Robust Reconstruction from Chopped and Nodded Images

In ground based infrared imaging a well-known technique to reduce the influence of thermal and background noise is chopping and nodding, where four different signals of the same object are recorded from which the object is reconstructed numerically. Since noise in the data can severely affect the reconstruction, regularization algorithms have to be implemented. In this paper we propose to combine iterative reconstruction algorithms with robust statistical methods. Moreover, we study the use of multiple chopped data sets with different chopping amplitudes and the according numerical reconstruction algorithm. Numerical simulations show robustness of the proposed methods with respect to noisy data.