Research Of Panorama Image Mosaic Algorithm Based On SIFT Feature

Image stitching is one of the most important research areas among computer vision and image processing. It's a task used to match and register a group of images having spatial overlap into a reference surface, and then use blending method to form a seamless panoramic image, which contains the image sequence information and resolves the problem of high resolution and wide viewing angles. Because the image mosaics technology has solved the contradictory problems between the field of vision and the resolution,it has many applications, such as remote sensing image processing, medical image processing, digital video compression, motion analysis, virtual reality and so on.First, the present research status, application domain and the broad application prospect of mosaics are introduced in this paper. Later, the basis steps and key techniques of image mosaic are discussed.The common image matching and image blending methods are introduced,and their performance as well as the existing problems is analyzd.Based on the former research, this paper describes an automatic panorama generation method using SIFT. Firstly, it extracts and purifies SIFT feature from each image. In addition to finishing the Image stitching of several disorder images even including noise images, it supplies a probabilistic model to verify the image matches, and meanwhile process the noise images. Later,this paper linkes all the images adopting cylinder projection model.Lastly, the paper uses weighted average method to realize image blending in order to have a more satisfied panorama image.In this paper,an improved SIFT feature-based approach is proposed whose goal is to get longer match time and overcome the setback of wrong match points in the traditional SIFT feature-based algorithm.In the part of SIFT feature extraction,it chooses three sample groups while constructing Gaussian pyramid.In the part of SIFT feature matching,in order to improve search efficiency it uses BBF search algorithm based on Kd-Tree search algorithm.In the part of SIFT feature purification,this paper uses modified RANSAC algorithm to eliminate redundancies and wrong match parts. Experimental results indicate that this algorithm is as robust as SIFT,meanwhile, the efficiency and precision are improved.Furthermore, this method reduces image collection condition and enhances adaptability in image mosaic.