Medical image segmentation is a hotspot in the field of image segmentation, and there are many segmentation methods. As a method of image segmentation, pulse coupled neural network (PCNN) has excellent segmentation effect. Of course, it also reduces the efficiency and effect of segmentation because of the complexity of parameter setting and the need for manual setting. This paper presents a method of searching simplified PCNN parameters by using Harris Hawks optimization (HHO) algorithm. For one thing the number of parameters of PCNN is reduced without affecting the segmentation effect, for another the corresponding parameters of PCNN are searched quickly and accurately by intelligent optimization algorithm. Then, image entropy (H) and mutual information entropy (MI) are introduced as fitness functions. The performance of HHO-PCNN is compared with WOA-PCNN, SCA-PCNN, SSA-PCNN, PSO-PCNN, GWO-PCNN, MVO-PCNN, Otsu and K-means by performance indicators (UM, CM, Precision, Recall, and Dice). The experimental results verify the superiority of this method in image segmentation.

医学图像分割是图像分割领域的热点，并且有许多分割方法。作为图像分割的一种方法，脉冲耦合神经网络（PCNN）具有出色的分割效果。当然，由于参数设置的复杂性和手动设置的需要，它还会降低分割的效率和效果。本文提出了一种使用哈里斯·霍克斯优化（HHO）算法搜索简化的PCNN参数的方法。一方面，在不影响分割效果的情况下减少了PCNN的参数数量；另一方面，通过智能优化算法快速准确地搜索了PCNN的相应参数。然后，引入图像熵（H）和互信息熵（MI）作为适应度函数。将HHO-PCNN的性能与WOA-PCNN，SCA-PCNN，SSA-PCNN，PSO-PCNN，GWO-PCNN，MVO-PCNN，Otsu和K-means的性能指标（UM，CM，Precision，Recall，和骰子）。实验结果证明了该方法在图像分割中的优越性。