Research On Steel Strip Image Segmentation Algorithm Based On Particle Swarm Optimization

A method based on particle swarm optimization (PSO) for steel strip image segmentation was presented. Considered the traditional markov method is hard to get good effect in global optimization solution, the particle swarm optimization is used to enhance search capacity in the multi-dimensional space and determine the parameters of markov random field to optimize the objective function which comes from the random field. The method is compared with the classical simulated annealing algorithm. The segmentation effect is quantitative assessed by pixel dispersion, coincidence degree and area of detesting. Results show that the proposed algorithm performs better than the traditional algorithm in the three aspects. It can rapidly get the better segmentation result with satisfactory noise rejection and edge preserving. The robustness to noise and the smoothness are remarkably improved.