CN112730269A - Aflatoxin intelligent detection method based on deep learning - Google Patents
Aflatoxin intelligent detection method based on deep learning Download PDFInfo
- Publication number
- CN112730269A CN112730269A CN202011433931.2A CN202011433931A CN112730269A CN 112730269 A CN112730269 A CN 112730269A CN 202011433931 A CN202011433931 A CN 202011433931A CN 112730269 A CN112730269 A CN 112730269A
- Authority
- CN
- China
- Prior art keywords
- aflatoxin
- accuracy
- deep learning
- network
- grain
- 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
- 229930195730 Aflatoxin Natural products 0.000 title claims abstract description 23
- XWIYFDMXXLINPU-UHFFFAOYSA-N Aflatoxin G Chemical compound O=C1OCCC2=C1C(=O)OC1=C2C(OC)=CC2=C1C1C=COC1O2 XWIYFDMXXLINPU-UHFFFAOYSA-N 0.000 title claims abstract description 23
- 239000005409 aflatoxin Substances 0.000 title claims abstract description 23
- 238000013135 deep learning Methods 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 title claims abstract description 16
- 238000001228 spectrum Methods 0.000 claims abstract description 17
- 239000003053 toxin Substances 0.000 claims abstract description 9
- 231100000765 toxin Toxicity 0.000 claims abstract description 9
- 238000004811 liquid chromatography Methods 0.000 claims abstract description 4
- 230000004913 activation Effects 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 23
- 238000000034 method Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 238000012795 verification Methods 0.000 claims description 7
- 241000228197 Aspergillus flavus Species 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 claims description 2
- 238000013145 classification model Methods 0.000 abstract 1
- 239000010410 layer Substances 0.000 description 14
- 238000013527 convolutional neural network Methods 0.000 description 5
- 235000017060 Arachis glabrata Nutrition 0.000 description 2
- 244000105624 Arachis hypogaea Species 0.000 description 2
- 235000010777 Arachis hypogaea Nutrition 0.000 description 2
- 235000018262 Arachis monticola Nutrition 0.000 description 2
- 240000008042 Zea mays Species 0.000 description 2
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 2
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 235000005822 corn Nutrition 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 235000020232 peanut Nutrition 0.000 description 2
- 239000002344 surface layer Substances 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 238000002965 ELISA Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000002306 biochemical method Methods 0.000 description 1
- 230000000711 cancerogenic effect Effects 0.000 description 1
- 231100000357 carcinogen Toxicity 0.000 description 1
- 239000003183 carcinogenic agent Substances 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000004128 high performance liquid chromatography Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000004809 thin layer chromatography Methods 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20216—Image averaging
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to an intelligent aflatoxin detection method based on deep learning, which comprises the steps of firstly, obtaining hyperspectral data of single mildewed grains, calibrating the toxin content of each grain by using a liquid chromatography, then, building a one-dimensional deep learning network, defining the number and structure of network layers, importing the average spectrum of each grain, continuously adjusting the network hyper-parameters, and selecting the hyper-parameters with the highest accuracy as a classification model.
Description
Technical Field
The invention discloses an aflatoxin intelligent detection method based on deep learning, and belongs to the field of food safety artificial intelligent detection.
Background
Aspergillus flavus is classified as a naturally occurring carcinogen by the World Health Organization (WHO) cancer research institution, and the detection method thereof is developed from the beginning mainly by biochemical methods such as thin layer chromatography to the common application of various methods such as high performance liquid chromatography, microcolumn method, enzyme-linked immunosorbent assay and the like, but the methods need long-time experiments by professionals.
The deep learning is a neural network system based on convolution operation, has strong calculation power and self-learning capability, has high accuracy and high calculation speed, is well known in the field of digital image processing at first, and can well process two-dimensional image data classification tasks. A proper network structure can be constructed according to different classification targets and data characteristics thereof, and the hyper-parameters of the neural network need to be adjusted according to a target data set.
The hyperspectral imaging technology can collect images of hundreds of wave bands of an object, different performances of different substances under different wave band spectrum signals can be drawn into a curve related to the spectrum wave bands and the spectrum values by each pixel, each pixel is classified according to the difference of the curve and can be realized through a one-dimensional convolutional neural network (1D-CNN), toxin identification is carried out by analyzing whether the spectrum characteristics of aflatoxin exist in a single pixel in the hyperspectral images, the toxin content can be calculated by accumulation, and the aflatoxin has the distribution characteristics of a surface layer and a shallow surface layer, so that deep learning and hyperspectral are matched, and the defects of complex operation, long time and high detection cost of the traditional method can be overcome.
Disclosure of Invention
The invention mainly aims to overcome the defects of the problems and discloses an aflatoxin intelligent method based on 1D-CNN deep learning.
The technical scheme of the invention is an aflatoxin intelligent detection method based on deep learning, which comprises the following steps:
step 1: acquiring hyperspectral data of the front side and the back side of a single mildewed grain;
step 2: obtaining a binary image of the grains and the background by Gaussian clustering by using the spectrum obtained in the step 1, and calculating an average spectrum of single-sided grains;
and step 3: adding the hyperspectral data of the front side and the back side of the single grain obtained in the step 2, and taking an average value as the spectral data of the single grain;
and 4, step 4: calibrating the toxin content of the single grain in the step 1 by using a liquid chromatography to obtain a label for judging whether the aflatoxin of the single grain exceeds the standard or not;
and 5: normalizing the input data;
step 6: building a one-dimensional deep learning network, and defining the number of network layers and structure;
and 7: introducing the average spectrum and toxin content of each grain, continuously adjusting network hyper-parameters, and recording the accuracy of each training network model;
and 8: and selecting the optimal parameters of the hyper-parameters to configure the optimal parameters into a final hyperspectral aspergillus flavus sorting network.
Wherein:
(1) the specific steps of the step 2 are as follows:
step 2-1: setting the clustering number to be 2 by using a Gaussian clustering method for the whole hyperspectral cube, and accurately classifying the background and the grains;
step 2-2: expressing the grain picture by using a binary image, wherein a white area represents the grain, and a black area represents the background;
step 2-3: and taking the average spectrum of the white area as the hyperspectral data of the single-sided grains.
(2) The concrete steps of the step 6 are as follows:
step 6-1: establishing a 7-layer one-dimensional information convolution model: each of the first four layers is a convolution layer with a largest pooling layer, and the last three layers are full-connection layers;
step 6-2: keeping the basic network structure in the step 6-1 unchanged, and setting initial values of three hyper-parameters of iteration times, learning rate and activation function. The number of iterations is set to 30, the learning rate is set to 0.005, and the activation function is set to relu.
(3) The specific steps of the step 7 are as follows:
step 7-1: the number of fixed iterations is 30 and the learning rate is 0.005. And replacing the activation function with sigmoid and tanh, importing the activation function into a data training network, and recording the accuracy of the verification set. Recording the accuracy of the activation functions of relu, sigmoid and tanh;
step 7-2: the fixed learning rate is 0.005 and the activation function is relu. And replacing the iteration times with 50 and 100, importing the data training network and recording the accuracy of the verification set. Recording the accuracy rates of 30, 50 and 100 of the iteration times;
and 7-3: the fixed number of iterations is 30 and the activation function is relu. The learning rate was replaced with 0.0005, 0.00005, and 0.000005, the data training network was imported and the validation set accuracy was recorded. Accuracy rates of replacement of learning rates by 0.0005, 0.00005, and 0.000005 are recorded;
and 7-4: and applying the model with the highest target identification accuracy to the aflatoxin of the grains according to the parameters with the highest iteration times, learning rate and the highest accuracy in the activation function.
A1D-CNN deep learning model in the aflatoxin intelligent method based on 1D-CNN deep learning can be one of Alexnet, VGG-16 or models.
Drawings
FIG. 1 is a general schematic diagram of an aflatoxin intelligent detection method based on deep learning.
FIG. 2 shows the result of aflatoxin detection of peanut kernels.
FIG. 3 shows the result of aflatoxin detection of corn kernels.
Detailed Description
The following describes specific embodiments of the present apparatus and method with reference to fig. 1-3, wherein fig. 2 and 3 show the results of detection of peanut and corn kernels using Alexnet and VGG-16 models, respectively:
step 1: acquiring hyperspectral data of the front side and the back side of a single mildewed grain;
step 2: obtaining a binary image of the grains and the background by Gaussian clustering by using the spectrum obtained in the step 1, and calculating an average spectrum of single-sided grains;
and step 3: adding the hyperspectral data of the front side and the back side of the single grain obtained in the step 2, and taking an average value as the spectral data of the single grain;
and 4, step 4: calibrating the toxin content of the single grain in the step 1 by using a liquid chromatography to obtain a label for judging whether the aflatoxin of the single grain exceeds the standard or not;
and 5: normalizing the input data;
step 6: building a one-dimensional deep learning network, and defining the number of network layers and structure;
and 7: introducing the average spectrum and toxin content of each grain, continuously adjusting network hyper-parameters, and recording the accuracy of each training network model;
and 8: and selecting the optimal parameters of the hyper-parameters to configure the optimal parameters into a final hyperspectral aspergillus flavus sorting network.
Wherein:
(1) the specific steps of the step 2 are as follows:
step 2-1: setting the clustering number to be 2 by using a Gaussian clustering method for the whole hyperspectral cube, and accurately classifying the background and the grains;
step 2-2: expressing the grain picture by using a binary image, wherein a white area represents the grain, and a black area represents the background;
step 2-3: and taking the average spectrum of the white area as the hyperspectral data of the single-sided grains.
(2) The concrete steps of the step 6 are as follows:
step 6-1: establishing a 7-layer one-dimensional information convolution model: each of the first four layers is a convolution layer with a largest pooling layer, and the last three layers are full-connection layers;
step 6-2: keeping the basic network structure in the step 6-1 unchanged, and setting initial values of three hyper-parameters of iteration times, learning rate and activation function. The number of iterations is set to 30, the learning rate is set to 0.005, and the activation function is set to relu.
(3) The specific steps of the step 7 are as follows:
step 7-1: the number of fixed iterations is 30 and the learning rate is 0.005. And replacing the activation function with sigmoid and tanh, importing the activation function into a data training network, and recording the accuracy of the verification set. Recording the accuracy of the activation functions of relu, sigmoid and tanh;
step 7-2: the fixed learning rate is 0.005 and the activation function is relu. And replacing the iteration times with 50 and 100, importing the data training network and recording the accuracy of the verification set. Recording the accuracy rates of 30, 50 and 100 of the iteration times;
and 7-3: the fixed number of iterations is 30 and the activation function is relu. The learning rate was replaced with 0.0005, 0.00005, and 0.000005, the data training network was imported and the validation set accuracy was recorded. Accuracy rates of replacement of learning rates by 0.0005, 0.00005, and 0.000005 are recorded;
and 7-4: and applying the model with the highest target identification accuracy to the aflatoxin of the grains according to the parameters with the highest iteration times, learning rate and the highest accuracy in the activation function.
Claims (5)
1. An aflatoxin intelligent detection method based on deep learning is characterized by comprising the following steps:
step 1: acquiring hyperspectral data of the front side and the back side of a single mildewed grain;
step 2: obtaining a binary image of the grains and the background by Gaussian clustering by using the spectrum obtained in the step 1, and calculating an average spectrum of single-sided grains;
and step 3: adding the hyperspectral data of the front side and the back side of the single grain obtained in the step 2, and taking an average value as the spectral data of the single grain;
and 4, step 4: calibrating the toxin content of the single grain in the step 1 by using a liquid chromatography to obtain a label for judging whether the aflatoxin of the single grain exceeds the standard or not;
and 5: normalizing the input data;
step 6: building a one-dimensional deep learning network, and defining the number of network layers and structure;
and 7: introducing the average spectrum and toxin content of each grain, continuously adjusting network hyper-parameters, and recording the accuracy of each training network model;
and 8: and selecting the optimal parameters of the hyper-parameters to configure the optimal parameters into a final hyperspectral aspergillus flavus sorting network.
2. The intelligent aflatoxin detection method based on deep learning as claimed in claim 1, which is characterized in that the specific steps in the step 2 are as follows:
step 2-1: setting the clustering number to be 2 by using a Gaussian clustering method for the whole hyperspectral cube, and accurately classifying the background and the grains;
step 2-2: expressing the grain picture by using a binary image, wherein a white area represents the grain, and a black area represents the background;
step 2-3: and taking the average spectrum of the white area as the hyperspectral data of the single-sided grains.
3. The intelligent aflatoxin detection method based on deep learning as claimed in claim 1, which is characterized in that the specific steps of step 6 are as follows:
step 6-1: establishing a one-dimensional information convolution model with a fixed layer number: the front N-1 layers are convolution layers with the largest pooling layer, and the rear three layers are full-connection layers;
step 6-2: keeping the basic network structure in the step 6-1 unchanged, and setting initial values of three hyper-parameters of iteration times, learning rate and activation function.
4. The intelligent aflatoxin detection method based on deep learning as claimed in claim 1, which is characterized in that the specific steps of step 7 are as follows:
step 7-1: replacing the activation function with fixed iteration times and unchanged initial values of the learning rate, importing the activation function into a data training network, recording the accuracy of a verification set, and recording the accuracy of different activation functions;
step 7-2: the initial value of the fixed learning rate is unchanged, the activation function is unchanged, the iteration times are replaced by importing a data training network, and the accuracy rate of a verification set is recorded;
and 7-3: the initial value of the fixed iteration times is unchanged, the activation function is unchanged, the learning rate is replaced, the learning rate is led into a data training network, the accuracy rate of a verification set is recorded, and the accuracy rates under different learning rates are recorded;
and 7-4: and using the model with the highest target identification accuracy for aflatoxin classification of the grains according to the parameters with the highest iteration times, learning rate and accuracy in the activation function.
5. The method for intelligently detecting aflatoxin based on deep learning of claim 1, wherein the deep learning model can be one of Alexnet, VGG-16 or model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011433931.2A CN112730269B (en) | 2020-12-10 | 2020-12-10 | Aflatoxin intelligent detection method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011433931.2A CN112730269B (en) | 2020-12-10 | 2020-12-10 | Aflatoxin intelligent detection method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112730269A true CN112730269A (en) | 2021-04-30 |
| CN112730269B CN112730269B (en) | 2022-09-13 |
Family
ID=75598775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011433931.2A Active CN112730269B (en) | 2020-12-10 | 2020-12-10 | Aflatoxin intelligent detection method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112730269B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118090699A (en) * | 2024-01-23 | 2024-05-28 | 浙江工业大学 | Aflatoxin detection method combining differential Raman spectrum and SE-Res2Net |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120061586A1 (en) * | 2010-09-10 | 2012-03-15 | Haibo Yao | Method and detection system for detection of aflatoxin in corn with fluorescence spectra |
| CN102636581A (en) * | 2011-12-26 | 2012-08-15 | 湖北省农业科学院农业质量标准与检测技术研究所 | Liquid chromatogram and fluorescence method for simultaneously detecting aflatoxin B1, ochratoxin A, zearalenone and citrinin in grains |
| CN103575694A (en) * | 2013-11-11 | 2014-02-12 | 东北农业大学 | Quick detection method for aflatoxin B1 of rice based on near infrared spectroscopy |
| CN104598886A (en) * | 2015-01-23 | 2015-05-06 | 中国矿业大学(北京) | Method for identifying moldy peanuts by using near-infrared high-spectrum image |
| CN104914052A (en) * | 2015-05-12 | 2015-09-16 | 江苏大学 | Spectral imaging detection method of moldy peanut |
| CN108985237A (en) * | 2018-07-20 | 2018-12-11 | 安徽农业大学 | A kind of detection method and its system of the wheat scab based on depth mixing |
| CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
| TW201925777A (en) * | 2017-12-08 | 2019-07-01 | 艾思博生物科技股份有限公司 | Non-invasive testing method for aflatoxin, food inspection server, and food inspection system |
| CN110298396A (en) * | 2019-06-25 | 2019-10-01 | 北京工业大学 | Hyperspectral image classification method based on deep learning multiple features fusion |
| CN110470834A (en) * | 2019-09-22 | 2019-11-19 | 河南工业大学 | A kind of aflatoxin B1 detection method of content based on magnetic signal |
-
2020
- 2020-12-10 CN CN202011433931.2A patent/CN112730269B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120061586A1 (en) * | 2010-09-10 | 2012-03-15 | Haibo Yao | Method and detection system for detection of aflatoxin in corn with fluorescence spectra |
| CN102636581A (en) * | 2011-12-26 | 2012-08-15 | 湖北省农业科学院农业质量标准与检测技术研究所 | Liquid chromatogram and fluorescence method for simultaneously detecting aflatoxin B1, ochratoxin A, zearalenone and citrinin in grains |
| CN103575694A (en) * | 2013-11-11 | 2014-02-12 | 东北农业大学 | Quick detection method for aflatoxin B1 of rice based on near infrared spectroscopy |
| CN104598886A (en) * | 2015-01-23 | 2015-05-06 | 中国矿业大学(北京) | Method for identifying moldy peanuts by using near-infrared high-spectrum image |
| CN104914052A (en) * | 2015-05-12 | 2015-09-16 | 江苏大学 | Spectral imaging detection method of moldy peanut |
| TW201925777A (en) * | 2017-12-08 | 2019-07-01 | 艾思博生物科技股份有限公司 | Non-invasive testing method for aflatoxin, food inspection server, and food inspection system |
| CN108985237A (en) * | 2018-07-20 | 2018-12-11 | 安徽农业大学 | A kind of detection method and its system of the wheat scab based on depth mixing |
| CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
| CN110298396A (en) * | 2019-06-25 | 2019-10-01 | 北京工业大学 | Hyperspectral image classification method based on deep learning multiple features fusion |
| CN110470834A (en) * | 2019-09-22 | 2019-11-19 | 河南工业大学 | A kind of aflatoxin B1 detection method of content based on magnetic signal |
Non-Patent Citations (5)
| Title |
|---|
| BA TUAN LE: "Application of deep learning and near infrared spectroscopy in cereal", 《JOURNAL PRE-PROOF》 * |
| F.R. BERTANI ET AL.: "Optical detection of aflatoxins B in grained almonds using fluorescence spectroscopy and machine learning algorithms", 《JOURNAL PRE-PROOF》 * |
| JIYUE GAO ET AL.: "Aflatoxin rapid detection based on hyperspectral with 1D-convolution neural network in the pixel level", 《JOURNAL PRE-PROOFS》 * |
| Z. HAN AND J. GAO: "Pixel-level aflatoxin detecting based on deep learning and hyperspectral imaging", 《COMPUTERS AND ELECTRONICS IN AGRICULTURE》 * |
| 韩仲志，刘杰: "高光谱亚像元分解预测花生中的黄曲霉毒素 B1高光谱亚像元分解预测花生中的黄曲霉毒素 B1高光谱亚像元分解预测花生中的黄曲霉毒素 B1高光谱亚像元分解预测花生中的黄曲霉毒素 B1", 《中国食品学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118090699A (en) * | 2024-01-23 | 2024-05-28 | 浙江工业大学 | Aflatoxin detection method combining differential Raman spectrum and SE-Res2Net |
| CN118090699B (en) * | 2024-01-23 | 2024-09-24 | 浙江工业大学 | Aflatoxin detection method combining differential Raman spectrum and SE-Res2Net |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112730269B (en) | 2022-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | Automatic diagnosis of rice diseases using deep learning | |
| CN103593670B (en) | A kind of copper plate/strip detection method of surface flaw based on online limit of sequence learning machine | |
| Pasquet-Itam et al. | Deep learning approach for classifying, detecting and predicting photometric redshifts of quasars in the Sloan Digital Sky Survey stripe 82 | |
| CN107316300B (en) | Tire X-ray defect detection method based on deep convolutional neural network | |
| CN107451565B (en) | Semi-supervised small sample deep learning image mode classification and identification method | |
| CN107609503A (en) | Intelligent cancerous tumor cell identifying system and method, cloud platform, server, computer | |
| Zhao et al. | SEV‐Net: Residual network embedded with attention mechanism for plant disease severity detection | |
| Weng et al. | Reflectance images of effective wavelengths from hyperspectral imaging for identification of Fusarium head blight-infected wheat kernels combined with a residual attention convolution neural network | |
| CN107145830A (en) | Hyperspectral image classification method with depth belief network is strengthened based on spatial information | |
| CN112308152B (en) | Hyperspectral image ground object classification method based on spectrum segmentation and homogeneous region detection | |
| CN104966085A (en) | Remote sensing image region-of-interest detection method based on multi-significant-feature fusion | |
| Lv et al. | A visual identification method for the apple growth forms in the orchard | |
| CN109598711B (en) | Thermal image defect extraction method based on feature mining and neural network | |
| CN113191178B (en) | Underwater sound target identification method based on auditory perception feature deep learning | |
| CN107665352A (en) | A kind of pearl sorting technique based on multichannel residual error network | |
| Reddy et al. | Optimized convolutional neural network model for plant species identification from leaf images using computer vision | |
| CN113052216B (en) | Oil spill hyperspectral image detection method based on two-way graph U-NET convolutional network | |
| CN112730269B (en) | Aflatoxin intelligent detection method based on deep learning | |
| Shantkumari et al. | Grape leaf image classification based on machine learning technique for accurate leaf disease detection | |
| CN111145145A (en) | Image surface defect detection method based on MobileNet | |
| CN113610035A (en) | Rice tillering stage weed segmentation and identification method based on improved coding and decoding network | |
| CN115810191A (en) | Pathological cell classification method based on multi-attention fusion and high-precision segmentation network | |
| CN110688966B (en) | Semantic guidance pedestrian re-recognition method | |
| Ramamoorthy et al. | Reliable and accurate plant leaf disease detection with treatment suggestions using enhanced deep learning techniques | |
| CN109145770B (en) | Automatic wheat spider counting method based on combination of multi-scale feature fusion network and positioning model |
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 |