Drones, automatic counting tools, and artificial neural networks in wildlife population censusing

Abstract

The use of a drone to count the flock sizes of 33 species of waterbirds during the breeding and non‐breeding periods was investigated. In 96% of 343 cases, drone counting was successful. 18.8% of non‐breeding birds and 3.6% of breeding birds exhibited adverse reactions: the former birds were flushed, whereas the latter attempted to attack the drone. The automatic counting of birds was best done with ImageJ/Fiji microbiology software – the average counting rate was 100 birds in 64 s. Machine learning using neural network algorithms proved to be an effective and quick way of counting birds – 100 birds in 7 s. However, the preparation of images and machine learning time is time‐consuming, so this method is recommended only for large data sets and large bird assemblages. The responsible study of wildlife using a drone should only be carried out by persons experienced in the biology and behavior of the target animals.

Understanding the effects of seasonal variation in prey availability on prey switching by large carnivores

“…Estimating seasonal herbivore abundance would provide insights into potential interannual variation in predator–prey dynamics. While this is a challenging undertaking for a large study area with high densities of wildlife, this may be possible in the future with technological and computational advancements (Marchowski, 2021).…”

Existing and emerging uses of drones in restoration ecology

“…Many other studies have used drones to successfully count wildlife, from waterbirds to crocodiles (Aubert et al, 2021; Marchowski, 2021). Counting wildlife is an important facet of restoration ecology––for example, it can be part of measuring the composition, functionality and complexity of ecosystems prior to, during and following a restoration intervention.…”

Counting animals in aerial images with a density map estimation model

“…Hurford ( 2017 ) introduced image processing software, ImageJ to assist in counting birds, but its practical use was limited since the software cannot handle complex environments, which is common in various ecological contexts. To alleviate this problem, Marchowski ( 2021 ) preprocessed images with a denoising neural network (Buchholz et al, 2020 ) before counting with ImageJ, which then makes the counting accuracy highly dependent upon the performance of the denoising process. Object‐based image analysis (OBIA) has also been popular in previous ecological field studies (Afán et al, 2018 ; Chrétien et al, 2016 ; Lhoest et al, 2015 ; Rush et al, 2018 ).…”