This title appears in the Scientific Report :
2017
Please use the identifier:
http://dx.doi.org/10.1109/ICMLA.2016.0133 in citations.
Please use the identifier: http://hdl.handle.net/2128/13829 in citations.
Automatic Object Detection using DBSCAN for Counting Intoxicated Flies in the FLORIDA Assay
Automatic Object Detection using DBSCAN for Counting Intoxicated Flies in the FLORIDA Assay
In this paper, we propose an instrumentation andcomputer vision pipeline that allows automatic object detectionon images taken from multiple experimental set ups. We demon-strate the approach by autonomously counting intoxicated fliesin the FLORIDA assay. The assay measures the effect of ethanolexpo...
Saved in:
Personal Name(s): | Bodenstein, Christian (Corresponding author) |
---|---|
Götz, Markus / Jansen, Annika / Scholz, Henrike / Riedel, Morris | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Published in: |
ISBN 978-1-5090-6167-9 |
Imprint: |
IEEE
2017
|
Physical Description: |
746 - 751 |
DOI: |
10.1109/ICMLA.2016.0133 |
Conference: | 15th IEEE International Conference on Machine Learning and Applications, Anaheim (USA), 2016-12-18 - 2016-12-20 |
Document Type: |
Contribution to a book Contribution to a conference proceedings |
Research Program: |
Data-Intensive Science and Federated Computing |
Link: |
Restricted OpenAccess |
Publikationsportal JuSER |
Please use the identifier: http://hdl.handle.net/2128/13829 in citations.
In this paper, we propose an instrumentation andcomputer vision pipeline that allows automatic object detectionon images taken from multiple experimental set ups. We demon-strate the approach by autonomously counting intoxicated fliesin the FLORIDA assay. The assay measures the effect of ethanolexposure onto the ability of a vinegar fly Drosophila melanogasterto right itself. The analysis consists of a three-step approach.First, obtaining an image of a large set of individual experiments,second, identify areas containing a single experiment, and third,discover the searched objects within the experiment. For theanalysis we facilitate well-known computer vision and machinelearning algorithms—namely color segmentation, threshold imag-ing and DBSCAN. The automation of the experiment enables anunprecedented reproducibility and consistency, while significantlydecreasing the manual labor. |