Quantifying camouflage: how to predict detectability from appearance

Abstract

BackgroundQuantifying the conspicuousness of objects against particular backgrounds is key to understanding the evolution and adaptive value of animal coloration, and in designing effective camouflage. Quantifying detectability can reveal how colour patterns affect survival, how animals’ appearances influence habitat preferences, and how receiver visual systems work. Advances in calibrated digital imaging are enabling the capture of objective visual information, but it remains unclear which methods are best for measuring detectability. Numerous descriptions and models of appearance have been used to infer the detectability of animals, but these models are rarely empirically validated or directly compared to one another. We compared the performance of human ‘predators’ to a bank of contemporary methods for quantifying the appearance of camouflaged prey. Background matching was assessed using several established methods, including sophisticated feature-based pattern analysis, granularity approaches and a range of luminance and contrast difference measures. Disruptive coloration is a further camouflage strategy where high contrast patterns disrupt they prey’s tell-tale outline, making it more difficult to detect. Disruptive camouflage has been studied intensely over the past decade, yet defining and measuring it have proven far more problematic. We assessed how well existing disruptive coloration measures predicted capture times. Additionally, we developed a new method for measuring edge disruption based on an understanding of sensory processing and the way in which false edges are thought to interfere with animal outlines.ResultsOur novel measure of disruptive coloration was the best predictor of capture times overall, highlighting the importance of false edges in concealment over and above pattern or luminance matching.ConclusionsThe efficacy of our new method for measuring disruptive camouflage together with its biological plausibility and computational efficiency represents a substantial advance in our understanding of the measurement, mechanism and definition of disruptive camouflage. Our study also provides the first test of the efficacy of many established methods for quantifying how conspicuous animals are against particular backgrounds. The validation of these methods opens up new lines of investigation surrounding the form and function of different types of camouflage, and may apply more broadly to the evolution of any visual signal.Electronic supplementary materialThe online version of this article (doi:10.1186/s12862-016-0854-2) contains supplementary material, which is available to authorized users.

Gray plumage color is more cryptic than brown in snowy landscapes in a resident color polymorphic bird

“…The use of human observers via online “games” has been recently used in camouflage studies in birds and their eggs (Troscianko, Wilson‐Aggarwal, Griffiths, Spottiswoode, & Stevens, ), as well as crabs (Nokelainen, Maynes, Mynott, Price, & Stevens, ). Similar citizen science approach has also previously been used to study disruptive coloration with, for example, computer‐generated moth images (Fraser, Callahan, Klassen, & Sherratt, ), and detection and learning of camouflage strategies (Troscianko, Lown, Hughes, & Stevens, ) and to quantify the appearance of camouflaged prey (Troscianko, Skelhorn, & Stevens, ). We designed an online study website where a randomly chosen series of pictures was presented for each volunteer participant.…”

A computer vision for animal ecology

“…Image features have been primarily used to study the evolutionary ecology of animal coloration (Stoddard, Kilner, & Town, ), shape (Lavy et al., ) and patterning (Levy, Lerner, & Shashar, ). Compared to human review, computer vision provides a more consistent way to score animal appearance across images by using non‐RBG colour spaces, such as HSV or YChCr, which are less sensitive to changes in illumination and other image artefacts (Kühl & Burghardt, ; Troscianko, Skelhorn, & Stevens, ). By comparing image features, computer vision can be used to study animal camouflage (Tankus & Yeshurun, ) and biomimicry (Yang, Wang, Liang, & Møller, ).…”

Camouflage

“…Experiments with humans searching for animal shapes indicate that edge enhancement interferes with object recognition, not only detection (Sharman, Moncrieff & Lovell, ). If the contrasting colour patches intersect the body's edge, the continuity of the signal at the true outline is reduced, and this form of disruptive coloration has been shown to be more effective against both birds and humans than background matching alone (Cuthill et al ., ; Schaefer & Stobbe, ; Stevens et al ., ; Fraser et al ., ; Webster, Godin & Sherratt, ; Troscianko, Skelhorn & Stevens, ). The same effect can be used to disguise features other than the outline by having highly salient false edges that run across different body parts, something Cott called ‘coincident disruptive coloration’.…”