中文摘要：

脉冲耦合神经网络在图像处理中需要通过多次调整参数才能获得比较好的图像处理效果．为此提出了一种结合免疫克隆算法的PCNN参数自动判定算法，提出的方法将PCNN的参数设置f＊7题转化为基于免疫克隆算法的参数优化问题，将图像熵值作为免疫克隆算法最佳适应度的评判依据，从而达到PCNN在图像处理中参数自适应调整的目的．同时，在免疫克隆算法中引入自适应算予和梯度操作来影响部分抗体的进化，以保持群体的活性，打破群体信息陷入局部最优的状况．实验表明，提出的算法实现了PCNN参数的自适应调整，图像分割效果优于比较的多种图像分割算法．

英文摘要：

In the field of image processing, pulse coupled neural network { PCNN） has to adjust the parameters repeatedly to obtain the good performace. For this reason, we present a novel algorithm to automatically determine PCNN parameters by combining with im- mune algorithm. Aimming at determining PCNN parameter automatically, the proposed method transforms PCNN parameter setting problem into the parameter optimization problem of immune algorithm and takes image entropy as the evaluation basis of the fitness. Meanwhile, in the immune algorithm, the adaptive operator and gradient operation are adopted to affect the evolution of some anti- bodies, so that it can keep the activity of the group and prevent falling into local optimum. Experiments show that the proposed meth- od achieves the adaptive parameter adjustment of PCNN and yields the better segmentation performance than other methods.