Applications of Pulse-Coupled Neural Network in Image Segmentation for Liver Hydatid
TIAN Xianghua1, HAMIT Murat1, ZHU Congxu2, KONG Dewei3
1. College of Medical Engineer Technology, Xinjiang Medical University, Urumqi 830011, China;2. School of Information Science & Engineering, Central South University, Changsha 410083, China;3. Department of Radiology, The First Affiliated Hospital, Xinjiang Medical University, Urumqi 830054, China
摘要在应用脉冲耦合神经网络模型分割图像的研究中,确定模型的参数是一个难点问题,其中连接系数β在脉冲耦合神经网络中起着重要的作用。本文使用最小交叉熵D(P, Q; t )和标准差,简化了脉冲耦合神经网络模型的连接系数β的估计公式,该方法可以自动确定并简化脉冲耦合神经网络模型的连接系数β。实验结果表明,该方法对肝包虫医学图像的分割效果显著,能获得较好的视觉结果并具有较强的普适性。
Abstract:In the application of pulse-coupled neural network(PCNN),it is difficult to set the parameter of pulse-coupled neural network. In the pulse-coupled neural network model, the linking coefficient β plays an important role. According to the relative entropy D(P, Q: t) and standard deviation, educe an estimate formula that to set the linking coefficient β automatically in the simplified pulse-coupled neural network model. Experimental results show that the algorithm gets a better visual effect when dealing liver hydatid disease image,showing great adaptability.