OBJECTS RECOGNITION BY MEANS OF PROJECTIVE INVARIANTS CONSIDERING CORNER-POINTS.
Vicente,M.A., Gil,P.*, Reinoso,O., Torres,F.*
Department of Engineering, Division of Systems Engineering and Automatic.
Miguel Hernandez University, Avda. del
Ferrocarril, s/n
03202 Elche
Spain
o.reinoso@umh.es,
suni@umh.es, http://lorca.umh.es/
*Department of Physics, Systems Engineering and
Theory of the Signal .
University of Alicante, Crtra. San Vicente, s/n
03080 Alicante
Spain
pgil@disc.ua.es, medina@disc.ua.es http://www.disclab.ua.es/gava/ing
ABSTRACT
This paper presents an object recognition
technique based on projective geometry for industrial pieces that satisfy
geometric properties. First at all, we consider some methods of corner
detection which are useful for the extraction of interest points in digital
images. For object recognition by means of projective invariants, an excessive
number of points to be processed supposes a greater complexity of the algorithm
We present a method that allows to reduce the points extracted by different
corner detection techniques, based on the elimination of non-significant
points, using the estimation of the straight lines that contain those points. Secondly,
these groups of points are then used to build projective invariants which allow
us to distinguish one object from another. Experiments with different pieces
and real images in grey-scale show the validity of this approach.
Keywords: object recognition, corner
detector, projective invariants, projective geometry.