DSpace Arşivi :: by Yazar "Al-Shalfan, Khaled" değerine göre listeleniyor

Listeleniyor 1 - 1 / 1

Corner detection based on uncalibrated images
(Maltepe Üniversitesi, 2009) Al-Shalfan, Khaled
Precise determination of object corners in an image is very important in applications of robotics and computer vision, such as pattern recognition and 3-D reconstruction. The corners of a polygonal object plane, e.g. roof, wall, etc. in an image, can be determined by detecting image corners bounding the plane edges. An automatic system for locating image corners is likely to produce many corners do not represent corners of a polygonal object, and so despite many studies in the field of boundary recognition, the question of whether the detected corner in the image of a 3-D scene corresponds to an object point still merits further investigation. The aim of the research presented in this paper is to propose an image system capable of detecting image corners and then show the ones that correspond to polygonal object planes for a variety of types of subject. A review of the existing literature suggested that a two-camera approach based on uncalibrated images would offer the most flexible potential solution. The method is based on rectified images obtained from a pair of uncalibrated images utilizing the epipolar constraint, and is illustrated with images of several scenes captured using a digital camera.