Abstract: The Pulse Coupled Neural Network(PCNN) is a new artificial neural networks model, and several PCNN structures for image segmentation are proposed depending on the model’s potential. But it isn’t a trivial task to define the relative parameters properly in the research of the theories and the applications of PCNN. As a contribution to this research field, this paper presents a new method for image segmentation based on differential evolution algorithm. Differential evolution algorithm as a new evolutionary algorithm can accomplish the automatic search of target parameter with its superior characteristic. Application of key parameters’ automatic setting in image segmentation, and the comparison between this segmentation result and other segmentation methods, the correctness and advancement of this case are verified.

Key words: image segmentation, Pulse Coupled Neural Network(PCNN), differential evolution algorithm

摘要： 脉冲耦合神经网络(PCNN)是一种新型神经网络，可以应用于图像分割。然而在对PCNN的研究应用中，其模型参数的合理确定是个难点，这在很大程度上限制了PCNN的应用。针对这一问题，提出一种基于微分进化的PCNN图像分割方法。该方法使用微分进化算法来实现脉冲耦合神经网络参数的自动设定，并通过将其应用于图像分割，将分割结果与其他优秀分割方法比较，从而验证了该方案的正确性与可行性。

关键词: 图像分割, 脉冲耦合神经网络, 微分进化算法

CLC Number: 

• TP338