基于特征的hausdorff距离图像配准

针对图像配准问题,提出了基于Harris及SIFT(Scale-invariant feature transform)特征的Hausdorff距离方法来实现图像配准。首先利用harris角点检测和SIFT特征提取参考图像和待配准图像的角点,通过两种方法获得的角点在融合之后获得更大的角点搜索范围,再利用相似一致性匹配原则剔除错误角点,进而通过改进的Hausdorff距离算法完成图像的配准操作。结果证明,改进算法比传统Hausdorff距离算法运行时间更短,算法时间降低约45%,具有较强的抗噪声能力和旋转鲁棒性,提高了图像配准的效率和精确性。

The Hausdorff Distance Image Regisitration Based on Features

Aiming at the problem of image registration. This paper proposes Harris and SIFT（ Scale-invariant feature transform ）features to the Hausdorff distance image regisitration based on the achieve the image registration. The angular points of reference image and the registered image are extracted by using Harris corner detection and SIFT feature extraction. Searching range of angular point is extended after fusing the angular points obtained through the two methods, then the error points are eliminated by using similarity matching principle. It can be used to realize image registration according to the improved Hausdorff distance algorithm. The experimental results demonstrate that the running time of improved algorithm decreases 45% when compared with traditional Hausdorff distance algorithm, and has stronger anti-noise ability and rotating robustness, it improves the efficiency and accuracy of registration.