Piecewise Sparse Approximation via Threshold P_OMP

The sparse representation of the signal has been studied extensively in recent years. In reality, lots of signals have their own structure. Inspired by the application of signal decomposition and surface reconstruction with scattered data, we consider a special type of signal that can be expressed by the union of several sparse vectors. We define this form of signal as piecewise sparse signal. In order to recover piecewise sparse signal adaptively, we propose a threshold piecewise orthogonal matching pursuit (TP_OMP) algorithm equipped with thresholding strategy to overcome the shortcomings of P_OMP in this paper. The effectiveness of the proposed scheme is verified by the simulations of surface reconstruction.