Satellite Cloud Image Registration by Combining Curvature Shape Representation with Particle Swarm Optimization

A new feature point extraction algorithm which is used to match the satellite cloud image feature point is proposed. The new algorithm is proposed by combining corner detection with curvature scale space. The new algorithm can accurately extract the satellite cloud image corner points in different positions and directions. In order to accurately match the corner points of two source images, an overall restricted condition, which combines angle difference, gray level difference, relative distance and normalized correlation coefficient of the two matched corner points, is used to improve the matching accuracy. Finally, particle swarm optimization algorithm is used to obtain the optimal registration parameters. The optimal registration parameters are used to accurately match the two source images. The experimental results show that the proposed algorithm can accurately match the satellite cloud images and better than traditional image registration methods.