Method, system for odor visual expression based on electronic nose technology, and storage medium

A method for odor visual expression based on electronic nose technology, comprising:

• acquiring category information of an odor to be identified based on the electronic nose technology;
• determining a demand information of the odor to be identified according to the category information;
• collecting corresponding relevant data according to the demand information so as to construct a database;
• constructing a structured knowledge map centered on odor identification according to the database;
• and
• converting the structured knowledge map into a visual node-link graph to display its own characteristics, related characteristics, subordinate keywords of the characteristics, and subordinate keywords of the related characteristics, so as to facilitate users to have more intuitive understanding of the odor,
• wherein the acquiring category information of the odor to be identified based on the electronic nose technology specifically comprises:
• collecting a number of different odor sample data in a gas collection chamber by an electronic nose system to form a covariance matrix with n rows and m columns, where n is a dimension of the collected data and m is a number of odor samples, the electronic nose system comprises a sensor array unit, a signal preprocessing unit and a pattern recognition unit, the category information refers to a category that a trained model predicts the odor belongs to in pattern recognition, and the training process refers to a process that the electronic nose system needs to be trained before it performs identification, that is, a process of using known odor categories and their characteristics for classification;
• calculating eigenvalues and eigenvectors of the covariance matrix;
• selecting eigenvectors of q dimensions with largest eigenvalues to form a projection matrix of q rows and m columns;
• training a classification model with data features obtained by multiplying each odor feature by the projection matrix, wherein the odor feature of each odor refers to sample data collected for each odor, which is a data matrix with n rows and 1 column;
• and
• collecting data of the odor to be identified so as to be input into the trained classification model to obtain the category information,
• wherein the demand information comprises odor categories and subordinate keywords of the odor categories;
• the odor categories comprises special classification, gas composition, odor descriptors, odor color, and odor volatility;
• the subordinate keywords of the special classification comprises:
• an aromatic odor, a resin odor, a mint odor, sweetness, a chemical odor, a popcorn burned odor, a lemon odor, a non-orange fruit odor, a pungent odor, and a rotten odor;
• and the gas composition comprises a first component, a second component, and a third component,
• wherein the collecting corresponding relevant data according to the demand information so as to construct the database specifically comprises:
• crawling network data;
• processing crawled network data so as to obtain target information;
• and
• saving the calculated target information to form a database, and
• wherein the processing crawled network data so as to obtain the target information specifically comprises calculating the crawled network data with a Connectionist Text Proposal Network (CTPN) text detection algorithm, a Convolutional Recurrent Network (CRNN) text recognition algorithm and a Single Shot Detector (SSD) image detection algorithm so as to get the target information.