USING IMAGE ANALYSIS FOR CALCULATING THE OVERALL COVERAGE OF UNDERWATER VEGETATION IN BALTIC SEA REGION

摘要

This paper describes underwater vegetation monitoring by means of image analysis. The objective of this paper is to determine the overall coverage (i.e. the percentage of the sea bottom covered by macro-vegetation.) The segments are extracted with RGB-filters and classified by means of Hopfield neural networks from underwater photos and videos. Test results are verified with the opinion of a hydrobiologists. The results show that in case of sparse vegetation the classification accuracy is comparable with that of a human expert. Based on the experimental results new future work directions are determined.