An Algorithm of Gross Error Elimination in Image Matchingfor Large Rotation Angle Images

This paper has proposed an gross error elimination algorithm for image matching based on the imaging surface transformation. Image matching practice which embedded this algorithm in the coarse to fine the matching strategy showes that this algorithm can effective-ly control matching error and correct matching parameters. Experiments with large rotation angle images, such as low-altitude images and close-range images prove that the method can greatly reduce the gross errors in the matching results and ensure the quality and efficiency of image matching.