SURF and RANSAC: A Conglomerative Approach to Object Recognition

In this paper, an object recognition system [7] has been developed that uses SURF(Speeded-Up Robust Features) and RANSAC(Random Sample Consensus) algorithms to identify a series of real-life objects in a given scene using their 2-D images. SURF algorithm has been used for feature detection, extraction and matching. The features are invariant to image scaling, translation, and rotation, and partially invariant to illumination changes and affine or 3D projection[2]. RANSAC algorithm has been used to filter out the results obtained by the SURF algorithm and remove the outliers. Ten different objects have been successfully recognized using this system