CN114113471A - Method and system for detecting food freshness of artificial nose refrigerator based on machine learning - Google Patents
Method and system for detecting food freshness of artificial nose refrigerator based on machine learning Download PDFInfo
- Publication number
- CN114113471A CN114113471A CN202111312254.3A CN202111312254A CN114113471A CN 114113471 A CN114113471 A CN 114113471A CN 202111312254 A CN202111312254 A CN 202111312254A CN 114113471 A CN114113471 A CN 114113471A
- Authority
- CN
- China
- Prior art keywords
- test
- refrigerator
- sample set
- food
- train
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 53
- 238000012360 testing method Methods 0.000 claims abstract description 44
- 230000004044 response Effects 0.000 claims abstract description 28
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 230000009467 reduction Effects 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 238000004458 analytical method Methods 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000013145 classification model Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000007635 classification algorithm Methods 0.000 claims description 7
- 238000000513 principal component analysis Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 238000007637 random forest analysis Methods 0.000 claims description 2
- 238000012706 support-vector machine Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000006866 deterioration Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000001953 sensory effect Effects 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009920 food preservation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
- G01N33/0009—General constructional details of gas analysers, e.g. portable test equipment
- G01N33/0062—General constructional details of gas analysers, e.g. portable test equipment concerning the measuring method or the display, e.g. intermittent measurement or digital display
Landscapes
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medicinal Chemistry (AREA)
- Food Science & Technology (AREA)
- Combustion & Propulsion (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Cold Air Circulating Systems And Constructional Details In Refrigerators (AREA)
Abstract
An artificial nose refrigerator food freshness detection method and system based on machine learning belong to the field of refrigerator food detection. Obtaining an original response data set X of food volatile gas in the refrigerator; then carrying out median filtering processing to obtain a sample set Xm(ii) a And (3) mixing the sample sets: 1 division into training sets XtrainAnd test set Xtest(ii) a Carrying out standardization processing on the training set and the test set; for the processed training set Xtrain_stdAnd test set Xtest_stdFeature dimensionality reduction is carried out to generate a new feature subset Xtrain_pcaAnd Xtest_pca(ii) a Training set X using machine learning algorithmtrain_pcaTraining to obtain optimal parameters of model, and testing set Xtest_pcaAnd (4) identifying the freshness classification of the food, and verifying the prediction performance of the model. The method can quickly predict the freshness of the food in the refrigerator, reduce the hardware cost of the system through a software algorithm and improve the prediction precision.
Description
Technical Field
The invention relates to the field of refrigerator food detection, in particular to an artificial nose refrigerator food freshness detection method and system based on machine learning.
Background
The freshness of food directly affects the health and safety of people, and is a big matter related to the national civilization. Monitoring food freshness is a necessary means for ensuring food safety. In daily life, a refrigerator is a common low-temperature food preservation method. With time, the food stored in the refrigerator is also subject to decay and deterioration. Meanwhile, due to the relatively sealed environment inside the refrigerator, people can hardly find the deterioration of food in time. Therefore, the development of a rapid and accurate detection technology for the freshness of food in the refrigerator is significant.
The existing food freshness detection method mainly depends on expensive analytical instruments, has high cost, complex operation and long analysis time, requires strict laboratory environment, damages samples and is inconvenient for daily use. The traditional method for sensory evaluation is mainly completed by trained professionals, is easily influenced by external factors such as age, sex, mood and experience, has strong subjectivity and poor repeatability, and is easy to cause olfactory fatigue. It is therefore necessary to develop a rapid and non-destructive food freshness detection method.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an artificial nose refrigerator food freshness detection method and system based on machine learning, which utilize a sensor array in an artificial nose to collect fingerprint information of refrigerator food volatile gas, and analyze data by combining a detection method of machine learning, thereby achieving the purpose of detecting the freshness of the refrigerator food.
In order to solve the problems of the prior art, the invention adopts the technical scheme that:
a method for detecting food freshness of an artificial nose refrigerator based on machine learning comprises the following steps:
s1: obtaining an original response data set X of food volatile gas in the refrigerator by utilizing a sensor array;
s2: carrying out median filtering processing on the original response data set X to obtain a sample set Xm;
S3: sample set XmAccording to the following steps of 3: 1 into a training sample set XtrainAnd test sample set Xtest;
S4: first pair training sample set XtrainNormalizing the data, and normalizing the test sample set X according to the statistical characteristics of the training sample settestCarrying out standardization treatment;
s5: for the training sample set X after the standardization processingtrain_stdAnd test sample set Xtest_stdFeature dimensionality reduction is carried out to generate a new feature subset Xtrain_pcaAnd Xtest_pcaThe calculation amount of the system is reduced;
s6: training sample set X after dimensionality reduction by using machine learning classification algorithmtrain_pcaTraining to obtain optimal parameters of the model, and then testing the sample set Xtest_pcaAnd classifying and identifying the freshness of the food sample in the refrigerator.
Further, step S2 is specifically: replacing the original response value of the sensor at a certain moment by the median value of all response values in a neighborhood window of the point, as shown in formula (1):
Xm(i)=Median{X(i-n),…X(i),…X(i+n)} (1)
wherein: x (i) is the original response value of the sensor at a certain moment, the length L of a neighborhood window is 2n +1, n is a positive integer, Xm(i) Is a filtered response value.
Further, in step S4, the normalization process specifically includes the steps of:
S42: computing a training sample set XtrainStandard deviation σ of middle feature;
s43: for training sample set XtrainAnd test sample set XtestThe following formula (2) is used for the normalization process:
wherein, XtrainAnd XtestFiltered response data, X, for the training sample set and the test sample set, respectivelytrain_stdFor the normalized data of the training sample set, XtestAnd _isthe data after the test sample set is standardized.
Further, in the step S5, the dimensionality reduction algorithm adopts Principal Component Analysis (PCA), and the specific steps are as follows:
s51: solving a covariance matrix C of the sample data normalized in the step 4;
s52: calculating an eigenvalue of the covariance matrix C and a corresponding eigenvector;
s53: the eigenvalues are arranged from big to small in sequence, and the variance contribution rate E of each eigenvalue is calculatedjAs shown in formula (3):
wherein λ isjFor the value of the j-th characteristic,is the accumulated sum of all the characteristic values;
s54: setting the threshold of the accumulated variance contribution rate of the principal components to 95%, selecting eigenvectors corresponding to the first eigenvalues with the variance contribution rate larger than the threshold to form a mapping matrix W, and obtaining a new low-dimensional feature subset X through the mapping matrix Wtrain_pcaAnd Xtest_pcaTherefore, the key information is kept, and the dimensionality of training data is reduced, so that the training efficiency and the scheduling performance of the classification model are indirectly improved.
Further, the machine learning classification algorithm in step S6 is one of an extreme gradient boosting algorithm XGBoost, K-nearest neighbor KNN, random forest RF, support vector machine SVM, or BP neural network BPNN, and the mesh search is used to perform hyper-parameter tuning in the model training process to obtain the optimal value of each hyper-parameter, provide the optimal parameter combination for the classification model, and maximize the prediction performance of the classification model.
The system utilizing the method for detecting the food freshness of the refrigerator with the artificial nose based on the machine learning comprises the artificial nose equipment, a display module and an upper computer, wherein the artificial nose equipment is mainly used for collecting gas fingerprint information of food in the refrigerator, then transmitting the information to the upper computer for analysis, receiving an analysis result of the upper computer and feeding the analysis result back to the display module; the display module comprises green, yellow and red lamps which respectively represent that the freshness of food in the refrigerator is in a fresh state, a sub-fresh state and a rotten state and are used for displaying the detection result of the freshness of the food in the refrigerator; the artificial nose equipment comprises a sensor array, a controller and a wireless module, wherein the sensor array comprises a temperature and humidity sensor and a plurality of gas sensors responding to food volatile gas in the refrigerator; the controller is connected with the sensor array, the wireless module and the display module; the wireless module sends response data X acquired by the sensor array through collecting refrigerator food volatile gas to the upper computer, receives an analysis result of the upper computer and feeds the analysis result back to the control module.
Has the advantages that:
compared with the prior art, the method and the system for detecting the food freshness of the refrigerator with the artificial nose based on the machine learning have the advantages that fingerprint information of volatile gas of the refrigerator food is collected by the aid of the sensor array in the artificial nose, the freshness of the refrigerator food is detected by the aid of the machine learning algorithm in the upper computer, the problem that the freshness of the refrigerator food cannot be rapidly and efficiently detected without damage is effectively solved, meanwhile, the non-linear problem can be well solved by the aid of the machine learning method, the refrigerator food freshness prediction accuracy is improved, and the defects that a traditional sensory prediction method has strong subjectivity and a method for judging the food freshness by using a threshold value are overcome. The detection system provided by the invention has the functions of automatically acquiring, processing and uploading data, so that the hardware cost of the system is reduced, and the efficiency and the prediction precision of refrigerator food freshness analysis are improved.
Drawings
FIG. 1 is a schematic structural diagram of a system of the method for detecting food freshness of an artificial nose refrigerator based on machine learning according to the present invention;
FIG. 2 is a flow chart of a method for detecting food freshness in an artificial nose refrigerator based on machine learning according to an embodiment;
FIG. 3 is a comparison graph of response data before and after median filtering in one embodiment;
FIG. 4 is a diagram of the results of the PCA extraction in one embodiment;
FIG. 5 is a diagram of a classification confusion matrix implementing a machine learning classification algorithm.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
As shown in fig. 1, an artificial nose refrigerator food freshness detection system based on machine learning comprises an artificial nose device, a display module and an upper computer, wherein the artificial nose device comprises a sensor array, a controller and a wireless module, and the sensor array comprises a temperature and humidity sensor and a plurality of gas sensors responding to food volatile gas in a refrigerator; the controller is connected with the sensor array, the wireless module and the display module; the wireless module sends response data X acquired by the sensor array through collecting refrigerator food volatile gas to the upper computer, receives an analysis result of the upper computer and feeds the analysis result back to the control module; the display module comprises green, yellow and red lamps which respectively represent that the freshness of food in the refrigerator is in a fresh state, a sub-fresh state and a rotten state and are used for displaying the detection result of the freshness of the food in the refrigerator; the host computer mainly realizes the detection of food freshness in the refrigerator, sends the detection result to artificial nose equipment, and then shows through display module.
The sensor array in the artificial nose device in this embodiment includes 12 gas sensors (6 types, 2 each type) and 1 temperature and humidity sensor, and its working process is as follows: before the sensor array works, the sensor array is preheated for 30min, the working voltage is 5V, the sampling frequency is 1Hz, signal data output by the sensor array are sent to a controller in real time, and the controller sends the signal data to an upper computer to be detected through a machine learning algorithm.
As shown in fig. 2, the method for detecting food freshness of the artificial nose refrigerator based on machine learning by using the system comprises the following steps:
s1: obtaining an original response data set X of food volatile gas in the refrigerator by utilizing a sensor array;
s2: carrying out median filtering processing on the original response data set X to obtain a sample set Xm;
S3: sample set XmAccording to the following steps of 3: 1 into a training sample set XtrainAnd test sample set Xtest;
S4: first pair training sample set XtrainNormalizing the data, and normalizing the test sample set X according to the statistical characteristics of the training sample settestCarrying out standardization treatment;
s5: for the training sample set X after the standardization processingtrain_stdAnd test sample set Xtest_stdFeature dimensionality reduction is carried out to generate a new feature subset Xtrain_pcaAnd Xtest_pcaThe calculation amount of the system is reduced;
s6: training sample set X after dimensionality reduction by using machine learning classification algorithmtrain_pcaTraining to obtain optimal parameters of the model, and then testing the sample set Xtest_pcaAnd classifying and identifying the freshness of the food sample in the refrigerator.
The detection method of the present invention is described below with a specific embodiment, where the environment of the operation of the present embodiment is an implementation environment of an experiment, and the implementation environment is written and implemented on a Dell 3681 computer, Windows10, intel i5 processor, 8G run memory, Pycharm2019, python3.7, scinit-learnt 0.21.3, and xgbost 1.4.2.
Step S1: obtaining a raw response data matrix X of food volatile gas in the refrigerator by using a sensor array:
wherein, [ x ]i,1,xi,2,…,xi,j,…,xi,n]Measuring a sample data formed by food volatile gas in the refrigerator for a plurality of gas sensors, wherein m is the number of times of measuring the sample by the plurality of gas sensors; m is 600, n is 12;
step S2 specifically includes: replacing the original response value of the sensor at a certain moment by the median value of all response values in a neighborhood window of the point, as shown in formula (1):
Xm(i)=Median{X(i-n),…X(i),…X(i+n)} (1)
wherein: x (i) is the original response value of the sensor at a certain moment, the length L of a neighborhood window is 2n +1, n is 21, and Xm(i) For the filtered response value, as shown in fig. 3, the median filter has a good filtering effect on the impulse noise, and a smoother response curve can be obtained after the median filter;
step S3: sample data set XmAccording to the following steps of 3: 1 into a training sample set XtrainAnd test sample set XtestWherein the proportion of the training sample set accounts for 75 percent, and the proportion of the testing sample set accounts for 25 percent;
in step S4, the normalization process specifically includes the steps of:
S42: calculating a standard deviation sigma of the features in the training sample set;
s43: for training sample set WtrainAnd test sample set XtestThe following formula (2) is used for the normalization process:
wherein, XtrainAnd XtestFiltered response data, X, for the training sample set and the test sample set, respectivelytrain_stdFor the normalized data of the training sample set, Xtest_stdThe processed data was normalized for the test sample set.
In the step S5, the dimensionality reduction algorithm adopts Principal Component Analysis (PCA) to reduce dimensionality, and the specific steps are as follows:
s51: solving a covariance matrix C of the sample data normalized in the step 4;
s52: calculating an eigenvalue of the covariance matrix C and a corresponding eigenvector;
s53: the eigenvalues are arranged from big to small in sequence, and the variance contribution rate E of each eigenvalue is calculatedjAs shown in formula (3):
wherein λ isjFor the value of the j-th characteristic,is the accumulated sum of all the characteristic values;
s54: setting the threshold of the accumulated variance contribution rate of the principal components to 95%, selecting eigenvectors corresponding to the first eigenvalues with the variance contribution rate larger than the threshold to form a mapping matrix W, and obtaining a new low-dimensional feature subset X through the mapping matrix Wtrain_pcaAnd Xtest_pcaTherefore, the key information is kept, and the dimensionality of training data is reduced, so that the training efficiency and the scheduling performance of the classification model are indirectly improved. The feature extraction result is shown in fig. 4, and the first 3 principal components with the cumulative variance contribution rate of more than 95% are taken as a new feature subset. The principal component contribution statistics are shown in table 1:
TABLE 1 principal Components contribution ratio
In this embodiment, the machine learning classification algorithm of step S6 is an extreme gradient boosting algorithm XGBoost, and the super-parameter tuning is performed by using grid search in the model training process: in the model, a gbtree model is selected as a weak evaluator, the number n _ estimators of the hyper-parameter weak evaluators is set to be [1,100], the maximum depth max _ depth of the weak evaluator is set to be [1,20], the learning rate learning _ rate in integration is set to be [0,1], the penalty term gamma of complexity is set to be [0,5], the proportion colsample _ byneve of random sampling characteristics is set to be [0,1] when one layer of the tree is generated, the proportion colsample _ bynede of random sampling characteristics is set to be [0,1] when one leaf node is generated, and the parameter reg _ lambda of an L2 regular term is set to be [0,2 ]; the parameters are used as grid search parameters for training the model, the optimal values of the hyper-parameters are obtained after the hyper-parameters are subjected to grid-based search optimization, and optimal parameter combinations are provided for a subsequent decision model, so that the prediction performance of the classification model is maximized.
In the experiment, 0.75 × 600 samples are used for sample training, and super-parameter tuning is performed through grid search to obtain an optimal parameter combination, so that the XGboost model in the optimal parameter combination is used as a machine learning classification model for completing training. Then, selecting samples with the total sample number ratio of 0.25 as a test set to be analyzed and predicted under the trained machine learning classification model structure, wherein the experimental result is shown in fig. 5, a classification confusion matrix of the XGboost decision model on the test set is shown, 50 samples in three categories of fresh, sub-fresh and rotten are provided, and all 50 fresh samples are predicted correctly; there were 3 prediction errors for 50 sub-fresh samples; the total recognition accuracy of 5 prediction errors in 50 rotten samples reaches 94.67%.
The result shows that the method and the system for detecting the food freshness of the artificial nose refrigerator based on machine learning can quickly and efficiently detect the food freshness in the refrigerator without damage, reduce the hardware cost of the system on one hand, improve the efficiency and the prediction precision of the food freshness analysis of the refrigerator on the other hand, and have better comprehensive performance.
The above description is only a few of the preferred embodiments of the present application and is not intended to limit the present application, which may be modified and varied by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (6)
1. A refrigerator food freshness detection method based on machine learning is characterized by comprising the following steps:
s1: obtaining an original response data set X of food volatile gas in the refrigerator by utilizing a sensor array;
s2: carrying out median filtering processing on the original response data set X to obtain a sample set Xm;
S3: sample set XmRandomly dividing the training sample set into training sample sets X according to the ratio of 3: 1trainAnd test sample set Xtest;
S4: first pair training sample set XtrainNormalizing the data, and normalizing the test sample set X according to the statistical characteristics of the training sample settestCarrying out standardization treatment;
s5: for the training sample set X after the standardization processingtrain_stdAnd test sample set Xtest_stdFeature dimensionality reduction is carried out to generate a new feature subset Xtrain_pcaAnd Xtest_pcaThe calculation amount of the system is reduced;
s6: training sample set X after dimensionality reduction by using machine learning classification algorithmtrain_pcaTraining to obtain optimal parameters of the model, and testing sample set Xtest_pcaAnd classifying and identifying the freshness of the food samples of the refrigerator, and verifying the prediction performance of the model.
2. The method for detecting the freshness of food in a refrigerator based on machine learning according to claim 1, wherein the step S2 is specifically as follows: replacing the original response value of the sensor at a certain moment by the median value of all response values in a neighborhood window of the point, as shown in formula (1):
Xm(i)=Median{X(i-n),…X(i),…X(i+n)} (1)
wherein: x (i) is the original response value of the sensor at a certain moment, the length L of a neighborhood window is 2n +1, n is a positive integer, Xm(i) Is a filtered response value.
3. The method for detecting food freshness of a refrigerator based on machine learning according to claim 1, wherein in step S4, the standardization process specifically comprises the following steps:
S42: calculating a standard deviation sigma of the features in the training sample set;
s43: for training sample set XtrainAnd test sample set XtestThe following formula (2) is used for the normalization process:
wherein, XtrainAnd XtestFiltered response data, X, for the training sample set and the test sample set, respectivelytrain_stdFor the normalized data of the training sample set, Xtest_stdThe processed data was normalized for the test sample set.
4. The refrigerator food freshness detection method based on machine learning according to claim 1, wherein the dimensionality reduction algorithm in step S5 adopts Principal Component Analysis (PCA), and comprises the following steps:
s51: solving a covariance matrix C of the sample data normalized in the step 4;
s52: calculating an eigenvalue of the covariance matrix C and a corresponding eigenvector;
s53: the eigenvalues are arranged from big to small in sequence, and the variance contribution rate of each eigenvalue is calculatedEjAs shown in formula (3):
wherein λ isjFor the value of the j-th characteristic,is the accumulated sum of all the characteristic values;
s54: setting the threshold of the accumulated variance contribution rate of the principal components to 95%, selecting eigenvectors corresponding to the first eigenvalues with the variance contribution rate larger than the threshold to form a mapping matrix W, and obtaining a new low-dimensional feature subset X through the mapping matrix Wtrain_pcaAnd Xtest_pcaTherefore, the key information is kept, and the dimensionality of training data is reduced, so that the training efficiency and the scheduling performance of the classification model are indirectly improved.
5. The refrigerator food freshness detection method based on machine learning according to claim 4, characterized in that the machine learning classification algorithm in step S6 is one of extreme gradient boosting algorithm XGboost, K neighbor KNN, random forest RF, support vector machine SVM or BP neural network BPNN, and during model training, grid search is used to perform hyper-parameter tuning to obtain optimal values of each hyper-parameter, so as to provide optimal parameter combinations for the classification model and maximize the prediction performance of the classification model.
6. The system for detecting the food freshness of the refrigerator with the artificial nose based on the machine learning of claim 1 is characterized by comprising an artificial nose device, a display module and an upper computer, wherein the artificial nose device is mainly used for collecting gas fingerprint information of food in the refrigerator, transmitting the gas fingerprint information to the upper computer for analysis, receiving an analysis result of the upper computer and feeding the analysis result back to the display module; the display module comprises green, yellow and red lamps which respectively represent that the freshness of food in the refrigerator is in a fresh state, a sub-fresh state and a rotten state and are used for displaying the detection result of the freshness of the food in the refrigerator; the artificial nose equipment comprises a sensor array, a controller and a wireless module, wherein the sensor array comprises a temperature and humidity sensor and 6 gas sensors which respond to food volatile gas in the refrigerator; the controller is connected with the sensor array, the wireless module and the display module; the wireless module sends response data X acquired by the sensor array through collecting refrigerator food volatile gas to the upper computer, receives an analysis result of the upper computer and feeds the analysis result back to the control module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111312254.3A CN114113471A (en) | 2021-11-08 | 2021-11-08 | Method and system for detecting food freshness of artificial nose refrigerator based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111312254.3A CN114113471A (en) | 2021-11-08 | 2021-11-08 | Method and system for detecting food freshness of artificial nose refrigerator based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114113471A true CN114113471A (en) | 2022-03-01 |
Family
ID=80381114
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111312254.3A Pending CN114113471A (en) | 2021-11-08 | 2021-11-08 | Method and system for detecting food freshness of artificial nose refrigerator based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114113471A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114739109A (en) * | 2022-04-27 | 2022-07-12 | 长虹美菱股份有限公司 | Control method of fresh-keeping refrigerator |
CN114970675A (en) * | 2022-04-22 | 2022-08-30 | 合肥工业大学 | Artificial nose refrigerator food freshness detection system and method based on feature selection |
CN117589951A (en) * | 2023-12-08 | 2024-02-23 | 山东工商学院 | Fresh food freshness detection method |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104330441A (en) * | 2014-09-30 | 2015-02-04 | 山东国家农产品现代物流工程技术研究中心 | Method and system for determining fish meat quality change |
CN104792826A (en) * | 2015-03-23 | 2015-07-22 | 中国农业大学 | System and method for detecting milk freshness based on electronic nose |
CN106568907A (en) * | 2016-11-07 | 2017-04-19 | 常熟理工学院 | Chinese mitten crab freshness damage-free detection method based on semi-supervised identification projection |
CN106996676A (en) * | 2016-01-25 | 2017-08-01 | 希姆通信息技术(上海)有限公司 | Refrigerator management system |
US20170363348A1 (en) * | 2016-02-20 | 2017-12-21 | Philip Bogrash | System and method of refrigerator content tracking |
WO2018077121A1 (en) * | 2016-10-24 | 2018-05-03 | 合肥美的智能科技有限公司 | Method for recognizing target object in image, method for recognizing food article in refrigerator and system |
CN108062570A (en) * | 2017-12-25 | 2018-05-22 | 重庆大学 | A kind of pattern recognition system for screening lung cancer |
CN109959765A (en) * | 2019-04-01 | 2019-07-02 | 中国农业大学 | Salmon freshness detection system and method |
CN110927217A (en) * | 2019-11-22 | 2020-03-27 | 苏州慧闻纳米科技有限公司 | Food freshness identification method based on electronic nose system and electronic nose system |
CN111595907A (en) * | 2020-06-02 | 2020-08-28 | 山东农业大学 | Method for identifying organophosphorus pesticide in tea tree leaves and diagnosing content of organophosphorus pesticide in tea tree leaves based on electronic nose technology |
CN111855757A (en) * | 2020-07-21 | 2020-10-30 | 广西壮族自治区亚热带作物研究所(广西亚热带农产品加工研究所) | Electronic nose-based Liupao tea old fragrance identification method |
CN112903919A (en) * | 2021-01-25 | 2021-06-04 | 上海应用技术大学 | Sea crab safety detection and identification method and system |
-
2021
- 2021-11-08 CN CN202111312254.3A patent/CN114113471A/en active Pending
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104330441A (en) * | 2014-09-30 | 2015-02-04 | 山东国家农产品现代物流工程技术研究中心 | Method and system for determining fish meat quality change |
CN104792826A (en) * | 2015-03-23 | 2015-07-22 | 中国农业大学 | System and method for detecting milk freshness based on electronic nose |
CN106996676A (en) * | 2016-01-25 | 2017-08-01 | 希姆通信息技术(上海)有限公司 | Refrigerator management system |
US20170363348A1 (en) * | 2016-02-20 | 2017-12-21 | Philip Bogrash | System and method of refrigerator content tracking |
WO2018077121A1 (en) * | 2016-10-24 | 2018-05-03 | 合肥美的智能科技有限公司 | Method for recognizing target object in image, method for recognizing food article in refrigerator and system |
CN106568907A (en) * | 2016-11-07 | 2017-04-19 | 常熟理工学院 | Chinese mitten crab freshness damage-free detection method based on semi-supervised identification projection |
CN108062570A (en) * | 2017-12-25 | 2018-05-22 | 重庆大学 | A kind of pattern recognition system for screening lung cancer |
CN109959765A (en) * | 2019-04-01 | 2019-07-02 | 中国农业大学 | Salmon freshness detection system and method |
CN110927217A (en) * | 2019-11-22 | 2020-03-27 | 苏州慧闻纳米科技有限公司 | Food freshness identification method based on electronic nose system and electronic nose system |
CN111595907A (en) * | 2020-06-02 | 2020-08-28 | 山东农业大学 | Method for identifying organophosphorus pesticide in tea tree leaves and diagnosing content of organophosphorus pesticide in tea tree leaves based on electronic nose technology |
CN111855757A (en) * | 2020-07-21 | 2020-10-30 | 广西壮族自治区亚热带作物研究所(广西亚热带农产品加工研究所) | Electronic nose-based Liupao tea old fragrance identification method |
CN112903919A (en) * | 2021-01-25 | 2021-06-04 | 上海应用技术大学 | Sea crab safety detection and identification method and system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114970675A (en) * | 2022-04-22 | 2022-08-30 | 合肥工业大学 | Artificial nose refrigerator food freshness detection system and method based on feature selection |
CN114970675B (en) * | 2022-04-22 | 2024-03-29 | 合肥工业大学 | Method for detecting food freshness of artificial nose refrigerator based on feature selection |
CN114739109A (en) * | 2022-04-27 | 2022-07-12 | 长虹美菱股份有限公司 | Control method of fresh-keeping refrigerator |
CN117589951A (en) * | 2023-12-08 | 2024-02-23 | 山东工商学院 | Fresh food freshness detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114113471A (en) | Method and system for detecting food freshness of artificial nose refrigerator based on machine learning | |
US11467570B2 (en) | Anomalous sound detection apparatus, anomaly model learning apparatus, anomaly detection apparatus, anomalous sound detection method, anomalous sound generation apparatus, anomalous data generation apparatus, anomalous sound generation method and program | |
CN116625438B (en) | Gas pipe network safety on-line monitoring system and method thereof | |
US11144576B2 (en) | Target class feature model | |
Zhang et al. | Fault detection and diagnosis of chemical process using enhanced KECA | |
CN116559598B (en) | Smart distribution network fault positioning method and system | |
CN109623489B (en) | Improved machine tool health state evaluation method and numerical control machine tool | |
CN113866455A (en) | Bridge acceleration monitoring data anomaly detection method, system and device based on deep learning | |
CN114118219A (en) | Data-driven real-time abnormal detection method for health state of long-term power-on equipment | |
CN114970675A (en) | Artificial nose refrigerator food freshness detection system and method based on feature selection | |
CN110956331A (en) | Method, system and device for predicting operation state of digital factory | |
CN117056678B (en) | Machine pump equipment operation fault diagnosis method and device based on small sample | |
CN117782198A (en) | Highway electromechanical equipment operation monitoring method and system based on cloud edge architecture | |
CN116858822A (en) | Quantitative analysis method for sulfadiazine in water based on machine learning and Raman spectrum | |
CN114580472B (en) | Large-scale equipment fault prediction method with repeated cause and effect and attention in industrial internet | |
CN115994327A (en) | Equipment fault diagnosis method and device based on edge calculation | |
CN114936600A (en) | Document abnormity monitoring method, device, equipment and storage medium | |
CN112347162A (en) | Multivariate time sequence data rule mining method based on online learning | |
CN116226767B (en) | Automatic diagnosis method for experimental data of power system | |
CN116740900B (en) | SVM-based power construction early warning method and system | |
Guo et al. | Detection and analysis of wheat storage year based on electronic tongue and DWT-IPSO-LSSVM algorithm | |
CN112801173B (en) | Lettuce near infrared spectrum classification method based on QR fuzzy discriminant analysis | |
CN112070761B (en) | Prawn freshness nondestructive testing method based on deep learning | |
CN112378942B (en) | White spirit grade classification and identification method based on nuclear magnetic resonance fingerprint | |
CN112541554B (en) | Multi-mode process monitoring method and system based on time constraint and nuclear sparse representation |
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 |