Length estimation of fish detected as non-occluded using a smartphone application and deep learning techniques
Y. Shibata, Y. Iwahara, M. Manano, A. Kanaya, R. Sone, S. Tamura, N. Kakuta, T. Nishino, A. Ishihara, S. Kugai
This paper proposes a deep learning method to estimate fish length from images using a smartphone app, achieving less than 3% bias across five species.
bioRxiv · Mar 12, 2023 · PreprintEcology