Merlin Bird Photo ID can identify species of birds just from the photos


Merlin Bird Photo ID

Cornell University together with Visipedia develops a site that has amazing ability. Addressed in AllAboutBirds.org/photoID, the site that's made by Cornell Lab of Ornithology has smart capabilities that are generated by artificial intelligence technology.

Thanks to this technology, the site is able to recognize the species of birds from the photos that are uploaded by users. Named Merlin Bird Photo ID, this site is a breakthrough in the world of computer vision. This can happen because of the hard work of researchers group from Cornell Tech and the California Institute of Technology.

In total, the site is able to recognize 400 species of birds. The only drawback is, for now this site is only able to identify the birds that are found in the United States and Canada.

Researchers have an opportunity to explain the workings and results of this technology in the Computer Vision and Pattern Recognition (CVPR) in Boston on June 8, 2015.

Merlin Bird Photo ID

This site can be used easily, when users upload photos to the site, they have to make a box-shaped area in the main image and then click on the bird's tail, beak and eyes. The site will also ask users to include details of where and when the photos were taken.

Once all required information is inputted, the site will show off its ability, the site will search, shuffle and analyze a number of photos taken by bird watchers, including database of 70 million images at ebird.org and try to find some pictures that fit. As a result, the site will display a list of the most suitable with the uploaded photos along with the voice of the bird.

A professor from Computer Science at Cornell Tech said that the technology in the Merlin Bird Photo ID is capable of processing images more efficiently than humans. This tool is organizing, indexing and matching the visual information in the form of colors and shapes.

The basis of this technology is accompanied with a learning machine, so the more it is used, the smarter it will be. Looking ahead, the researchers want to bring the same technology in the form of application that can be used by the users in mobile devices for free.