基于PCNN和最大相关准则的图像分割算法  

Image Segmentation Method Based on PCNN and Two-dimension MCC

作　　者：邓灵博[1] 

机构地区：[1]解放军炮兵学院,合肥230031

出　　处：《弹箭与制导学报》2008年第3期237-239,共3页Journal of Projectiles,Rockets,Missiles and Guidance

摘　　要：提出了一种脉冲耦合神经网络和最大相关准则相结合的算法来对图像进行分割。该算法运用最大相关准则进行最优分割,与文献[2]的算法相比,由于涉及较少的对数运算,因此实时性较好。The algorithm that segments images with pulse-coupled neural networks （PCNN） and two-dimension maximum correlation criterion（MCC） was presented. There is less logarithmic operation in two-dimension MCC than in twodimension entropy in the reference [2]. Therefore, the segmentation algorithm in this paper is faster and has good realtime performance.

关 键 词：图像分割 脉冲耦合神经网络 最大相关准则 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]