Automatic waterline registration of a ship model in waves
Tello Ruiz, M.; Geerts, S.; Delefortrie, G.; Vantorre, M. (2013). Automatic waterline registration of a ship model in waves, in: 3rd International conference on advacend model measurement technologies for the maritime industry 2013 (AMT'13): Proceedings. pp. [1-15]
In: (2013). 3rd International conference on advacend model measurement technologies for the maritime industry 2013 (AMT'13): Proceedings. [S.n.]: [s.l.]. , meer
|
Beschikbaar in | Auteurs |
|
Documenttype: Congresbijdrage
|
Abstract |
Investigation of problems concerning the instantaneous waterline of sea-going vessels requires consideration of several effects, such as squat, bow wave dynamics, wave diffraction and radiation, as well as wave induced motions. Due to the complexity of such problems, (captive) model tests can be considered as a suitable research method, provided that the instantaneous waterline can be determined. Thus, the accuracy of the estimation of the issues involved depends on the proper registration of the waterline. This is usually carried out with image processing techniques, which, if performed manually, mostly result in a tedious work if large series of model tests are involved. A typical example of such issues is the seakeeping behaviour of so-called estuary (or fluvio-maritime) vessels, i.e. inland vessels that can be utilised for short sea voyages on well-defined trajectories in favourable wave conditions. Besides the strength requirements, the seakeeping results must comply with prescribed values to avoid slamming, green water or flooding of the holds. To investigate these problems, a systematic series of captive model tests was carried out with a 1/25 scale model of an estuary container carrier in the towing tank for manoeuvring in shallow water (co-operation Flanders Hydraulics Research – Ghent University). The tests were carried out in regular waves with different combinations of wave amplitudes and periods, ship speeds and heading angles. During the tests, images of the instantaneous waterline were recorded by an experimental test setup consisting of four black light tubes, a fluorescent coating on the ship model and a computer controlled digital camera. The large amount of images has to be processed automatically, incorporating new challenges and additional issues such as the varying ship position in waves, causing variations in pixel colour and brightness. The present paper intends to describe a method for identifying the waterline profile based on image processing analysis integrating methods such as correlation analysis and image registration technique. |
|