DocumentCode :
2228363
Title :
Testing a new algorithm for automatic processing of bathymetric data
Author :
Canepa, Gaetano ; Bergem, Oddbjørn
Author_Institution :
SACLANT Undersea Res. Centre, La Spezia, Italy
Volume :
1
fYear :
1998
fDate :
28 Sep-1 Oct 1998
Firstpage :
313
Abstract :
Accurate and fast automatic processing of multibeam data is necessary to obtain a bathymetric map of a sea area with minimal operator intervention. The authors present an automated processing technique which effectively eliminates bad data (outliers) and generates a map of the seafloor in one step. The authors emphasise the testing of the algorithm on real data. The main characteristics of the algorithm are: production of a triangulated map of uniform accuracy irrespective of seafloor features, a map resolution which depends on the local data noise amplitude, automatic elimination of outliers, intrinsic modularity of the program that can be easily split in a multi-processor environment, and low computing cost even on large data files. The algorithm parameters are defined on the basis of their effect on the results. The resulting map can be finely tuned according to the parameters. Some of the features of the algorithm are shown in relation to the characteristics of different data sets. The results of the tests are shown with respect to accuracy and run-time requirements
Keywords :
bathymetry; geophysical signal processing; geophysical techniques; oceanographic techniques; sonar imaging; algorithm; automatic; bad data elimination; bathymetry; geophysical measurement technique; multibeam data; ocean; outlier; seafloor topography; sonar imaging; triangulated map; Automatic testing; Costs; Iterative algorithms; Noise level; Oceanographic techniques; Production; Robustness; Sea floor; Surface fitting; Working environment noise;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
OCEANS '98 Conference Proceedings
Conference_Location :
Nice
Print_ISBN :
0-7803-5045-6
Type :
conf
DOI :
10.1109/OCEANS.1998.725758
Filename :
725758
Link To Document :
بازگشت