摘要
提供了一种快速稳健的正则化超分辨率图像复原算法。该算法从Roberts交叉梯度算子的角度构造正则项,从自适应性的角度确定正则参数。用1-范数度量数据逼近项和正则项。算法分析和实验结果表明该算法具有好的稳健性,并且计算速度较快。
This paper offers a fast and robust regularized super-resolution image reconstruction algorithm.The algorithm chooses regularization item based on the Roberts cross gradient operator and chooses the regularization parameter adaptively.The algorithm adopts 1-norm to measure data approximation item and regularization item.Algorithm analysis and simulation results confirm that our algorithm has good robustness and fast computation.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第4期76-78,共3页
Computer Engineering and Applications
关键词
图像复原
超分辨率
正则化
image restoration, super-resolution, regularization
