View in EDS

Impact of deep learning and post-processing algorithms performances on biodiversity metrics assessed on videos.

Saved in:
Bibliographic Details
Title: Impact of deep learning and post-processing algorithms performances on biodiversity metrics assessed on videos.
Authors: Fleuré V; MARBEC, University Montpellier, CNRS, Ifremer, IRD, Montpellier, France.; ZooParc de Beauval & Beauval Nature, Saint-Aignan, France., Planolles K; MARBEC, University Montpellier, CNRS, Ifremer, IRD, Montpellier, France.; Research-team ICAR, LIRMM, University Montpellier, CNRS, Montpellier, France., Claverie T; UMR ENTROPIE, IRD, IFREMER, CNRS, Univ La Réunion, Saint Denis, Réunion, France., Mulot B; ZooParc de Beauval & Beauval Nature, Saint-Aignan, France., Villéger S; MARBEC, University Montpellier, CNRS, Ifremer, IRD, Montpellier, France.
Source: PloS one [PLoS One] 2025 Aug 11; Vol. 20 (8), pp. e0327577. Date of Electronic Publication: 2025 Aug 11 (Print Publication: 2025).
Publication Type: Journal Article
Language: English
Journal Info: Publisher: Public Library of Science Country of Publication: United States NLM ID: 101285081 Publication Model: eCollection Cited Medium: Internet ISSN: 1932-6203 (Electronic) Linking ISSN: 19326203 NLM ISO Abbreviation: PLoS One Subsets: MEDLINE
Imprint Name(s): Original Publication: San Francisco, CA : Public Library of Science
MeSH Terms: Biodiversity* , Deep Learning* , Algorithms* , Video Recording* , Image Processing, Computer-Assisted*/methods, Animals ; Fishes ; Ecosystem
Abstract: Assessing the escalating biodiversity crisis, driven by climate change, habitat destruction, and exploitation, necessitates efficient monitoring strategies to assess species presence and abundance across diverse habitats. Video-based surveys using remote cameras are a promising, non-invasive way to collect valuable data in various environments. Yet, the analysis of recorded videos remains challenging due to time and expertise constraints. Recent advances in deep learning models have enhanced image processing capabilities in both object detection and classification. However, the impacts on models' performances and usage on assessment of biodiversity metrics on videos is yet to be assessed. This study evaluates the impacts of video processing rates, detection and identification model performance, and post-processing algorithms on the accuracy of biodiversity metrics, using simulated remote videos of fish communities and 14,406 simulated automated processing pipelines. We found that a processing rate of one image per second minimizes errors while ensuring detection of all species. However, even near-perfect detection (both recall and precision of 0.99) and identification (accuracy of 0.99) models resulted in overestimation of total abundance, species richness and species diversity due to false positives. We reveal that post-processing model outputs using a confidence threshold approach (i.e., to discard most erroneous predictions while also discarding a smaller proportion of correct predictions) is the most efficient method to accurately estimate biodiversity from videos.
(Copyright: © 2025 Fleuré et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)
Competing Interests: The authors have declared that no competing interests exist.
References: PLoS One. 2022 Feb 2;17(2):e0263377. (PMID: 35108340)
Nature. 2015 May 28;521(7553):436-44. (PMID: 26017442)
Sci Rep. 2020 Jul 3;10(1):10972. (PMID: 32620873)
Ecol Lett. 2022 Dec;25(12):2753-2775. (PMID: 36264848)
Biol Rev Camb Philos Soc. 2020 Dec;95(6):1706-1719. (PMID: 32648358)
Conserv Biol. 2021 Feb;35(1):88-100. (PMID: 32297655)
Nat Ecol Evol. 2017 Jun 22;1(7):176. (PMID: 28812589)
PLoS Biol. 2013;11(5):e1001569. (PMID: 23723735)
PLoS One. 2016 Mar 02;11(3):e0149701. (PMID: 26934587)
Entry Date(s): Date Created: 20250811 Date Completed: 20250811 Latest Revision: 20250814
Update Code: 20250814
PubMed Central ID: PMC12338835
DOI: 10.1371/journal.pone.0327577
PMID: 40788894
Database: MEDLINE
Description
Abstract:Assessing the escalating biodiversity crisis, driven by climate change, habitat destruction, and exploitation, necessitates efficient monitoring strategies to assess species presence and abundance across diverse habitats. Video-based surveys using remote cameras are a promising, non-invasive way to collect valuable data in various environments. Yet, the analysis of recorded videos remains challenging due to time and expertise constraints. Recent advances in deep learning models have enhanced image processing capabilities in both object detection and classification. However, the impacts on models' performances and usage on assessment of biodiversity metrics on videos is yet to be assessed. This study evaluates the impacts of video processing rates, detection and identification model performance, and post-processing algorithms on the accuracy of biodiversity metrics, using simulated remote videos of fish communities and 14,406 simulated automated processing pipelines. We found that a processing rate of one image per second minimizes errors while ensuring detection of all species. However, even near-perfect detection (both recall and precision of 0.99) and identification (accuracy of 0.99) models resulted in overestimation of total abundance, species richness and species diversity due to false positives. We reveal that post-processing model outputs using a confidence threshold approach (i.e., to discard most erroneous predictions while also discarding a smaller proportion of correct predictions) is the most efficient method to accurately estimate biodiversity from videos.<br /> (Copyright: © 2025 Fleuré et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)
ISSN:1932-6203
DOI:10.1371/journal.pone.0327577