Demystifying image-based machine learning: a practical guide to automated analysis of field imagery using modern machine learning tools

Frontiers in Marine Science

Marine Science

Eliana H. Bower¹  Salma Valladares¹  Isabela L. Guevara¹  Byron T. Belcher¹  Brayan Vaca¹  Mohamad H. Saleh¹  Anjana Manjunath¹  Zulekha Khokhar¹  Stella A. Hein²  Andrew M. Hein³  Benjamin Burford⁴  Samuel Nelson⁴  Simone Olivetti⁴  Maria Rosa Celis⁴  Ashkaan K. Fahimipour⁵  Kakani Katija⁶  Eric Orenstein⁶ 

[1] AI for the Ocean program, University of California Santa Cruz, Santa Cruz, CA, United States;AI for the Ocean program, University of California Santa Cruz, Santa Cruz, CA, United States;Cornell University, College of Agriculture and Life Sciences, Ithaca, NY, United States;AI for the Ocean program, University of California Santa Cruz, Santa Cruz, CA, United States;Department of Computational Biology, Cornell University, Ithaca, NY, United States;AI for the Ocean program, University of California Santa Cruz, Santa Cruz, CA, United States;Institute of Marine Sciences, University of California Santa Cruz, Santa Cruz, CA, United States;AI for the Ocean program, University of California Santa Cruz, Santa Cruz, CA, United States;Institute of Marine Sciences, University of California Santa Cruz, Santa Cruz, CA, United States;Florida Atlantic University, Department of Biology, Boca Raton, FL, United States;Monterey Bay Aquarium Research Institute, Research and Development, Moss Landing, CA, United States;

关键词: machine learning;    image analysis;    deep neural network;    underwater imagery;    computer vision;    artificial intelligence;    distribution shift;   

DOI  :  10.3389/fmars.2023.1157370

 received in 2023-02-08, accepted in 2023-05-19,  发布年份 2023

来源: Frontiers

【 摘 要 】

Image-based machine learning methods are becoming among the most widely-used forms of data analysis across science, technology, engineering, and industry. These methods are powerful because they can rapidly and automatically extract rich contextual and spatial information from images, a process that has historically required a large amount of human labor. A wide range of recent scientific applications have demonstrated the potential of these methods to change how researchers study the ocean. However, despite their promise, machine learning tools are still under-exploited in many domains including species and environmental monitoring, biodiversity surveys, fisheries abundance and size estimation, rare event and species detection, the study of animal behavior, and citizen science. Our objective in this article is to provide an approachable, end-to-end guide to help researchers apply image-based machine learning methods effectively to their own research problems. Using a case study, we describe how to prepare data, train and deploy models, and overcome common issues that can cause models to underperform. Importantly, we discuss how to diagnose problems that can cause poor model performance on new imagery to build robust tools that can vastly accelerate data acquisition in the marine realm. Code to perform analyses is provided at https://github.com/heinsense2/AIO_CaseStudy.

【 授权许可】

Unknown   
Copyright © 2023 Belcher, Bower, Burford, Celis, Fahimipour, Guevara, Katija, Khokhar, Manjunath, Nelson, Olivetti, Orenstein, Saleh, Vaca, Valladares, Hein and Hein