| 基于L2Boost的低阶核回归迭代去噪算法 |
| |
| 引用本文： | 许志宏,王沛. 基于L2Boost的低阶核回归迭代去噪算法[J]. 上海电机学院学报, 2011, 14(1): 44-49 |
| |
| 作者姓名： | 许志宏 王沛 |
| |
| 作者单位： | 上海师范大学,信息与机电工程学院,上海,200234 |
| |
| 摘 要： | 针对常用去噪算法易对图像边角纹理区域造成模糊的事实,通过引入机器学习理论,提出了基于L2Boost的低阶核回归迭代去噪算法。该算法应用低阶(零阶或一阶)的高斯自适应权重去噪方法对含噪图像依次迭代B次得到B个估计,然后将B个估计组合起来作为最后的去噪结果。数值实验显示了方法的优越性。
|
| 关 键 词： | L2Boost回归算法 低阶 自适应高斯核 图像去噪 |
A Novel L2 Boost-based Low-order Kernel Regression Denoising Method |
| |
| XU Zhihong,WANG Pei. A Novel L2 Boost-based Low-order Kernel Regression Denoising Method[J]. JOurnal of Shanghai Dianji University, 2011, 14(1): 44-49 |
| |
| Authors: | XU Zhihong WANG Pei |
| |
| Affiliation: | XU Zhihong,WANG Pei(College of Information,Mechanical and Electrical Engineering,Shanghai Normal University,Shanghai 200234,China) |
| |
| Abstract: | As most existing denoising approaches cause edge blurring,we propose a novel low-order kernel regression denoising method based on L_2boost.It is a procedure of iterative residual fitting.The output is a sum of the fits,each of which is obtained by low-order(zeroth or first order) data driven Gaussian based kernel regression.Numeric experiments show its strength. |
| |
| Keywords: | L_2Boost kernel regression low-order data-driven image denoising |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
