改进的多尺度压缩感知方案

Improved multiscale compressed sensing scheme

传统多尺度压缩感知方案只对设定的小波变换系数采样,不用于采样的小波系数直接置零,造成重建图像边缘粗糙,影响图像重建质量。针对该问题,提出了一种改进的多尺度压缩感知方案。新方案在原方案获得的重建图像基础上,通过轮廓波变换对重建图像进行图像插值,在保持重建得到的小波系数不变的同时估计出原来被直接置零的小波系数,从而改进了图像重建质量。通过仿真实验证明了改进的多尺度压缩感知方案的有效性,并且与传统多尺度压缩感知方案相比能得到效果更好的重建图像。

According to the traditional multiscale compressed sensing （MCS） scheme,only a few of wavelet coefficients are sampled and the others are set to zero, which results in the coarse edges of the reconstruction images and low reconstruction accuracy. In order to overcome this conundrum, an improved multiscale compressed sensing （IMCS） scheme is proposed. Based on the reconstruction image achieved by MCS,the IMCS scheme interpolates the image with contourlet transform. In this way, the sampled wavelet coefficients are kept while the unsampled wavelet coefficients are estimated. The edges of the images reconstructed by IMCS become smoother and the mosaic effect is reduced. The validity of IMCS is proved by the experiments which show that IMCS has better reconstruction accuracy compared with MCS.