35
Solution:
Teaching an animal a task, like sign language or maze training, was usually thought to only be possible in larger brained animals. However, research of the worker bee has given techies a whole new host of ideas.
The worker bee,“ she,” is an altruistic being. She goes out to collect nectar from flowers and returns to the hive to feed the entire colony of bees. So when researchers sought to train bees to do complex tasks, like recognize 3D facial images, there was an obvious benefit.
Problem:
“ There have been a lot of difficulties in producing algorithms that can reliably recognize faces when there is a change in viewpoint. For example, a field test of an artificial intelligence face recognition system at a major airport in the United States produced so many false positives that the system was practically not usable in crowded situations – which is when it is needed to work best. Knowing how biological systems cope with these types of visual challenges can potentially provide some insights for software developers about new or novel types of approach.”( From an interview with Australian bee researcher Adrian Dyer, available from the Humboldt Foundation)
Once the bee’ s honey stomach was full, she would return to the hive, deposit her nectar … and go back to the experiment. That’ s a quick turnaround time over say, a rat, who has to learn through food … and then gets full.
Tech researchers have begun to examine the bee’ s vision, which sees one pixel per ommatidia( or single eye), and are looking at pairing that pixel with an algorithm relating it back to“ center command” of a larger computer system to be ingested and processed to form whole images.
Like nectar, these millions of small images will be reformulated into a larger image pattern and used in everything from door locks to security controls.… pretty cool stuff from a furry little bee, eh?
The way a bee sees could be the next advancement in the way humans use facial recognition technologies!