We introduce a novel video processing method called
Detect+Track that significantly enhances the accuracy
and robustness of animal tracking. We use a range
of computer vision algorithms (e.g. object detector
and tracker, optical flow, parallel plane homology)
and computational geometry techniques (e.g. Voronoi
tessellation) to analyse the movement behaviour of
fish in response to experimental stimuli. We show
our method overcomes some of the limitations of
existing tools and provides a more reliable solution
for complex experimental conditions. Our method
was developed using a behavioural experiment which
involved tracking a fish’s trajectory through a field of
obstacles. This problem motivated our development
of a set of tools that: (a) measure an animal’s trajectory,
(b) record obstacle position, and (c) detect when the
fish passed through "virtual gates" between adjacent
obstacles and/or the aquarium wall. Our workflow
is divided into several discrete steps, and provides
a set of modular software building blocks that can
be adapted to analyse other experimental designs.
A detailed tutorial is provided, together with all the
data and code required to reproduce our results and
enable future innovations in behavioral tracking and
analysis.
motion analysis
,tracking
,detection
,animal behaviour
,animal movement
