Title :
Feature extraction and selection of neural network
Author :
Chengdong, Wu ; Feng, Gao ; Shaohua, Ma
Author_Institution :
Shenyang Archit. & Civil Eng. Inst., Shenyang, China
Abstract :
This paper presents the feature extraction and feature selection methods for the wood veneer inspection application using neural network classifiers. The paper emphatically describes the statistical method of feature extraction and feature selection. The intra-class variation, inter-class variation and feature correlation are introduced to measure the discriminatory power of the wood veneer image
Keywords :
automatic optical inspection; correlation methods; feature extraction; neural nets; pattern classification; statistical analysis; wood processing; correlation method; feature extraction; feature selection; neural network; pattern classification; statistical method; wood veneer inspection; Feature extraction; Finite impulse response filter; Neural networks; Power measurement; Statistical analysis;
Conference_Titel :
Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress on
Conference_Location :
Hefei
Print_ISBN :
0-7803-5995-X
DOI :
10.1109/WCICA.2000.863410
