This study proposes an online seafloor obstacle detection method using sonar scanlines for submarine cable construction application. It is necessary to find out the obstacles, such as natural or linear artificial reefs and out of service cables or pipelines on the seabed along the proposed corridors of cable route for future route design jobs. And sonar system is almost the popular tool for this kind of survey. The sidescan sonar (SSS) produces an image representation by the backscattering and reflection of signals to show the seabed morphology and features. A seafloor feature searching and detecting method in the article using sonar scanlines instead of sonar images is developed to reduce the computation and speed up the processing. There are three steps of the procedure. Firstly, a pre-processing step is applied to filter the data and to reduce computation time. Then, a detector is designed to detect the obstacle by a scanline and represent obstacles with different classification. Finally, every individual scanlines are located and corrected on an obstacle map for further application. An experimental result conducted on a combined seafloor with muddy, sandy and rocky seabed was demonstrated to show every individual obstacle locations and its track on the chart automatically. The proposed detecting method implement online because of it provide efficient than traditional image processing method. The map of obstacles on the proposed corridors is useful for future route design, clearance and pre-lay grapnel run jobs. The online detection not only evaluates the efficient of route survey but also saving the funding of submarine construction.
[1]
J. M. Bell,et al.
Simulation and analysis of synthetic sidescan sonar images
,
1997
.
[2]
J. Dunlop,et al.
Statistical modelling of sidescan sonar images
,
1997,
Oceans '97. MTS/IEEE Conference Proceedings.
[3]
John Perry Fish,et al.
Sound underwater images : 日本語版 : サイドスキャンソーナーデータの作成と解析のガイドブック
,
2003
.
[4]
Larry H. Matthies,et al.
Integration of sonar and stereo range data using a grid-based representation
,
1988,
Proceedings. 1988 IEEE International Conference on Robotics and Automation.
[5]
Lawrence R. Rabiner,et al.
A pattern recognition approach to voiced-unvoiced-silence classification with applications to speech recognition
,
1976
.