A Method of Fire and Smoke Detection Based on Surendra Background and Gray Bitmap Plane Algorithm

摘要

Fire and smoke detection is an important technology in the fire monitoring system. This paper presents a method of fire and smoke detection based on Surendra background and gray bitmap plane algorithm. In our method, the dynamic background based on Otsu adaptive threshold and Surendra background first is modeled, and then the moving target is extracted by the continuous cumulative average of moving target area and the mathematical morphology processing method, the smoke area is extracted by color feature, and finally the fire area is extracted by OHTA color space and gray bitmap plane algorithm. The experimental results show that our method can effectively detect the fire area in the moving vehicle scene, the red plastic bag waving scene, the red balloon fluttering scene, or the moving vehicle similar to fire color.