A Rectification Algorithm for Distorted Images from Cone Surface

This paper presents a novel approach on how to rectify the distorted images of the document on extensible surface. The surface of the document is modeled by a cone surface. By the geometry of camera image formation, differential geometry and projection geometry, the transformation equations to map the points on the surface in the 3D scene to the points on the image plane are achieved, and then the distorted images are rectified. Experiments are done by analyzing photographic methodology of images on cone surface in search of rectification algorithm with MATLAB program. The proposed method needs no auxiliary device. Experimental results are presented to demonstrate the feasibility and the application of the method.