Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Kathryn Hollowood - Roberts Wesleyan College
Co-Author(s): Olatide Omojaro, Georgia Institute of Technology, Atlanta, GA Dalwinderjeet Kular and Eraldo Ribeiro, Florida Institute of Technology, Melbourne, FL
A major concern of modern society is environmental conservation. Pressures such as littering, pollution, and climate change deteriorate natural habitats and harm the environment. In order to stop this deterioration, we need some way to track changes to the environment. We need a bio-indicator. Frogs make excellent bio-indicators because they have a permeable skin which allows them absorb water Oxygen, things they need to survive. However, when they live in polluted areas, they also absorb toxins due to this pollution. As the pollution rates rise, the frog populations decline. Citizen scientists have been monitoring these frog populations by recording their calls for years. However, this requires expensive equipment, training, and a large time commitment. The focus of this paper is the automatic classification of frog calls using computer programs. Features were extracted from the audio data and then classified using either Locality Sensitive Hashing or Gaussian Mixture Modeling. Tests were performed on a data set consisting of fifteen different species produced promising classification results for the Gaussian Mixture Model approach.
Funder Acknowledgement(s): National Science Foundation Grant no. 1263011; National Science Foundation Grant no. 1152306
Faculty Advisor: Eraldo Ribeiro,