Sign in
Back
Using the BPSO algorithm in image segmentation for dynamic thresholding
Conference proceeding

Using the BPSO algorithm in image segmentation for dynamic thresholding

L. Djerou, H. Dehimi, N. Khelil and M. Batouche
2009 Fourth International on Conference on Bio-Inspired Computing, pp.1-6
10/2009
DOI: https://doi.org/10.1109/BICTA.2009.5338061

Abstract

Biomedical optical imaging Computer science Genetic algorithms Image analysis Image processing Image segmentation Nonlinear optics Optical character recognition software Particle swarm optimization Text analysis
A new dynamic Multilevel Thresholding method (DMTBPSO), based on Binary Particle Swarm Optimization and Otsu's method, is proposed. This approach can automatically determine the appropriate number of thresholds as well as the adequate threshold values with saving a significant amount of computing time which is independent from the number of thresholds. The proposed method is validated by illustrative examples, comparison with the Otsu's method shows its efficiency.

Metrics

1 Record Views

Details