A lot of biodiversity research relies on being able to recognise and catalogue the species in a given ecosystem. That’s a task made slightly easier when spiders are involved thanks to researchers at the University of Las Palmas de Gran Canaria in Spain who’ve developed algorithms that can automatically recognise an arachnid based on the design of the web it spins.
In a similar approach to how the authorities can identify you using your fingerprints, the researchers have developed custom software and algorithms that will analyse a spider’s web, pinpoint its most important identifying features, compare them with a database, and then report back the exact species with a remarkable 99.96 success per cent rate. Besides making spider research considerably more efficient, it would be fantastic if the researchers turned this into an app that could instantly determine if a spider was poisonous or not — assuming it doesn’t fall into that 0.04 error percentile.
[ScienceDirect via Wired UK]
Picture: TSpider/Shutterstock