Video smoke detection
Feiniu Yuan, PhD, Professor, yfn@ustc.edu
You can down load these videos and the demo software of video smoke detection freely only for research purposes. Any commerical use is not allowed before we agree.
Dr. Yuan has implemented and alway been improving the video smoke detection software since he was a post-doc at the State Key Lab of Fire Science, University of Science and Technology of China from 2004. So the software at least has 20 different methods for video smoke detection. Of course, some of methods are not mature. Even Dr. Yuan does not remember some of methods and how to config the parameters of those methods and operate them. So for easy operation of the software, all the parameters are specified in advance, please press ctrl+D to open a dialog to select an AVI video for video smoke detection using Dr. Yuan's lastest method, which is based on EOH, LBP, and Adaboost, and so on.
Usage of the demo software: (1) Please down load the package of the demo software, unzip the package into a directory, and double click the file of vsd2.0.exe in the directory to run the software. (2) Press ctrl+D to open a dialog to select an AVI video for video smoke detection using an Adaboost algorithm with partition features. (3) You can generate and e-mail your request code to yfn@ustc.edu to obtain an activate code for 30-day trial
Notes: (1) If the demo can not display a video, i.e., black images for all frames, it may mean that your computer does not properly install a corresponding codec that was used to compress the video, so you need to install one. (2) I used the Xvid codec to compress all the smoke and non-smoke videos at this webpage for your testing. So if you download one of the videos for testing, you can download the Xvid codec from www.xvid.org and install it on your computer.
The results of smoke detection can be shown in videos at YouTube:
More resources will be uploaded soon, ...
Acknowledgements: Smoke videos were recorded at the State Key Lab of Fire Science, University of Science and Technology of China when Dr. Yuan was a post-doc at the lab. The traffic video was recorded by a PhD student of Prof. ZHOU Heqin, who was also the PhD supervisor of Dr. Yuan. Other videos were recorded by Dr. Yuan.