Detection of Defects in Glass: Using Image Processing

Glass defects which result into poor quality are a major reason of embarrassment for manufacturers. It is an extremely tedious process to manually inspect very large size glasses. The manual inspection process is slow, time-consuming and prone to human error. In this research work, the analyses and methodology employed to detect the defects in the glass sheets seek to address this need, using image processing technique because of its higher precision and speed to overcome many of these disadvantages and offer manufacturers an opportunity to significantly improve quality and reduce costs. The implementation of defect detection methodology has two main phases, namely the visibility test and segmentation phase. The visibility test enables the selection of an appropriate color space followed by the region-based active contour segmentation technique for final detection of the defect. The results show that the technique when applied to glass industry enables detection of any kind of major defects like surface defects, foreign materials, etc that can be present in the glass sheet providing good quality inspection with reasonable accuracy.