Semi-sparsity Priors for Image Structure Analysis and Extraction
This addresses a fundamental problem in image processing and computer vision for applications requiring clean structural extraction, though it appears incremental as it builds on existing regularization and textural analysis models.
The authors tackled the problem of image structure-texture decomposition by proposing a generalized semi-sparse regularization framework, which effectively decouples structures from textures while avoiding staircase artifacts and handling oscillatory patterns, with experimental results showing comparable or superior performance to cutting-edge methods.
Image structure-texture decomposition is a long-standing and fundamental problem in both image processing and computer vision fields. In this paper, we propose a generalized semi-sparse regularization framework for image structural analysis and extraction, which allows us to decouple the underlying image structures from complicated textural backgrounds. Combining with different textural analysis models, such a regularization receives favorable properties differing from many traditional methods. We demonstrate that it is not only capable of preserving image structures without introducing notorious staircase artifacts in polynomial-smoothing surfaces but is also applicable for decomposing image textures with strong oscillatory patterns. Moreover, we also introduce an efficient numerical solution based on an alternating direction method of multipliers (ADMM) algorithm, which gives rise to a simple and maneuverable way for image structure-texture decomposition. The versatility of the proposed method is finally verified by a series of experimental results with the capability of producing comparable or superior image decomposition results against cutting-edge methods.