Next Article in Journal
Remotely Sensed Urban Surface Ecological Index (RSUSEI): An Analytical Framework for Assessing the Surface Ecological Status in Urban Environments
Previous Article in Journal
Alfalfa Yield Prediction Using UAV-Based Hyperspectral Imagery and Ensemble Learning
Article

Using Deep Learning to Count Albatrosses from Space: Assessing Results in Light of Ground Truth Uncertainty

1
School of Computing Sciences, University of East Anglia, Norwich NR4 7TJ, UK
2
Mapping and Geographic Information Centre, British Antarctic Survey, Cambridge CB3 0ET, UK
*
Author to whom correspondence should be addressed.
Remote Sens. 2020, 12(12), 2026; https://doi.org/10.3390/rs12122026
Received: 15 May 2020 / Revised: 19 June 2020 / Accepted: 21 June 2020 / Published: 24 June 2020
Many wildlife species inhabit inaccessible environments, limiting researchers ability to conduct essential population surveys. Recently, very high resolution (sub-metre) satellite imagery has enabled remote monitoring of certain species directly from space; however, manual analysis of the imagery is time-consuming, expensive and subjective. State-of-the-art deep learning approaches can automate this process; however, often image datasets are small, and uncertainty in ground truth labels can affect supervised training schemes and the interpretation of errors. In this paper, we investigate these challenges by conducting both manual and automated counts of nesting Wandering Albatrosses on four separate islands, captured by the 31 cm resolution WorldView-3 sensor. We collect counts from six observers, and train a convolutional neural network (U-Net) using leave-one-island-out cross-validation and different combinations of ground truth labels. We show that (1) interobserver variation in manual counts is significant and differs between the four islands, (2) the small dataset can limit the networks ability to generalise to unseen imagery and (3) the choice of ground truth labels can have a significant impact on our assessment of network performance. Our final results show the network detects albatrosses as accurately as human observers for two of the islands, while in the other two misclassifications are largely caused by the presence of noise, cloud cover and habitat, which was not present in the training dataset. While the results show promise, we stress the importance of considering these factors for any study where data is limited and observer confidence is variable. View Full-Text
Keywords: WorldView-3; convolutional neural network; VHR satellite imagery; wildlife monitoring; observer uncertainty; Wandering Albatross WorldView-3; convolutional neural network; VHR satellite imagery; wildlife monitoring; observer uncertainty; Wandering Albatross
Show Figures

Figure 1

MDPI and ACS Style

Bowler, E.; Fretwell, P.T.; French, G.; Mackiewicz, M. Using Deep Learning to Count Albatrosses from Space: Assessing Results in Light of Ground Truth Uncertainty. Remote Sens. 2020, 12, 2026. https://doi.org/10.3390/rs12122026

AMA Style

Bowler E, Fretwell PT, French G, Mackiewicz M. Using Deep Learning to Count Albatrosses from Space: Assessing Results in Light of Ground Truth Uncertainty. Remote Sensing. 2020; 12(12):2026. https://doi.org/10.3390/rs12122026

Chicago/Turabian Style

Bowler, Ellen, Peter T. Fretwell, Geoffrey French, and Michal Mackiewicz. 2020. "Using Deep Learning to Count Albatrosses from Space: Assessing Results in Light of Ground Truth Uncertainty" Remote Sensing 12, no. 12: 2026. https://doi.org/10.3390/rs12122026

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop