Aerial Image Mosaic Algorithms Research Based on Global Mosaic
Longxian Chu · Jisuanji fangzhen · 2012
Research mosaic problem of aerial images to improve the accuracy of the image mosaic.When the overlapping part proportion in aerial images is not large,the traditional local image mosaicing algorithm can not implement the image mosaic effectively with smaller overlap areas in the aerial images,resulting in the low accuracy of aerial image mosaicing.This paper proposed an algorithm based on global vision.Through using the SIFT algorithm to extract SIFT feature points in the image,according to the principle of the minimum position error,the SIFT feature points matching of two images was finished.Then using the whole optimized method to optimize the finished mosaic image.Experiments show that the method can use the information of the whole image and accurately realize image mosaic.