Early recognition of smoke in digital video

Authors:
Julia Åhlén;Stefan Seipel
Affiliations:
University of Gävle, Department of Building, Energy and Environmental Engineering, Gävle, Swedem;University of Gävle, Department of Building, Energy and Environmental Engineering, Gävle, Swedem
Venue:
ECS'10/ECCTD'10/ECCOM'10/ECCS'10 Proceedings of the European conference of systems, and European conference of circuits technology and devices, and European conference of communications, and European conference on Computer science
Year:
2010

Citing 3
Cited 0

Application of Panoramic Annular Lens for Motion Analysis Tasks: Surveillance and Smoke Detection

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 4
Computer vision based method for real-time fire and flame detection

Pattern Recognition Letters
A fast accumulative motion orientation model based on integral image for video smoke detection

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a method for direct smoke detection from video without enhancement pre-processing steps. Smoke is characterized by transparency, gray color and irregularities in motion, which are hard to describe with the basic image features. A method for robust smoke description using a color balancing algorithm and turbulence calculation is presented in this work. Background extraction is used as a first step in processing. All moving objects are candidates for smoke. We make use of Gray World algorithm and compare the results with the original video sequence in order to extract image features within some particular gray scale interval. As a last step we calculate shape complexity of turbulent phenomena and apply it to the incoming video stream. As a result we extract only smoke from the video. Features such shadows, illumination changes and people will not be mistaken for smoke by the algorithm. This method gives an early indication of smoke in the observed scene.