Particle/Object Tracker

[Miss Judit KIBEDI]

Dear all,

I am trying to use automated tracking to analyse the motion of field
crickets (responding to sound), as opposed to manual tracking which
though serves the purpose is quite tedious for the # of
frames/sequence and # of sequences I am analysing.
I may be wrong but for my purposes it seems Object Tracker is the most
appropriate.
Unfortunately, I can't quite get the program to recognize the cricket
as the object to track.
I have followed the steps provided in Ross Lang's test stack for
tracking termites and have tried to apply the same to my own sequences
of a moving cricket.
However, during conversion from the greyscale to binary image sequence
(where background is subtracted from the image stack) the cricket is
not picked up.
When I look at adjusting the threshold, the cricket has disappeared
entirely, instead of being the only part of the image sequence to
remain, as happens in the termites example.
The only difference I can think of between my own images and that
provided in the test stack of termites, is that my cricket is black
(termites = white) and the background is white (termite background =
grey).
Does anyone know if this contrast is the issue? Or is it that the
outline of the object (cricket) needs to be well defined? Or any other
suggestions?
Has anyone applied automated tracking on insects (or other macro-
organisms)?


Judit Kibedi
PhD Student
[hidden email]
School of Biomedical Sciences
Physiology & Pharmacology
UQ - St Lucia
Brisbane 4072
Australia
ph1: [phone]
ph2: [phone]
mob: [phone]