This work describes a new method for correcting underwater images acquired under natural illumination condition and targeted to photogrammetry for 3D object reconstruction. In particular, a comparison between the new approach and a previous one from literature is presented. The new approach demonstrated to be a feasible and robust solution to improve quality, colour and contrast in underwater imaging by means of a three-step procedure. The focus of the paper is on 3D reconstruction of small-sized corals in shallow water. The colour of pre-processed images, with a visible green-blue colour cast, are enhanced and the higher images quality eases the whole object reconstruction acquired in a poor visible scene. The higher performance of the proposed method is experimentally demonstrated.
Underwater image pre-processing for photogrammetric application / Napolitano, R.; Chiariotti, P.; Tomasini, E. P.. - (2020), pp. 287-291. (Intervento presentato al convegno 2019 IMEKO TC19 International Workshop on Metrology for the Sea: Learning to Measure Sea Health Parameters, MetroSea 2019 tenutosi a ita nel 2019).
Underwater image pre-processing for photogrammetric application
Napolitano R.
;Chiariotti P.;Tomasini E. P.
2020-01-01
Abstract
This work describes a new method for correcting underwater images acquired under natural illumination condition and targeted to photogrammetry for 3D object reconstruction. In particular, a comparison between the new approach and a previous one from literature is presented. The new approach demonstrated to be a feasible and robust solution to improve quality, colour and contrast in underwater imaging by means of a three-step procedure. The focus of the paper is on 3D reconstruction of small-sized corals in shallow water. The colour of pre-processed images, with a visible green-blue colour cast, are enhanced and the higher images quality eases the whole object reconstruction acquired in a poor visible scene. The higher performance of the proposed method is experimentally demonstrated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.