DocumentCode :
3184736
Title :
Control chart pattern classification using fourier descriptors and neural networks
Author :
Phokharatkul, Pisit ; Phaiboon, Supachai
Author_Institution :
Dept. of Comput. Eng., Mahidol Univ., Nakornpathom, Thailand
fYear :
2011
fDate :
8-10 Aug. 2011
Firstpage :
4587
Lastpage :
4590
Abstract :
This paper presents the method of Fourier descriptors and neural networks developed for control chart pattern analysis. The pattern analysis is important to achieve appropriate control and to produce high quality products. This paper also investigates the use of features extracted from Fourier descriptors as the Fourier coefficient components. The Fourier coefficients used to train the neural networks for classifying patterns. Thus, the networks were able to identify the classes. This research concluded the extracted features to improve the performance of the number of Fourier coefficients for neural network training. Experimental results and comparisons based on simulated and unknown data show that the proposed approach performs better than the symbol-sequence histogram with neural network approach.
Keywords :
Fourier transforms; control charts; control engineering computing; neural nets; pattern classification; Fourier coefficient component; Fourier descriptor; control chart pattern classification; neural network; Artificial neural networks; Control charts; Feature extraction; Pattern recognition; Process control; Training; Fourier descriptors; control charts; neural networks; pattern classification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC), 2011 2nd International Conference on
Conference_Location :
Deng Leng
Print_ISBN :
978-1-4577-0535-9
Type :
conf
DOI :
10.1109/AIMSEC.2011.6011169
Filename :
6011169
Link To Document :
بازگشت