Image Auto-registration on Harris-Laplace Features

In order to overcome the difficulty of image automatic registration with different rotations and resolutions, this paper proposes a new method for automatic image registration. In the method, Harris-Laplace detector is applied to extract image features, the scale space projection based on scale-invariance of Harris-Laplace is proposed, then, a feature space on the projection information in the direction of feature main orientation is constructed, which makes feature space rotation-invariant and resolution-invariant. In addition, the feature matching adopts feature space k-d tree distance matching to combine with RANSAC, which implements feature matching effectively and satisfactorily. The experiments and comparison analyses show that the method in this study can accomplish automatic image registration accurately in the condition of different resolutions and rotations.