Insight into how bees see may help improve artificial intelligence systemsJanuary 24th, 2009 - 4:27 pm ICT by ANI
Washington, January 24 (ANI): A study revealing honeybees can learn to recognise human faces even when seen from different viewpoints has raised the possibility of improving artificial intelligence systems and computer programs for facial recognition.
Adrian Dyer, a leading bee researcher at Australia’’s Monash University, said that his research could be applied in the areas of new technology, particularly the development of imaging systems.
“What we have shown is that the bee brain, which contains less than 1 million neurons, is actually very good at learning to master complex tasks. Computer and imaging technology programmers who are working on solving complex visual recognition tasks using minimal hardware resources will find this research useful,” he said.
“Most current artificial intelligence (AI) recognition systems perform poorly at reliably recognising faces from different viewpoints. However the bees have shown they can recognise novel views of rotated faces using a mechanism of interpolating or image averaging previously learnt views,” he added.
The findings show that despite the highly constrained neural resources of the insects, their ability has evolved so that they”re able to process complex visual recognition tasks.
For their study, the researchers individually trained different groups of free flying bees with a sugar reward for making correct choices, or alternatively the bees were punished with a bitter tasting solution for incorrect choices.
Faces were presented on a vertical screen, and the bees slowly learnt to fly to the correct target faces.
The researchers said that over the course of a day, a bee brain learned a complex task.
When they later tested the bees in non-rewarded tests, only those insects that had experience multiple views faces at both 0 and 60 degrees were able to solve a novel rotational angle of 30 degrees.
Dr Dyer said that the discovery could help answer a fundamental question about how brains solve complex image rotational problems by either image averaging or mentally rotating previously learnt views.
“Bee brains clearly use image interpolation to solve the problem. In other words, bees that had learnt what a particular face looked like from two different viewpoints could then recognise a novel view of this target face. However, bees that had only learnt a single view could not recognise novel views,” he said.
“The relationships between different components of the object often dramatically change when viewed from different angles but it is amazing to find the bees” brains have evolved clever mechanisms for problem solving which may help develop improved models for AI face recognition systems,” he added.
The study, performed over two years in Australia and Germany, has been published in the journal PLoS ONE. (ANI)
- Bees' face recognition mechanism can improve artificial intelligence systems - Jan 27, 2009
- Bees may help improve robotic vision, says study - May 14, 2012
- Artificial bee eye could improve visual navigation in mobile robots - Aug 07, 2010
- How do bees recognise human faces? - Jan 29, 2010
- Bees can recognise human faces - Jan 30, 2010
- Bees beat computers in solving complex problems - Oct 26, 2010
- 'New vision' tweaks drones' ability to fly precisely - Dec 13, 2011
- Soon, robots that ask for help to recognise objects - Sep 18, 2009
- Honey bee? No, aircraft on autopilot! - Dec 02, 2010
- Scientists gain insight into 'expertise' - Jul 24, 2012
- How we combine faces and voices to recognise people - Mar 10, 2011
- Google conjures artificial brain with billion connections - Jun 26, 2012
- Paper wasps have flair for recalling faces - Dec 04, 2011
- Tiny insect brains can perform remarkable feats of mental gymnastics - Nov 18, 2009
- Some honey bees tend to take risks - Mar 11, 2012
Tags: artificial intelligence systems, correct choices, correct target, course of a day, facial recognition, fundamenta, hardware resources, honeybees, human faces, imaging systems, imaging technology, incorrect choices, interpolating, minimal hardware, monash university, novel views, recognition systems, recognition tasks, vertical screen, visual recognition