Ever saw some birds and wondered what species the bird are? Well, now there’s an app for that! Almost! The Cornell Lab of Ornithology and Visipedia are collaborating to develop computer vision technology to identify birds in photos. They have released a beta product called Merlin, where right now, they can identify through a photo (and some hints for better results, like where was this picture taken, if you know) about 400 North American birds! We agree, that’s not a lot, but wow, that is something! Hopefully soon, they will be able to identify a lot more than 400 birds!
Merlin can recognize 400 bird species, including the Blackburnian warbler shown above.
To begin, you simply add a picture of the unknown bird in question to the web site, enter the location and date that the image was taken (you can leave the fields blank if you are not sure). You’ll be prompted to draw a box around the bird, click on its beak tip, eye, and tail tip. Merlin does the rest, querying the eBird.org citizen science database’s 70 million entries to present you with the most probable species matches, together with pictures and sounds.
The site is still in beta, and the developers are the process of making a smartphone app. But that’s no reason you shouldn’t give it a go now: Because the picture identifier uses machine learning techniques, it has the potential to improve the more we use it. Imagine seeing any bird anywhere, just take out your smart phone and zap, you’ll know the name of the bird, species, life, age, favorite foods etc etc.. well, that reality is now very close! A true gift for the bird-watchers I say! Just imagine a world where you could go on a hike and, utilizing your cell, quickly determine whether or not that plant, mushroom, spider or snake you came across was harmful or could be eating as food!
