Object recognition based on ORB and self-adaptive kernel clustering algorithm

In this paper, we investigate the problem about object recognition using image matching. SIFT feature matching algorithm is a certain degree of stability on the angle changes and affine transformation. There has been more stable match capacity on images used in the field of image matching. ORB is another choice for it needs less storage space and matching time than SIFT which extracts too many feature points. In this paper, we adopt ORB combine with Self-adaptive kernel clustering algorithm to implement object recognition. Compared with SIFT and ORB, test images are processed in advance to find the target region, so the incorrect matching will be restrained and targets in test images can be recognized in a large degree, especially in the images contains of deformed targets.