图像压缩感知中自适应二维投影梯度重构算法

Adaptive Two-Dimensional Projected Gradient Algorithm for Compressed Image Sensing

二维图像的压缩感知及重构大多利用一维信号压缩感知及重构方法实现,导致图像重构效率较低,重构算法复杂度高等缺点。二维随机投影及二维投影梯度重构算法有效地解决了这一问题。但在二维投影梯度重构算法中,不同图像不同采样率的重构中采用相同滤波阈值参数η的方案会降低图像重构质量。本文结合二维图像信号的纹理特性,提出了自适应二维投影梯度重构算法,该算法提出了一种双变量收缩阈值参数η在迭代重构过程中基于图像纹理信息的自适应计算公式。实验结果表明,自适应二维投影梯度重构算法比二维投影梯度重构算法在重构质量和视觉效果上都有所提升。

Most of existing two-dimensional compressed sensing and reconstruction methods for images are complemented by utilizing one-dimensional signals compressed sensing and reconstruction algorithms, which is inefficient and increases memory requirements. Two-dimensional random projection theory and two-dimensional projected gradient algorithm can overcome these disadvantages. However, fixed thresh- old parameter used in two-dimensional projected gradient algorithm for different images at different sam- pling rate may lead to poor reconstruction quality. Here, we propose an adaptive two-dimensional projec ted gradient algorithm based on image texture property. The parameter r/of bivariate shrinkage is calcu- lated according to image texture information during the iterative reconstruction process. Experimental re- sults show that compared with two-dimensional projected gradient algorithm, the proposed adaptive two- dimension projected gradient algorithm provides superior performance on both the image reconstruction quality and visual effect.