OpenSURF performance in windows phone 7

Abstract

OpenSURF is an algorithm to detect point correspondences between two images. OpenSURF is an integral part of computer vision and machine vision. The Algorithm aims is to find important region in images which can found under variety of image transformation. This algorithm can be used for augmented reality object recognition, or medical imaging. This research is to reach maximum performance augmented reality markerless application. Especially on Actual Mobile Guide for Tourist application. OpenSURF doesn't have library for windows phone so we have to porting from C++ to C#. The performance of OpenSURF is not much further than android library, it is 6,666 second(windows phone) versus 8.779 second(android) with 604 × 453 pixel picture.