Description

To reduce loss of life and property from fire, an early warning is an imperative. A fire detection system based on light detection and analysis is proposed in the paper. This system uses HSV and YCbCr color models with given conditions to separate orange, yellow, and high brightness light from background and ambient light. Fire growth is analyzed and calculated based on frame differences. The overall accuracy from the experiments has been greater than 90%. Flame plays a more and more important role in the fire recognition, and there are many scholars who do many in depth researches about flame detection and put forward a number of algorithms. For example, Healey achieves the flame recognition according to the flame color. Some researchers also take the shape of the fire flame into the criterion of flame recognition. flame contour’s Fourier transform coefficients as a template, recognize the Fourier coefficients of combustion process and determine the flame region . Though the recognition rate of the method is high, but it is demanding environment and in is not suitable for the situation of small power combustion or stable burning. Li J takes the first-order matrix of the flame as a criterion , and his experiments show that the first-order matrix of the flame is not easily influenced by external environment or burning material, and its rate of change is bigger than the non-flame region, but in complex lighting conditions, the first-order matrix changing rate of non-flame image area is also a great moment, and this makes the algorithm difficult to achieve the right identification.