Presentation 2009-01-23
Study of the Feature Values for Detection of Smoke by Image Recognition
Kentaro IWAMOTO, Hironori INOUE, Toshihisa TANAKA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We study feature values for the detection of smoke using image recognition. In a previous work, feature values based on statistics and differential values are extracted from a time sequence of grayscale (luminance) images. In this paper, we apply this feature values to the color components of both an input image and its edge image. Then, by simulation we find the best combination of feature values. In addition, we study parameters of SVM which provide better performance in recognition.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) image recognition / support vector machine / machine learning / feature selection
Paper # SIP2008-162,RCS2008-210
Date of Issue

Conference Information
Committee RCS
Conference Date 2009/1/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Radio Communication Systems (RCS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Study of the Feature Values for Detection of Smoke by Image Recognition
Sub Title (in English)
Keyword(1) image recognition
Keyword(2) support vector machine
Keyword(3) machine learning
Keyword(4) feature selection
1st Author's Name Kentaro IWAMOTO
1st Author's Affiliation Department of Electrical and Electronic Engineering, Tokyo University of Agriculture and Technology()
2nd Author's Name Hironori INOUE
2nd Author's Affiliation Department of Electrical and Electronic Engineering, Tokyo University of Agriculture and Technology
3rd Author's Name Toshihisa TANAKA
3rd Author's Affiliation Department of Electrical and Electronic Engineering, Tokyo University of Agriculture and Technology
Date 2009-01-23
Paper # SIP2008-162,RCS2008-210
Volume (vol) vol.108
Number (no) 391
Page pp.pp.-
#Pages 6
Date of Issue