2D Defect Detection at Subpixel level

Quality control is of almost important in any industry. During the production process, many damaged or undesirable products are also produced. These kinds of products are unfit for use and deploying into the market and thereby need to be sorted out. Today, several machine vision software and cameras are available in the market. The software available is so effective that ready-made applications for automated inspection already exist. The user needs to give only the inputs to the software. In this paper solution is given to find the accuracy of defect at micron level. The application is implemented as Dynamic Link Library (DLL), developed using Microsoft Visual C++. The key idea behind it is that each point in the template point set can be exactly represented by an affine combination of its neighboring points. In this paper resolution of image is increased using interpolation. Using 2nd derivative pixel location of defect at micron level is calculated. As we move to subpixel the computation time is increased by using parallel programming this time can be reduced.