An image segmentation algorithm of snake model based on firefly algorithm

作者：Du Xiaogang^*; Dang Jianwu; Wang Yangping

来源：International Review on Computers and Software, 2012, 7(5): 2718-2723.

• Best value
• Centripetal force
• Firefly algorithms
• Image segmentation algorithm
• Initial contour
• Local extremes
• Object contour
• Object region
• Segmentation results
• Snake model

In order to solve the problem that the snake model isn't convergent to the concave edges of object region during image segmentation; a snake image segmentation algorithm based on the firefly algorithm is put forward in this paper. Firstly, the traditional snake model is improved by adding the centripetal force, and it has the characteristics of convergence to the concave edges of object region; Secondly, the initial contour of object is converged coarsely using the improved Snake model; Finally, the firefly algorithm which has the good ability of finding the global best value is adopted for the imprecise convergence contour to search the real edges of the object region and complete the process of image segmentation. The experimental results show that this algorithm can overcome the shortcoming that the snake model is easy to fall into local extremes and not convergent to the concave edges of object region and the accuracy of the segmentation results is also improved.

• 出版日期2012