基于多字典 L1/2 正则化的超分辨率重建算法

为详细表达图像高频细节信息, 提高重建图像质量, 提出了一种基于多字典 L1 /2 正则化的超分辨 率重建算法。 该算法在稀疏重建字典对训练阶段, 为有效提取低分辨率图像边缘、 纹理等特征细节信息, 采用改进的一阶二阶导数方法对低分辨率图像进行特征提取; 而在图像重建阶段, 为解决基于 L1 正则模 型得到的解时常不够稀疏, 重建图像质量有待进一步提高的问题, 采用 L1 /2 范数代替 L1 范数构建超分 辨率重建模型。 实验表明, 与现有算法相比较, 该算法可更好地表达图像细节部分信息, 并能提高图像 的重建质量。

Super-Resolution Reconstruction Based on L1/2 Regularization of Multi Component Dictionary

In order to express the high frequency minutia information of th e image in detail, and improve the quality of reconstructed image, a super resol ution reconstruction algorithm is applied based on multi dictionary L1/2 regular ization.In the dictionary training phase, in order to effectively extract the i nformation of feature detail of edge and texture of low resolution image, the mo dified first or second order method is used to extract feature for low resolutio n image.In the stage of image reconstruction, because of the problems that the solution based on L1 regular model is usually not sparse enough and the quality of the reconstructed image needs to be further improved, L1/2 norm is employed to substitute L1 norm to establish the super resolution reconstruction model.The experiment shows that the present algorithm compared with the exist ing algorithms can better express the section information of the image d etails and improve the quality of image reconstruction.