Automated Detection and Tracking of Adult Pacific Lampreys in Underwater Video Collected at Snake and Columbia River Fishways

Abstract

To reduce staffing demands and enhance observational accuracy, we designed, implemented, and tested a computerized system for processing underwater video clips captured by static cameras and removing “quiet” frames in which no activity is detected. When the system detects activity, it tracks and counts the moving object. The implementation uses an adaptive background-subtraction algorithm for detection and motion prediction for tracking. The system can be used to reduce the amount of video that must be reviewed by personnel, and it can produce total fish passage counts through the monitored area. The automated system reduced the total number of video hours requiring review by an average of 87.5% for count window videos and 83.5% for videos of picketed leads and lamprey passage orifices. The software detected 98.6% of the 144 Pacific Lampreys Entosphenus tridentatus that were observed in 185 h of raw video. Because the system is fully automated, monitoring requirements are negligible, and the cost reduction for fish monitoring is proportional to the number of quiet frames removed.

Received March 27, 2013; accepted September 23, 2013

Published online January 30, 2014

ACKNOWLEDGMENTS

This project was funded by the U.S. Army Corps of Engineers, Walla Walla District (Contract Number W912EF-08-D-0007). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the supporting agency. William Abernathy (University of California, Davis) helped to prepare the manuscript.