Applying computer‐aided photo‐identification to messy datasets: a case study of Thornicroft's giraffe (Giraffa camelopardalis thornicrofti)

Abstract

Digital photography enables researchers to rapidly compile large quantities of data from individually identifiable animals, and computer software improves the management of such large datasets while aiding the identification process. Wild-ID software has performed well with uniform datasets controlling for angle and portion of the animal photographed; however, few datasets are collected under such controlled conditions. We examined the effectiveness of Wild-ID in identifying individual Thornicroft's giraffe from a dataset of photographs (n = 552) collected opportunistically in the Luangwa Valley, Zambia from March to October 2009. We assessed the programme's accuracy in correctly identifying individuals and the effect of five image quality factors on identification success: blurriness, background type and complexity, amount of sky and the presence of other giraffe. The programme correctly identified individuals in 71.6% of photographs. Background complexity was the only significant variable affecting identification success and removing background imagery reduced identification error by 52.8% (from 28.4 to 13.4%). Our results indicate higher levels of error than previously reported for Wild-ID. However, they also suggest the programme is an effective tool for quickly identifying individuals in large field datasets, especially if photograph backgrounds are removed beforehand and postanalysis visual verification is performed. 13.4%). Nos r esultats indiquent un taux d'erreur sup erieur a ce qui etait pr ec edemment indiqu e pour Wild-ID, mais ils sugg erent aussi que ce programme est un outil efficace pour identifier rapidement des individus dans un vaste ensemble de donn ees de terrain pour autant que le fond des photos soit enlev e avant analyse et qu'une v erification visuelle postanalyse soit r ealis ee.

Social media reveals consistently disproportionate tourism pressure on a threatened marine vertebrate

“…Wildlife watching is intrinsically linked with taking photographs, which creates opportunities to collect important baseline data for ecological research at unprecedented spatial and temporal scales (Dickinson et al, 2012;Toivonen et al, 2019). Photo-identification is a minimally invasive approach used to identify unique individuals in a given population, and has become widely adopted by the scientific community, from research on seadragons to giraffes (Martin-Smith, 2011;Halloran et al, 2015). This approach is also providing opportunities for the public to become directly involved in projects as citizen scientists, by contributing their photographs and increasing the quantity of available data (Holmberg et al, 2008).…”

A newt does not change its spots: using pattern mapping for the identification of individuals in large populations of newt species

“…Wild-ID has already been used with great success in various animal taxa Bendik et al 2013;Cross et al 2014;Elgue et al 2014;Halloran et al 2015). This lack of species-specificity, and the encouraging results reported here and in these previous studies, indicate that Wild-ID can be successfully applied in a wide range of animals that exhibit natural markings, such as many amphibian species.…”

“…However, recent technological advances in digital photography and pattern recognition algorithms (Arzoumanian et al 2005;Van Tienhoven et al 2007;Gamble et al 2008;Bolger et al 2012) have enabled field ecologists to create and analyse large photographic databases quickly and efficiently while minimizing effort and misidentification errors , thus allowing for larger sample sizes in CMR studies. In recent years pattern mapping has been employed in a wide and quite varied range of animals, including beetles (Caci et al 2013), seadragons (Martin-Smith 2011), lizards (Sreekar et al 2013), giraffes (Halloran et al 2015) and dolphins (Martinho et al 2014).…”

Gazing at a giraffe gyroscope: where are we going?

“…The individually distinctive coat patterns of giraffe provide scientists with a method for estimating population sizes and tracking individual movement patterns using photographic images combined with special software as a type of capture/recapture process (Bolger et al ., ; Halloran, Murdoch & Becker, ). Although the accuracy of the method is dependent upon background complexity (Halloran, Murdoch & Becker, ), the potential emerges for giraffe to use a similar system to modulate herd composition. Perhaps they can recognize each other at a distance.…”