Development of a machine learning technique for automatic analysis of seafloor image data: Case example Pogonophora coverage at mud volcanoes (bibtex)
by Andree Luedtke, Kerstin Jerosch, Otthein Herzog, Michael Schlueter
Abstract:
Digital image processing provides powerful tools for fast and precise analysis of large image data sets in marine and geoscientific applications. Facing the increasing amount of georeferenced image and video data acquired by underwater platforms such as Remotely Operated Vehicles (ROVs), means of automatic analysis of the acquired image data are required. A new and fast developing application is the combination of video imagery and mosaicking techniques for seafloor habitat mapping. In this article we introduce an approach for fully automatic detection and quantification of Pogonophora coverage in seafloor video mosaics from mud volcanoes. The automatic recognition is based on textural features extracted from the raw image data and classification using machine learning techniques. A classification accuracy of up to 98.86\% was achieved on the training data. The approach was extensively validated on a data set of more than 4000 seafloor video mosaics from the Haakon Mosby Mud Volcano.
Reference:
Andree Luedtke, Kerstin Jerosch, Otthein Herzog, Michael Schlueter, "Development of a machine learning technique for automatic analysis of seafloor image data: Case example Pogonophora coverage at mud volcanoes", In Computers & Geosciences, 2011. (accepted for publication)
Bibtex Entry:
@ARTICLE{Luedtke2011a,
  author = {Luedtke, Andree and Jerosch, Kerstin and Herzog, Otthein and Schlueter,
	Michael},
  title = {Development of a machine learning technique for automatic analysis
	of seafloor image data: Case example Pogonophora coverage at mud
	volcanoes},
  journal = {Computers {\&} Geosciences},
  year = {2011},
  note = {(accepted for publication)},
  abstract = {Digital image processing provides powerful tools for fast and precise
	analysis of large image data sets in marine and geoscientific applications.
	Facing the increasing amount of georeferenced image and video data
	acquired by underwater platforms such as Remotely Operated Vehicles
	(ROVs), means of automatic analysis of the acquired image data are
	required. A new and fast developing application is the combination
	of video imagery and mosaicking techniques for seafloor habitat mapping.
	In this article we introduce an approach for fully automatic detection
	and quantification of Pogonophora coverage in seafloor video mosaics
	from mud volcanoes. The automatic recognition is based on textural
	features extracted from the raw image data and classification using
	machine learning techniques. A classification accuracy of up to 98.86{\%}
	was achieved on the training data. The approach was extensively validated
	on a data set of more than 4000 seafloor video mosaics from the Haakon
	Mosby Mud Volcano.},
  keywords = {automatic image analysis,machine learning,supervised learning,image
	classification,Pogonophora recognition,Haakon Mosby Mud Volcano},
  owner = {pmania},
  timestamp = {2012.11.06}
}
Powered by bibtexbrowser