Digital field guide to trees and leaves

[Tom Rankin]

"THE traditional way to identify an unfamiliar tree is to pull out a field guide and search its pages for a matching description. One day people may pull out a smartphone instead, photographing a leaf from the mystery tree and then having the phone search for matching images in a database."

"Software to identify leaves by searching a field guide on a PC or a phone could be useful not only to hikers but also to scientists compiling data."

http://www.nytimes.com/2009/05/10/business/10novel.html?partner=rss&emc=rss

 

[arghman]

Interesting. This would be useful for narrowing down preliminary choices of species. As a definite answer it's probably not practical. The example used in the photo is kind of an ironic one: it's very easy to determine from leaf shape that a tree is an oak (Quercus sp.), but often very difficult to go to a particular species. Quercus rubra and Quercus velutina and Quercus coccinea (red, black, scarlet oak) have very similar leaves with variation in leaf type that blurs the line between species, so you have to go by bark and/or acorn type. Plus the oaks hybridize The only reason I know this is because I've been trying to find a clear example of Q. coccinea and have asked several professional botanists. And that's just an example for confusion/uncertainty in the Northeastern US... if all I know is that the tree is somewhere in the US there are lots of oak species and who knows which one it is.

Still the general idea has a lot of potential as a tool if done right. I'd love to have something that I can take into the field that gives me confidence levels for various species based on location data and pattern matching algorithms for leaf shape, flower color/petals, and other visual cues.

edit: p.s. if you want to get qualitatively and quantitatively nitpicky about leaves, see the Manual of Leaf Architecture.

 

[Tom Rankin]

Woodn't you know it!

LEAF it to someone like you to nit-'pick' this post!