Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks

Moreni, Mael; Theau, Jerome; Foucher, Samuel

doi:10.3390/geomatics1010004

Submit to this Journal Review for this Journal Edit a Special Issue

Article Menu

...

Open AccessArticle

Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks

Mael Moreni

^1,2,3,*

Jerome Theau

^1,3

and

Samuel Foucher

Department of Applied Geomatics, Université de Sherbrooke, Sherbrooke, QC J1K 2R1, Canada

Computer Research Institute of Montréal, Montréal, QC H3T 1J4, Canada

Quebec Centre for Biodiversity Science (QCBS), Stewart Biology, McGill University, Montréal, QC H3A 1B1, Canada

Author to whom correspondence should be addressed.

Geomatics 2021, 1(1), 34-49; https://0-doi-org.brum.beds.ac.uk/10.3390/geomatics1010004

Received: 17 December 2020 / Revised: 11 January 2021 / Accepted: 12 January 2021 / Published: 15 January 2021

View Full-Text Download PDF

Browse Figures

Citation Export

Abstract

The combination of unmanned aerial vehicles (UAV) with deep learning models has the capacity to replace manned aircrafts for wildlife surveys. However, the scarcity of animals in the wild often leads to highly unbalanced, large datasets for which even a good detection method can return a large amount of false detections. Our objectives in this paper were to design a training method that would reduce training time, decrease the number of false positives and alleviate the fine-tuning effort of an image classifier in a context of animal surveys. We acquired two highly unbalanced datasets of deer images with a UAV and trained a Resnet-18 classifier using hard-negative mining and a series of recent techniques. Our method achieved sub-decimal false positive rates on two test sets (1 false positive per 19,162 and 213,312 negatives respectively), while training on small but relevant fractions of the data. The resulting training times were therefore significantly shorter than they would have been using the whole datasets. This high level of efficiency was achieved with little tuning effort and using simple techniques. We believe this parsimonious approach to dealing with highly unbalanced, large datasets could be particularly useful to projects with either limited resources or extremely large datasets. View Full-Text

Keywords: unmanned aerial vehicles; convolutional neural network; wildlife survey; remote sensing; deep learning; conservation; hard-negative mining

▼ Show Figures

Figure 1

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

Share and Cite

MDPI and ACS Style

Moreni, M.; Theau, J.; Foucher, S. Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks. Geomatics 2021, 1, 34-49. https://0-doi-org.brum.beds.ac.uk/10.3390/geomatics1010004

AMA Style

Moreni M, Theau J, Foucher S. Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks. Geomatics. 2021; 1(1):34-49. https://0-doi-org.brum.beds.ac.uk/10.3390/geomatics1010004

Chicago/Turabian Style

Moreni, Mael; Theau, Jerome; Foucher, Samuel. 2021. "Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks" Geomatics 1, no. 1: 34-49. https://0-doi-org.brum.beds.ac.uk/10.3390/geomatics1010004

Find Other Styles

Article Metrics

Abstract views Pdf views Html views

Article Menu

Train Fast While Reducing False Positives: Improving Animal Classification Performance Using Convolutional Neural Networks

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Article Access Map by Country/Region

Further Information

Guidelines

MDPI Initiatives

Follow MDPI