| 基于脉冲耦合神经网络的红外图像增强 |
| |
| 引用本文： | 陆佳佳,方亮,叶玉堂,杨先明,成志强.基于脉冲耦合神经网络的红外图像增强[J].光电工程,2007,34(2):50-54. |
| |
| 作者姓名： | 陆佳佳 方亮 叶玉堂 杨先明 成志强 |
| |
| 作者单位： | 电子科技大学,光电信息学院,四川,成都,610054 |
| |
| 摘 要： | 在分析了脉冲耦合神经网络的工作机理和行为特性后,指出可以利用神经元的点火-熄灭特性对图像进行增强.为了区分神经元的点火方式,提出一种根据链接矩阵判定神经元点火方式的方法,并利用自然点火和捕获点火建立了能使图像得到增强的非线性映射.文中对算法参数的设置及其对增强图像的影响做了详细地讨论,实验结果表明该算法不仅能使图像的对比度和亮度得到适当的增强,而且能够有效地抑制图像中的椒盐噪声,尤其适用于对比度和亮度都较低的红外图像.
|
| 关 键 词： | 红外图像 脉冲耦合神经网络 自然点火 捕获点火 图像增强 |
| 文章编号： | 1003-501X(2007)02-0050-05 |
| 收稿时间： | 2006/3/10 |
| 修稿时间： | 2006-03-10 |
Infrared image enhancement based on pulse-coupled neural network |
| |
| LU Jia-jia,FANG Liang,YE Yu-tang,YANG Xian-ming,CHENG Zhi-qiang.Infrared image enhancement based on pulse-coupled neural network[J].Opto-Electronic Engineering,2007,34(2):50-54. |
| |
| Authors: | LU Jia-jia FANG Liang YE Yu-tang YANG Xian-ming CHENG Zhi-qiang |
| |
| Affiliation: | School of Opto-electronic Information, University of Electronic Science and Technology of China, Chengdu 610054, China |
| |
| Abstract: | By analyzing the mechanism and dynamical feature of Pulse-coupled Neural Network(PCNN) ,it was concluded that fire-extinguishing characteristic of neuron could be applied in image enhancement.In order to distinguish different fire mode,a method based on the linking map was proposed to distinguish neuron fire mode.According to nature fire and capture fire of neurons,a nonlinear mapping was established to enhance image.In this paper,Parameter setting and the influence of the parameters on image enhancement was discussed in detail.The experiment results show that the algorithm can enhance contrast and brightness of image properly and suppress noise effectively,especially for the infrared image of low contrast and brightness. |
| |
| Keywords: | Infrared images Pulse-coupled neural network Nature fire Capture fire Image enhancement |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
