Image Analysis Based Fish Tail Beat Frequency Estimation for Fishway Efficiency

Abstract

In this paper, we propose image analysis based methods for estimating fish tail beat frequency, which is an indicator of fish energy consumption at fish passage structures. For this purpose, average magnitude difference and autocorrelation function based periodicity detection techniques are utilized. Actual fish images are acquired using a visible range camera installed in a brush type fish pass in Ikizdere River, near Rize, Turkey, which is very rich in biodiversity. Results show that image analysis based periodicity detection methods can be used for fishway efficiency evaluation purposes. To the best of authors' knowledge, this is the first study that automatically estimates fish tail beat frequency using image analysis. The findings of this study are expected to have implications for fish monitoring and fishway design.