Image mosaic creation is one of the most developing area in recent underwater research. Automatic definition of mosaics of the seabed deals with the problem of image motion estimation in a robust and automatic way. This paper presents a fully automatic feature-based approach to combine a sequence of images into a single 2D seamless mosaic (panorama). In this context, calibrated images have been acquired by a monocular camera carried by a ROV looking at the seabed. The major contribution of this work is the implementation of a robust scheme for the definition of correct correspondences between adjacent frames. In order to restrict the computational workload during the registration activity, the paper suggests a new strategy for the evaluation of a reliable homography between adjacent frames, estimated by using few correspondences properly selected from a wide set. Experimental results validate the performances of the proposed scheme.

A fully automated approach for underwater mosaicking

INDIVERI, GIOVANNI
2006-01-01

Abstract

Image mosaic creation is one of the most developing area in recent underwater research. Automatic definition of mosaics of the seabed deals with the problem of image motion estimation in a robust and automatic way. This paper presents a fully automatic feature-based approach to combine a sequence of images into a single 2D seamless mosaic (panorama). In this context, calibrated images have been acquired by a monocular camera carried by a ROV looking at the seabed. The major contribution of this work is the implementation of a robust scheme for the definition of correct correspondences between adjacent frames. In order to restrict the computational workload during the registration activity, the paper suggests a new strategy for the evaluation of a reliable homography between adjacent frames, estimated by using few correspondences properly selected from a wide set. Experimental results validate the performances of the proposed scheme.
2006
978-142440115-4
1424401151
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1021686
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 14
  • ???jsp.display-item.citation.isi??? 0
social impact