欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出一种新的基于双层脉冲耦合神经网络（PCNN）的自适应图像分割算法。双层PCNN的前级以简化PCNN模型为基础,获得区域生长的种子;后级采用区域生长机制,征募区域内灰度相似像素,完成前级种子的生长。新算法PCNN的关键参数可自适应更新,避免了传统PCNN参数设置难的问题;区域生长机制强化了PCNN的区域特性。实验结果表明,新算法所得分割图像的区域连通性及边缘规整性得到进一步提高,发挥了PCNN应用于图像分割的优越性。
英文摘要：
A novel adaptive algorithm for image segmentation based on the dual-level pulse coupled neural networks（PCNN） is proposed.For the dual-level PCNN,the first level is based on the simplified PCNN model to obtain the region seeds;the next level adopts the region growing strategy,and recruits the pixels which have similar gray level to the seeds to achieve the growth of the regions.The sensitive parameters of the PCNN can be tuned adaptively,which can overcome the limitation of the parameter setting.Moreover,the region growing strategy strengthens the region characteristics of PCNN.Experimental results show that the proposed algorithm can improve the region connectivity and the edge regularity of the segmented image,and the advantages of PCNN for image segmentation are developed.
同期刊论文项目
同项目期刊论文
位置：