Figure 3 | Scientific Reports

Figure 3

From: Metric-guided regularisation parameter selection for statistical iterative reconstruction in computed tomography

Figure 3

The histogram entropy of the numerical simulation during reconstruction. It is evaluated close to the image edges over 500 iterations for 23 regularisation parameters. The white lines represent the parameter optimisation. The steps show a change of parameter after a stabilisation interval of 35 iterations. The two lines entering the plot from the bottom started at the initial values of \({\lambda }_{{\rm{start}}}\in [1,3]\) outside the displayed range. They approach the final parameter by increasing with \({\lambda }_{{\rm{stronger}}}=2{\lambda }_{{\rm{central}}}\) (see Eq. (6)) over several optimisation steps. During the course of optimisation, the tuning method approaches the minimum for all initial parameters.

Back to article page