Author(s): Yasin Yildirim; B. Ugur Toreyin; Serhat Kucukali
Linked Author(s): Serhat Kucukali
Keywords: No Keywords
Abstract: Measurement of fish movement is an important task for evaluating fish passage efficiency. In this work, we propose a computer vision based fish detection and movement analysis framework. The proposed approach detects fish and estimates the fish tail beat frequency automatically, while the fish trajectory is obtained manually. The framework is tested using visible-range videos captured by cameras installed in a physical model and a prototype of brush fish pass. Results demonstrate the efficacy of the proposed approach to quantify fish behavior in various flow fields. The framework may be beneficial for rehabilitation of existing fish pass structures.
Year: 2018