Hi Tom and Frank Have you heard of the tiny antennae radar system that Rothamsted Research Institute (www.iacr.ac.uk) have used to track the movement of butterflies in fields? Depends how big your Dutch bees are and whether, as you say, a greenhouse structure interferes with radar.
Regarding automatic video tracking software (e.g. Ethovision by Noldus, a Dutch company!), you really have to make the observation 'arena' work for the tracking system rather than making the tracking system work in your greenhouse arena. So you need to simply the captured image as much as possible to get the most out of the software. Examples might include shooting from above on low lying flowers, arranging flowers in neat and discrete groups that the software can readily identify as separate zones of interest, removing all other clutter, maximising contrast in the video (e.g. putting white sheets on the floor, playing with the lighting - really make the bees and flowers stand out). You could also try filming under UV light with the right kind of camera - perhaps even dusting bees in a fluorescent powder to aid their detection (different colours for different bees?). Why not combine the RFID with video tracking? Plate readers under the flowers and RFID tags on the bees, at the hive and perhaps around holes in sheet partitions bisecting the greenhouse into sections (to give an idea of routes). It may be useful to contact Bill Budenburg at Tracksys UK (www.tracksys.co.uk) who is developing a RFID/video combination system with rodent researchers at Liverpool University. I think redesigning the layout of the greenhouse is key to whatever system you go for - 40m sq may be a bit ambitious, especially for video. Good luck Nic ______________________________________________________ Dr Nic Malone www.nicmalone.com ______________________________________________________ On 12 Jun 2007, at 11:48, Langevelde, Frank van wrote: > Hi all, > We are looking for methods to track bumblebees in a relatively small > room/greenhouse (about 40 by 40 m). We would like to record the > location > of a bumblebee every second as precise as possible, including its > visits > to flowers. We are aware of the previous use of harmonic radar and > have > investigated the use of RFID tags. Standard RFID readers, however, do > not have the required range to follow bumblebees during flight, > although > we could use them to register visits to flowers. We are not quite sure > as to whether harmonic radar will work within greenhouses. We have > also > considered the use of cameras and video tracking software, but the > resolution of the cameras makes any recognition of a bumblebee in a 40 > by 40 m greenhouse very difficult. > > Do you have any suggestions what we could use? > > Many thanks, > > Tom Huisman and Frank van Langevelde > > > Resource Ecology Group > Department of Environmental Sciences > Wageningen University > P.O. Box 47 > 6700 AA Wageningen > The Netherlands > http://www.reg.wur.nl/UK/
