Intelligent Video-based Smoke Detection

• Main Authors: Jian-Wei Huang, 黃建瑋
• Other Authors: Wei-Chang Du
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/38564911909343106983

碩士 === 義守大學 === 資訊工程學系 === 102 === In order to have a healthy living environment, smoking areas are often set up in certain places. However, a few people still like to smoke in public places with no-smoking signs and then many non-smokers have possible dangers of second-hand smoke. With the purpose of protecting the rights and interests of non-smokers, this thesis is focused on a real-time video surveillance system which is able to detect cigarette smoke through analyzing cigarette smoke color features from successive images taken by the surveillance cameras. In combination with the function of face detection to automatically identify smoking behavior, it is expected to judge whether there is any smoking behavior conducted in public places.