Rapid literature mapping on the recent use of machine learning for wildlife imagery - Research article

Aug 4, 2026

Upload by Himbo

Machine learning, especially deep learning, is reshaping how wildlife imagery is processed, greatly accelerating detection, counting, and classification of animals and their behaviours. Systematic surveys of this application remain scarce, motivating a rapid review and bibliometric mapping. The study examines usage across vertebrate species, image sources, study locations, algorithm types, outcomes, reporting quality and openness, author affiliations, and journal types. Results highlight growing convolutional neural network adoption, limited code sharing, and regional concentration, with recommendations for broader collaboration to improve conservation insights.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Introduction

Background

Literature mapping approaches

Methods

Rapid review and bibliometric mapping

Results

Coverage across species, image types, locations, algorithms

Outcomes, reporting quality, openness

Geographic patterns and collaborations

Discussion and recommendations

Increasing collaboration and sharing approaches

Peer Community Journal Section: Ecology RESEARCH ARTICLE Published 2023-04-11 Cite as Shinichi Nakagawa, Malgorzata Lagisz, Roxane Francis, Jessica Tam, Xun Li, Andrew Elphinstone, Neil R. Jordan, Justine K. O’Brien, Benjamin J. Pitcher, Monique Van Sluys, Arcot Sowmya and Richard T. Kingsford (2023) Rapid literature mapping on the recent use of machine learning for wildlife imagery , Peer Community Journal, 3: e35 . Correspondence [s.nakagawa@unsw.edu.au](s.nakagawa@unsw.edu.au)[m.lagisz@unsw.edu.au](m.lagisz@unsw.edu.au) Peer-review Peer reviewed and recommended by PCI Ecology, [https://doi.org/10.24072/pci](https://doi.org/10.24072/pci). ecology.100513 This article is licensed under the Creative Commons Attribution 4 .0 License. Rapid literature mapping on the recent use of machine learning for wildlife imagery Shinichi Nakagawa , \#, 1 , Malgorzata Lagisz , \#, 1 , Roxane Francis , 2 , Jessica Tam , 2 , Xun Li , 3 , Andrew Elphinstone4 , Neil R. Jordan , 2,4 , Justine K. O’Brien , 2,4 , Benjamin J. Pitcher ,4,5 , Monique Van Sluys4 , Arcot Sowmya , 3 , and Richard T. Kingsford , 2 Volume 3 (2023), article e35 [https://doi.org/10.24072/pcjournal.261](https://doi.org/10.24072/pcjournal.261) Abstract Machine (especially deep) learning algorithms are changing the way wildlife imagery is processed. They dramatically speed up the time to detect, count, and classify animals and their behaviours. Yet, we currently have very few systematic literature surveys on its use in wildlife imagery. Through a literature survey (a‘rapid’ review) and bibliometric mapping, we explored its use across: 1) species (vertebrates), 2) image types (e.g., camera traps, or drones), 3) study locations, 4) alternative machine learning algorithms, 5) outcomes (e.g., recognition, classiﬁ cation, or tracking), 6) reporting quality and openness, 7) author aﬃliation, and 8) publication journal types. We found that an increasing number of studies used convolutional neural networks (i.e., deep learning) . Typically, studies have focused on large charismatic or iconic mammalian species. An increasing number of studies have been published in ecology-speciﬁc journals indicating the uptake of deep learning to transform the detection, classiﬁcation and tracking of wildlife. Sharing of code was limited, with only 20% of studies providing links to analysis code. Much of the published research and focus on animals came from India, China, Australia, or the USA. There were relatively few collaborations across countries. Given the power of machine learning, we recommend increasing collaboration and sharing approaches to utilise increasing amounts of wildlife imagery more rapidly and transform and improve understanding of wildlife behaviour and conservation. Our survey, augmented with bibliometric analyses, provides valuable signposts for future studies to resolve and address shortcomings, gaps, and biases. 1 UNSW Data Science Hub, Evolution & Ecology Research Centre and School of Biological, Earth and Environmental Sciences, UNSW, Sydney, NSW 2052, Australia, 2 Centre for Ecosystem Science and School of Biological, Earth and Environmental Sciences, UNSW, Sydney, NSW 2052, Australia, 3School of Computer Science and Engineering, UNSW Sydney, NSW 2052, Australia, 4Taronga Institute of Science and Learning, Taronga Conservation Society Australia, Mosman, NSW 2088, Australia, 5School of Natural Sciences, Macquarie University, Sydney, NSW 2109, Australia, \# Equal contribution C E NT R E MERSENNE Peer Community Journal is a member of the Centre Mersenne for Open Scientiﬁc Publishing [http:// www.centre-mersenne.org/](http:// www.centre-mersenne.org/) e-ISSN 2804-3871 Introduction Background Camera-trap, surveillance-video, and drone imagery are producing a deluge of digital data on wildlife (Koh & Wich, 2012; Meek et al., 2014; Allan et al., 2018; Weinstein, 2018; Tuia et al., 2022; Besson et al. 2022) . Processing these

FAQs

What is the purpose of the rapid review and bibliometric mapping in this paper?

To systematically examine how machine learning is currently used with wildlife imagery by combining a rapid literature review with bibliometric mapping across multiple dimensions.

Which machine learning approach is increasingly used for wildlife imagery detection and classification?

Convolutional neural networks (deep learning) are highlighted as the approach with an increasing number of studies.

What limitation did the authors observe regarding sharing of research code?

Code sharing was limited, with only about 20% of studies providing links to analysis code.

Aug 4, 2026

Upload by Himbo