ALL ABOUT BIRDS: IDENTIFICATION & BIRDSONG with Cornell Lab of Ornithology


The outstanding  Cornell Laboratory of Ornithology site has a comprehensive bird ID section. I recently posted about the new ID resource MERLIN for which they are inviting user-testing to help them perfect it. There are also 4 excellent videos concentrating on the principles ways in which birds can best be ID’d: size & shape; colour pattern; behaviour; and habitat

I haven’t yet cracked embedding the 4 videos so while the backroom boffins in the cerebral cortex puzzle it out (alongside ‘The Purpose of Hornets – What & Why?”) the best thing is to give a direct link to the website feature

CLICK LOGO ===>>> 

The same link will offer you the chance to download 5 bird songs – and now a second set of 5, part of the Great Backyard Bird Count currently underway – see details below. You may need to click on the images to make them legible…

‘MERLIN’ – NEW ONLINE BIRD ID TOOL FROM CORNELL LAB FOR ORNITHOLOGY


STOP PRESS January 2014

Cornell are now re-promoting ‘MERLIN’ with a free App and a video to go with it.

 has produced a new proactive bird identification gizmo called MERLIN (CLICK for direct link). They are trying to build up a user-friendly ID ‘wizard’ using the sort of variable descriptions that people like me use to describe birds they don’t recognise. Perhaps we’ve all been there – “well, it was a medium-size greyish bird, but I think it had white under the wings. Or maybe a lighter grey. And a sort of white streak on its head. Actually the bird was more bluey-grey…”etc. Merlin seeks to iron out the variations using AI, by showing a bird and asking a number of questions to get users to describe the colouring  they are looking at. I tried it with a teal, and it worked first time.

Gradually, the input of descriptions for each bird will be analysed, so that future users are more like to get a correct ID based on their description, even if others might describe the bird differently. With any luck it will also improve the chance to ID that pretty bird seen fleetingly at a distance. It’s worth trying this out even if you are a serious birder, because each ‘attempt’ adds to the picture. And anyway, there’s a very slight element of a game here – will the computer get it right?