Abstract | ||
---|---|---|
Digital images could be degraded by a variety of blur during the image acquisition (i.e. relative motion of cameras, electronic noise, capturing defocus, and so on). Blurring images can be computationally modeled as the result of a convolution process with the corresponding blur kernel and thus, image deblurring can be regarded as a deconvolution operation. In this paper, we explore to deblur images by approximating blind deconvolutions using a deep neural network. Different deep neural network structures are investigated to evaluate their deblurring capabilities, which contributes to the optimal design of a network architecture. It is found that shallow and narrow networks are not capable of handling complex motion blur. We thus, present a deep network with 20 layers to cope with text image blur. In addition, a novel network structure with Sequential Highway Connections (SHC) is leveraged to gain superior convergence. The experiment results demonstrate the state-of-the-art performance of the proposed framework with the higher visual quality of the delurred images. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1007/s11042-019-7251-y | Multimedia Tools and Applications |
Keywords | Field | DocType |
Text Deblurring, Convolutional Neural Network (CNN), Blind deconvolution, Short connection | Computer vision,Pattern recognition,Deblurring,Blind deconvolution,Computer science,Convolution,Deconvolution,Motion blur,Network architecture,Digital image,Artificial intelligence,Artificial neural network | Journal |
Volume | Issue | ISSN |
78 | 13 | 1380-7501 |
Citations | PageRank | References |
2 | 0.42 | 10 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jianhan Mei | 1 | 2 | 0.76 |
Ziming Wu | 2 | 29 | 3.47 |
Xiang Chen | 3 | 139 | 30.34 |
Yu Qiao | 4 | 2267 | 152.01 |
Henghui Ding | 5 | 36 | 10.78 |
Xudong Jiang | 6 | 1885 | 117.85 |