Let’s face it, Shazam is a world-class invention. Got a song stuck in your head and have no idea what it is? Shazam. Want to prove a mate wrong about who sings a particular song? Shazam. Want to impress a date with your extensive music knowledge? Shazam.
Of course, it’s not much use outside of the music world, but all of that’s about to change, because there’s now a version of Shazam for spider and snake lovers – or for those who maybe don’t love them but certainly want to know which ones to avoid.
Essentially, the app – aptly called Critterpedia – will help the everyday user identify different species of spiders and snakes using just a single photo, in the same way Shazam helps you identify different songs using a single sound clip.
All you need to do is upload an image of a specimen to the app, which will then use an AI-powered algorithm to identify and classify the creature before providing information on the family or species – including its threat level.
In doing so, the developers of Critterpedia hope to provide greater awareness of different wildlife species to potentially help save both human and animal lives, something that will no doubt prove extremely helpful considering there are at least 2,000 species of spider in Australia and 170 species of snake.
While more than 90% of those are unlikely to pose any serious threat to people, there are some – including two types of spider and 12 types of snake – that are venomous and that can kill people.
As it’s difficult to tell just by looking at them which are potentially deadly, having an app such as Critterpedia handy will be helpful in more ways than one.
And having been created by Australia’s National Science Agency, CSIRO, in collaboration with a pair of local engineers, it’s fair to say we’re in safe hands.
Dr Matt Adcock, project lead and researcher at Data61, CSIRO’s data research arm, said in a statement:
The visual differences between two species can sometimes be quite subtle, and so a great deal of training data is needed to adequately identify critters.
We’ve started off with an enormous amount of images sourced from zoological experts collaborating with Critterpedia, and have developed a suite of tools to help semi-automatically label these images, verify the information, and cross check with other data sources.
Critterpedia users will also be able to contribute to the app by submitting their own photos to help train the machine learning engine, which will help expand the identification system.
‘The intent is to form (consensual) user generated images into datasets of all animals and to extend our AI training with the team to eventually include many more species,’ said Nic Scarce who, along with partner Murray, came up with the idea for the app.
The program is yet to be released, although if you like the sound of it you can currently sign up to become a Phase 1 tester – which lets you download a beta version of the app and submit wildlife photos to help train the algorithm.