Abstract | ||
---|---|---|
In this paper, we introduce a class of variational models for the restoration of ultrasound images corrupted by noise. The proposed models involve the convex or nonconvex total generalized variation regularization. The total generalized variation regularization ameliorates the staircasing artifacts that appear in the restored images of total variation based models. Incorporating total generalized variation regularization with nonconvexity helps preserve edges in the restored images. To realize the proposed convex model, we adopt the alternating direction method of multipliers, and the iteratively reweighted $$\\ell _1$$ℓ1 algorithm is employed to handle the nonconvex model. These methods result in fast and efficient optimization algorithms for solving our models. Numerical experiments demonstrate that the proposed models are superior to other state-of-the-art models. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/s10915-017-0357-3 | J. Sci. Comput. |
Keywords | Field | DocType |
Ultrasound image denoising, Total generalized variation, Nonconvex regularization, Alternating direction method of multipliers, Iteratively reweighted algorithm | Noise reduction,Mathematical optimization,Regular polygon,Regularization (mathematics),Total variation denoising,Optimization algorithm,Total generalized variation,Mathematics | Journal |
Volume | Issue | ISSN |
72 | 1 | 1573-7691 |
Citations | PageRank | References |
2 | 0.39 | 30 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Myeongmin Kang | 1 | 29 | 4.54 |
Myungjoo Kang | 2 | 332 | 52.48 |
Miyoun Jung | 3 | 125 | 10.72 |