Абстрактный

Evaluation of Boundary Detection Methods Based on a Pair of Images

Khalid A. Al-Shalfan

Local discontinuities in image intensity fall into three categories: points, lines, and edges. Discontinuous in image amplitude are important primitive characteristics of an image that carry information about object borders. Detection methods of image discontinuities are principal approaches to image segmentation and identification of objects in a scene. This leads to locating intersection boundaries that represent a 3D point in the scene correspond to an object point. In this paper some of edge detect filters have been evaluated based on detecting image intersection points and then show the ones that correspond to polygonal object planes for a variety of types of subject. The proposed algorithm starts up with obtaining image corresponding object intersection points after been detected. Epipolar rectification then applied on the pair of images using the detected corresponding object intersection points. This technique is illustrated with a pair of uncalibrated images of several scenes, utilizing the epipolar constraint.