提出了一种新的结合特征点与极谐变换(Polar Harmonic Transform,PHT)的图像拼接算法。利用Harris角检测器提取图像中的特征点,计算特征点圆形邻域的PHT特征矢量,并通过计算欧氏距离作为特征点匹配的依据提取出初始特征点对。根据几何变换模型剔除伪匹配对,利用正确映射模型计算出变换参数,采用加权平均法得到图像的拼接结果。实验结果表明了算法的有效性。
A new image mosaic method is proposed by combining feature point and Polar Harmonic Transform(PHT). The feature points are extracted from the image using Harris detector, and the PHT feature vectors are computed over the circular neigh-boring area around the feature points. The initial feature point pairs are obtained by estimating the Euclidean distance between the PHT feature vectors. The spurious feature point pairs are removed according to the geometric transform model, and the transform parameters are computed through the correct mapping mode. Image mosaic is achieved using the weighed average method. Experimental results show the efficiencies of the proposed scheme.