Sensors · 2013 · 107 citations · 30 references
ComputingMarine BiologyMarine SciencesMaritime ArchaeologySurveyingUnderwater Archaeology3D Imaging3D ReconstructionStereo VisionComparative Data AnalysisReal-world ApplicationStereo ImagingUnderwater 3D Reconstruction
Underwater archaeology and marine biology require close‑range 3D data of in‑situ objects, and 3D imaging techniques are widely used for such acquisitions. This study preliminarily compares active and passive 3D imaging methods under turbidity to evaluate their ability to acquire 3D points. The authors compared the two techniques in a laboratory tank with clay‑induced turbidity using a common stereo setup and assessed accuracy and point‑density performance. The comparison provides a reference for future evaluations of 3D imaging techniques and algorithms.
In some application fields, such as underwater archaeology or marine biology, there is the need to collect three-dimensional, close-range data from objects that cannot be removed from their site. In particular, 3D imaging techniques are widely employed for close-range acquisitions in underwater environment. In this work we have compared in water two 3D imaging techniques based on active and passive approaches, respectively, and whole-field acquisition. The comparison is performed under poor visibility conditions, produced in the laboratory by suspending different quantities of clay in a water tank. For a fair comparison, a stereo configuration has been adopted for both the techniques, using the same setup, working distance, calibration, and objects. At the moment, the proposed setup is not suitable for real world applications, but it allowed us to conduct a preliminary analysis on the performances of the two techniques and to understand their capability to acquire 3D points in presence of turbidity. The performances have been evaluated in terms of accuracy and density of the acquired 3D points. Our results can be used as a reference for further comparisons in the analysis of other 3D techniques and algorithms.
30
Accurate, Dense, and Robust Multi-View Stereopsis
Yasutaka Furukawa, Jean Ponce · 2007 · 697 citations
Yoav Y. Schechner, N. Karpel · 2004 · 334 citations