Abstract

We present a fast parameter sensitivity analysis by combin- ing recent developments from uncertainty quantification with image pro- cessing operators. The approach is not based on a sampling strategy, instead we combine the polynomial chaos expansion and stochastic fi- nite elements with PDE-based image processing operators. With our approach and a moderate number of parameters in the models the full sensitivity analysis is obtained at the cost of a few Monte Carlo runs. To demonstrate the efficiency and simplicity of the approach we show a parameter sensitivity analysis for Perona-Malik diffusion, random walker and Ambrosio-Tortorelli segmentation, and discontinuity-preserving op- tical flow computation.