中文摘要：

为了克服传统的压缩感知重构中正交小波方向选择性差的局限性,针对图像信号方向性决定了需要在不同纹理区域选择滤波器以使变换后信号能量更加稀疏,提出一种基于自适应方向提升稀疏表示的重构方法。重构时,在每次迭代更新后,根据图像信号的纹理特征选择不同强度方向和信号光滑度的小波基,使得变换后信号能量分布更加集中,并利用小波域阈值处理方法解决信号的重构噪声问题。实验结果表明,该算法提高了重构图像的峰值信噪比和视觉效果,保护了图像的细节。

英文摘要：

To overcome the poor directional selectivity of orthogonal wavelet bases,signal energy is made more sparse by selecting the filters in different texture regions for image signal direction,an image compressed sensing reconstruction algorithm based on an adaptive directional lifting sparse representation is presented.After each iterative refinement in the reconstruction,the signal energy distribution is more concentrated by selecting wavelet bases with different intensities of the direction and signal smoothness according to the texture characteristics of the image signal,and a wavelet thresholding method is used to realize signal reconstruction denoising.Experimental results show that the proposed algorithm improves the peak signal-to-noise ratio and visual quality,and protects image details.