The traditional image restoration method only uses the original image data as a dictionary to make sparse representation of the pending blocks, which leads to the poor adaptation of the dictionary and the blurred image of the restoration. And only the effective information around the restored block is used for sparse coding, without considering the characteristics of image blocks, and the prior knowledge is limited. Therefore, in the big data environment, a new method of image restoration based on structural coefficient propagation is proposed. The clustering method is used to divide the image into several small area image blocks with similar structures, classify the images according to the features, and train the different feature types of the image blocks and their corresponding adaptive dictionaries. According to the characteristics of the restored image blocks, the restoration order is determined through the sparse structural propagation analysis, and the image restoration is achieved by sparse coding. The design method is programmed, and the image restoration in big data environment is realized by designing the system. Experimental results show that the proposed method can effectively restore images and has high quality and efficiency.

1. [1] H. Kutscha and D. J. Verschuur, “The utilization of the double focal transformation for sparse data representation and data reconstruction,” Geophysical Prospecting, Vol.64, No.6, pp. 1498-1515, 2016.
2. [2] W. Dong, G. Shi, Y. Ma et al., “Image Restoration via Simultaneous Sparse Coding: Where Structured Sparsity Meets Gaussian Scale Mixture,” Int. J. of Computer Vision, Vol.114, No.2-3, pp. 217-232, 2015.
3. [3] J. Liu, T. Z. Huang, I. W. Selesnick et al., “Image restoration using total variation with overlapping group sparsity,” Information Sciences, Vol.295, No.C, pp. 232-246, 2015.
4. [4] A. Li, D. Chen, G. Sun et al., “Sparse representation-based image restoration via nonlocal supervised coding,” Optical Review, Vol.23, No.5, pp. 776-783, 2016.
5. [5] L. K. Liu, S. Chan, and T. Nguyen, “Depth Reconstruction from Sparse Samples: Representation, Algorithm, and Sampling,” IEEE Trans. on Image Processing, Vol.24, No.6, pp. 1983-1996, 2015.
6. [6] X. He, J. Fan, and Z. Zheng, “Image Restoration with l2-type edge-continuous overlapping group sparsity,” Pattern Recognition Letters, Vol.68, No.P1, pp. 211-216, 2015.
7. [7] J. Duan, Z. Pan, B. Zhang et al., “Fast algorithm for color texture image restoration using the non-local CTV model,” J. of Global Optimization, Vol.62, No.4, pp. 853-876, 2015.
8. [8] Q. J. Wu, Y. Sun, and L. Zhao, “Research and simulation of sparse representation of deep image super-resolution reconstruction,” Computer simulation, Vol.34, No.5, pp. 234-237, 2017.
9. [9] G. Zhang and N. Kingsbury, “Variational Bayesian image restoration with group-sparse modeling of wavelet coefficients,” Digital Signal Processing, Vol.47, No.C, pp. 157-168, 2015.
10. [10] L. Zhang and W. Zuo, “Image Restoration: From Sparse and Low-Rank Priors to Deep Priors [Lecture Notes],” IEEE Signal Processing Magazine, Vol.34, No.5, pp. 172-179, 2017.
11. [11] D. Q. Chen and Y. Zhou, “Wavelet Frame Based Image Restoration via Combined Sparsity and Nonlocal Prior of Coefficients,” J. of Scientific Computing, Vol.66, No.1, pp. 196-224, 2016.
12. [12] M. Kallel, R. Aboulaich, A. Habbal et al., “A Nash-game approach to joint image restoration and segmentation,” Applied Mathematical Modelling, Vol.38, No.11-12, pp. 3038-3053, 2014.