BIBBA NEWS

BIBBA has co-operated with UK University of Hull on a project to automate the process of wing morphometry. The process works in a manner similar to that used to detect abnormal cells in medical samples

Many BIBBA members utilise wing morphometry to determine the sub species of Apis mellifera when they are selecting breeder colonies for queen rearing. By examining the wing venation and measuring the dimensions of particular cells and the location of nodes in relation to one another, an excellent indication of the sub species can be determined. The data from say 30 wings is normally plotted on a scatter graph using a spreadsheet program like Microsoft Excel. To some, this sounds complicated but in reality it is a very simple, although time consuming, process. Everybody who uses the technique finds it a valuable resource in the development of native bees in areas where hybrids exist.

BIBBA was responsible for bringing the technique to the attention of beekeepers in the UK and Ireland by various publications. The tools and explanation of morphometry techniques are detailed in “Breeding Techniques and Selection for Breeding of the Honeybee” (£5.50 inc UK p&p) , also “Breeding Better Bees using simple methods” (£4.00 including UK p&p) both published by BIBBA and available from postal sales

At present, the measurement of cubital index (CI) and discoidal shift (DS) in Apis mellifera is carried out by projecting an image of 15 forewings from a 35mm slide projector and performing the measurements manually.  Whilst this is accurate, it is slow and time consuming. 

Claire Chamaillard’s Masters degree project at Hull University was to develop a computer program to automate the measurements from the 35mm slide containing 15 honey bee wings which is scanned into the computer using a 35mm computer slide scanner.  The program finds each wing on the scanned image and produces 15 individual images; these are then separately analysed by following the wing veins to find the correct locations on each wing.  The program then determines DS and CI and stores the results for each wing for further analysis.  The overall effect is to take a slide of 15 wings and produce a table of 15 values of DS and CI.  It is also possible for the user to exclude any wings that are not perfect from the image to allow for any potential problems that may occur (e.g. dirt or serious wing misalignment).  The computer processing time is typically less than 2 minutes per slide.

The computer program is currently being completed and will run on any reasonable speed PC with a 35mm slide scanner such as the Canoscan 2700F.

The project is only one of a number of projects in the environmental bioinformatics laboratory in the School of Engineering that are related to image processing and analysis of insect and animal sounds.  Examples include - automated identification of insect species from their sounds, discrimination  of quarantine and non-quarantine fungal diseases using image processing and a recently started project to measure the stress level of pigs in transport using biotelemetry.

BIBBA members are excited by the breakthrough and look forward to using the computer software . Other co-operations and honey bee projects have been discussed.

M.Sc. Supervisor for Claire Chamaillard is:

Dr David Chesmore, FRES, MIAgrE
School of Engineering
Faculty of Science
University of Hull
Hull, HU6 7RX
England
Tel: +44 1482 465062  Fax: +44 1482 466664 
e.d.chesmore@eng.hull.ac.uk