Features of translucent materials and products defects detection with support of optical system

Abstract

The translucent materials and products defects detection by optical methods are published in the article. The methods are described on quality assessment of sheet glass. There is an application approach based on complex feature. The recognition algorithm is based on three main feature types of defect: brightness, color (primary features) and geometrical or features of shape (secondary). In the document main (algorithmic) and technical sides of creation of machine vision system are described.