Header menu link for other important links
X
An efficient interactive segmentation algorithm using color correction for underwater images
Published in Springer Science and Business Media LLC
2019
Abstract
Image segmentation is one of the basic strategies in image analytics. It is used for pattern detection, photo montage and object recognition applications. In underwater images light scrambling, suspended particles that exhibit in the water, marine snow, and color deviations etc. are some of the issues for image acquisition. Due to these issues, all the segmentation algorithms invariably produce poor results in the case of underwater images. We propose an efficient algorithm for extracting the foreground region from the underwater images based on a region of interest. Our segmentation procedure for underwater images includes color correction, contrast enhancement followed by an interactive GrabCut algorithm. Inclusion of color correction and contrast enhancement minimizes the number of touch ups required to enhance the segmentation when compared to GrabCut algorithm without these preprocessing steps. Our algorithm is efficient in the sense of time, when compared to the other algorithms for the same purpose. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.
About the journal
JournalData powered by TypesetWireless Networks
PublisherData powered by TypesetSpringer Science and Business Media LLC
ISSN1022-0038
Open Access0