Apparatus of detecting smoke using subband energy ratio in wavelet transform domain

摘要

A method and apparatus for detecting smoke using a sub-band energy ratio in a wavelet transform domain. The smoke detection method according to the present invention includes the steps of converting an RGB input image into a brightness image I (t), comparing a difference between a current brightness image I (t) and a previous brightness image I (t-1) generating a binary image G1 composed of a background image and an object image, divided into blocks of a predetermined number of MxN image corresponding to the object image of the binary image G1, and a segmented block step wavelet transform to DC sub-band energy E calculating _L and high frequency sub-band energy E_H, and the DC sub-band energy E_L and the high frequency sub-band energy E ratio and the DC sub-band energy E_L and the high frequency sub-band energy E_H of the image in sequential between _H And detecting smoke in the object image based on the respective ratios. According to the present invention, it is possible to reliably discriminate between moving objects and smoke by using the energy values of sub-bands of the wavelet coefficients obtained by using the wavelet transform and the relationship therebetween.