A Block Components Average Algorithm for Compressed Sensing Based Tomography

The constrained total variation minimization has been developed successfully for image reconstruction in computerized tomography. The convergence of the block cyclic projection and Cimmino algorithms for compressed sensing based tomography have been proved recently. In this paper, a block component averaging algorithm incorporation to the total variation minimization in the compressed sensing framework is proposed. The convergence of the new algorithm is derived in certain case. An example is given to illustrate its convergence behavior.