An image sorting algorithm based on SIFT feature matching

Abstract

The panoramic image mosaic technology means images with partial overlapping area are composited to a 360-degree circular image which describes the information of a certain scene. A hotspot of this technology is how to determine the image order accurately and fast when the number of images is large. In this paper, we present a novel image sorting algorithm based on SIFT(Scale-Invariant Features Transform). In the proposed method, matching feature is firstly characterized by calculating the number of matching points between images to find the nearest image of a certain one. After calculating the matching points one by one, an accurate picture order could be finally obtained. The results show the effectiveness of the proposed algorithm.