An optimization method for electronic nose feature selection based on multi-kernel fisher discriminant analysis

Abstract

The invention discloses an electronic nose signal feature selection optimization method based on multi-kernel Fisher discriminant analysis. Firstly, a sample feature matrix is ​​obtained, parameters are initialized and basic kernel functions are constructed according to the parameters, and then a composite kernel matrix is ​​calculated based on the basic kernel matrix, and a composite kernel function is calculated. The projection of the kernel matrix in the high-level feature space, and then send the projection to the classifier for pattern recognition, determine the kernel function with the highest recognition rate, and finally calculate the projection of the new sample matrix in the feature space based on the kernel function and use it as an electronic nose The signal is used as the input of the classifier for pattern recognition. Its remarkable effect is: it overcomes the problem of poor data discrimination after the single kernel function method realizes high-dimensional projection, solves the redundancy between sensors, optimizes the sensor array and reduces the data dimensionality, and improves the accuracy of the electronic nose signal. recognition rate, thus providing useful guidance for doctors to choose appropriate treatment methods.

Description

technical field [0001] The invention relates to the technical field of electronic nose signal processing, in particular to an electronic nose feature selection optimization method based on multi-core Fisher discriminant analysis. Background technique [0002] An electronic nose is an electronic system that uses the response map of a gas sensor array to identify odors, and it can continuously and real-time monitor the odor status of a specific location within hours, days or even months. [0003] Medical electronic nose is a special electronic nose system, which can realize the diagnosis of disease or wound infection by detecting the gas exhaled by the patient or the gas in the head space of the wound. It has short response time, fast detection speed, low cost, simple and convenient operation, and has the advantages of artificial intelligence, so it has gained wide attention and application. [0004] After the electronic nose feature extraction, it needs to be used as the inp...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method for processing electronic nose signals based on multi-kernel Fisher discriminant analysis. Using this method for feature selection optimization can solve the problem of inconsistency in data discrimination after high-dimensional projection using a single kernel function method. Optimizing the problem, and addressing redundancy between sensors

Images