Compressed Sensing Image Reconstruction Using Efficient Algorithm

Compressed sensing is a new information sampling theory and it’s done for acquiring sparse (or) compressible data with much fewer measurements. This is particularly important for some imaging applications such as magnetic resonance image or in astronomy. In many practical situations, the noise behavior is impulsive and that the probability density function has very complex calculation than Gaussian. This motivates a number of impulsive noise suppression methods. Therefore, a new method is called robust CS is applied, following the principle of robust statistics which is using a convex but quadratic cost function on the residuals. By using a robust cost function on the residuals, we are able to suppress large outliers in the measurement noise. It also shows that an iterative algorithm can be developed under minimization– majorization framework and have established a theoretical guarantee on the improvement of the upper bound of the recovery error. The proposed method can be used to improve Cs recovery depend upon an inspection of the residuals for impulsiveness