Seabed Image Mosaicing for Benthic Species Counting

Counting benthic species from sea floor imagery is useful for evaluating the state of the local ecosystem. These images being taken by digital still Cameras mounted on ROVs (Remote Operating Vehicles) are used to capture images of animals on the seabed. The overlap between images must be detected in order to eliminate the possibility of counting the same individual more than once. In order to prevent such an error, a feature-based Mosaicing method using SIFT has been employed to find these overlapping regions. This paper will present some initial results on creating a mosaic from images of the seabed for counting purpose. Key WordsImage Processing, Image Registration, Image Mosaicing, Homography Transformation.