| 基于调和模型的快速神经网络图像复原算法 |
| |
| 引用本文： | 吴亚东,孙世新.基于调和模型的快速神经网络图像复原算法[J].计算机应用研究,2007,24(6):158-160. |
| |
| 作者姓名： | 吴亚东 孙世新 |
| |
| 作者单位： | 1. 西南科技大学,计算机科学与技术学院,四川,绵阳,621010;电子科技大学,计算机科学与工程学院,四川,成都,610054
2. 电子科技大学,计算机科学与工程学院,四川,成都,610054 |
| |
| 摘 要： | 针对传统神经网络图像复原算法在复原过程中模糊图像边缘,收敛速度慢等不足,提出一种基于调和模型的快速神经网络图像复原算法.在该算法中,图像复原模型的正则化项采用调和模型,并在每次网络状态更新时引入最陡下降方法,使得网络能量迅速减小.实验表明,提出的算法能够很好复原图像的边缘特征,并具有快速收敛等优点.
|
| 关 键 词： | 图像复原 神经网络 调和模型 去模糊 原模型 快速收敛 神经网络 图像复原 算法 Harmonic Model Based Neural Network Algorithm Image Restoration 边缘特征 实验 能量 方法 最陡下降 状态更新 正则化 收敛速度 图像边缘 模糊 |
| 文章编号： | 1001-3695（2007）06-0158-03 |
| 修稿时间： | 2006-03-17 |
Fast Image Restoration Algorithm Using Neural Network Based on Harmonic Model |
| |
| WU Ya dong,SUN Shi xin.Fast Image Restoration Algorithm Using Neural Network Based on Harmonic Model[J].Application Research of Computers,2007,24(6):158-160. |
| |
| Authors: | WU Ya dong SUN Shi xin |
| |
| Affiliation: | (1.School of Computer Science & Technology, Southwest University of Science & Technology, Mianyang Sichuan 621010, China;2.School of Computer Science & Engineering, University of Electronic Science & Technology of China, Chengdu Sichuan 610054, China) |
| |
| Abstract: | Traditional image restoration algorithm using neural network has some shortcomings was fond in experiments,such as blurring edges,low speed of convergence.A fast image restoration algorithm using neural network based on harmonic model was given.In this algorithm,this paper used harmonic model as regular term in image restoration model,and introduced the steepest descent method in update rule to accelerate the convergence of network.Experiment results show that the proposed algorithm could restore the degraded image while preserving the edge in a fast speed. |
| |
| Keywords: | image restoration neural network harmonic model image deblurring |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
