Image edge detection for Furnace Flame Based on Section Mutation Glowworm Swarm Algorithm

Abstract

In view of the gray change of furnace flame, the problem of edge detection cannot be detected by the traditional edge detection algorithm. This paper proposes the phase variation of firefly algorithm to detect image edge of power plant boiler furnace flame, the algorithm adopts the method of phase variation of firefly algorithm is improved so that it can be bound out of the local extreme point, converge to the global extremism. Firstly, the gradient value matrix of gray scale image is obtained by first order derivative of gray level image matrix, and then using the phase variation firefly algorithm, which is based on the search area to find other firefly, and to realize the maximum gradient of search image. Edge detection simulation results show that the algorithm not only can detect the image edge accurately, but also improve the convergence speed and has good adaptability, therefore its application in image edge detection is feasible and effective