ITR: Pattern Recognition for Ecological Science and Environmental Monitoring

Principal Investigator
Thomas G. Dietterich (Oregon State University), Linda Shapiro, David A. Lytle (Oregon State University), Andrew R. Moldenke (Oregon State University), Robert K. Paasch (Oregon State University)

Sponsor(s)
Oregon State University (flow through from NSF)

Award Period
09/01/2003 - 08/31/2008

Abstract
Many problems in ecological science and environmental monitoring would benefit from inexpensive, automated methods for obtaining population counts of insects and other microfauna. Existing methods for obtaining such population counts require manual identification by human experts, which is a too costly to permit the routine use of these data for large-scale scientific and monitoring projects. A multi-disciplinary team of computer scientists, entomologists, and electrical and mechanical engineers will develop general-purpose pattern recognition algorithms for the identification and classification of microfuna and mechanical devices for handling and photographing specimens. The methods and devices will be developed and tested on three important scientific, environmental, and agricultural problems: (a) water quality monitoring in streams (by recognizing and counting stonefly larvae in stream substrates. (b) measurement and characterization of forest beetle biodiversity (by recognizing and counting adult beetles in soil and leaf litter), and (c) measurement of the pollination servicees provided by wild bees and prospecting for new, agriculturally-useful bee species (by recognizing and counting adult bees in agricultural fields).

Updates or corrections to this page should be sent to gheaton@u.washington.edu.