A fast image reconstruction algorithm for compressed sensing-based atomic force microscopy

The idea of compressed sensing (CS) can be applied to atomic force microscopy (AFM) to reduce the amount of data that needs to be sampled for accurate image reconstruction. The data sampling strategy and measurement matrix design in AFM have been discussed in previous work. However, standard CS image recovery needs to solve a large size convex optimization problem, which requires a lot of computational resources in terms of both time and memory. In this paper, we propose a new variant of the Matching Pursuit (MP) algorithm for image reconstruction based on CS in AFM. With this algorithm, the computational time and memory space for image reconstruction can be reduced significantly with only a small loss in image quality. The proposed algorithm is demonstrated through MATLAB simulation.