Abstract
We propose an algorithm for object recognition based on clustering of vectors in the space of coefficients of affine transformations obtained as a result of the formation of hypotheses about the correspondence of segments of contours of a reference image and an input image approximated by linear segments. The results of numerical studies using a collection of images from New York University show that the proposed algorithm has a higher efficiency than an algorithm based on invariant moments or an algorithm for the invariant-to-scale comparison of singular points.
© 2017 Optical Society of America
PDF Article
More Like This
Cited By
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Login to access Optica Member Subscription