"

Cookies ussage consent

Our site saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing our site without changing the browser settings you grant us permission to store that information on your device.

I agree, do not show this message again.

Ultrasonic C scan image evaluation system for electric contacts with model and feat ure selection

CHUNGUO FEI1,* , RUI WANG1, WEIXING CHEN1, PENGPENG WANG1, QINKUN LIU2

Affiliation

  1. College of Aeronautical Automation , Civil Aviation University of China , Tianjin, China, 300300
  2. Xiamen Reliance M&E Engineering Co. Ltd , Beijing, China, 100044

Abstract

In this p aper, in order to improve the traditional evaluation of electric contact, the novel ultrasonic C scan image evaluation system is built by combining the S upport V ector M achine (SVM), the M utative S cale C haos O ptimization (MSCO) and the G enetic A lgorithm (GA )). The SVM is used as a classifier, the MSCO is employed as a model selector to optimize the parameters of SVM and the GA is applied as a feature selector to get rid of redundant and irrelevant features for this system. In the experiment, the grayscale histograms are extracted as the feature dataset from the ultrasonic C-scan images. Then the novel system can classify the electric contacts into three categories with high efficiency and accuracy by using a certain feature samples when the 10-fold cross validation procedure is applied: qualified product, unqualified product and repairable product. Some comparisons have been made to show that the model selection and the feature selection are very important for improving the average evaluation accuracy rate.

Keywords

Ultrasonic C scan image evaluation system ; support vector machine ( SVM )); model selection ; feature selection.

Submitted at: Oct. 15, 2014
Accepted at: May 7, 2015

Citation

CHUNGUO FEI, RUI WANG, WEIXING CHEN, PENGPENG WANG, QINKUN LIU, Ultrasonic C scan image evaluation system for electric contacts with model and feat ure selection, Journal of Optoelectronics and Advanced Materials Vol. 17, Iss. 5-6, pp. 806-815 (2015)