一种改进的局部线性嵌入超分辨率重建算法

An Improved Super-resolution Reconstruction Algorithm with Locally Linear Embedding

提出了一种改进的局部线性嵌入超分辨率重建算法。该算法着重对局部线性嵌入超分辨率重建算法三个方面做了改进:特征选取,用图像块的DCT系数来取代图像块的1阶、2阶梯度作为图像块的特征描述,可以减弱噪声的影响;邻近块的数目,根据图像块与周围图像块的关系自适应的选取邻近块的数目,可以避免将距离较远的块选为邻近块;样本库的训练过程,用高分辨率图像与低分辨率图像的残差图像作为高分辨率图像的训练样本,这样既可以避免低频分量的干扰,又可以减少在计算过程中的平滑次数。实验结果表明这种改进的算法比原算法的重建效果有了较大程度的提高:PSNR提高4.07 dB,SSIM提高0.065 4;比稀疏重建算法PSNR提高0.62 dB,SSIM提高0.006 6,而且用DCT系数作为图像块的特征表示,每一个图像块所需要提取的特征数比用1阶、2阶梯度减少了四分之三,降低了算法的复杂度。

An improved super resolution reconstruction algorithm is proposed with locally linear embed- ding. The improvement includes three aspects. Firstly, the DCT coefficients of low resolution image patches are taken as the feature representative instead of the first order and second order gradients, which will re- duce the effect of noise. Secondly, the number of adjacent blocks is chosen adaptively according to the re- lationship between the input low resolution image patch and its neighbors ,which will avoid the possibility of choosing a distant patch as neighbor. Thirdly, the training sample of the high resolution image is taken as the residual image resulting from the difference between the high resolution image and the correspond- ing low resolution one. This can not only avoid the disturbance of low frequency components, but also re- duce the number of smoothness computation. The experimental results have shown that the improved algo- rithm can achieve a better reconstruction effect with improved PSNR of 4.07 dB and improved SSIM of 0.0654 compared to the existing LLE algorithm, and improved PSNR of 0.62 dB and improved SSIM of 0. 0066 compared to the sparse representation algorithm. In addition, using DCT coefficients as the feature representative reduces the computational complexity in that the number of the extracted features needed is only a quarter of that using the first order and second order gradients.