A semi‐automated camera trap distance sampling approach for population density estimation

Abstract

Camera traps have become important tools for the monitoring of animal populations. However, the study‐specific estimation of animal detection probabilities is key if unbiased abundance estimates of unmarked species are to be obtained. Since this process can be very time‐consuming, we developed the first semi‐automated workflow for animals of any size and shape to estimate detection probabilities and population densities. In order to obtain observation distances, a deep learning algorithm is used to create relative depth images that are calibrated with a small set of reference photos for each location, with distances then extracted for animals automatically detected by MegaDetector 4.0. Animal detection by MegaDetector was generally independent of the distance to the camera trap for 10 animal species at two different study sites. If an animal was detected both manually and automatically, the difference in the distance estimates was often minimal at a distance about 4 m from the camera trap. The difference increased approximately linearly for larger distances. Nonetheless, population density estimates based on manual and semi‐automated camera trap distance sampling workflows did not differ significantly. Our results show that a readily available software for semi‐automated distance estimation can reliably be used within a camera trap distance sampling workflow, reducing the time required for data processing, by >13‐fold. This greatly improves the accessibility of camera trap distance sampling for wildlife research and management.

Efficient data collection for camera trap‐based density estimation: A preliminary assessment

“…This allows OCTDE to be applied efficiently in difficult environments, such as marshlands or mountainous setups, where the placing of markers is not feasible. Furthermore, the simplification of setting up a camera site even when compared to methods requiring calibration (Haucke et al, 2022;Henrich et al, 2023;Wearn et al, 2022) makes the process less cumbersome, which can increase community engagement (Wiggins & Crowston, 2011). This can help solve the recent issue of volunteers becoming increasingly hard to enlist in citizen science projects (Willi et al, 2019), which is-otherwise-a promising approach to tackling the problems of data collection in large-scale camera trap studies (Hsing et al, 2022;McShea et al, 2016;Swanson et al, 2015).…”

“…Common to the more recent of these methods is the requirement to estimate the distances and angles from the camera trap to the locations of detected animals. Various methods have been developed for this, such as tracking animal movements on the images based on nearby landmarks (Rowcliffe et al, 2011) or calibration imagery (Henrich et al, 2023;Wearn et al, 2022), using a physical cane grid (Caravaggi et al, 2016), or using poles along the midline of the field of view (FOV; Hofmeester et al, 2017;Mason et al, 2022).…”

“…Recent works addressing the efficiency of distance estimation include the use of virtual distance grids laid over the landscape (McKaughan et al, 2023) and a semi-automated distance estimation method requiring calibration for each camera site (Haucke et al, 2022;Henrich et al, 2023). Meanwhile, the fully AUtomated DIstance esTimation (AUDIT) method (Johanns et al, 2022) simplifies both processes of camera site preparation and distance estimation using machine learning algorithms.…”

Action Detection for Wildlife Monitoring with Camera Traps Based on Segmentation with Filtering of Tracklets (SWIFT) and Mask-Guided Action Recognition (MAROON)

“…Camera traps have become an ubiquitous tool in ecology and conservation that offer a reliable, minimally invasive and visual means of surveying wildlife [2][3][4]. Over the last few decades, camera traps have been adopted for various ecological tasks, including abundance estimation [5][6][7], the quantification of species diversity [8], the detection of rare species [9], the investigation of animal activity patterns [10], and the analysis of species replacement processes [11]. Automatic analysis using artificial intelligence is absolutely necessary to deal with the vast amount of collected camera trap data [12,13].…”