(404f) Automatic Grading of TFT-LCD Glass Substrates Based On Machine Vision

The outer appearance of a product is often one of important quality attributes for certain products that is mainly used for display or used as an exterior part of other products [1,2]. TFT-LCD (Thin Film Transistor – Liquid Crystal Display) glass substrates can be one of the representative cases. In such cases, visual quality (i.e., visual appearance) as well as the physical or mechanical quality attributes has to be controlled or maintained.

A main difficulty in this type of problems arises from the fact that prior information or knowledge is very limited: no class labels are often available nor is prior knowledge about important aspects of visual appearance often not available. This limitedness of prior information as well as the stochastic nature of the visual appearance of the products makes it much more difficult to apply machine vision to the problems.

This work presents an industrial application of a new machine vision methodology [3] to manufacturing of TFT-LCD glass substrates. Careful observation and interpretations of the results from the methodology enable one to automatically monitor the visual quality of the products. Other issues such as such as imbalanced training data that are frequently encountered in industrial applications will be addressed as well.

Literature cited

(1) Liu, J., MacGregor, J. F., “Modeling and Optimization of Product Appearance: Application to Injection-Molded Plastic Panels,” Industrial & Engineering Chemistry Research, 44(13), pp. 4687-4696, 2005

(2) Liu, J., MacGregor, J. F., “Estimation and Monitoring of Product Aesthetics: Application to Manufacturing of Engineered Stone” Countertops,” Machine Vision and Applications, 16(6), pp. 374-383, 2006

(3) Liu, J. and Han, C., “Wavelet texture anlaysis in process industries,” Korean Journal of Chemical Engineering. Accepted, 2011