| 基于Lorentzian范数的图像超分辨率重建 |
| |
| 引用本文： | 解大鹏,王培康.基于Lorentzian范数的图像超分辨率重建[J].电子技术,2010,37(11):1-3. |
| |
| 作者姓名： | 解大鹏 王培康 |
| |
| 作者单位： | 中国科学技术大学信息与科学技术学院; |
| |
| 摘 要： | 本文在MAP随机正则化技术估计框架下,提出了一种基于Lorentzian范数估计和自适应核回归正则项的最小化代价函数。此算法对不同假设类型的噪声模型不敏感,鲁棒性较好。实验结果证明了本文方法不仅能有效提高图像清晰度,且与其它方法相比,去噪能力更强,边缘保持较好。
|
| 关 键 词： | 超分辨率重建 Lorentzian范数 正则化项 自适应核回归 鲁棒性 |
Lorentzian-norm-based Image Super-resolution Reconstruction |
| |
| Xie Dapeng,Wang Peikang.Lorentzian-norm-based Image Super-resolution Reconstruction[J].Electronic Technology,2010,37(11):1-3. |
| |
| Authors: | Xie Dapeng Wang Peikang |
| |
| Institution: | Xie Dapeng Wang Peikang(School of Information Science and Technology,University of Science and Technology of China) |
| |
| Abstract: | In this paper,we propose an approach using Lorentzian norm estimator and adaptive kernel regression regularization term based on the stochastic regularization technique of MAP estimation by minimizing a cost function.This algorithm is not sensitive different assumed noise models and has good robustness.The experiment results demonstrate the Lorentzian-norm-based Image Super-resolution Reconstruction approach can not only effectively improve the definition of the images,but also keep the edges much cleaner a... |
| |
| Keywords: | super-resolution reconstruction Lorentzian-norm regularization term adaptive kernel regression robustness |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
