A new approach to vision-based fire detection using statistical features and bayes classifier
SEAL'12 Proceedings of the 9th international conference on Simulated Evolution and Learning
| Hi-index | 0.00 |
This paper proposes a model for detecting fire captured in video data by combining the methods of correlation coefficient, Gaussian Mixture Model - GMM and turbulent analysis. The method of correlation efficient is used to determine movement objects. We use GMM to cluster fire-colored pixel in the RGB space. The objective of turbulent analysis is to detect the flame of fire. A model built on three above methods will be presented and the experimental results are discussed in Section III.