Description and Recognition of 3D Objects Using MEGI Model and Extended Spherical Correlation

Abstract

Description and recognition of object is one of the central consideration of research of the computer vision. It is the key issue that how to represent 3D objects on machine for recognition of them. Researchers of computer vision are used to employ EGI model, but it's not able to express concave objects. In this paper, MEGI model and coefficient of extended spherical correlation have been proposed. MEGI model is an extended EGI model to represent concave objects. Extended spherical correlation is the measure to recognize object using MEGI model. It has been demonstrated that this model is able to recognize 3D objects which is included concave one and to distinguish objects using a part of MEGI which makes from range data.