Automated classification of zooplankton for a towed imaging system

Most studies on zooplankton image processing are focused on global texture feature. In this paper, an efficient method for classifying zooplankton images by combination of local binary pattern (LBP) is demonstrated. The classification performance of plankton image with global and local features have been evaluated and the overall performance of combination of these features for zooplankton identification shows that by combining conventional global features with local binary pattern texture feature have the similar classification performance for 6 taxonomic groups.