Using very-high-resolution satellite imagery and deep learning to detect and count African elephants in heterogeneous landscapes - Dataset

This repository contains a dataset of satellite images for African Elephant (Loxodonta africana) detection. The dataset is from “Using very‐high‐resolution satellite imagery and deep learning to detect and count African elephants in heterogeneous landscapes” Remote Sensing in Ecology and Conservation. The dataset consists of 600x600 sub-images extracted from World-View-3 and Word-View-4 satellite images (c) Maxar Technology acquired between 2014 and 2019 in Addo Elephant National Park in South Africa. Each sub-image is manually labelled with bounding boxes around individual elephants.

The data is split into train and test sets. Labels are provided in the csv files with filenames referring to the images in the corresponding image folders. 

 How to cite:

Duporge, I., Isupova, O., Reece, S., Macdonald, D.W. and Wang, T., 2021. Using very‐high‐resolution satellite imagery and deep learning to detect and count African elephants in heterogeneous landscapes. Remote Sensing in Ecology and Conservation, 7(3), pp.369-381.