A New Flame Image Edge Detection Algorithm for Reconstruction of the Internal Temperature Field in the Furnace

For a coal-fired boiler, it is always a difficult problem to reconstruct the internal temperature field. Due to the complex environment inside the furnace, the traditional reconstruction methods are not suitable for the furnace, so the current mainstream methods are based on the characteristics of the furnace internal flame image to reconstruct the temperature field. But the flame image inside the furnace is easily affected by external factors, especially the background. Therefore, it is very important to detect the edge of the flame image and extract the separate flame image for the subsequent temperature field reconstruction. For this purpose, a new edge detection method is proposed in this paper. First, a new rule is designed for color image to gray image. The rule can improve the gray difference between flame target and background area. Then the threshold segmentation algorithm is used to preprocess the flame gray image, and a new weighted edge image calculation method is designed to capture more details of the flame edge. Finally, an adaptive double threshold algorithm is used to remove the false edge in the flame edge. The simulation results show that the method is effective and can detect the continuous flame edge very well. It is of great practical value for the temperature field reconstruction in the furnace.