Colorimetric classification for video-camera acquired images

M.Quadri

Università degli Studi di Milano, A.A. 1991-92, degree thesis

Thesis responsible: Dott. A.Bondi, I.B.M. Semea S.p.A.
University responsible: Prof. G. Degli Antoni, Università Statale di Milano

ABSTRACT

The objective of the project has been to study and develop a system able of classifying objects on the basis of the chromatic distribution on their surface. The application field is the one of production process control: the typical situation is the one of homogeneous products to be divided in groups based on their color. For this kind of application the system must be able of discriminating objects also if they show only small chromatic differences and of classifying them based on criteria working on non-predefined classes but on classes developed dynamically while the objects are analyzed by the acquisition system.

The development of this system has required, mainly, the study of an image analysis technique to collect colorimetric data necessary in order to evaluate color distribution on the surfaces of the objects and the development of techniques to extract from this information numerical parameters to be used in the classification using cluster analysis algorithms.


[ ESO | VLT software | G.Chiozzi's Home Page ]


© gchiozzi@eso.org

Last modified: Thu May 21 22:12:15 METDST 1998 ;