Periodic Pattern Inspection Using Convolution Masks

A two-dimensional (2-D) convolution mask is proposed for detecting the location of irregularities and defects in a periodic two-dimensional signal or image. In this approach, defects on a 2-D test image is automatically filtered out by a 2-D convolution mask implementing the spatial filter. This approach is reliable and computationally efficient. Besides, unlike the traditional die-to-die approach, it can avoid all troubles in a database image, a scaling or alignment procedure.

• Type: Conference
• Year: 2007
• Where: MVA
• Authors: Y. S. Weng, Ming-Hwei Perng
• Keywords: 2-D Convolution Mask, Computer Vision, Convolution Mask, MVA 2007, Periodic Two-dimensional Signal