2D Otsu Segmentation Algorithm Based on Simulated Annealing Genetic Algorithm for Iced-Cable Images

Abstract

In order to distinguish the ice-covered cables from their backgrounds in low contrast images, 2D Otsu algorithm based on simulated annealing genetic algorithm was proposed. Firstly, the principles of 2D Otsu method were introduced as well as its advantages and disadvantages. Secondly, the principles of genetic algorithm and its application in searching the optimal solution of 2D Otsu were discussed. And then the improvement of genetic algorithm by simulated annealing was implemented, which can overcome the disadvantages of genetic algorithm and obtain better results. Finally, experimental results show that the proposed algorithm can obtain better segmentation results and quicker searching speed to low contrast iced-cable images gotten in snow weather. It has good practicability in the on-line monitoring of electricity transmission wires.