Producing Panoramas Of Long Scenes Based On Graph Cut

The multi-viewpoint panorama of long scenes is produced from a relatively sparse set of images captured with a camera that is moved along a long scene. This image can be well resolve the structural problems of long scene that the ordinary camera and wide-angle camera images can not solve. This paper describes how to construct the multi-viewpoint panorama of long scenes with the method of Graph Cut. The first Step includes the basic theory of graph cuts, how to solve the problem of minimizing energy function in the basic forms and how to Structure the Graph for Graph Cut; the second Step presents Some preparatory work before The multi-viewpoint panorama of long scenes produced. These works include the Exposure Compensation of the Source image, reconstruct the 3D point cloud of the Source image and Project the Source image to the Relative position of Picture Surface using the camera matrix of Source image. The final step is making the Several quality of multi-viewpoint panorama of long scenes expressed by the form of Energy function, and make the Energy function minimized by the Fast Approximate Energy Minimization which is Created by boykov. Then the Purpose of Selecting the most appropriate pixel value for every pixel in the Picture Surface will be Reach.In the part of Selecting the most appropriate pixel value for every pixel in the Picture Surface, the matter is consided Not only in the point of Computational Mathematics, but also in the point of arts. The Several features of effective multi-viewpoint images is inspired by the work of artist Michael Koller. So, the multi-viewpoint panorama of long scenes produced in this paper is a perfect work not only in Mathematical Theory, but also in arts.