¹ School of Information science and Engineer, Central South University, Changsha 410073, China)
² School of Information Science and Engineering, Jishou University, Jishou 416000, China

---

Abstract—In the domain of image processing, PCNN (Pulse Coupled Neural Network) need to adjust parameters time after time to obtain the better performance. To this end, we propose a novel PCNN parameters automatic decision algorithm based on immune algorithm. The proposed method transforms PCNN parameters setting problem into parameters optimization based on immune algorithm. It takes image entropy as the evaluation basis of the best fitness of immune algorithm so that PCNN parameters can be adjusted adaptively. Meanwhile, in order to break the condition that population information fall into local optimum, the proposed method introduces gradient information to affect the evolution of antibody to keep the population activity. Experiment results show that the proposed method realizes the adaptive adjustment of PCNN parameters and yields the better segmentation performance than many existing methods.

Index Terms—pulse couple neural network; Immune Algorithm; Image Segmentation; Parameter optimization； Image Entropy; Fitness function

[PDF]

Cite: Jianfeng Li, Beiji Zou, Lei Ding, and Xu Gao, " Image Segmentation with PCNN Model and Immune Algorithm," Journal of Computers vol. 8, no. 9, pp. 2429-2436, 2013.