Abstract—This paper proposed a new patch-wise image
inpainting algorithm based on sparse representation with
related dictionaries. For each target patch, i.e., the patch with
corrupted parts, a related dictionary is defined, which consists
of the patches in the uncorrupted area having a similar
histogram with that of the patch. The related dictionary
contains the most similar patches for target patch. That ensure
the accuracy of the inpainting process. The target patch then can
be inpainted by a sparse representation of the patches in the
related dictionary. Experimental results are given to show the
effectiveness of the proposed algorithm.
Index Terms—Image inpainting, sparse representation, histogram, related dictionaries.
Qiaoqiao Li is with the Graduate School of Systems Science and Technology, Akita Prefectural University, Akita, 015-0055, Japan (e-mail: D18S005@akita-pu.ac.jp).
Guoyue Chen, Xingguo Zhang, Kazuki Saruta, and Yuki Terata are with the Dept. of Electronics and Information Systems, Akita Prefectural University, Akita, 015-0055, Japan (e-mail: email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org).
Cite: Qiaoqiao Li, Guoyue Chen, Xingguo Zhang, Kazuki Saruta, and Yuki Terata, "Image Inpainting Based on Related Dictionary Constructed by Histogram," International Journal of Machine Learning and Computing vol. 8, no. 5, pp. 477-482, 2018.