30 September 2011 Fire flame detection using color segmentation and space-time analysis
Author Affiliations +
Proceedings Volume 8285, International Conference on Graphic and Image Processing (ICGIP 2011); 82851J (2011) https://doi.org/10.1117/12.913421
Event: 2011 International Conference on Graphic and Image Processing, 2011, Cairo, Egypt
Abstract
This paper presents a fire flame detection using CCTV cameras based on image processing. The scheme relies on color segmentation and space-time analysis. The segmentation is performed to extract fire-like-color regions in an image. Many methods are benchmarked against each other to find the best for practical CCTV camera. After that, the space-time analysis is used to recognized fire behavior. A space-time window is generated from contour of the threshold image. Feature extraction is done in Fourier domain of the window. Neural network is used for behavior recognition. The system will be shown to be practical and robust.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Miti Ruchanurucks, Miti Ruchanurucks, Praphin Saengngoen, Praphin Saengngoen, Theeraphat Sajjawiso, Theeraphat Sajjawiso, } "Fire flame detection using color segmentation and space-time analysis", Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 82851J (30 September 2011); doi: 10.1117/12.913421; https://doi.org/10.1117/12.913421
PROCEEDINGS
7 PAGES


SHARE
RELATED CONTENT

Webcam classification using simple features
Proceedings of SPIE (March 11 2015)
Arabic sign language recognition based on HOG descriptor
Proceedings of SPIE (February 07 2017)
A new method for feature extraction and matching using the...
Proceedings of SPIE (November 01 2004)
Machine vision techniques for rose grading
Proceedings of SPIE (November 28 1993)
Identification of simple objects in image sequences
Proceedings of SPIE (August 16 1994)

Back to Top