1: \begin{abstract}

2: 

3: Primal-dual splitting involving proximity operators in order to

4: be able to find some approximation to the minimizer 

5: for a general form of

6: Tikhonov type functional is in the focus of this work. 

7: This approximation is produced by a pair of 

8: iterative variational regularization procedures.

9: 

10: Under the assumption of some variational source condition (VSC), 

11: total error estimation both in the iterative sense

12: and in the continuous sense has been analysed separately.

13: Rates of convergence 

14: will be obtained in terms some concave and positive definite index function. 

15: Of the choice of the penalty term, we are interested in Bregman distance

16: penalization associated with the non-smooth

17: total variation (TV) functional. Furthermore, following up the lower and bounds

18: defined for the regularization parameter, some deterministic choice of the 

19: regularization parameter is given explicitly. 

20: It is in the emphasis of this work that the regularization

21: parameter obeys {\em Morozov`s discrepancy principle} (MDP) 

22: in order for the stability analysis of regularized solution.

23: 

24: In the computerized environment, the algorithms are verified as iterative

25: regularization methods by applying it to an atmospheric 

26: tomography problem named as {\em GPS-Tomography}.

27: Apart from this 3-D tomographic inverse

28: problem, we also apply the algorithms to some 2-D 

29: conventional tomographic image reconstruction 

30: problems in order to be able test algorithms` capability of capturing the 

31: details and observe that algorithms behave as iterative regularization

32: procedures.

33: 

34: \bigskip

35: \textbf{Keywords.}

36: {iterative regularization, primal dual algorithm, Bregman iteration, 

37: total variation}

38: \end{abstract}

39: