| 预估计最小平方误差初始值方法改善交替映射算法性能 |
| |
| 引用本文： | 单宝堂,沈庭芝,崔玉红,赵鹏,胡宇. 预估计最小平方误差初始值方法改善交替映射算法性能[J]. 北京理工大学学报, 2007, 27(5): 436-440 |
| |
| 作者姓名： | 单宝堂 沈庭芝 崔玉红 赵鹏 胡宇 |
| |
| 作者单位： | 北京理工大学,信息科学技术学院电子工程系,北京,100081;装备指挥技术学院,信息装备系,北京,101416;北京理工大学,信息科学技术学院电子工程系,北京,100081;装备指挥技术学院,信息装备系,北京,101416 |
| |
| 摘 要： | 改进交替映射算法以提高反马赛克插值的峰值信噪比(PSNR)性能.采用预估计最小平方误差思想,改进交替映射插值算法的初始值估计,计算改进算法和其它典型反马赛克方法的峰值信噪比,采用典型测试图片作为测试源进行比较.结果表明,改进算法的峰值信噪比高于其它典型插值算法的峰值信噪比,插值恢复图像的伪像也少于其它算法.改进的交替映射算法具有最好的峰值信噪比和最少的插值伪像,说明改变初始值方法可以提高交替映射算法的性能.
|
| 关 键 词： | 颜色滤波阵列 反马赛克 预估计最小平方误差 |
| 文章编号： | 1001-0645(2007)05-0436-05 |
| 收稿时间： | 2006-12-07 |
| 修稿时间： | 2006-12-07 |
Improving the Performance of Alternating Projection Demosaicking Algorithm by Pre-Estimating Minimum Square Error Initialization |
| |
| SHAN Bao-tang,SHEN Ting-zhi,CUI Yu-hong,ZHAO Peng and HU Yu. Improving the Performance of Alternating Projection Demosaicking Algorithm by Pre-Estimating Minimum Square Error Initialization[J]. Journal of Beijing Institute of Technology(Natural Science Edition), 2007, 27(5): 436-440 |
| |
| Authors: | SHAN Bao-tang SHEN Ting-zhi CUI Yu-hong ZHAO Peng HU Yu |
| |
| Affiliation: | Department of Electronic Engineering,School of Information Science and Technology,Beijing Institute of Technology,Beijing 100081,China; Department of Information Equipment,the Academy of Equipment Command & Technology,Beijing 101416,China;Department of Electronic Engineering,School of Information Science and Technology,Beijing Institute of Technology,Beijing 100081,China;Department of Information Equipment,the Academy of Equipment Command & Technology,Beijing 101416,China;Department of Electronic Engineering,School of Information Science and Technology,Beijing Institute of Technology,Beijing 100081,China;Department of Electronic Engineering,School of Information Science and Technology,Beijing Institute of Technology,Beijing 100081,China |
| |
| Abstract: | To improve PSNR performance of alternating projection demosaicking algorithm,based on pre-estimating the minimum square error,the estimated initialization of alternating projection algorithm was meliorated.On the aspects of computed PSNR and tested typical testing pictures,the improved algorithm was compared with other typical demosaicking approaches.PSNR of the improved algorithm was better than other typical methods,and the interpolated artifact was found to be less.The improved algorithm is with best PSNR and the least interpolated artifact.The performance of alternating projection algorithm can be improved by mending the initialization. |
| |
| Keywords: | color filter array demosaicking pre-estimating minimum square error |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《北京理工大学学报》浏览原始摘要信息 |
|
点击此处可从《北京理工大学学报》下载全文 |
|
