Personal tools

Skip to content. | Skip to navigation

You are here: Home / Our People / Students / Hamed Bagheri

Hamed Bagheri

Strategies for filtering incorrect matches in seabed image mosaicing

Hamed Bagheri
Current image JPEG image — 17 KB

Ray Gosine

Engineering and Applied Science, Memorial University of Newfoundland

09/2009 - 07/2012

Marine Biodiversity, MB-08: Image Analysis and Object Recognition Tools

Population statistics have a wide range of applications in the fishery industry, oceanographic research (e.g. population studies, habitat analysis), as well as for the oil and gas industry (e.g. population monitoring for environmental impact assessment). Most Remotely Operated Vehicles (ROV) or Autonomous Underwater Vehicles (AUV) image acquisition transects produce overlap between successive or adjacent images, such that individual animals could appear in several images, which could yield inaccurate counts. In order to eliminate the possibility of counting the same animal more than once, the overlap between images must be detected. We developed a feature-bases mosaicing algorithm that uses the Scale Invariant Feature Transform (SIFT) in which feature descriptors of images are extracted and appropriate correspondences are found and matched by computing the standardized Euclidean distance between descriptor vectors. We present a new strategy for fining correct correspondences and discarding incorrect matches from the background using spatial clustering and standardized Euclidean distance for computing an adaptive threshold value used by the second-best match method. Results are provided to validate the proof of concept for our strategy. 

Document Actions