Sparse Analysis Recovery via Iterative Cosupport Detection Estimation

Cosparse analysis model (CAM) provides a new signal processing paradigm for recovering cosparse signals with respect to given operator from the undersampled linear measurements in context of emerging theory compressed sensing (CS). The sparse recovery/cosparse recovery is key one brought up by this paradigm. In paper, we propose family pursuit algorithms problem when obey model, termed as iterative cosupport detection estimation (ICDE). ICDE an algorithmic framework, which alternates between detecting set unknown true and estimating underlying solving truncated on detected cosupport. Further, effective implementations equipped efficient thresholding strategy detection. Empirical performance comparisons show that favorable comparison state-of-the-art algorithms. Source code has been made publicly available Github: https://github.com/songhp/ICDE.