摘要
脉冲耦合神经网络(Pu lse Coup led N eura l N etw ork,PCNN)具有良好的脉冲传播特性,在图像分割中得到了广泛应用。针对其需要人机交互通过实验确定其相关参数,实时性差等问题,改进了标准的PCNN模型,提出了一种基于改进型脉冲耦合神经网络的图像分割方法。仿真结果表明,该方法实时性好、自适应性强,分割出的目标轮廓清楚,细节更多。
For its good property of pulse burst, Pulse Coupled Neural Network is widely used in image segmentation. For its problems when PCNN is used in image segmentation, such as, its parameter is decided by experiment and its real time ability is bad. The standard model of PCNN is improved and an approach for image segmentation based on improved PCNN is proposed. Experiment results show that the method is adaptive and its real time ability is good, target contour are clear and details are more.
出处
《弹箭与制导学报》
CSCD
北大核心
2006年第1期126-128,131,共4页
Journal of Projectiles,Rockets,Missiles and Guidance
基金
国家部委预研项目(51405030104BQ0171)
关键词
脉冲耦合神经网络
图像分割
图像熵
阈值
pulse coupled neural network
image segmentation
image entropy
threshold
