Automatically Extracting Drosophila Courtship Behaviour Statistics from Video

Many laboratories work with the fruit fly (Drosophila melanogaster) as the ideal model organism for performing genetic studies. The short life cycle, low maintenance, and controllable population size make it ideal for efficient analysis. However, dealing with the shear volume of assays in a consistent, robust and reliable manner is a major challenge. In this work we present an approach to improve and automate this process using Computer Vision based techniques. We show how this can be performed to extract relevant behavioural statistics which can then be compared favourably against expert human annotations.

[1]  Frank Dellaert,et al.  MCMC-based particle filtering for tracking a variable number of interacting targets , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  S.N. Fry,et al.  A 6000 Hz Computer Vision System for Real-Time Wing Beat Analysis of Drosophila , 2006, The First IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics, 2006. BioRob 2006..

[3]  L P Noldus,et al.  EthoVision: A versatile video tracking system for automation of behavioral experiments , 2001, Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc.

[4]  Leopoldo Altamirano Robles,et al.  Automatization of the mexican fruit fly activity recognition process using 3D and gray-level features , 2006, ACST.

[5]  Á. Háy,et al.  Sex , 1980 .

[6]  K. O’Dell,et al.  The voyeurs’ guide to Drosophila melanogaster courtship , 2003, Behavioural Processes.

[7]  M. Heisenberg,et al.  Evolutionary significance of courtship conditioning in Drosophila melanogaster , 2002, Animal Behaviour.

[8]  Michael H Dickinson,et al.  The influence of visual landscape on the free flight behavior of the fruit fly Drosophila melanogaster. , 2002, The Journal of experimental biology.

[9]  Sergio A. Velastin,et al.  People tracking in surveillance applications , 2006, Image Vis. Comput..