Automatic fish segmentation on vertical slot fishways using SOM neural networks

Authors:
Álvaro Rodriguez;Juan R. Rabuñal;María Bermúdez;Alejandro Pazos
Affiliations:
Dept. of Information and Communications Technologies, Faculty of Informatics, University of A Coruña, A Coruña, Spain;Center of Technological Innovation in Construction and Civil Engineering (CITEEC), University of A Coruña, A Coruña, Spain;Dept. of Hydraulic Engineering, ETSECCP, University of A Coruña, A Coruña, Spain;Dept. of Information and Communications Technologies, Faculty of Informatics, University of A Coruña, A Coruña, Spain
Venue:
IWANN'13 Proceedings of the 12th international conference on Artificial Neural Networks: advances in computational intelligence - Volume Part I
Year:
2013

Citing 5
Cited 0

Colour image segmentation by modular neural network

Pattern Recognition Letters
Particle Filter-Based Predictive Tracking for Robust Fish Counting

SIBGRAPI '05 Proceedings of the XVIII Brazilian Symposium on Computer Graphics and Image Processing
Object tracking: A survey

ACM Computing Surveys (CSUR)
Automatic segmentation of overlapping fish using shape priors

SCIA'07 Proceedings of the 15th Scandinavian conference on Image analysis
Color clustering and learning for image segmentation based on neural networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Vertical slot fishways are hydraulic structures which allow the upstream migration of fish through obstructions in rivers. The appropriate design of these should consider the behavior and biological variables of the target fish species and currently existing mechanisms to measure the behavior of the fish in these assays, such as direct observation or placement of sensors on the specimens, are impractical or unduly affect the animal behavior. This paper studies the application of Artificial Neural Networks to the problem of automatic fish segmentation in vertical slot fishways. In particular, SOM Neural Networks have been used to detect fishes using visual information sampled by an underwater camera system. A ground true dataset was designed with experts and different approaches were tested providing promising results.