The Daily Bulletin Archive

Please join the Biology Department and Great Lakes Center for their next seminar, "Development and Application of a Robot-Assisted Computer Vision System to Support Ecological Monitoring in the Great Lakes," presented by Peter Esselman, Ph.D., research fisheries biologist with the U.S. Geological Survey Great Lakes Science Center, on Monday, December 2, from 3:00 to 3:50 p.m. in Technology Building 160. All are welcome.

Abstract
The benthification of Great Lakes food webs, facilitated by dreissenid mussel population expansion, places new emphasis on the need to quantify benthic ecosystem attributes including habitats, mussels, round goby, Cladophora, and other attributes. The U.S. Geological Survey has been working collaboratively with partner institutions to develop a robot-assisted computer vision system that includes mobile technologies to deploy camera systems close to the lake bed, as well as algorithms to receive and condition acquired imagery and automatically quantify features of interest. Data collected during the 2018 and 2019 field campaigns with autonomous underwater vehicles and a scuba-assisted camera system were used to support algorithm development for substrate classification, 3-D reconstruction and structural characterization of the lake bed, analysis of multi-scale structural characteristics of bottom habitats, round goby presence, Cladophora presence and volume, and dreissenid mussel presence. Initial findings from these algorithm development processes will be presented with suggestions for needed future refinements. The implications of these new capabilities to support spatially targeted management of Great Lakes habitats and biological resources will be discussed.