An Analytical Study of SIFT and SURF in Image Registration
Vivek Gupta, Kanchan Cecil · 2014
This paper present point matching using Speeded-Up Robust Features (SURF) and scale invariant feature transform (SIFT) and comparison between them and removing outliers using Restricted Spatial Order Constrains (RSOC) in image registration. Surf outperforms previously proposed schemes with respect to repeatability, distinctiveness, and robustness Computed and compared much faster. This is achieved by relying on integral images for image Convolution; by building on the strength of the leading existing detectors and descriptors. RSOC is proposed to remove outliers for registering image with monotonous back ground, simple patterns, low overlapping areas, and large affine transform. Based on adjacent spatial order, an affine invariant descriptor is defined in RSOC. In order to eliminate dubious matches, a filtering strategy is designed. The strategy integrates two way spatial order constraints and two decision criteria restrictions, i.e. the stability and accuracy of transformation error.