Foodborne pathogens are the most common food safety hazards, and the traditional detection methods of foodborne pathogens are cumbersome, long waiting time and slow efficiency. This paper studies two common foodborne pathogenic bacteria, Brucella and Escherichia coli. LightGBM algorithm combined with Principal Component Analysis (PCA) was used to analyze Raman spectrum sample data to solve the problem of classification and detection of foodborne pathogens. The results show that LightGBM algorithm has excellent detection rate. Compared with traditional machine learning algorithm models such as Decision Tree, Random Forest and XGBoost, LightGBM algorithm has the following advantages of low memory consumption and model training speed fastly, and the model accuracy rate reaches 91.23%.
The Raman spectral data feature is generally the Raman wavelength of the sample, and there is a correlation between the feature attributes. Too many features will lead to weak generalization ability of the model, so a Recursive Feature Elimination (RFE) dimensionality reduction method combined with BP neural network is proposed to classify the Raman spectrum of the COVID-19. Firstly, the collected serum Raman spectral data of the population were processed, the maximum and minimum standard scaling method (Min-Max), the Savitzky-Golay smoothing filter method, and then the recursive feature elimination (RFE-RF) based on the random forest base model and two different dimensionality reduction methods of PCA reduce the dimensionality of Raman spectral data and classify them through the BP neural network algorithm model. The experimental results show that the RFE-RF dimensionality reduction method can improve the accuracy of the classification algorithm, providing a new idea for the detection of the COVID-19, with high accuracy, and the classification accuracy of the model is 92.47%
The accurate classification of foodborne pathogenic bacteria is an important measure to solve the food safety problem in China. Compared with the traditional spectral classification method of foodborne pathogenic bacteria, Raman spectral classification has the characteristics of high flexibility, wide range and high efficiency. This paper, by using common foodborne pathogenic bacteria as the research object, we collected article 11 kinds of pathogenic bacteria of 132 spectra data. And after preprocessing the obtained data, principal component analysis (PCA) and linear discriminant analysis (LDA) were used to extract the main feature information of the spectral data. Then, based on the continuity characteristics of the spectral data, a Raman spectral classification model of recurrent neural network (RNN) was proposed. For each RNN neuron, the model always shares its parameters and has the characteristic of memory, so it has a great advantage in learning sequential information. The experimental results show that the classification accuracy of the model is as high as 96%, higher than the traditional machine learning classification methods such as decision tree and logistic regression.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.