ALGORITHM FOR complex SMOKE AND FLAME detection based on video surveillance systems data analysis

Abstract

An algorithm to detect the main signs of fire for video surveillance systems, which allows the detection of smoke, open flames, and their combination on dynamic images is proposed. A distinctive feature of the algorithm is the use of motion estimation, as a common feature, as well as the application of the algorithm of common approaches in the processing of video sequences. A set of experimental results are presented in the paper.

Keywords

• анализ движения
• цветовая сегментация
• вейвлет-преобразование
• контурный анализ