idBee - Automatic Bee Identification from Wing Venation

About 35% of the world's crops and 80% of the world’s flowers are pollinated by bees! Most bees that are farmed for pollination are European Honeybees, whose physique and behavior makes them very efficient. But native bees can share in the pollination effort, and some may be more efficient.

Recently, honeybees have been dying off at an extraordinarily high rate, and no one is quite sure why. Researchers call it Colony Collapse Disorder, and the problem is serious because of our dependence on honeybees to pollinate food crops. While many are researching the cause of Colony Collapse Disorder, research into native bees may uncover more productive alternatives to the European honeybee. Finding the cause and potential native alternatives involve tracking wild bee populations and habits.

There are many species of bees, more than 500 in Wisconsin alone, but it's not easy to tell which species an individual belongs to.

While bee species identification is essential to research, identifying the species of a bee can be expensive and time-consuming. Since there are few experts who can reliably distinguish many different species, bees must be captured and sent to an expert, which may take several months. Bee research can be excruciatingly slow.

Rather than capturing, killing, and sending the bees off to be catalogued, imagine an iPhone app that lets graduate students and researchers identify bees in the field. One could simply take a photo of a bee and instantly record its species and location. Research could be conducted much faster, and identified bees could be released back to nature.

That's the problem that the idBee project at UW-Madison is trying to solve. We currently have a system capable of identifying over 20 bee species from wing photos with an accuracy above 90%.

We are in the process of adapting our system to run on this website. So, in the near future, researchers can upload photos of wings for identification or for increasing the system accuracy and/or number of species.


Further Reading
An Automated Approach to Bee Identification from Wing Venation by Christopher Hall
Article in UW-Madison ECE Innovations Magazine by Mark Riechers
UW-Madison ECE News article by Mark Riechers

Get Involved
We need many more photos of bee wings with known species to increase the number of species the system supports.
If interested, please email Mark Lenz at mmlenz at wisc domain.

Team
William Sethares (PI)
Mark Lenz
Chuck Hatt
Chris Hall (now at Sandia)