Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Assessing the performances of an object recognition technique based on correlation and edge enhancement

Not Accessible

Your library or personal account may give you access

Abstract

Sea mines are larger and much more expensive than land mines, but they may be well disseminated inside troubled waters. Thus, it is primordial to propose robust image recognition techniques which are adequate for the context of troubled waters. A recognition technique based on optical correlation has been developed for this specific purpose. A set of images of an object (assumed to represent a sea mine) was produced with a rotating camera around it. Then, these images were processed with an algorithm based on the correlation method to recognize the object. To improve the correlator performances we have proposed a correction that consists in emphasizing the contour of the object. For that purpose, the phase contrast method was used. To test the performance of this recognition technique several scenarios were considered, e.g. objects in air, mist, and placed in clear and troubled water. The latter case simulates the underwater mine situation.

© 2015 Optical Society of America

PDF Article
This paper was not presented at the conference

More Like This
Imaging polarimetry and correlation techniques for target detection

M. Dubreuil, A. Alfalou, and C. Brosseau
FW3A.30 Frontiers in Optics (FiO) 2012

Digital holography based 3D object recognition under varying light illumination using photon counting imaging

Dhirendra Kumar and Naveen K. Nishchal
DTh1A.6 Digital Holography and Three-Dimensional Imaging (DH) 2015

Digital Holography based Three-dimensional Object Recognition using Binary Differential Joint Transform Correlator

Dhirendra Kumar and Naveen K. Nishchal
T3A.61 International Conference on Fibre Optics and Photonics (Photonics) 2014

References

You do not have subscription access to this journal. Citation lists with outbound citation 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

Select as filters


Select Topics Cancel
© Copyright 2022 | Optica Publishing Group. All Rights Reserved