Application of Normalized Cross Correlation to Image Registration

Image correspondence and registration techniques have gained popularity in recent times due to advancement of utilization in digital media and its storage. The main problem associated with image processing is when it is applied to fields like robotic vision and machine vision. The problem is due to clutter, i.e. the same frame with different objects has to be matched. Hence there has been need for efficient techniques of Image Registration. This led to development of feature extraction techniques and template matching techniques. The normalized cross correlation technique is one of them. A classical solution for matching two image patches is to use the cross-correlation coefficient. This works well if there is a lot of structure within the patches, but not so well if the patches are close to uniform. This means that some patches are matched with more confidence than others. By estimating this uncertainty more weight can be put on the confident matches than those that are more uncertain. All the simulations have been performed using MATLAB tool.