1: \begin{abstract}

2: 	Signals and images with discontinuities appear in many problems in such diverse areas as

3: 	biology, medicine, mechanics, and electrical engineering. 

4: 	The concrete data are often discrete, indirect and noisy measurements

5: 	of some quantities describing the signal under consideration. 

6: 	A frequent task is to find the segments of the signal or image which corresponds to finding the discontinuities or jumps in the data.

7: 	Methods based on minimizing the piecewise constant Mumford-Shah functional 

8: 	--whose discretized version is known as Potts functional-- 

9: 	are advantageous in this scenario, in particular, in connection with segmentation.	

10: 	However, due to their non-convexity, minimization of such functionals is challenging.

11: 	In this paper we propose a new iterative minimization strategy for

12: 	the multivariate Potts functional dealing with indirect, noisy measurements.

13: 	We provide a convergence analysis and underpin our findings with numerical experiments.  

14: \end{abstract}

15: