Study On Design And Implementation Of Panarama Stiching System

Panoramic image technology is a kind of virtual reality technology that can expand the view scope of people's eyes. The generation of panorama image involves technology of image mosaic and image fusion. Image mosaic stitches many local images that come from the same scene into a smooth mosaic image with wide view angle. Image fusion processes redundant data between images so as to improve the reliability of the image, and deal with the complementarity between images in order to improve image clarity.Image mosaic is an important branch in image processing field; for all aspects of image mosaic, this paper first optimizes the feature points so as to remove high redundant feature points, and improve the detection speed and quality of features; then analyses relationship between image features and image registration in order to overcome shortcomings of the current image mosaic algorithm, and at last kernel regression is used in image mosaic, thus makes an improvement to the existing algorithm.This paper analyses image mosaic based on corner features, researches technology of image mosaic and image fusion, and makes improvements based on lacks of different algorithms. After considering the speed and accuracy of image mosaic, this paper presents an image mosaic algorithm based on kernel regression, and improves the speed, accuracy and robustness. Experiment results show that the proposed method can fulfill the image mosaic effectively.