A new method, developed by scientists at the University of Bristol, can determine the optimal coloring pattern to cover an object in order to make it as visible or hidden as possible, in any given environment.
And doing so can help provide information about the evolution of animal staining.
The authors, all members of the University’s CamoLab, believe their new method, called The Camouflage Machine, can be very useful for evolutionary biologists trying to understand how coloration evolved and can also explain the trade-offs between concealment or color visible, for example warning staining or sexual samples.
Explaining the motivation for the research, Professor Innes Cuthill said, “We all know that animals like leopards and jaguars have stunning camouflage, but is it optimal?
“Has evolution been limited by the way patterns develop in mammals? Are differences in the point pattern between species accidental in history or do they reflect differences in habitat? The camouflage machine allows us to investigate these. issues “.
Conventionally, setting the optimal colors for camouflage or signage presents biologists with a serious challenge, because the number of potential color combinations and patterns to be tested is often enormous.
Through a mix of genetic algorithms and deep learning, the authors have demonstrated how the detection times of millions of multi-colored patterns can be predicted using only a modest amount of data acquired from human observers.
The lead author, Dr. John Fennell, of Bristol Veterinary School, added: “The method is based on showing images of objects in various backgrounds to people and asking them to press a button as soon as they see the object.
“During the experiment, a genetic algorithm reduces the range of colors and patterns to those that are harder or easier to see, depending on whether we would like to find coloration for camouflage or visibility.
“Multiple deep neural networks are formed using human data. These networks, used in conjunction with a genetic algorithm, can be considered artificial ‘participants’ who can quickly evaluate all combinations of color patterns; which human participants would take years to do. . “
In the new research, published in the journal Evolution, the authors demonstrate the effectiveness of the method in different environments. The effectiveness of coloration is critically influenced by the visual system of the observer and scientists show that optimal camouflage is very different depending on whether the viewer has full color vision or, for example, red-green blind color ( as is the case with most non-human mammals). The method allows scientists to evaluate the full range of colors available for a given animal taxon.
One of the co-authors, Dr. Laszlo Talas, also of Bristol Veterinary School, said: “If we consider all the colors and patterns available to mammals, from light beige to tiger orange, including grayscale colors. and patterns that can vary across a wide range of dots, lines, and spots, it is simply impossible to thoroughly test all potential combinations using actual observers.
“The camouflage machine not only allows us to test all the coloring we see in nature, but also a possible coloration that, for some reason, evolution didn’t notice.”
Senior Research Professor Nick Scott-Samuel, of the Bristol School of Psychological Sciences, said: “The potential to apply the method in the human domain has not been lost on us.”
This was reinforced by his colleague Roland Baddeley, who said: “We are excited about the huge number of potential applications, from making road signs or cyclists as visible as possible, to minimizing the visibility of the necessary, but visually unpleasant, infrastructure of the modern world (such as telephone poles, electric pylons, and wind farms).
“While it’s unlikely we can make them disappear,” it should be relatively straightforward, using The Camouflage Machine, to minimize the visual intrusiveness of these things in a given environment. Something would probably agree with everyone. ”
Animals reduce the symmetry of their marks to improve camouflage
John G. Fennell et al. The camouflage machine: optimization of protective coloring through deep learning with genetic algorithms, Evolution (2021). DOI: 10.1111 / evo.14162
Provided by the University of Bristol
Citation: New research shows how to find the best camouflage for an arbitrary environment (2021, January 22) recovered on January 22, 2021 at https://phys.org/news/2021-01-camouflage-arbitrary- environment.html
This document is subject to copyright. Apart from any fair treatment for private study or research purposes, no part may be reproduced without written permission. Content is provided for informational purposes only.