Defect classification of TFT-LCD with bag of visual words approach

Author

Wei Huang ; Hongtao Lu

Author_Institution

Shanghai Jiao Tong Univ., Shanghai, China

fYear

2012

fDate

27-29 Sept. 2012

Firstpage

167

Lastpage

170

Abstract

A defect classification algorithm with bag of visual words approach for thin film transistor liquid crystal display (TFT-LCD) manufacturing is proposed in this paper. Color and SIFT features are introduced to describe defect region. Visual words vocabularies are learnt separated for each features. The two features are separately coded in bag of visual words and combined by multiple chi-square kernel SVM. Classifier performances with different parameters are compared in experiments. Finally a good enough classifier for identifying 5 classes of defect is achieved.

Keywords

automatic optical inspection; feature extraction; flaw detection; image classification; image colour analysis; liquid crystal displays; production engineering computing; support vector machines; thin film transistors; SIFT features; TFT-LCD defect classification; TFT-LCD manufacturing; bag of visual words approach; classifier performance; color features; defect classification algorithm; defect region classification; liquid crystal display manufacturing; multiple chi-square kernel SVM; thin film transistor; visual words vocabularies; Image color analysis; Inspection; Kernel; Support vector machines; Thin film transistors; Visualization; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Information and Automation for Sustainability (ICIAfS), 2012 IEEE 6th International Conference on

Conference_Location

Beijing

Print_ISBN

978-1-4673-1976-8

Type

conf

DOI

10.1109/ICIAFS.2012.6419899

Filename

6419899