Insects, like praying mantises, don’t view the world in the same way that people do. While a human has 85 billion neurons in their brains, insects typically have less than a million. However, a praying mantis has a somewhat unique form of vision that allows it to be a highly effective predator which may help robots see in 3-D.
The vision of a praying mantis differs from the 3-D capabilities associated with primates and many other animals, researchers at the University of Newcastle in Britain found out.
“Praying mantises are really specialized visual predators,” stated Vivek Nityananda, an expert on animal behavior at the Institute of Neuroscience at the University of Newcastle. They operate as ambush predators, sitting completely still until the movement of their prey signals them to strike.
Unlike most other insects, a praying mantis’ eyes face forward. This arrangement in vertebrates gives the ability to sense depth or see in 3-D.
Earlier research suggested that a praying mantis used stereopsis, a form of 3-D vision known as stereo vision. The two eyes perceive portions of the visual field differently, which, according to Nityananda, lets them compare “the slightly different views of each eye to be able to work out how far things are from you.”
To learn more about the vision of praying mantises, according to a report by the Washington Post, Nityananda and his fellow scientists attached lenses to the faces of the insects using beeswax. The beeswax acted as a harmless glue, ensuring the insects weren’t harmed.
The glasses featured different colors lenses on each side, similar to old-fashioned red/blue 3-D glasses people use. Praying mantises were sat in front of a screen and, thanks to the colored lenses, researchers could project a different image onto each eye.
The images considered of a moving target dot against a field of polka-dots. The resulting 3-D motion was perceived by the praying mantises as a real object, leading them to attack.
“Mantises can see depth if there’s a moving object,” said Nityananda.
The target dots were manipulated in a way that the human eye wouldn’t be able to detect, thanks to a reliance on the ability to meld two static images into a single coherent picture.
Mantises seem to rely more on how the image is changing and not necessarily matching up the details between both eyes.
“For the mantis, it looks like [targets] have to be moving. But they don’t have to be matching,” said Nityananda.
The research, if ultimately successful, may create a new approach to vision in robots, moving it away from the human paradigm and more toward how mantises view the world, making them more adept at perceiving depth.