A two-fly tracker that solves occlusions by dynamic programming: computational analysis of Drosophila courtship behaviour

Abstract

This paper introduces a two-fly tracker which focuses on an approach to model and to solve occlusions as an optimization problem. Automated tracking of genetic model organisms is gaining importance since geneticists and neuroscientists have biological tools to systematically study the connection between genes, neurons and behaviour by performing large-scale behavioural experiments. This paper is about a fly tracker that provides automated quantification for such functional behaviour studies on Drosophila courtship behaviour. It enables measurement and visualization of behavioural differences in genetically modified fly pairs. The developed system provides solutions for all major challenges that were identified: arena detection, segmentation, quality control, resolving occlusions, resolving heading and detection of behaviour events. Among all challenges especially resolving occlusions turned out to be of particular importance and huge effort was invested to resolve that particular problem. Our tests show that our system is capable to identify flies through an entire video with an accuracy of 99.97%. This result is achieved by combining different types of local methods and modeling the global identity assignment as an optimization problem.

Algoritmos Para Detecção Automática De Múltiplos Objetos E Segmentos Em Animais De Laboratório

“…Tais experimentos podem possuir mais de um animal no ambiente, estes estudos revelam importantes características sobre o comportamento social [1] . Além disto, a detecção de movimentos discretos de segmentos de um mesmo animal é essencial em estudos sobre sistemas locomotores [2].…”

Targeted manipulation of neuronal activity in behaving adult flies

“…Therefore, machine vision-based behavioral tracking algorithms have been developed to improve annotation consistency, throughput rate, and quantitative analysis of behavior (Anderson and Perona 2014;Egnor and Branson 2016). Such algorithms track multiple statistics of the trajectories of flies and/or their body parts through time (e.g., translational speed, angular speed, or distance to another fly), which can be used to define classifiers for particular behaviors (Dankert et al 2009;Branson et al 2009;Robie et al 2010;Straw et al 2011;Tsai and Huang 2012;Donelson et al 2012;Schusterreiter and Grossmann 2013;Ardekani et al 2013;Bidaye et al 2014;Dell et al 2014;Berman et al 2014). These statistics can also be fed into supervised machine-learning algorithms, such as the Janelia Automatic Animal Behavior Annotator (JAABA), where the researcher can train new behavior classifiers by manually annotating a small amount of video based on their own intuition about the behavior (Branson et al 2009;Kabra et al 2013).…”