sugar_dms

Automated image reconstruction and contour detection

SUGAR DMS is an image recontruction and contour detection matlab package. It performs the minimization of the Discrete Mumford-Shah (D-MS) functional, which enforces constraints related to smoothing over image and sparsity over contours to obtain a piecewise smooth reconstructed image and sparse estimated contours from observed noisy images. A Stein-like strategy providing optimal hyperparameters $\beta$ and $\lambda$ is designed, based on the minimization of an unbiased quadratic risk estimate. Efficient and automated minimization of the risk estimate relies on an unbiased estimate of the risk's gradient with respect to the hyperparameters.

If you use this code, please cite:

C.-G. Lucas, B. Pascal, N. Pustelnik and P. Abry (2022). Hyperparameter selection for Discrete Mumford-Shah, Signal, Image and Video Processing, 17(5), 1897-1904