Bayesian Image Restoration Using a Large-Scale Total Patch Variation Prior

<table class="table-group" id="tab4"><tr><td><table class="table"><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td align="left">Wiener filtering: simu: 58.16, real: 64.48</td><td align="left"> TV prior res: simu: 214.96, real: 228.92</td></tr><tr><td align="left">Huber prior res: simu: 270.52, real: 292.36</td><td align="left"> ForWaRD res: simu: 2128.27, real: 2411.15</td></tr><tr><td align="left"> SA prior res: simu: 2128.27, real: 2382.43</td><td align="left"> LS-TPV prior res1: simu: 2885.61, real: 3058.07</td></tr><tr><td align="left">LS-TPV prior res2: simu: 910.48, real: 954.48</td><td align="left"> LS-TPV prior res3: simu: 58.61, real: 64.45</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

CPU times (seconds) needed for different restorations for real data experiment.

Mathematical Problems in Engineering

tab4

Table 4

Table 4: Bayesian Image Restoration Using a Large-Scale Total Patch Variation Prior 