Merlin was launched in January 2014 as a free bird identification resource covering the 285 most common birds in the US and Canada (now up to 400 species, with more on the way). Now, 2.5 years later, Merlin has reached more than 1,000,000 users! If you aren’t familiar with Merlin, it lets you describe simple features of birds that you see, and, like magic, tells you what you’re looking at. Merlin users are opening the app nearly a million times monthly, logging 18 million identified birds each year! It is a great way to engage people with the natural world, and an invaluable resource for anyone interested in birds. Learn more here.
In the coming months, Merlin’s coverage will continue to expand beyond the U.S. and Canada—and you can help make this happen! Each sighting that you enter in eBird, and each photo that you add to the Macaulay Library, contributes to the data resources powering Merlin. We are working to build automated photo identification into the app, which will offer ID assistance in the field.
To add a species to the automated photo ID feature, we need at least 150 images to train the system (and more is better). Training a computer to identify birds is a lot like training people to do the same. Broadly speaking, there are two steps. First, you need to detect the bird. Second, you need to identify it. With this in mind, and we are currently working on a computer vision project to train Merlin to detect birds in an image. You can help with this effort by making quick notes on birds in images. If this sounds of interest, give MerlinVision a spin—it’s a great way to see recently uploaded photos while helping improve Merlin at the same time.