1: \begin{abstract}

2: Affine matrix rank minimization problem is a fundamental problem with a lot of important applications in many fields. It is well known that this problem is

3: combinatorial and NP-hard in general.

4: %Therefore, it is important to choose the suitable substitution for this matrix rank minimization problem.

5: In this paper, a continuous promoting low rank non-convex fraction function is studied to replace the rank function in this NP-hard problem. Inspired by our former work in

6: compressed sensing, an iterative singular value thresholding algorithm is proposed to solve the regularization transformed affine matrix rank minimization

7: problem. For different $a>0$, we could get a much better result by adjusting the different value of $a$, which is one of the advantages for the iterative

8: singular value thresholding algorithm compared with some state-of-art methods. Some convergence results are established and numerical experiments show that this

9: thresholding algorithm is feasible for solving the regularization transformed affine matrix rank minimization problem. Moreover, we proved that the value of the

10: regularization parameter $\lambda>0$ can not be chosen too large. Indeed, there exists $\bar{\lambda}>0$ such that the

11: optimal solution of the regularization transformed affine matrix rank minimization problem is equal to zero for any $\lambda>\bar{\lambda}$. Numerical experiments

12: on matrix completion problems show that our method performs powerful in finding a low-rank matrix and the numerical experiments about image inpainting problems

13: show that our algorithm has better performances than some state-of-art methods.

14: 

15: \keywords{Affine matrix rank minimization \and Low-rank \and Matrix completion \and Fraction function \and Iterative singular value thresholding algorithm \and Image inpainting}

16: %\PACS{PACS code1 \and PACS code2 \and more}

17: \subclass{90C26 \and 90C27 \and 90C59}

18: \end{abstract}