Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network
Aiming at the problems that the traditional optical flow method is not suitable for gas and liquid image detection,this paper proposes a method which uses the optimal mass transmission optical flow as a low dimensional descriptor of the complex process for fire and smoke detection. The detection pro...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | zho |
Published: |
Harbin University of Science and Technology Publications
2017-02-01
|
Series: | Journal of Harbin University of Science and Technology |
Subjects: |
id |
doaj-96ffb7a41eac425fa162edf4102d699e |
---|---|
record_format |
Article |
spelling |
doaj-96ffb7a41eac425fa162edf4102d699e2020-11-24T22:33:41ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832017-02-01869010.15938/j.jhust.2017.01.015Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural NetworkWANG Wei-bing XU QieHAN Zai-boAiming at the problems that the traditional optical flow method is not suitable for gas and liquid image detection,this paper proposes a method which uses the optimal mass transmission optical flow as a low dimensional descriptor of the complex process for fire and smoke detection. The detection process can be abstracted into a problem about the supervised Bayesian classification of spatio-temporal neighborhood pixels;feature vectors are composed of the optimal mass transmission optical flow and R,G,B color channels and the single hidden layer neural network classifier are employed. Finally,we determine the pixel belongs to the flame or belongs to the smoke by the analysis the pixel probability. Experiments show that the proposed method successfully distinguishes smoke and the color-similar cloud,also distinguish between the flame and the flame color-similar background,and has strong robustness.optimal mass transmission; neural network; video detection; supervised Bayesian classification |
collection |
DOAJ |
language |
zho |
format |
Article |
sources |
DOAJ |
author |
WANG Wei-bing XU Qie HAN Zai-bo |
spellingShingle |
WANG Wei-bing XU Qie HAN Zai-bo Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network Journal of Harbin University of Science and Technology optimal mass transmission; neural network; video detection; supervised Bayesian classification |
author_facet |
WANG Wei-bing XU Qie HAN Zai-bo |
author_sort |
WANG Wei-bing |
title |
Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network |
title_short |
Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network |
title_full |
Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network |
title_fullStr |
Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network |
title_full_unstemmed |
Fire and Smoke Detection Transmission Optical Flow Based on the Optimal Mass Method and Neural Network |
title_sort |
fire and smoke detection transmission optical flow based on the optimal mass method and neural network |
publisher |
Harbin University of Science and Technology Publications |
series |
Journal of Harbin University of Science and Technology |
issn |
1007-2683 |
publishDate |
2017-02-01 |
description |
Aiming at the problems that the traditional optical flow method is not suitable for gas and liquid image detection,this paper proposes a method which uses the optimal mass transmission optical flow as a low dimensional descriptor of the complex process for fire and smoke detection. The detection process can be abstracted into a problem about the supervised Bayesian classification of spatio-temporal neighborhood pixels;feature vectors are composed of the optimal mass transmission optical flow and R,G,B color channels and the single hidden layer neural network classifier are employed. Finally,we determine the pixel belongs to the flame or belongs to the smoke by the analysis the pixel probability. Experiments show that the proposed method successfully distinguishes smoke and the color-similar cloud,also distinguish between the flame and the flame color-similar background,and has strong robustness. |
topic |
optimal mass transmission; neural network; video detection; supervised Bayesian classification |
work_keys_str_mv |
AT wangweibing fireandsmokedetectiontransmissionopticalflowbasedontheoptimalmassmethodandneuralnetwork AT xuqie fireandsmokedetectiontransmissionopticalflowbasedontheoptimalmassmethodandneuralnetwork AT hanzaibo fireandsmokedetectiontransmissionopticalflowbasedontheoptimalmassmethodandneuralnetwork |
_version_ |
1725729897262874624 |