Bridging the Gap between Laboratory and Field Experiments in American Eel Detection Using Transfer Learning and Convolutional Neural Network

Journal Article · 07 January 2020 · Proceedings of the Annual Hawaii International Conference on System Sciences

DOI: https://doi.org/10.24251/hicss.2020.116 · OSTI ID:1756052

Yin, Tianzhixi (Tim) ^[1]; Zang, Xiaoqin ^[2]; Hou, Zhangshuan (Jason) ^[2]; Jacobson, Paul T. ^[3]; Mueller, Robert P. ^[2]; Deng, Zhiqun (Daniel) ^[2]

---

1. Pacific Northwest National Lab. (PNNL), Richland, WA (United States); Electric Power Research Institute
2. Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
3. Electric Power Research Inst. (EPRI), Knoxville, TN (United States)

An automatic system that utilizes data analytics and machine learning to identify adult American eel in data obtained by imaging sonars is created in this study. Wavelet transform has been applied to de-noise the ARIS sonar data and a convolutional neural network model has been built to classify eels and non-eel objects. Because of the unbalanced amounts of data in laboratory and field experiments, a transfer learning strategy is implemented to fine-tune the convolutional neural network model so that it performs well for both the laboratory and field data. The proposed system can provide important information to develop mitigation strategies for safe passage of out-migrating eels at hydroelectric facilities.

• Research Organization: Electric Power Research Inst. (EPRI), Knoxville, TN (United States)
• Sponsoring Organization: USDOE Office of Energy Efficiency and Renewable Energy (EERE)
• Grant/Contract Number: EE0008341
• OSTI ID: 1756052
• Journal Information: Proceedings of the Annual Hawaii International Conference on System Sciences, Journal Name: Proceedings of the Annual Hawaii International Conference on System Sciences Vol. 2020; ISSN 2572-6862
• Publisher: University of Hawaii at Manoa LibraryCopyright Statement
• Country of Publication: United States
• Language: English