Computer Vision-based Fire Detection using Enhanced Chromatic Segmentation and Optical Flow Model
Kaliyev D. Shvets O. Györök G.
2023Budapest Tech Polytechnical Institution
Acta Polytechnica Hungarica
2023#20Issue 627 - 45 pp.
Forests are one of the most important natural resources in the world. However, the occurrence of forest fires will burn plants and kill animals. Emergency incidents and events of fires can be dangerous and require quick and accurate decision-making. The use of computer vision for fire detection can provide an efficient solution to deal with these situations. We propose a combined method for detecting fire from a video sequence in monitoring and early fire detection operations. The method is based on motion detection methods, chromatic analysis and image segmentation. To improve the efficiency of the system, image pre-processing algorithms are proposed, and optical flow methods are used to detect the motion in fire video frames. We calculate the growth rate of the fire to reduce false-alarms. The proposed method has been tested on a very large dataset of fire videos captured by drones. It is assumed that the algorithm program is run on a computer that receives data from the camera of the drone that scans the required area. Experimental results demonstrate the effectiveness of our method while keeping their precision compatible with the existing methods.
computer vision , early fire detection , image processing
Text of the article Перейти на текст статьи
D. Serikbayev East Kazakhstan Technical University, Faculty of Information Technology and Intelligent Systems, A. K. Protazanov Str. 69, Ust-Kamenogorsk, 070004, Kazakhstan
Óbuda University, Alba Regia Technical Faculty, Budai út 45, Székesfehérvár, H-8000, Hungary
D. Serikbayev East Kazakhstan Technical University
Óbuda University
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026