Image stitching technology based on SIFT, FLANN, and RPOSAC algorithms

For the traditional SIFT algorithm extracts a large number of feature points, it takes a lot of time for each matching in the image stitching process, and the matching effect is not good when there is a scale change or rotation change in the image. An image stitching method based on SIFT, FLANN and PROSAC algorithms is proposed. Firstly, the collected images are preprocessed, and the SIFT algorithm is used to extract the features. The FLANN fast nearest neighbor search package is used to match the nearest neighbor feature points, and the progressive consistent sampling ( PROSAC ) algorithm is used to eliminate the mismatched feature points. Finally, the gradual-in and gradual-out method is used for image fusion to obtain the mosaic image. Experiments show that this method realizes the stitching of multiple images and achieves good results.