中文摘要：

针对人眼对图像低频成分敏感而对高频细节相对不敏感的特点，在分析图像子带能量分布特性的基础上，提出了一种基于子带能量分布的图像自适应加权压缩采样方法．该方法通过提取图像分块离散余弦变换各系数子带的能量，进而用于对测量过程进行加权修正，最终实现了对不同子带系数有区别的压缩投影．比较同类研究结果表明：提出的加权压缩采样方法在有效减少测量维数或提高重构图像的峰值信噪比和主观视觉效果，以及在降低计算复杂度方面均有更好的表现．

英文摘要：

Considering that human eyes are more sensitive to low frequency components than to high frequency ones, a novel adaptive compressive sampling method for image representation is proposed based on subband energy distribution characteristics. Each subband＇s energy is obtained by using block DCT, which is used to modify the measurement procedure, and consequently achieve the compressive image projection with the differentiated subband coefficients. Comparison results with the similar work demonstrate that the proposed adaptive weighted compressive sampling method could not only efficiently reduce the computational complexity, but also considerably decrease the measurement rate and/or enhance the recovery image quality in both PSNR and subjective visual quality.