摘要
文章结合人类视觉系统(HVS)对图像各个区域敏感度不同这一特性,对通常的脉冲耦合神经网络模型(PC-NN--PulseCoupledNeuralNetwork)进行了改进,分析了改进模型的特性及其参数优化原理,提出了一种基于这种改进PCNN的图像分割算法。该算法可根据像素周边区域的灰度梯度大小发放不同值的脉冲,从而自适应地将图像分为多个不同等级的高低信息区域,较好地仿真了人类视觉系统特性。并将该算法应用于图像压缩,在压缩比和重建图像主观视觉感知质量上均达到了较好的性能。
Based on the property of Human Vision System(HVS)that human eye's sensitivity to an image varies with different regions of an image where different regions correspond to different informative area of the image,Pulse Coupled Neural Network(PCNN)model is modified for image segmentation.The modified PCNN stimulated by an input image has outputs of pulses with many pulse values other than only two according to the local intensity variation of pixels in the put image.This results in segmentation of the image with respect to the local information delivered by the image.The proposed algorithm is applied to image compression and performs well in both compression rate and subjective percep-tual quality of the reconstructed image.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第8期7-8,44,共3页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(编号:60071026)
部委预研跨行业基金资助项目(编号:00J1.4.4.DZ0106)
信息处理与智能控制教育部重点实验室开放基金资助项目(编号:TKLJ0005)
关键词
脉冲耦合神经网络
图像分割
图像信息
图像压缩
Pulse-Coupled Neural Networks,Image Segmentation,Image Information,Image compression
