Optimal derivative filters with well-distributed based image mosaic algorithm

An automatic image mosaic method based on the optimal derivative filters with well-distributed was proposed. Firstly interest points in an image are detected by Harris operator with optimal derivative filters, delaunay triangulation is used to reject the spurious feature points pair and estimate the parameters of geometric transform model. After the geometric transform of input images, the two images were fused and the image stitching was finished. The experimental results show that the proposed algorithm produces an improvement in mosaic accuracy, efficiency and robustness.