Multi-level Image Thresholding based on Improved Fireworks Algorithm

Aiming at achieving the optimal multi-level thresholding quickly and effectively for the image segmentation, this paper proposes an improved fireworks algorithm based image segmentation method. The proposed method transforms the multi-level thresholding problem into a multivariate combinational optimization problem and then improved the fireworks algorithm for a better image segmentation. Meanwhile, the Otsu method is adopted as the fitness function, and the global search and the local search in the improved fireworks algorithm method is utilized to achieve multi-level thresholding concurrently and efficiently. The experimental results show that the improved fireworks algorithm based image segmentation method can significantly improve the segmentation efficiency comparing with other swarm intelligence algorithm based image segmentation methods.