Optimizing colour for camouflage and visibility using deep learning: the effects of the environment and the observer's visual system

Abstract

<jats:p>Avoiding detection can provide significant survival advantages for prey, predators, or the military; conversely, maximizing visibility would be useful for signalling. One simple determinant of detectability is an animal's colour relative to its environment. But identifying the optimal colour to minimize (or maximize) detectability in a given natural environment is complex, partly because of the nature of the perceptual space. Here for the first time, using image processing techniques to embed targets into realistic environments together with psychophysics to estimate detectability and deep neural networks to interpolate between sampled colours, we propose a method to identify the optimal colour that either minimizes or maximizes visibility. We apply our approach in two natural environments (temperate forest and semi-arid desert) and show how a comparatively small number of samples can be used to predict robustly the most and least effective colours for camouflage. To illustrate how our approach can be generalized to other non-human visual systems, we also identify the optimum colours for concealment and visibility when viewed by simulated red–green colour-blind dichromats, typical for non-human mammals. Contrasting the results from these visual systems sheds light on why some predators seem, at least to humans, to have colouring that would appear detrimental to ambush hunting. We found that for simulated dichromatic observers, colour strongly affected detection time for both environments. In contrast, trichromatic observers were more effective at breaking camouflage.</jats:p>

The effect of pelage, background, and distance on predator detection and the evolution of primate color vision

“…The role of pelage coloration on camouflage has been investigated, mainly, by phylogenetic comparative analyses (e.g., Caro et al, 2017) or visual perception models (e.g., Pessoa et al, 2014). Most experimental studies on this matter have used artificial targets as surrogates of potential predators (e.g., Fennell et al, 2019); therefore, empirical data with natural targets (such as carnivore's pelages) are lacking. Here, we experimentally investigated how different taxidermized carnivore mammals (each with specific pelage coloration patterns) are visually detected by humans of varied color vision phenotypes in three different complex natural scenarios, and at different detection distances.…”

Mathematical modeling of the protective coloration of animals with usage of parameters of diversity and evenness

“…The protective mechanisms of animal coloration are very complex and are systemic in nature. This implies the use of certain approaches to mathematical modeling procedures for their study (Turing, 1952;Murray, 1981a,b;Murray and Maini, 1986;Fennell et al, 2019). This paper aims to study the abilities of usage of new approaches described below.…”

“…1 Literature Analysis and Statement of the Problem For a long time, the coloration of animals has been of interest to representatives of dierent branches of science (Turing, 1952;Murray, 1981a,b;Murray and Maini, 1986;Endler and Mappes, 2017;Fennell et al, 2019). They investigate multiple aspects of this problem and specied objects of the study.…”

Camouflage in arid environments: the case of Sahara-Sahel desert rodents

“…To sum up, we predict that more mobile animals, and those with large home ranges spanning heterogenous environments, should adopt a more generalist and/or compromise camouflage strategy. Less mobile, more sedentary or less competitive species should rely on more specialized camouflage (Merilaita et al 2017, Fennell et al 2019, Hughes et al 2019. Animals having larger home ranges would more likely have a need to cross several different types of micro-habitats in comparison to less mobile individuals/species with smaller ranges.…”