectance spectral image of apple sample, R-corrected diffuse reflectance spectral image.
S22, carrying out vectorization processing on the image after black and white calibration to obtain an image description curve, after a target image is defined, generating a specific interested mask image, and controlling a sample image processing area;
and S23, performing image cutting on the image after the mask processing, wherein the spectral values of the masked spectral regions except the interested part are all 0.
S24, obtaining the preprocessed 90 apple spectral images, wherein an apple sample is divided into 4 faces when the sample is irradiated by a hyperspectral classifier because the apple is a spheroidal fruit, 5 interesting regions with the numbers of 1-5 are respectively obtained on each face, the size of each interesting region is 300 pixel points, the integral spectral mean value of 5 surfaces of each sample is obtained, and finally 90 x 256 dimensional apple data is obtained;
s3 selection of characteristic wave band
Respectively selecting 3 kinds of sensitive wave bands corresponding to sweet, sour and salty basic tastes, determining the number of the optimal sensitive wave bands by using a variation coefficient, and then determining the correlation degree between taste information and a spectral image of an apple sample by using a gray correlation scale (GRA) method;
s4 construction of apple taste visual model
Adopting SVM to predict the concentration value of each discrete point of taste information, wherein the hyperplane function, RBF kernel function and regression function formulas (2), (3) and (4) are respectively as follows:
K(x，x′)＝exp(g Px-x′P2) (2)
f(x)＝wφ(x)+b (3)
in the formula, w is a normal vector of a hyperplane, phi (x) is a nonlinear mapping function, b is an offset, and g is a width coefficient;
s5, taste visual presentation
After the GS-SVM model is established, predicted spectrum values under 10 wave bands are input point by point, taste prediction output values under all point spectrum values are calculated and recorded as K (i, j), the taste values are converted into corresponding actual color values, and the apple taste information is visually presented.
Preferably, the gray correlation method specifically includes the following steps: firstly, converting multi-dimension data into a uniform dimensionless form, defining a reference number sequence as a taste information number sequence, and comparing the number sequences to spectral information number sequences under each wave band of 380-; then, the grey correlation coefficient of each single taste information array and the spectrum information array under the full wave band is obtained; and finally, solving the grey correlation degree of each wave band.
Preferably, the step S4 is to optimize the parameters c and g based on a Genetic Algorithm (GA) in which the maximum genetic algebra is 100, the initial population number is 20, the search range of the parameter c is 0 to 100, and g is 0 to 100, and a Grid Search (GS) method; in the grid search method, parameter optimization is performed at intervals of 0.5, and the search range of the parameters c and g is 2-10To 210。
Preferably, in the step S4, in the process of building the model, when mapping the spectral data and the taste information, the model is built after calculating the mean value of the spectral values in each band; wherein 2/3 spectral data samples are selected as a training set and 1/3 spectral data samples are selected as a prediction set by using a Kennard-Stone method, and visual analysis of spectral-taste information is realized based on GS-SVM and GA-SVM.
The invention combines the advantages of apple sample on hyperspectral single-point difference with the electronic tongue taste integral information detection technology, thereby realizing the visualization of apple taste map information and providing a more accurate analysis method for the analysis of apple taste.
Drawings
Fig. 1 is a spectral image description curve in the embodiment of the present invention.
FIG. 2 is a vector mask image in an embodiment of the present invention.
Fig. 3 is a region-of-interest spectral plot in an embodiment of the present invention.
FIG. 4 is a graph illustrating variation of coefficient of variation according to an embodiment of the present invention
FIG. 5 is a diagram illustrating a parameter optimization process according to an embodiment of the present invention;
wherein, (a) is a genetic algorithm parameter optimization process; (b) the optimization process for the grid search parameters.
Fig. 6 is a graph of the GS-SVM-based taste visualization results in an embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described in detail below with reference to examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Examples
In the embodiment, 90 superior fruits are selected according to the national standard GB/T10651-2008 by selecting the Akxose heart apples which are picked in the same batch and in the same place and have regular shape, uniform size and no defects or pollutants as research objects; the method comprises the following steps:
s1 obtaining taste data based on electronic tongue apple sample
After numbering the samples, the samples are placed at the normal temperature of 20 ℃ (± 2 ℃), and stored for 24 hours with the relative humidity of 55% (± 5%), and the temperature and the humidity are kept basically unchanged. Sequentially measuring 90 samples according to the serial number sequence, cleaning, peeling, juicing and filtering each apple sample to obtain 40mL of supernatant, and respectively placing the supernatant in 2 pure measuring cups to be measured. Before the test is started, the sensor is firstly cleaned in positive and negative cleaning solution for 90s, cleaned in reference solution for 120s after the test is finished, cleaned in another reference solution for 120s continuously, and the balance of the sensor is reset to zero for 30 s. And after the balance is achieved, starting detection, wherein the test time is 30s, and automatically entering a cleaning step after each measurement is finished. The electronic tongue detection is carried out to obtain 90X 3 dimensional data of three tastes (sugar degree, sour taste and salty taste) at 30 seconds.
S2, obtaining and preprocessing apple sample data based on hyperspectral technology
In the hyperspectral image acquisition process, part of noise information is mixed due to the difference of the intensity distribution of the light source under each wave band and the influence of the dark current noise of the camera. The noise information can affect the quality of the hyperspectral image, and further affect the precision and stability of a hyperspectral image qualitative or quantitative analysis model. Therefore, the hyperspectral image is corrected by adopting a black and white calibration method to eliminate the influence of noise, and the formula (1) shows.
In the formula, RdDark image, RwDiffuse reflection image of whiteboard, Rs-original diffuse reflectance spectral image of apple sample, R-corrected diffuse reflectance spectral image.
Vectorizing the image after the black and white calibration to obtain an image description curve as shown in fig. 1, after the target image is defined, generating a specific interested mask image as shown in fig. 2, and controlling a sample image processing area. Because the acquired image edge contains more spectral noise, the difficulty of post-processing data is increased, the image after mask processing is subjected to image cutting, and other spectral values of the masked spectral region except the interested part are all 0.
And acquiring the preprocessed 90 apple spectral images. Wherein, because the apple is the quasi-spherical fruit, so divide into 4 faces with an apple sample when using the hyperspectral sorter to shine the sample to respectively take 5 region of interest (every region of interest size is about 300 pixel points) of serial number 1-5 on each face. The average spectral values of the 5 interesting regions on each cross section are shown in fig. 3, and the spectral average of the 5 surface integrals of each sample is obtained, so that 90 × 256 dimensional apple data is finally obtained. S3 selection of characteristic wave band
The spectral image reflects the difference of internal quality of the apple, and the electronic tongue obtains taste information through the free functional groups in the apple sample. Therefore, the grey correlation degree (GRA) method is adopted to determine the correlation degree between the two, and 3 sensitive wave bands corresponding to the sweet, sour and salty basic tastes are respectively selected. Before grey correlation analysis is carried out, the number of the optimal sensitive wave bands is determined by using the coefficient of variation. As the overall coefficient of variation is larger, the correlation between the variables is lower, and the influence of the number of characteristic bands spaced at 2 intervals on the coefficient of variation is examined, and as can be seen from fig. 4, the number of characteristic bands has the highest coefficient of variation at 10 points. In the grey correlation method, firstly, the multi-dimensional data is converted into a uniform dimensionless form, the reference number sequence is defined as a taste information number sequence, and the number sequence is compared with the spectral information number sequence under each wave band of 380-. Then, a gray correlation coefficient between each single taste information sequence and the spectrum information sequence in the full-band is obtained. And finally, solving the grey correlation degree of each wave band. The results are shown in Table 1, where the top 10 characteristic bands with higher relevance values are found for each single taste message.
TABLE 1 characteristic bands of the respective Pretaste 10
S4 construction of apple taste visual model
The spectrum and taste information data have discreteness, and local linear and nonlinear relations are presented between the spectrum and the taste information data, the SVM main idea is to establish a regression hyperplane as a decision surface, and utilize a kernel function to map multi-dimensional data to a high-dimensional space to make the multi-dimensional data linear as much as possible so as to solve the local nonlinearity of original data and finally make the distance from all data in a set to the hyperplane nearest. On the basis of the method, SVM is adopted to predict the concentration value of each discrete point of taste information, wherein the hyperplane function, RBF kernel function and regression function formulas (2), (3) and (4) are respectively as follows. The SVR modeling process result depends on parameters c and g, and correct and effective parameter selection has good regression performance on the support vector machine. Thus, the parameters c and g are optimized herein based on the methods of Genetic Algorithm (GA) and Grid Search (GS), as shown in fig. 5. In the genetic algorithm, the maximum genetic algebra is 100, the initial population number is 20, the search range of the parameter c is 0 to 100, and g is 0 to 100. In the grid search method, parameter optimization is performed at intervals of 0.5, and the search range of the parameters c and g is 2-10To 210. In the process of establishing the model, the taste of the whole sample is detected by the experimentAnd (4) obtaining the mean value of the spectral values under each wave band and then establishing a model when mapping the spectral data and the taste information. Wherein 2/3 spectral data samples are selected as a training set and 1/3 spectral data samples are selected as a prediction set by using a Kennard-Stone method, and visual analysis of spectral-taste information is realized based on GS-SVM and GA-SVM. The result shows that in the parameter optimization process of the GA-SVM, the genetic algorithm rapidly screens out the optimal modeling parameter c of 6.3859, g of 83.0159 and the parameter optimization process of the GS-SVM under the condition that the optimal fitness Mse is 0.01667, and the grid type search screens out the optimal modeling parameter c of 22.6274 and g of 0.0019531 under the condition that the optimal fitness Mse is 0.0037346, so that the GS-SVM is selected as the model for solving the apple taste value.
K(x，x′)＝exp(g Px-x′P2) (2)
f(x)＝wφ(x)+b (3)
Wherein w is the normal vector of the hyperplane, phi (x) -nonlinear mapping function, b-offset, and g-width coefficient
TABLE 2 c and g results of the optimization procedure
S5, taste visual presentation
After the GS-SVM model is established, predicted spectral values under 10 wave bands are input point by point, taste prediction output values under all point spectral values are calculated and recorded as K (i, j), taste values are converted into corresponding actual color values, apple taste information is displayed visually, a sweet taste, sour taste and salty taste single-point taste distribution diagram is shown in figure 6, and the legend gradually thickens from dark blue to brick red taste. As can be seen from the figure, the akxose apple is mainly sweet and sour, has very light salty taste, and has the most intense and wide distribution of sweet taste. Sweet taste is mainly distributed on two sides of an equator, and the central axis part is sweet; the distribution position of the sour taste is similar to the sweet taste; the salty taste is more uniformly distributed on the equatorial section.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be construed as the protection scope of the present invention.
Claims (4)
1. An apple taste map information visualization method is characterized by comprising the following steps:
s1 obtaining taste data based on electronic tongue apple sample
S11, selecting apples with regular shape, uniform size and no defects or pollutants, which are picked in the same batch and same place, and selecting 90 superior fruits according to the national standard GB/T10651-2008;
s12, numbering the samples, placing the samples at the normal temperature of 20 +/-2 ℃, storing the samples for 24 hours at the relative humidity of 55 +/-5%, and keeping the temperature and the humidity unchanged;
s13, cleaning, peeling, juicing and filtering each apple sample to obtain 40mL of supernatant, respectively placing the supernatant into 2 pure measuring cups, and sequentially detecting by an electronic tongue according to the numbering sequence to obtain three taste data of 90 multiplied by 3 sugar degree, sour taste and taste at 30 seconds; before the test starts, the sensor is firstly cleaned in positive and negative cleaning solution for 90s, cleaned in reference solution for 120s after the test is finished, cleaned in another reference solution for 120s continuously, and the balance of the sensor returns to zero for 30 s; after the balance is achieved, starting to detect, wherein the test time is 30s, and automatically entering a cleaning step after each measurement is finished;
s2, obtaining and preprocessing apple sample data based on hyperspectral technology
S21, correcting the hyperspectral image by adopting a black and white calibration method according to the following formula so as to eliminate the influence of noise:
in the formula, RdDark image, RwDiffuse reflection image of whiteboard, RS-an original diffuse reflectance spectral image of an apple sample, R-corrected diffuse reflectance spectral image;
s22, carrying out vectorization processing on the image after black and white calibration to obtain an image description curve, after a target image is defined, generating a specific interested mask image, and controlling a sample image processing area;
s23, performing image cutting on the image subjected to mask processing, wherein the spectral values of the masked spectral regions except the interested part are 0;
s24, obtaining the preprocessed 90 apple spectral images, wherein an apple sample is divided into 4 faces when the sample is irradiated by a hyperspectral classifier because the apple is a spheroidal fruit, 5 interesting regions with the numbers of 1-5 are respectively obtained on each face, the size of each interesting region is 300 pixel points, the integral spectral mean value of 5 surfaces of each sample is obtained, and finally 90 x 256 dimensional apple data is obtained;
s3 selection of characteristic wave band
Respectively selecting 3 kinds of sensitive wave bands corresponding to sweet, sour and salty basic tastes, determining the number of the optimal sensitive wave bands by using a variation coefficient, and then determining the correlation degree between taste information and a spectral image of an apple sample by using a gray correlation scale (GRA) method;
s4 construction of apple taste visual model
Adopting SVM to predict the concentration value of each discrete point of taste information, wherein the hyperplane function, RBF kernel function and regression function formulas (2), (3) and (4) are respectively as follows:
K(x，x′)＝exp(g Px-x′P2) (2)
f(x)＝ωφ(x)+b (3)
in the formula, w is a normal vector of a hyperplane, phi (x) is a nonlinear mapping function, b is an offset, and g is a width coefficient;
s5, taste visual presentation
After the GS-SVM model is established, predicted spectrum values under 10 wave bands are input point by point, taste prediction output values under all point spectrum values are calculated and recorded as K (i, j), the taste values are converted into corresponding actual color values, and the apple taste information is visually presented.
2. The method of claim 1, wherein said method of visualizing information about taste profiles of apples,
the grey correlation method specifically comprises the following steps: firstly, converting multi-dimension data into a uniform dimensionless form, defining a reference number sequence as a taste information number sequence, and comparing the number sequences to spectral information number sequences under each wave band of 380-; then, the grey correlation coefficient of each single taste information array and the spectrum information array under the full wave band is obtained; and finally, solving the grey correlation degree of each wave band.
3. The method of claim 1, wherein the step S4 is based on Genetic Algorithm (GA) in which the maximum genetic algebra is 100, the initial population number is 20, the search range of parameter c is 0 to 100, and the Grid Search (GS) is 0 to 100 to optimize the parameters c and g; in the grid search method, parameter optimization is performed at intervals of 0.5, and the search range of the parameters c and g is 2-10To 210。
4. The method of claim 1, wherein in step S4, during the modeling process, the spectral data and taste information are mapped, and the model is built after the mean value of the spectral values in each band is obtained; wherein 2/3 spectral data samples are selected as a training set and 1/3 spectral data samples are selected as a prediction set by using a Kennard-Stone method, and visual analysis of spectral-taste information 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 CN108760655A (en) | 2018-11-06 |
| CN108760655B true 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) |
Families Citing this family (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 |
| CN110763698B (en) * | 2019-10-12 | 2022-01-14 | 仲恺农业工程学院 | Hyperspectral citrus leaf disease identification method based on characteristic wavelength |
| SG10201911636PA (en) * | 2019-12-04 | 2020-03-30 | Teapasar Pte Ltd | System and method for non-destructive rapid food profiling using artificial intelligence |
| CN111289516B (en) * | 2020-03-26 | 2021-10-08 | 中国农业大学 | Method and device for detecting amino acid content in plant leaves |
| CN111289446B (en) * | 2020-03-30 | 2021-06-29 | 天津工业大学 | A kind of detection method and system of complex solution component concentration |
| 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 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 | Band Selection Method for Remote Sensing Hyperspectral Imagery 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 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7151606B2 (en) * | 2001-07-09 | 2006-12-19 | Mitsui Mining & Smelting Co., Ltd. | Method for evaluation of vegetables and fruits and evaluation apparatus therefor |
| CN201021922Y (en) * | 2007-02-13 | 2008-02-13 | 江西农业大学 | A device for non-destructive testing of fruit quality |
-
2018
- 2018-04-28 CN CN201810435827.3A patent/CN108760655B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 | Band Selection Method for Remote Sensing Hyperspectral Imagery 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 (3)
| Title |
|---|
| 寒富苹果品质无损检测光谱信息在线分析;宫元娟 等;《沈阳农业大学学报》;20141231;第45卷(第6期);第708-713页 * |
| 支持向量回归机的参数择优算法;张偲敏 等;《中国科技信息》;20150731(第13期);第26-27页 * |
| 高光谱图像技术检测苹果外部损伤;刘晶晶 等;《中国食品学报》;20180131;第18卷(第1期);第278-284页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108760655A (en) | 2018-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108760655B (en) | Apple taste map information visualization method | |
| Zhang et al. | Growth monitoring of greenhouse lettuce based on a convolutional neural network | |
| Sunli et al. | Non‐destructive detection for mold colonies in rice based on hyperspectra and GWO‐SVR | |
| Sun et al. | Hyperspectral imaging detection of decayed honey peaches based on their chlorophyll content | |
| CN111968080B (en) | A detection method for internal and external quality of Feicheng peach based on hyperspectral and deep learning | |
| Xie et al. | Detection of early blight and late blight diseases on tomato leaves using hyperspectral imaging | |
| CN110132862B (en) | Construction method and application of specific disease index for wheat scab detection | |
| CN103278464B (en) | Flesh of fish detection method and device | |
| Kim et al. | Classification of grapefruit peel diseases using color texture feature analysis | |
| Chen et al. | Nondestructive measurement of total volatile basic nitrogen (TVB-N) content in salted pork in jelly using a hyperspectral imaging technique combined with efficient hypercube processing algorithms | |
| CN108956604A (en) | A method of Eriocheir sinensis quality is identified based on hyper-spectral image technique | |
| CN105021535B (en) | A kind of paddy content of fatty acid lossless detection method and system | |
| CN114136920A (en) | A method for identification of single-grain hybrid rice seed species based on hyperspectral | |
| Yu et al. | Rapid and visual measurement of fat content in peanuts by using the hyperspectral imaging technique with chemometrics | |
| CN116297236A (en) | A method and device for identifying the vigor of single corn seeds based on hyperspectral | |
| CN113065230B (en) | High-spectrum inversion model for establishing rice leaf SPAD based on optimized spectral index | |
| CN103278467A (en) | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf | |
| Bu et al. | ResNet incorporating the fusion data of RGB & hyperspectral images improves classification accuracy of vegetable soybean freshness | |
| CN114965346B (en) | Kiwi fruit quality detection method based on deep learning and hyperspectral imaging technology | |
| Tian et al. | Early contamination warning of Aflatoxin B1 in stored maize based on the dynamic change of catalase activity and data fusion of hyperspectral images | |
| Yu et al. | Nondestructive testing of external defects on Nanguo pear | |
| Lin et al. | Prediction of protein content in rice using a near-infrared imaging system as diagnostic technique | |
| Yao et al. | On-line monitoring of egg freshness using a portable NIR spectrometer combined with deep learning algorithm | |
| CN118470540A (en) | Network model, construction method, and nondestructive identification method and system for gender of eggs | |
| Francis et al. | Development of a unified framework of low-rank approximation and deep neural networks for predicting the spatial variability of SSC inSpania'watermelons using vis/NIR hyperspectral imaging |
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 |