CN116223661B - Method for measuring content of allicin in garlic wastewater - Google Patents
Method for measuring content of allicin in garlic wastewater Download PDFInfo
- Publication number
- CN116223661B CN116223661B CN202310005127.1A CN202310005127A CN116223661B CN 116223661 B CN116223661 B CN 116223661B CN 202310005127 A CN202310005127 A CN 202310005127A CN 116223661 B CN116223661 B CN 116223661B
- Authority
- CN
- China
- Prior art keywords
- decoding
- allicin
- channel
- feature map
- characteristic
- 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
- JDLKFOPOAOFWQN-VIFPVBQESA-N Allicin Natural products C=CCS[S@](=O)CC=C JDLKFOPOAOFWQN-VIFPVBQESA-N 0.000 title claims abstract description 135
- JDLKFOPOAOFWQN-UHFFFAOYSA-N allicin Chemical compound C=CCSS(=O)CC=C JDLKFOPOAOFWQN-UHFFFAOYSA-N 0.000 title claims abstract description 135
- 235000010081 allicin Nutrition 0.000 title claims abstract description 135
- 235000004611 garlic Nutrition 0.000 title claims abstract description 135
- 239000002351 wastewater Substances 0.000 title claims abstract description 132
- 238000000034 method Methods 0.000 title claims abstract description 56
- 244000245420 ail Species 0.000 title 1
- 240000002234 Allium sativum Species 0.000 claims abstract description 134
- 239000007788 liquid Substances 0.000 claims abstract description 129
- 230000009467 reduction Effects 0.000 claims abstract description 75
- 238000009826 distribution Methods 0.000 claims abstract description 48
- 238000004811 liquid chromatography Methods 0.000 claims description 96
- 238000010586 diagram Methods 0.000 claims description 66
- 239000011159 matrix material Substances 0.000 claims description 52
- 238000003062 neural network model Methods 0.000 claims description 37
- 239000007791 liquid phase Substances 0.000 claims description 21
- 238000012937 correction Methods 0.000 claims description 20
- 238000013527 convolutional neural network Methods 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 abstract description 33
- 238000013135 deep learning Methods 0.000 abstract description 9
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000000605 extraction Methods 0.000 description 20
- 230000000694 effects Effects 0.000 description 12
- 230000006870 function Effects 0.000 description 11
- 230000008901 benefit Effects 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 7
- 238000005065 mining Methods 0.000 description 7
- CSCPPACGZOOCGX-UHFFFAOYSA-N Acetone Chemical compound CC(C)=O CSCPPACGZOOCGX-UHFFFAOYSA-N 0.000 description 6
- 238000004090 dissolution Methods 0.000 description 6
- 230000004927 fusion Effects 0.000 description 6
- VLKZOEOYAKHREP-UHFFFAOYSA-N n-Hexane Chemical compound CCCCCC VLKZOEOYAKHREP-UHFFFAOYSA-N 0.000 description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 238000007792 addition Methods 0.000 description 5
- 230000002708 enhancing effect Effects 0.000 description 5
- 238000005259 measurement Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 241000234282 Allium Species 0.000 description 3
- 230000000844 anti-bacterial effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000002481 ethanol extraction Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000002203 pretreatment Methods 0.000 description 3
- 108010092760 Alliin lyase Proteins 0.000 description 2
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- LCTONWCANYUPML-UHFFFAOYSA-N Pyruvic acid Chemical compound CC(=O)C(O)=O LCTONWCANYUPML-UHFFFAOYSA-N 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 2
- 244000223014 Syzygium aromaticum Species 0.000 description 2
- 235000016639 Syzygium aromaticum Nutrition 0.000 description 2
- 239000013060 biological fluid Substances 0.000 description 2
- 238000004061 bleaching Methods 0.000 description 2
- 238000004587 chromatography analysis Methods 0.000 description 2
- 238000003891 environmental analysis Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000004186 food analysis Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000012071 phase Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 244000151012 Allium neapolitanum Species 0.000 description 1
- 241000894006 Bacteria Species 0.000 description 1
- 208000030808 Clear cell renal carcinoma Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 235000001314 Nothoscordum inodorum Nutrition 0.000 description 1
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 230000001088 anti-asthma Effects 0.000 description 1
- 230000001093 anti-cancer Effects 0.000 description 1
- 230000000843 anti-fungal effect Effects 0.000 description 1
- 230000003064 anti-oxidating effect Effects 0.000 description 1
- 239000000924 antiasthmatic agent Substances 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000975 bioactive effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 239000002775 capsule Substances 0.000 description 1
- 206010073251 clear cell renal cell carcinoma Diseases 0.000 description 1
- 239000012230 colorless oil Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000002526 effect on cardiovascular system Effects 0.000 description 1
- 239000002955 immunomodulating agent Substances 0.000 description 1
- 230000002584 immunomodulator Effects 0.000 description 1
- 229940121354 immunomodulator Drugs 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 150000002632 lipids Chemical class 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 150000002898 organic sulfur compounds Chemical class 0.000 description 1
- 239000002243 precursor Substances 0.000 description 1
- 230000004224 protection Effects 0.000 description 1
- 229940107700 pyruvic acid Drugs 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 229910052717 sulfur Inorganic materials 0.000 description 1
- 239000011593 sulfur Substances 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 239000003039 volatile agent Substances 0.000 description 1
- 239000000341 volatile oil Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
- G01N30/8675—Evaluation, i.e. decoding of the signal into analytical information
- G01N30/8679—Target compound analysis, i.e. whereby a limited number of peaks is analysed
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
- G01N30/8696—Details of Software
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Chemical & Material Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Library & Information Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Automatic Analysis And Handling Materials Therefor (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
Abstract
The method is characterized in that an artificial intelligent detection technology based on deep learning is adopted, so that characteristic position information of the liquid chromatogram of the garlic wastewater in space and content related characteristic information on a channel are extracted after noise reduction is carried out on the liquid chromatogram of the garlic wastewater, characteristic distribution about the content of the allicin in the liquid chromatogram of the garlic wastewater is identified, and decoding regression is carried out to carry out content determination of the allicin. Therefore, the intelligent detection of the content of allicin in the garlic wastewater can be accurately performed.
Description
Technical Field
The application relates to the technical field of allicin detection, and in particular relates to a method for measuring the content of allicin in garlic wastewater.
Background
Allicin is a sulfur-containing volatile compound that is present in white garlic (allium) and other allium species and is considered to be one of the major bioactive organic sulfur compounds synthesized in allium species. Allicin is a colorless oil with low water solubility, which is not present in the whole garlic cloves, but is naturally produced from alliinase precursors together with pyruvic acid and ammonia by the action of alliinase, which produce allicin by contact when the garlic cloves are crushed or immersed. Allicin has various therapeutic effects such as cardiovascular protection, antioxidation, anticancer, antibacterial, antiasthmatic, immunomodulator, blood pressure lowering and blood lipid lowering etc. Allicin also has antibacterial and antifungal activity, and has strong antibacterial effect against a variety of gram-positive and gram-negative bacteria. Clinical researches show that the allicin has obvious path inhibition effect on treating human renal clear cell carcinoma, and the incidence and the size of tumors are obviously reduced. In recent years, various researches on allicin at home and abroad have been more and more intensive and extensive, and various garlic products such as garlic essential oil, garlic capsules and the like are also being used.
The conventional detection methods of allicin at present comprise NY/T1497-2007, NY/T1800-2009 and NY/T2643-2014. All these methods require pretreatment of the sample prior to detection, extraction of allicin therefrom, and subsequent determination of the content. Common pretreatment methods include water dissolution, acetone dissolution, ethanol extraction, n-hexane extraction, and the like. However, the treatment methods are not the most suitable means for garlic wastewater, are complex, have high manufacturing cost and are not suitable for daily detection of vast middle and small enterprises.
Thus, an optimized protocol for determining the content of allicin in garlic waste water is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a method for determining the content of allicin in garlic wastewater, which adopts an artificial intelligent detection technology based on deep learning, so that the characteristic position information of the garlic wastewater in space and the content-related characteristic information on a channel are extracted after the liquid chromatogram of the garlic wastewater is subjected to noise reduction, the characteristic distribution of the allicin content in the liquid chromatogram of the garlic wastewater is identified, and decoding regression is performed to determine the content of the allicin. Therefore, the intelligent detection of the content of allicin in the garlic wastewater can be accurately performed.
According to one aspect of the present application, there is provided a method for determining the content of allicin in garlic waste water, comprising: acquiring a liquid chromatogram of garlic wastewater to be detected; the liquid chromatogram is passed through a noise reduction module based on an automatic coder-decoder to obtain a noise-reduced liquid chromatogram; the liquid chromatogram after noise reduction is processed through a first convolution neural network model using spatial attention so as to obtain a spatial enhancement liquid chromatogram characteristic diagram; the liquid chromatogram after noise reduction is processed through a second convolution neural network model using the channel attention so as to obtain a channel enhanced liquid chromatogram characteristic diagram; fusing the space enhancement liquid chromatography feature map and the channel enhancement liquid chromatography feature map to obtain a decoding feature map; performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map; and performing decoding regression on the corrected decoding characteristic diagram through a decoder to obtain a decoding value for representing the content of allicin in the garlic wastewater.
In the above method for determining the content of allicin in garlic wastewater, the step of passing the liquid chromatogram through a noise reduction module based on an automatic codec to obtain a noise-reduced liquid chromatogram includes: inputting the liquid chromatogram to an encoder of the automatic codec-based noise reduction module, wherein the encoder performs explicit spatial encoding on the liquid chromatogram by using a convolution layer to obtain image features; and inputting the image features into a decoder of the noise reduction module based on the automatic coder-decoder, wherein the decoder uses a deconvolution layer to deconvolute the image features to obtain the noise-reduced liquid-phase chromatogram.
In the above method for determining the content of allicin in garlic wastewater, the image encoder of the automatic codec comprises at least one convolution layer, and the image decoder of the automatic codec comprises at least one deconvolution layer.
In the above method for determining the content of allicin in garlic wastewater, the step of obtaining a space-enhanced liquid chromatography characteristic map by using a first convolution neural network model of spatial attention from the denoised liquid chromatography comprises the following steps: performing depth convolution coding on the liquid phase chromatogram after noise reduction by using a convolution coding part of the first convolution neural network model to obtain an initial convolution characteristic map; inputting the initial convolution feature map into a spatial attention portion of the first convolution neural network model to obtain a spatial attention map; -passing said spatial attention map through a Softmax activation function to obtain a spatial attention profile; and calculating the position-wise point multiplication of the spatial attention characteristic map and the initial convolution characteristic map to obtain a spatial enhancement liquid chromatography characteristic map.
In the above method for determining the content of allicin in garlic wastewater, the step of obtaining a channel enhanced liquid chromatography characteristic map by using a second convolution neural network model of channel attention from the denoised liquid chromatography comprises the following steps: input data is processed in forward pass of layers using layers of the second convolutional neural network model: performing convolution processing based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; pooling the convolution feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the global average value of each feature matrix of the activated feature map along the channel dimension to obtain a channel feature vector; calculating the ratio of the eigenvalue of each position in the channel eigenvector relative to the weighted sum of the eigenvalues of all positions of the channel eigenvector to obtain a channel weighted eigenvector; taking the characteristic values of the positions of the channel weighted characteristic vectors as weights to perform point multiplication on the characteristic matrix of the activated characteristic map along the channel dimension to obtain a generated characteristic map; and the generated characteristic diagram output by the last layer of the second convolutional neural network model is the channel enhanced liquid chromatography characteristic diagram.
In the above method for determining the content of allicin in garlic waste water, the fusing the space enhancement liquid chromatography characteristic map and the channel enhancement liquid chromatography characteristic map to obtain a decoding characteristic map includes: fusing the spatially enhanced liquid chromatography signature and the channel enhanced liquid chromatography signature to obtain a decoding signature with the following formula; wherein, the formula is:
wherein F is d For the decoding feature map, F a F for the spatially enhanced liquid chromatography profile b For the channel enhanced liquid chromatography profile,elements representing positions of the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile corresponding to each other are added, and α and β are weighting parameters for controlling balance between the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile in the decoding profile.
In the above method for determining the content of allicin in garlic waste water, the performing feature distribution correction on the decoded feature map to obtain a corrected decoded feature map includes: performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map according to the following formula: wherein, the formula is:
Wherein M represents a diagonal matrix obtained by linear transformation of each feature matrix of the decoding feature map, M i,j Is the eigenvalue of the (i, j) th position of the diagonal matrix obtained by linear transformation of each eigenvector of the decoding eigenvector, and II is II 2 Is the two norms of the vector, andrepresenting multiplication of each value of the matrix by a predetermined value,/->Representing addition by position, M ′ Representing the corrected decoding feature mapEach feature matrix.
In the above method for determining the content of allicin in garlic waste water, the decoding regression of the corrected decoding characteristic map by a decoder is performed to obtain a decoded value for representing the content of allicin in garlic waste water, which comprises: performing a decoding regression on the corrected decoding feature map using the decoder in the following formula to obtain the decoded value; wherein, the formula is:wherein X is the corrected decoding profile, Y is the decoded values, W is a weight matrix,representing a matrix multiplication.
According to another aspect of the present application, there is provided a system for determining the content of allicin in garlic waste water, comprising: the data acquisition module is used for acquiring a liquid chromatogram of the garlic wastewater to be detected; the noise reduction module is used for enabling the liquid chromatogram to pass through the noise reduction module based on the automatic coder-decoder to obtain a liquid chromatogram after noise reduction; the spatial attention applying module is used for enabling the liquid chromatogram after noise reduction to obtain a spatial enhanced liquid chromatogram characteristic diagram through a first convolution neural network model using spatial attention; the channel attention applying module is used for obtaining a channel enhanced liquid chromatogram characteristic diagram through a second convolution neural network model using channel attention from the liquid chromatogram after noise reduction; the fusion module is used for fusing the space enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram to obtain a decoding characteristic diagram; the feature distribution correction module is used for carrying out feature distribution correction on the decoding feature map to obtain a corrected decoding feature map; and the decoding module is used for carrying out decoding regression on the corrected decoding characteristic diagram through a decoder so as to obtain a decoding value for representing the content of allicin in the garlic wastewater.
In the above system for determining the content of allicin in garlic wastewater, the noise reduction module is further configured to: inputting the liquid chromatogram to an encoder of the automatic codec-based noise reduction module, wherein the encoder performs explicit spatial encoding on the liquid chromatogram by using a convolution layer to obtain image features; and inputting the image features into a decoder of the noise reduction module based on the automatic coder-decoder, wherein the decoder uses a deconvolution layer to deconvolute the image features to obtain the noise-reduced liquid-phase chromatogram.
In the above system for determining the content of allicin in garlic waste water, the image encoder of the automatic codec comprises at least one convolution layer, and the image decoder of the automatic codec comprises at least one deconvolution layer.
In the above system for determining the content of allicin in garlic wastewater, the spatial attention application module is further configured to: performing depth convolution coding on the liquid phase chromatogram after noise reduction by using a convolution coding part of the first convolution neural network model to obtain an initial convolution characteristic map; inputting the initial convolution feature map into a spatial attention portion of the first convolution neural network model to obtain a spatial attention map; -passing said spatial attention map through a Softmax activation function to obtain a spatial attention profile; and calculating the position-wise point multiplication of the spatial attention characteristic map and the initial convolution characteristic map to obtain a spatial enhancement liquid chromatography characteristic map.
In the above system for determining the content of allicin in garlic wastewater, the channel attention application module is further configured to: input data is processed in forward pass of layers using layers of the second convolutional neural network model: performing convolution processing based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; pooling the convolution feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the global average value of each feature matrix of the activated feature map along the channel dimension to obtain a channel feature vector; calculating the ratio of the eigenvalue of each position in the channel eigenvector relative to the weighted sum of the eigenvalues of all positions of the channel eigenvector to obtain a channel weighted eigenvector; taking the characteristic values of the positions of the channel weighted characteristic vectors as weights to perform point multiplication on the characteristic matrix of the activated characteristic map along the channel dimension to obtain a generated characteristic map; and the generated characteristic diagram output by the last layer of the second convolutional neural network model is the channel enhanced liquid chromatography characteristic diagram.
In the above system for determining the content of allicin in garlic wastewater, the fusion module is further configured to: fusing the spatially enhanced liquid chromatography signature and the channel enhanced liquid chromatography signature to obtain a decoding signature with the following formula; wherein, the formula is:
wherein F is d For the decoding feature map, F a F for the spatially enhanced liquid chromatography profile b For the channel enhanced liquid chromatography profile,elements representing positions of the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile corresponding to each other are added, and α and β are weighting parameters for controlling balance between the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile in the decoding profile.
In the above system for determining the content of allicin in garlic wastewater, the characteristic distribution correction module is further configured to: performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map according to the following formula: wherein, the formula is:
wherein M represents a diagonal matrix obtained by linear transformation of each feature matrix of the decoding feature map, M i,j Is the decoding characteristic diagram Eigenvalues of (i, j) th positions of diagonal matrix obtained by linear transformation of each eigenvector 2 Is the two norms of the vector, andrepresenting multiplication of each value of the matrix by a predetermined value,/->Representing addition by position, M ′ Representing the respective feature matrices of the corrected decoded feature map.
In the above system for determining the content of allicin in garlic wastewater, the decoding module is further configured to: performing a decoding regression on the corrected decoding feature map using the decoder in the following formula to obtain the decoded value; wherein, the formula is:wherein X is the corrected decoding feature map, Y is the decoded values, W is a weight matrix,/and>representing a matrix multiplication.
Compared with the prior art, the method for measuring the content of the allicin in the garlic wastewater adopts an artificial intelligent detection technology based on deep learning, so that the characteristic position information of the garlic wastewater in space and the content associated characteristic information on a channel are extracted after the liquid chromatogram of the garlic wastewater is subjected to noise reduction, the characteristic distribution of the allicin content in the liquid chromatogram of the garlic wastewater is identified, and decoding regression is carried out to measure the content of the allicin. Therefore, the intelligent detection of the content of allicin in the garlic wastewater can be accurately performed.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is an application scenario diagram of a method for determining the content of allicin in garlic wastewater according to an embodiment of the present application.
Fig. 2 is a flow chart of a method for determining the content of allicin in garlic waste water according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a method for determining the content of allicin in garlic waste water according to an embodiment of the present application.
Fig. 4 is a flowchart of a method for determining the content of allicin in garlic wastewater according to an embodiment of the present application, wherein the liquid chromatogram after noise reduction is used to obtain a space enhancement liquid chromatogram by using a first convolution neural network model of space attention.
FIG. 5 is a block diagram of a system for determining the content of allicin in garlic waste water according to an embodiment of the present application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Summary of the application
As described above in the background art, the existing allicin detection method requires pretreatment of a sample before detection, extraction of allicin therefrom, and subsequent measurement of the content. Common pretreatment methods include water dissolution, acetone dissolution, ethanol extraction, n-hexane extraction, and the like. However, the treatment methods are not the most suitable means for garlic wastewater, are complex, have high manufacturing cost and are not suitable for daily detection of vast middle and small enterprises. Thus, an optimized protocol for determining the content of allicin in garlic waste water is desired.
At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
In recent years, deep learning and development of a neural network provide a new solution idea and scheme for detecting the content of allicin in garlic wastewater.
Accordingly, it is considered that the existing garlicin detection scheme is unsuitable for practical application due to complicated method and high cost. Since liquid chromatography is a chromatography using a liquid as a mobile phase, it is used to measure various ion contents due to its advantages of being fast, sensitive, good in selectivity and simultaneously measuring multiple components, and is widely used in various fields such as water, pulp and bleaching solutions, food analysis, biological fluids, steel and environmental analysis, etc. Therefore, in the technical scheme of the application, the method for analyzing the liquid chromatogram of the garlic wastewater can be adopted to measure the content of allicin in the wastewater. However, it is considered that since the liquid chromatogram of the garlic wastewater has much information, it is difficult to extract the information of the content of allicin effectively, and the liquid chromatogram is also subject to the interference of image noise, which in turn makes it difficult to measure the content of allicin.
Based on the above, in the technical scheme of the application, an artificial intelligent detection technology based on deep learning is adopted, so that after the liquid chromatogram of the garlic wastewater is subjected to noise reduction, spatial characteristic position information and content related characteristic information on a channel are extracted, so that the content characteristic distribution of allicin in the liquid chromatogram of the garlic wastewater is identified, and decoding regression is carried out to measure the content of the allicin. Therefore, the intelligent detection can be accurately carried out on the content of allicin in the garlic wastewater, so that the detection accuracy is improved while the allicin determination method is simplified.
Specifically, in the technical scheme of the application, firstly, a liquid chromatogram of garlic wastewater to be detected is obtained. Next, considering that a great deal of noise interference exists in the collected liquid chromatogram due to environmental factors in the process of collecting the liquid chromatogram of the garlic wastewater to be detected, the feature extraction in the liquid chromatogram image becomes difficult, so that in order to improve the capability of feature extraction to improve the accuracy of allicin content measurement, the liquid chromatogram is further subjected to a noise reduction module based on an automatic codec to obtain a noise-reduced liquid chromatogram. In particular, here, the image encoder of the automatic codec comprises at least one convolution layer, and the image decoder of the automatic codec comprises at least one deconvolution layer.
Further, the characteristic mining of the liquid chromatogram after noise reduction is performed by using a convolutional neural network model with excellent performance in terms of local implicit characteristic extraction of an image, particularly, considering that each local implicit characteristic of the liquid chromatogram after noise reduction has relevance, and the characteristic extraction of the liquid chromatogram after noise reduction is performed by focusing on the distribution information of important characteristics in space and the content characteristic relevance information on a channel. Therefore, in the technical scheme of the application, in order to improve the characteristic extraction effect on the liquid-phase chromatogram after noise reduction so as to accurately determine the content of allicin, a convolution neural network model with double attention is further used for characteristic mining of the liquid-phase chromatogram after noise reduction.
In particular, here, the dual attention mechanisms are a spatial attention mechanism and a channel attention mechanism. Namely, the liquid chromatogram after noise reduction is subjected to a first convolution neural network model of spatial attention so as to dig out characteristic information of important characteristics distributed on spatial positions in the liquid chromatogram after noise reduction, thereby obtaining a spatial enhancement liquid chromatogram characteristic diagram; and the noise-reduced liquid chromatogram is used for mining the content characteristic related information of the noise-reduced liquid chromatogram on the channel by using a second convolution neural network model of the channel attention, so that a channel-enhanced liquid chromatogram characteristic diagram is obtained. It should be appreciated that the image features extracted by the channel attention reflect the correlation and importance between feature channels, and the image features extracted by the spatial attention reflect the weights of the spatial dimension feature differences for suppressing or enhancing features at different spatial locations. The channel attention and the space attention can respectively pay attention to the characteristic content and the characteristic position in the image, the characteristic extraction effect of the network is improved to a certain extent, so that different types of effective information about the allicin content in the garlic wastewater can be captured in a large amount, the characteristic distinguishing learning capability can be effectively enhanced, and in the network training process, the task processing system is more focused on finding out the remarkable useful information related to the current output in the input image data, thereby improving the output quality, and the increasing attention module can bring continuous performance improvement.
And then fusing the space enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram to fuse important characteristic distribution information and content association characteristic distribution information about the allicin content in the liquid chromatography after noise reduction, and taking the important characteristic distribution information and the content association characteristic distribution information as decoding characteristic diagrams to obtain decoding values for representing the allicin content in the garlic wastewater through decoding regression in a decoder. Therefore, the intelligent detection can be carried out on the content of allicin in the garlic wastewater, so that the detection accuracy is improved while the allicin determination method is simplified.
Particularly, in the technical scheme of the application, when the spatial enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram are fused to obtain the decoding characteristic diagram, as the spatial enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram respectively strengthen characteristic association in an image spatial dimension and a model channel dimension, the inconsistency of convergence directions of respective characteristic distributions can lead to the monotonicity difference of the integral characteristic distribution of the decoding characteristic diagram obtained after fusion, thereby leading to the poor convergence effect of decoding regression of the decoding characteristic diagram decoder and influencing the accuracy of decoding values of the decoder.
Thereby, each feature matrix of the decoded feature map is subjected to a smooth maximum function approximation modulation expressed as:
m i,j is the eigenvalue of the diagonal matrix M obtained by linear transformation of each eigenvalue of the decoding eigenvector, II 2 Is the two norms of the vector, andrepresenting multiplying each value of the matrix by a predetermined value.
Here, by approximately defining the symbolized distance function with a smooth maximum function along the row and column dimensions of each feature matrix M of the decoding feature map, a relatively good union of convex optimizations of the high-dimensional manifolds characterized by each feature matrix M of the decoding feature map in the high-dimensional feature space can be achieved, and by modulating the structured feature distribution of each feature matrix M of the decoding feature map with it, a natural distribution transfer of the spatial feature variation from the intrinsic structure of the feature distribution to the feature space can be obtained, and the convex monotonicity retention of the feature expression of the high-dimensional manifolds of each feature matrix M of the decoding feature map is enhanced, thereby enhancing the distribution monotonicity of the decoding feature map as a whole, and further improving the convergence effect of the decoding regression by the decoding feature map decoder, and enhancing the accuracy of the decoding values of the decoder. Therefore, the intelligent detection can be accurately carried out on the content of allicin in the garlic wastewater, so that the detection accuracy is improved while the allicin determination method is simplified.
Based on the above, the application provides a method for measuring the content of allicin in garlic wastewater, which comprises the following steps: acquiring a liquid chromatogram of garlic wastewater to be detected; the liquid chromatogram is passed through a noise reduction module based on an automatic coder-decoder to obtain a noise-reduced liquid chromatogram; the liquid chromatogram after noise reduction is processed through a first convolution neural network model using spatial attention so as to obtain a spatial enhancement liquid chromatogram characteristic diagram; the liquid chromatogram after noise reduction is processed through a second convolution neural network model using the channel attention so as to obtain a channel enhanced liquid chromatogram characteristic diagram; fusing the space enhancement liquid chromatography feature map and the channel enhancement liquid chromatography feature map to obtain a decoding feature map; performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map; and carrying out decoding regression on the corrected decoding characteristic diagram through a decoder to obtain a decoding value for representing the content of allicin in the garlic wastewater.
Fig. 1 is an application scenario diagram of a method for determining the content of allicin in garlic wastewater according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, a liquid chromatogram of garlic wastewater to be detected is acquired using a chromatograph (e.g., se as shown in fig. 1). Further, the liquid chromatogram of the garlic waste water to be detected is input to a server (e.g., S as illustrated in fig. 1) in which an algorithm for measuring the content of allicin in the garlic waste water is disposed, wherein the server is capable of processing the liquid chromatogram of the garlic waste water to be detected based on the algorithm for measuring the content of allicin in the garlic waste water to obtain a decoded value for representing the content of allicin in the garlic waste water.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary method
Fig. 2 is a flow chart of a method for determining the content of allicin in garlic waste water according to an embodiment of the present application. As shown in fig. 2, the method for determining the content of allicin in garlic wastewater according to the embodiment of the application comprises the following steps: s110, acquiring a liquid chromatogram of garlic wastewater to be detected; s120, enabling the liquid chromatogram to pass through a noise reduction module based on an automatic coder-decoder to obtain a liquid chromatogram after noise reduction; s130, the liquid chromatogram after noise reduction is processed through a first convolution neural network model using spatial attention so as to obtain a spatial enhancement liquid chromatogram characteristic diagram; s140, the liquid chromatogram after noise reduction is processed through a second convolution neural network model using the attention of the channel to obtain a channel enhanced liquid chromatogram characteristic diagram; s150, fusing the space enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram to obtain a decoding characteristic diagram; s160, carrying out feature distribution correction on the decoding feature map to obtain a corrected decoding feature map; and S170, carrying out decoding regression on the corrected decoding characteristic diagram through a decoder to obtain a decoding value used for representing the content of allicin in the garlic wastewater.
Fig. 3 is a schematic diagram of a method for determining the content of allicin in garlic waste water according to an embodiment of the present application. In this architecture, as shown in fig. 3, first, a liquid chromatogram of garlic waste water to be detected is acquired. And then, the liquid chromatogram is passed through a noise reduction module based on an automatic coder-decoder to obtain a noise-reduced liquid chromatogram. And then, the liquid chromatogram after noise reduction is subjected to a first convolution neural network model using spatial attention so as to obtain a spatial enhancement liquid chromatogram characteristic diagram, and meanwhile, the liquid chromatogram after noise reduction is subjected to a second convolution neural network model using channel attention so as to obtain a channel enhancement liquid chromatogram characteristic diagram. Further, the spatially enhanced liquid chromatography profile and the channel enhanced liquid chromatography profile are fused to obtain a decoding profile. And then, carrying out feature distribution correction on the decoding feature map to obtain a corrected decoding feature map. And then, carrying out decoding regression on the corrected decoding characteristic diagram through a decoder to obtain a decoding value for representing the garlicin content in the garlic wastewater.
In step S110, a liquid chromatogram of the garlic waste water to be detected is acquired. As described above in the background art, the existing allicin detection method requires pretreatment of a sample before detection, extraction of allicin therefrom, and subsequent measurement of the content. Common pretreatment methods include water dissolution, acetone dissolution, ethanol extraction, n-hexane extraction, and the like. However, the treatment methods are not the most suitable means for garlic wastewater, are complex, have high manufacturing cost and are not suitable for daily detection of vast middle and small enterprises. Thus, an optimized protocol for determining the content of allicin in garlic waste water is desired.
Accordingly, it is considered that the existing garlicin detection scheme is unsuitable for practical application due to complicated method and high cost. Since liquid chromatography is a chromatography using a liquid as a mobile phase, it is used to measure various ion contents due to its advantages of being fast, sensitive, good in selectivity and simultaneously measuring multiple components, and is widely used in various fields such as water, pulp and bleaching solutions, food analysis, biological fluids, steel and environmental analysis, etc. Therefore, in the technical scheme of the application, the method for analyzing the liquid chromatogram of the garlic wastewater can be adopted to measure the content of allicin in the wastewater. However, it is considered that since the liquid chromatogram of the garlic wastewater has much information, it is difficult to extract the information of the content of allicin effectively, and the liquid chromatogram is also subject to the interference of image noise, which in turn makes it difficult to measure the content of allicin.
Based on the above, in the technical scheme of the application, an artificial intelligent detection technology based on deep learning is adopted, so that after the liquid chromatogram of the garlic wastewater is subjected to noise reduction, spatial characteristic position information and content related characteristic information on a channel are extracted, so that the content characteristic distribution of allicin in the liquid chromatogram of the garlic wastewater is identified, and decoding regression is carried out to measure the content of the allicin. Therefore, the intelligent detection can be accurately carried out on the content of allicin in the garlic wastewater, so that the detection accuracy is improved while the allicin determination method is simplified. Specifically, in the technical scheme of the application, firstly, a liquid chromatogram of garlic wastewater to be detected is obtained. Here, the liquid chromatogram of the garlic waste water to be detected may be acquired by a chromatograph.
In step S120, the liquid chromatogram is passed through a noise reduction module based on an automatic codec to obtain a noise reduced liquid chromatogram. Considering that in the process of collecting the liquid chromatogram of the garlic wastewater to be detected, a great deal of noise interference may exist in the collected liquid chromatogram due to environmental factors, so that feature extraction in the liquid chromatogram image becomes difficult, in order to improve the capability of feature extraction to improve the accuracy of allicin content measurement, the liquid chromatogram is further subjected to a noise reduction module based on an automatic codec to obtain a noise-reduced liquid chromatogram. In particular, here, the image encoder of the automatic codec comprises at least one convolution layer, and the image decoder of the automatic codec comprises at least one deconvolution layer.
Specifically, the auto-codec based noise reduction module first inputs the liquid-phase chromatogram to an encoder of the auto-codec based noise reduction module, wherein the encoder uses a convolutional layer to explicitly spatially encode the liquid-phase chromatogram to obtain image features. And then inputting the image features into a decoder of the noise reduction module based on the automatic coder-decoder, wherein the decoder uses a deconvolution layer to deconvolute the image features to obtain the noise-reduced liquid-phase chromatogram.
In step S130, the noise-reduced liquid chromatogram is passed through a first convolutional neural network model using spatial attention to obtain a spatially enhanced liquid chromatogram. That is, feature mining of the denoised liquid phase chromatogram is performed using a convolutional neural network model having excellent performance in terms of local implicit feature extraction of an image, and particularly, considering that there is a correlation between each of the local implicit features for the denoised liquid phase chromatogram, and distribution information of important features in space should be also focused when feature extraction of the denoised liquid phase chromatogram is performed. Therefore, in the technical scheme of the application, in order to improve the characteristic extraction effect on the liquid-phase chromatogram after noise reduction so as to accurately determine the content of allicin, a convolution neural network model with double attention is further used for characteristic mining of the liquid-phase chromatogram after noise reduction.
In particular, here, the dual attention mechanism includes a spatial attention mechanism. Namely, the liquid chromatogram after noise reduction is subjected to a first convolution neural network model of spatial attention so as to dig out characteristic information of important characteristics distributed on spatial positions in the liquid chromatogram after noise reduction, thereby obtaining a spatial enhancement liquid chromatogram characteristic diagram. It should be appreciated that the image features extracted by the spatial attention reflect the weights of the differences in spatial dimension features to suppress or enhance features at different spatial locations.
Fig. 4 is a flowchart of a method for determining the content of allicin in garlic wastewater according to an embodiment of the present application, wherein the liquid chromatogram after noise reduction is used to obtain a space enhancement liquid chromatogram by using a first convolution neural network model of space attention. As shown in fig. 4, the step of obtaining a space-enhanced liquid chromatogram by using a first convolution neural network model of space attention from the liquid chromatogram after noise reduction includes: s210, performing depth convolution coding on the liquid phase chromatogram after noise reduction by using a convolution coding part of the first convolution neural network model to obtain an initial convolution characteristic diagram; s220, inputting the initial convolution feature map into a spatial attention portion of the first convolution neural network model to obtain a spatial attention map; s230, the spatial attention is subjected to a Softmax activation function to obtain a spatial attention profile; and S240, calculating the position-wise point multiplication of the spatial attention characteristic map and the initial convolution characteristic map to obtain a spatial enhancement liquid chromatography characteristic map.
In step S140, the noise-reduced liquid chromatogram is passed through a second convolutional neural network model using channel attention to obtain a channel-enhanced liquid chromatogram. Similarly, considering that the liquid chromatogram after noise reduction has relevance among local implicit features, the content feature relevance information on the channel is also focused when the feature extraction of the liquid chromatogram after noise reduction is carried out. Therefore, in the technical scheme of the application, in order to improve the characteristic extraction effect on the liquid-phase chromatogram after noise reduction so as to accurately determine the content of allicin, a convolution neural network model with double attention is further used for characteristic mining of the liquid-phase chromatogram after noise reduction. In particular, here, the dual attention mechanism also includes a channel attention mechanism.
Namely, the liquid chromatogram after noise reduction is used for mining the content characteristic related information of the liquid chromatogram on the channel by using a second convolution neural network model of the channel attention, so that a channel enhanced liquid chromatogram characteristic diagram is obtained. It should be appreciated that the image features extracted by the channel attention reflect the correlation and importance between feature channels.
The channel attention and the space attention can respectively pay attention to the characteristic content and the characteristic position in the image, the characteristic extraction effect of the network is improved to a certain extent, so that different types of effective information about the allicin content in the garlic wastewater can be captured in a large amount, the characteristic distinguishing learning capability can be effectively enhanced, and in the network training process, the task processing system is more focused on finding out the remarkable useful information related to the current output in the input image data, thereby improving the output quality, and the increasing attention module can bring continuous performance improvement.
Specifically, in the above method for determining the content of allicin in garlic wastewater, the step of obtaining a channel enhanced liquid chromatography characteristic map by using a second convolution neural network model of channel attention from the denoised liquid chromatography comprises the following steps: input data is processed in forward pass of layers using layers of the second convolutional neural network model: performing convolution processing based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; pooling the convolution feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the global average value of each feature matrix of the activated feature map along the channel dimension to obtain a channel feature vector; calculating the ratio of the eigenvalue of each position in the channel eigenvector relative to the weighted sum of the eigenvalues of all positions of the channel eigenvector to obtain a channel weighted eigenvector; taking the characteristic values of the positions of the channel weighted characteristic vectors as weights to perform point multiplication on the characteristic matrix of the activated characteristic map along the channel dimension to obtain a generated characteristic map; and the generated characteristic diagram output by the last layer of the second convolutional neural network model is the channel enhanced liquid chromatography characteristic diagram.
In step S150, the spatially enhanced liquid chromatography profile and the channel enhanced liquid chromatography profile are fused to obtain a decoding profile. That is, the spatial enhancement liquid chromatography feature map and the channel enhancement liquid chromatography feature map are fused to fuse important feature distribution information and content association feature distribution information about the allicin content in the liquid chromatography after noise reduction, and the important feature distribution information and the content association feature distribution information are used as decoding feature maps.
Specifically, in the above method for determining the content of allicin in garlic wastewater, the fusing the space-enhanced liquid chromatography characteristic map and the channel-enhanced liquid chromatography characteristic map to obtain a decoding characteristic map includes: fusing the spatially enhanced liquid chromatography signature and the channel enhanced liquid chromatography signature to obtain a decoding signature with the following formula; wherein, the formula is:
wherein F is d For the decoding feature map, F a F for the spatially enhanced liquid chromatography profile b For the channel enhanced liquid chromatography profile,elements representing positions of the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile corresponding to each other are added, and α and β are weighting parameters for controlling balance between the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile in the decoding profile.
In step S160, the decoded feature map is subjected to feature distribution correction to obtain a corrected decoded feature map. Particularly, in the technical scheme of the application, when the spatial enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram are fused to obtain the decoding characteristic diagram, as the spatial enhancement liquid chromatography characteristic diagram and the channel enhancement liquid chromatography characteristic diagram respectively strengthen characteristic association in an image spatial dimension and a model channel dimension, the inconsistency of convergence directions of respective characteristic distributions can lead to the monotonicity difference of the integral characteristic distribution of the decoding characteristic diagram obtained after fusion, thereby leading to the poor convergence effect of decoding regression of the decoding characteristic diagram decoder and influencing the accuracy of decoding values of the decoder. Thereby, the respective feature matrices of the decoded feature map are subjected to a smooth maximum function approximation modulation.
Specifically, in the above method for determining the content of allicin in garlic waste water, the performing feature distribution correction on the decoded feature map to obtain a corrected decoded feature map includes: performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map according to the following formula: wherein, the formula is:
Wherein M represents a diagonal matrix obtained by linear transformation of each feature matrix of the decoding feature map, M i,j Is the eigenvalue of the (i, j) th position of the diagonal matrix obtained by linear transformation of each eigenvector of the decoding eigenvector, and II is II 2 Is the two norms of the vector, andrepresenting multiplication of each value of the matrix by a predetermined value,/->Representing addition by position, M ′ Representing the respective feature matrices of the corrected decoded feature map.
Here, by approximately defining the symbolized distance function with a smooth maximum function along the row and column dimensions of each feature matrix M of the decoding feature map, a relatively good union of convex optimizations of the high-dimensional manifolds characterized by each feature matrix M of the decoding feature map in the high-dimensional feature space can be achieved, and by modulating the structured feature distribution of each feature matrix M of the decoding feature map with it, a natural distribution transfer of the spatial feature variation from the intrinsic structure of the feature distribution to the feature space can be obtained, and the convex monotonicity retention of the feature expression of the high-dimensional manifolds of each feature matrix M of the decoding feature map is enhanced, thereby enhancing the distribution monotonicity of the decoding feature map as a whole, and further improving the convergence effect of the decoding regression by the decoding feature map decoder, and enhancing the accuracy of the decoding values of the decoder.
In step S170, the corrected decoding characteristic map is subjected to decoding regression by a decoder to obtain a decoded value representing the content of allicin in garlic waste water. Therefore, the intelligent detection can be accurately carried out on the content of allicin in the garlic wastewater, so that the detection accuracy is improved while the allicin determination method is simplified.
Specifically, in the method for determining the content of allicin in garlic waste water, the decoding regression of the corrected decoding characteristic diagram is performed by a decoder to obtain a decoded value for representing the content of allicin in garlic waste water, which comprises the following steps: performing a decoding regression on the corrected decoding feature map using the decoder in the following formula to obtain the decoded value; wherein, the formula is:wherein X is the corrected decoding feature map, Y is the decoded values, W is a weight matrix,/and>representing a matrix multiplication.
In summary, the method for determining the content of allicin in garlic wastewater according to the embodiment of the application is clarified, which adopts an artificial intelligence detection technology based on deep learning to extract spatial characteristic position information and content associated characteristic information on a channel after noise reduction of a liquid chromatogram of garlic wastewater, so as to identify the content characteristic distribution of allicin in the liquid chromatogram of garlic wastewater, and then perform decoding regression to determine the content of allicin. Therefore, the intelligent detection of the content of allicin in the garlic wastewater can be accurately performed.
Exemplary System
FIG. 5 is a block diagram of a system for determining the content of allicin in garlic waste water according to an embodiment of the present application. As shown in fig. 5, a system 100 for measuring the content of allicin in garlic wastewater according to an embodiment of the present application includes: a data acquisition module 110, configured to acquire a liquid chromatogram of garlic wastewater to be detected; the noise reduction module 120 is configured to pass the liquid chromatogram through a noise reduction module based on an automatic codec to obtain a noise-reduced liquid chromatogram; a spatial attention applying module 130, configured to apply the noise-reduced liquid chromatogram to a first convolutional neural network model using spatial attention to obtain a spatially enhanced liquid chromatogram; a channel attention applying module 140, configured to apply the noise-reduced liquid chromatogram to a second convolutional neural network model using channel attention to obtain a channel enhanced liquid chromatogram feature map; a fusion module 150, configured to fuse the spatial enhancement liquid chromatography feature map and the channel enhancement liquid chromatography feature map to obtain a decoding feature map; a feature distribution correction module 160, configured to perform feature distribution correction on the decoded feature map to obtain a corrected decoded feature map; and a decoding module 170, configured to perform decoding regression on the corrected decoding feature map through a decoder to obtain a decoded value that is used to represent the content of allicin in the garlic wastewater.
In one example, in the above system 100 for determining the content of allicin in garlic wastewater, the noise reduction module 120 is further configured to: inputting the liquid chromatogram to an encoder of the automatic codec-based noise reduction module, wherein the encoder performs explicit spatial encoding on the liquid chromatogram by using a convolution layer to obtain image features; and inputting the image features into a decoder of the noise reduction module based on the automatic coder-decoder, wherein the decoder uses a deconvolution layer to deconvolute the image features to obtain the noise-reduced liquid-phase chromatogram.
In one example, in the above system 100 for determining the amount of allicin in garlic waste water, the image encoder of the automatic codec includes at least one convolution layer, and the image decoder of the automatic codec includes at least one deconvolution layer.
In one example, in the above system 100 for determining the content of allicin in garlic waste water, the spatial attention application module 130 is further configured to: performing depth convolution coding on the liquid phase chromatogram after noise reduction by using a convolution coding part of the first convolution neural network model to obtain an initial convolution characteristic map; inputting the initial convolution feature map into a spatial attention portion of the first convolution neural network model to obtain a spatial attention map; -passing said spatial attention map through a Softmax activation function to obtain a spatial attention profile; and calculating the position-wise point multiplication of the spatial attention characteristic map and the initial convolution characteristic map to obtain a spatial enhancement liquid chromatography characteristic map.
In one example, in the above system 100 for determining the content of allicin in garlic waste water, the channel attention application module 140 is further configured to: input data is processed in forward pass of layers using layers of the second convolutional neural network model: performing convolution processing based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; pooling the convolution feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the global average value of each feature matrix of the activated feature map along the channel dimension to obtain a channel feature vector; calculating the ratio of the eigenvalue of each position in the channel eigenvector relative to the weighted sum of the eigenvalues of all positions of the channel eigenvector to obtain a channel weighted eigenvector; taking the characteristic values of the positions of the channel weighted characteristic vectors as weights to perform point multiplication on the characteristic matrix of the activated characteristic map along the channel dimension to obtain a generated characteristic map; and the generated characteristic diagram output by the last layer of the second convolutional neural network model is the channel enhanced liquid chromatography characteristic diagram.
In one example, in the above system 100 for determining the content of allicin in garlic waste water, the fusion module 150 is further configured to: fusing the spatially enhanced liquid chromatography signature and the channel enhanced liquid chromatography signature to obtain a decoding signature with the following formula; wherein, the formula is:
wherein F is d For the decoding feature map, F a F for the spatially enhanced liquid chromatography profile b For the channel enhanced liquid chromatography profile,elements representing positions of the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile corresponding to each other are added, and α and β are weighting parameters for controlling balance between the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile in the decoding profile.
In one example, in the above system 100 for determining the content of allicin in garlic waste water, the characteristic distribution correction module 160 is further configured to: performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map according to the following formula: wherein, the formula is:
wherein M represents a diagonal matrix obtained by linear transformation of each feature matrix of the decoding feature map, M i,j Is the eigenvalue of the (i, j) th position of the diagonal matrix obtained by linear transformation of each eigenvector of the decoding eigenvector, and II is II 2 Is the two norms of the vector, andrepresenting multiplication of each value of the matrix by a predetermined value,/->Representing addition by position, M ′ Representing the respective feature matrices of the corrected decoded feature map.
In one example, in the above system 100 for determining the content of allicin in garlic waste water, the decoding module 170 is further configured to: performing a decoding regression on the corrected decoding feature map using the decoder in the following formula to obtain the decoded value; wherein, the formula is:wherein X is the corrected decoding feature map, Y is the decoded values, W is a weight matrix,/and>representing a matrix multiplication.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described system 100 for measuring the content of allicin in garlic wastewater have been described in detail in the above description of the method for measuring the content of allicin in garlic wastewater with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.
As described above, the system 100 for measuring the content of allicin in garlic wastewater according to an embodiment of the present application may be implemented in various terminal devices, for example, a server for measuring the content of allicin in garlic wastewater, or the like. In one example, the system 100 for determining the content of allicin in garlic wastewater according to an embodiment of the present application may be integrated into a terminal device as a software module and/or hardware module. For example, the system 100 for determining the content of allicin in garlic waste water may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the system 100 for determining the amount of allicin in garlic waste water can also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the system for measuring the content of allicin 100 in garlic waste water and the terminal device may be separate devices, and the system for measuring the content of allicin 100 in garlic waste water may be connected to the terminal device through a wired and/or wireless network, and transmit interactive information according to a agreed data format.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (7)
1. A method for determining the content of allicin in garlic waste water, comprising the steps of: acquiring a liquid chromatogram of garlic wastewater to be detected; the liquid chromatogram is passed through a noise reduction module based on an automatic coder-decoder to obtain a noise-reduced liquid chromatogram; the liquid chromatogram after noise reduction is processed through a first convolution neural network model using spatial attention so as to obtain a spatial enhancement liquid chromatogram characteristic diagram; the liquid chromatogram after noise reduction is processed through a second convolution neural network model using the channel attention so as to obtain a channel enhanced liquid chromatogram characteristic diagram; fusing the space enhancement liquid chromatography feature map and the channel enhancement liquid chromatography feature map to obtain a decoding feature map; performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map; performing decoding regression on the corrected decoding characteristic diagram through a decoder to obtain a decoding value for representing the content of allicin in the garlic wastewater;
The step of obtaining the liquid chromatogram after noise reduction by the noise reduction module based on the automatic coder-decoder comprises the following steps: inputting the liquid chromatogram to an encoder of the automatic codec-based noise reduction module, wherein the encoder performs explicit spatial encoding on the liquid chromatogram by using a convolution layer to obtain image features; and inputting the image features into a decoder of the noise reduction module based on the automatic codec, wherein the decoder uses a deconvolution layer to deconvolute the image features to obtain the noise-reduced liquid-phase chromatogram.
2. The method of determining the amount of allicin in garlic waste water according to claim 1, wherein the image encoder of the automatic codec comprises at least one convolution layer and the image decoder of the automatic codec comprises at least one deconvolution layer.
3. The method for determining the content of allicin in garlic waste water according to claim 2, wherein the step of obtaining the space-enhanced liquid chromatography profile by using the first convolution neural network model of spatial attention on the liquid chromatography after noise reduction comprises the steps of: performing depth convolution coding on the liquid phase chromatogram after noise reduction by using a convolution coding part of the first convolution neural network model to obtain an initial convolution characteristic map; inputting the initial convolution feature map into a spatial attention portion of the first convolution neural network model to obtain a spatial attention map; -passing said spatial attention map through a Softmax activation function to obtain a spatial attention profile; and calculating the position-wise point multiplication of the spatial attention characteristic map and the initial convolution characteristic map to obtain a spatial enhancement liquid chromatography characteristic map.
4. A method for determining the content of allicin in garlic waste water according to claim 3, wherein the step of obtaining the channel enhancement liquid chromatogram by using the second convolution neural network model of channel attention from the denoised liquid chromatogram comprises the steps of: input data is processed in forward pass of layers using layers of the second convolutional neural network model: performing convolution processing based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; pooling the convolution feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the global average value of each feature matrix of the activated feature map along the channel dimension to obtain a channel feature vector; calculating the ratio of the eigenvalue of each position in the channel eigenvector relative to the weighted sum of the eigenvalues of all positions of the channel eigenvector to obtain a channel weighted eigenvector; taking the characteristic values of the positions of the channel weighted characteristic vectors as weights to carry out point multiplication on the characteristic matrix of the activated characteristic map along the channel dimension to obtain a generated characteristic map; and the generated characteristic diagram output by the last layer of the second convolutional neural network model is the channel enhanced liquid chromatography characteristic diagram.
5. The method of determining the amount of allicin in garlic waste water according to claim 4, wherein said fusing the space-enhanced liquid chromatography profile and the channel-enhanced liquid chromatography profile to obtain a decoded profile comprises: fusing the spatially enhanced liquid chromatography signature and the channel enhanced liquid chromatography signature to obtain a decoding signature with the following formula; wherein, the formula is:
wherein F is d For the decoding feature map, F a F for the spatially enhanced liquid chromatography profile b For the channel enhanced liquid chromatography profile,elements representing positions of the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile corresponding to each other are added, and α and β are weighting parameters for controlling balance between the spatial enhancement liquid chromatography profile and the channel enhancement liquid chromatography profile in the decoding profile.
6. The method for determining the allicin content in garlic waste water according to claim 5, wherein the performing the feature distribution correction on the decoded feature map to obtain a corrected decoded feature map comprises: performing feature distribution correction on the decoding feature map to obtain a corrected decoding feature map according to the following formula: wherein, the formula is:
Wherein M represents a diagonal matrix obtained by linear transformation of each feature matrix of the decoding feature map, M i,j Is the eigenvalue of the (i, j) th position of the diagonal matrix obtained by linear transformation of each eigenvector of the decoding eigenvector, and II is II 2 Is the two norms of the vector, andrepresenting multiplication of each value of the matrix by a predetermined value,/->Representing addition by position, M' represents the respective feature matrices of the corrected decoded feature map.
7. The method of determining the amount of allicin in garlic waste water according to claim 6, wherein said subjecting the corrected decoded signature to a decoding regression by a decoder to obtain a decoded value representing the amount of allicin in garlic waste water comprises: performing a decoding regression on the corrected decoding feature map using the decoder in the following formula to obtain the decoded value; wherein, the formula is:wherein X is the corrected decoding feature map, Y is the decoded values, W is a weight matrix,/and>representing a matrix multiplication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005127.1A CN116223661B (en) | 2023-01-04 | 2023-01-04 | Method for measuring content of allicin in garlic wastewater |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005127.1A CN116223661B (en) | 2023-01-04 | 2023-01-04 | Method for measuring content of allicin in garlic wastewater |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116223661A CN116223661A (en) | 2023-06-06 |
| CN116223661B true CN116223661B (en) | 2023-12-15 |
Family
ID=86586503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310005127.1A Active CN116223661B (en) | 2023-01-04 | 2023-01-04 | Method for measuring content of allicin in garlic wastewater |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116223661B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199233A (en) * | 2019-12-30 | 2020-05-26 | 四川大学 | Improved deep learning pornographic image identification method |
| CN113485261A (en) * | 2021-06-29 | 2021-10-08 | 西北师范大学 | CAEs-ACNN-based soft measurement modeling method |
| CN113889198A (en) * | 2021-09-24 | 2022-01-04 | 国网宁夏电力有限公司电力科学研究院 | Transformer fault diagnosis method and equipment based on oil chromatogram time-frequency domain information and residual error attention network |
| CN114492559A (en) * | 2021-11-24 | 2022-05-13 | 国网青海省电力公司海南供电公司 | Power equipment fault diagnosis method based on data time-frequency domain modeling |
| CN114511502A (en) * | 2021-12-30 | 2022-05-17 | 浙江大学 | Gastrointestinal endoscope image polyp detection system based on artificial intelligence, terminal and storage medium |
-
2023
- 2023-01-04 CN CN202310005127.1A patent/CN116223661B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199233A (en) * | 2019-12-30 | 2020-05-26 | 四川大学 | Improved deep learning pornographic image identification method |
| CN113485261A (en) * | 2021-06-29 | 2021-10-08 | 西北师范大学 | CAEs-ACNN-based soft measurement modeling method |
| CN113889198A (en) * | 2021-09-24 | 2022-01-04 | 国网宁夏电力有限公司电力科学研究院 | Transformer fault diagnosis method and equipment based on oil chromatogram time-frequency domain information and residual error attention network |
| CN114492559A (en) * | 2021-11-24 | 2022-05-13 | 国网青海省电力公司海南供电公司 | Power equipment fault diagnosis method based on data time-frequency domain modeling |
| CN114511502A (en) * | 2021-12-30 | 2022-05-17 | 浙江大学 | Gastrointestinal endoscope image polyp detection system based on artificial intelligence, terminal and storage medium |
Non-Patent Citations (16)
| Title |
|---|
| A new data processing strategy combined with a convolutional neural network for rapid and accurate prediction of geographical classifications of natural products;Bingwen Zhou等;Chemometrics and Intelligent Laboratory Systems;第227卷;第1-8页 * |
| Bingwen Zhou 等.A new data processing strategy combined with a convolutional neural network for rapid and accurate prediction of geographical classifications of natural products.Chemometrics and Intelligent Laboratory Systems.2022,第227卷第 1-8页. * |
| Channel attention convolutional neural network for Chinese baijiu detection with E-nose;Shanshan Zhang等;IEEE Sensors Journal;第21卷(第04期);第16170 – 16182页 * |
| Classification of organic and conventional olives using convolutional neural networks;Unluturk, MS等;Neural Computing & Applications;第33卷(第23期);第16733-16744页 * |
| Shanshan Zhang等.Channel attention convolutional neural network for Chinese baijiu detection with E-nose.IEEE Sensors Journal.2021,第21卷(第04期),第16170 – 16182页. * |
| Unluturk, MS等.Classification of organic and conventional olives using convolutional neural networks.Neural Computing & Applications.2021,第33卷(第23期),第16733-16744页. * |
| 刘航;汪西莉.基于注意力机制的遥感图像分割模型.激光与光电子学进展.2019,57(第04期),第170-180页. * |
| 基于卷积神经网络的多模态人脸活体检测算法研究与实现;李欣;CNKI 优秀硕士学位论文全文库(第08期);第 1-77页 * |
| 基于卷积神经网络的多模态人脸活体检测算法研究与实现;李欣;CNKI优秀硕士学位论文全文库(第08期);第1-77页 * |
| 基于卷积神经网络的水稻叶片病害识别;温鑫;CNKI 优秀硕士学位论文全文库(第03期);第1-66页 * |
| 基于卷积神经网络的水稻叶片病害识别;温鑫;CNKI优秀硕士学位论文全文库(第03期);第1-66页 * |
| 基于双注意力编码-解码器架构的视网膜血管分割;李天培;陈黎;;计算机科学;47(第05期);第166-171页 * |
| 基于注意力机制的遥感图像分割模型;刘航;汪西莉;;激光与光电子学进展;57(第04期);第170-180页 * |
| 李天培;陈黎.基于双注意力编码-解码器架构的视网膜血管分割.计算机科学.2019,47(第05期),第166-171页. * |
| 牛娜娜等.预处理工艺对黑蒜功能性成分、抗氧化活性影响及相关性研究.食品与发酵工业.2020,第47卷(第08期),第67-75页. * |
| 预处理工艺对黑蒜功能性成分、抗氧化活性影响及相关性研究;牛娜娜等;食品与发酵工业;第47卷(第08期);第67-75页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116223661A (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Checa et al. | Lipidomic data analysis: tutorial, practical guidelines and applications | |
| Chen et al. | Image‐denoising algorithm based on improved K‐singular value decomposition and atom optimization | |
| Liapikos et al. | Quantitative structure retention relationship (QSRR) modelling for Analytes’ retention prediction in LC-HRMS by applying different Machine Learning algorithms and evaluating their performance | |
| Matyushin et al. | Gas chromatographic retention index prediction using multimodal machine learning | |
| Zeng et al. | Chemometrics in comprehensive multidimensional separations | |
| CN115754107B (en) | Automatic sampling analysis system and method for lithium hexafluorophosphate preparation | |
| Myburgh et al. | Tracking translation invariance in CNNs | |
| Souihi et al. | MultiConditionRT: Predicting liquid chromatography retention time for emerging contaminants for a wide range of eluent compositions and stationary phases | |
| CN115754108B (en) | Acidity determination system and method for electronic grade hexafluorobutadiene | |
| CN116416248A (en) | Intelligent analysis system and method based on fluorescence microscope | |
| Xie et al. | Sliding-window based scale-frequency map for bird sound classification using 2d-and 3d-cnn | |
| CN113034408A (en) | Infrared thermal imaging deep learning image denoising method and device | |
| CN114547241A (en) | Small sample entity identification method and model combining character perception and sentence perception | |
| CN116223661B (en) | Method for measuring content of allicin in garlic wastewater | |
| Lopes Marques et al. | Modelling chromatographic behaviour as a function of pH and solvent composition in RPLC | |
| CN117690025A (en) | Method and system for detecting red tide based on CNN-converter spectrum reconstruction | |
| Buncher et al. | Survey2Survey: a deep learning generative model approach for cross-survey image mapping | |
| CN116874002A (en) | Automatic dosing system and method for sewage treatment equipment | |
| Zhang et al. | CAM R-CNN: End-to-end object detection with class activation maps | |
| CN115902004B (en) | Measurement device and measurement method for conductivity of degassed hydrogen | |
| CN117665734A (en) | Sea surface small target anomaly detection method and system | |
| CN116467485A (en) | Video image retrieval construction system and method thereof | |
| CN110689510A (en) | Sparse representation-based image fusion method introducing dictionary information | |
| CN113610817B (en) | Characteristic peak identification method, computing device and storage medium | |
| CN116106461A (en) | Method and device for predicting liquid chromatograph retention time based on deep graph network |
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 |