Scene matching based on using edge features

Geometrical and radiometrical corrections are important for scene matching applications. We suppose the application that there are no geometrical errors. In this case, Normalized Cross-Correlation (NCC) is commonly used method for scene matching. The problem of matching a pattern image (mask) to an image in this case, need to correction of radiometrical errors as illumination variations, noise, and blurring problems. In this paper we show that correlation between edge features of a mask and an image instead of those original versions, improves correlation value. Using edge detection methods, first we extract edges of both mask and image, and then match those using NCC and root mean square error (RMSE) methods. Simulation results confirm that according to using NCC and RMSE simultaneously, and according to using edge features for correlation, not only this method is a fast and real time method, but also it improves correlation value. Furthermore, it is a noise, blurring and specially illumination variations robustness method.