An Improved Fully-automatic Image Mosaic Algorithm
一种改进的全景图自动拼接算法

In this paper,an improved fully-automatic image mosaic algorithm is presented.Firstly,phase correlation is used to sort the unordered image sequence and roughly compute the translation offset between adjacent images,which speeds up corner match procedure and improves matching stability as well.Secondly,RANSAC algorithm is used to eliminate outliers to ensure effectiveness of the matched corner pairs.Then,Singular Value Decomposition-Least Square(SVDLS) method and Levenberg-Marquardt optimization are used to robustly determine the 8 parameters transform model,which aims to reduce the registration error and also improve the stability when handling very small number of matched corners.At the last of the algorithm,a multi-band blending technique is used to generate the final panorama.Invalid parameters are verified by the translation offset to make Levenberg-Marquardt optimization more successful and robust.The experimental results show that the proposed image mosaic algorithm exceeds existing ones at matching speed,stability,and it has less blur or ghost effect after blending, especially when there are noise,moving objects,repeated texture and small overlaps presented in the images.