Compressed sensing image reconstruction using intra prediction

Compressed sensing (CS) provides a general signal acquisition framework that enables the reconstruction of sparse signals from a small number of linear measurements. In this article we present a CS image reconstruction algorithm using intra prediction method based on block-based CS image framework. The current reconstruction block is firstly predicted by its surrounding reconstructed pixels, and then its prediction residual will be reconstructed. Because the sparsity level of prediction residual is higher than its original image block, the performance of our proposed CS image reconstruction algorithm is significantly superior to the traditional CS reconstruction algorithm. Furthermore, total variation model is also used to suppress the blocking artifacts caused by intra prediction and measurement noise. Experimental results also show the competitive performance with respect to peak signal-to-noise ratio and subjective visual quality.