摘要
介绍了具有侧抑制特性的中心-周边分离网络(CSSN),通过对一维信号处理效果的模拟分析,验证了CSSN对单幅图像或两幅图像间的对比度增强作用;在MIT融合结构的基础上,提出了一种新的基于CSSN的双通道融合结构,对微光与红外图像进行彩色融合处理后,证明能得到自然的色彩效果.
Centersurround shunting network (CSSN), based on lateral inhibition, is introduced. Through simulation using onedimensional signals, the properties of CSSN in enhancing contrast in a single image or between two images are testified. A new twochannel fusion architecture is proposed, built upon the MIT architecture. Analysis with lowlevel light image and infrared image shows that the new architecture proves to give better image contrast and requires a simpler arithmetic.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2003年第4期513-516,共4页
Transactions of Beijing Institute of Technology
基金
国家部委预研项目(03405050)
关键词
侧抑制
彩色图像
图像融合
夜视
微光
红外
lateral inhibition
color image
image fusion
night vision
low-light visible
infrared