55-6 Leafsnap and Other Real-Time Plant Identification Tools.

See more from this Division: ASA Section: Agronomic Production Systems
See more from this Session: Symposium--Weed ID: Can You Do It? A Robot Can
Monday, October 22, 2012: 3:45 PM
Duke Energy Convention Center, Room 206, Level 2
Share |

David Jacobs, Computer Science, University of Maryland, College Park, MD
I will describe the first mobile app for identifying plant species using automatic visual recognition. The system -- called Leafsnap -- identifies tree species from photographs of their leaves.  Key to this system are computer vision components for segmenting the leaf from an untextured background,  extracting features representing the curvature of the leaf's contour over scale, and identifying the species from a dataset of more than 250 trees. Results from the system are presented on real-world images, showing state-of-the-art performance on the Leafsnap Dataset -- the largest dataset of its kind.  Leafsnap has been used in biodiversity studies and in classrooms, and by hundreds of thousands of nature enthusiasts.
See more from this Division: ASA Section: Agronomic Production Systems
See more from this Session: Symposium--Weed ID: Can You Do It? A Robot Can