Unlock instant, AI-driven research and patent intelligence for your innovation.

Feature selection and array optimization of sensor array based on principal component analysis

A feature selection method, sensor array technology, applied to instruments, character and pattern recognition, computer components, etc., can solve problems such as poor effect

Active Publication Date: 2018-12-14
ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY
View PDF13 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since there is no essential connection between the size of the data variance and the effective classification information of the data, the dimensions with smaller data variance may also contain effective classification information, resulting in the existing method of feature selection for sensor arrays using PCA technology. Difference

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feature selection and array optimization of sensor array based on principal component analysis
  • Feature selection and array optimization of sensor array based on principal component analysis
  • Feature selection and array optimization of sensor array based on principal component analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0082] Example: There is an existing original sensor array comprising 10 sensors (the sensors are numbered 1 to 10 respectively), and it is necessary to identify fresh meat and spoiled meat through odor detection. To this end, a total of 600 data samples (including 300 fresh meat samples and 300 spoiled meat samples) were collected, and each data sample contained 10 sensor response curves. Four feature extraction methods including maximum value, peak area, maximum difference, and maximum slope are initially selected.

[0083] A: First evaluate the performance of various feature extraction methods, that is, use each feature extraction method to extract features separately and send them to the SVM classifier for the discrimination of fresh meat and spoiled meat. The best recognition results of each method are: the recognition rate of the maximum value method is 81%, the recognition rate of the peak area method is 75%, the recognition rate of the maximum difference method is 78%,...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a feature selection and array optimization of a sensor array based on principal component analysis, which comprises the following steps in turn: A, selecting initial features;B, forming an initial characteristic data set; C: calculating the importance of each dimension feature in the standardized initial feature data set; D, sorting the standardized initial feature to obtain a sorting list L; E, according to the sorted list L, selecting d corresponding characteristic data sets to be determined; F, respectively evaluating the recognition accuracy, and obtaining d recognition accuracy judgment result correspondingly; G, froming the d recognition accuracy judgment results obtained in the step F, and finding the highest accuracy judgment result, that is, the selected important features. The invention can further improve the detection performance of the sensor array, reduce the use cost of the sensor array, and greatly optimize the selection of the sensor array.

Description

technical field [0001] The invention relates to a sensor array feature selection and array optimization method, in particular to a sensor array feature selection and array optimization method based on principal component analysis. Background technique [0002] At present, when performing feature selection on sensor arrays, excellent features suitable for application scenarios can be found to improve the recognition performance and robustness of sensor array systems. Optimizing the sensor array can not only reduce the cost of the sensor array, but also further improve the performance of the sensor array. The current methods of feature selection for sensor arrays through principal component analysis (PCA) technology are to use PCA to transform the original sensor features, and perform dimensionality reduction and feature selection in the transform domain. This method only pays attention to the distribution of data variance, and believes that the data direction dimension with ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06K9/20
CPCG06V10/147G06F18/2135G06F18/214
Inventor 孙彤钱慎一张旭石永生
Owner ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY