Automated tracking of Drosophila specimens

Rubén Chao, Germán Macía-Vázquez , Eduardo Zalama Jaime Gómez-García-Bermejo and José Ramón Perán

Instituto de Tecnologías Avazadas de la Producción. University of Valladolid (Spain)

Abstract:

The fruit fly Drosophila Melanogaster has become a model organism in the study of the neurobiology and behaviour patterns. The analysis of the way the fly moves and its behaviour is of great scientific interest for research on aspects such as drug tolerance, aggression or ageing in humans. In this article a procedure for detecting, identifying and tracking numerous specimens of Drosophila by means of computer vision based, sensing system is presented. This procedure allows collecting dynamic information about each specimen at each time, and then characterizing quantitatively its behaviour. The proposed algorithm operates in three main steps: pre-processing step, detection and segmentation step, and tracking shape. The pre-processing and segmentation steps allow dealing with some limits of the image acquisition system and some visual artifacts (such as shadows and reflections). The improvements introduced in the tracking step allow the problems corresponding to identity loss and swaps, caused by the interaction between individual flies, to be solved efficiently. This way, a robust method that compares favourably to other existing methods is obtained.

Keywords: moving object sensing; computer vision; tracking; prediction methods.

Material:

· Program and libraries¹ : https:/github.com/RubenChao/tracking-drosophila

Raw Videos:

· Video 2

· Video 3

· Video 4

· Video 5

· Video 6

· Video 7

· Video 8

· Video Descriptions

Execution Example:

· Video example

¹ Code license GNU GPL v3

Program Inteface