Advertisement

DeepDeblur: text image recovery from blur to sharp

  • Jianhan Mei
  • Ziming Wu
  • Xiang Chen
  • Yu Qiao
  • Henghui Ding
  • Xudong JiangEmail author
Article
  • 22 Downloads

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.

Keywords

Text Deblurring Convolutional Neural Network (CNN) Blind deconvolution Short connection 

Notes

References

  1. 1.
    Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, Corrado GS, Davis A, Dean J, Devin M, Ghemawat S, Goodfellow I, Harp A, Irving G, Isard M, Jia Y, Jozefowicz R, Kaiser L, Kudlur M, Levenberg J, Mané D, Monga R, Moore S, Murray D, Olah C, Schuster M, Shlens J, Steiner B, Sutskever I, Talwar K, Tucker P, Vanhoucke V, Vasudevan V, Viégas F, Vinyals O, Warden P, Wattenberg M, Wicke M, Yu Y, Zheng X (2015) TensorFlow: Large-scale machine learning on heterogeneous systems. http://tensorflow.org/. Software available from tensorflow.org
  2. 2.
    Anat L, Yair W, Fredo D, William TF (2009) Understanding and evaluating blind deconvolution algorithms. In: International conference on computer vision and pattern recogintion (CVPR)Google Scholar
  3. 3.
    Boracchi G, Foi A (2011) Uniform motion blur in poissonian noise: blur/noise trade-off. In: IEEE transactions on image processing (TIP)Google Scholar
  4. 4.
    Boracchi G, Foi A (2012) Modeling the performance of image restoration from motion blur. In: IEEE transactions on image processing (TIP)Google Scholar
  5. 5.
    Chakrabarti A (2016) A neural approach to blind motion deblurring. In: Europeon conference on computer vision (ECCV), pp 221–235Google Scholar
  6. 6.
    Dilip K, Rob F (2009) Fast image deconvolution using hyper-laplacian priors. In: Conference and workshop on neural information processing systems (NIPS)Google Scholar
  7. 7.
    Dilip K, Terence T, Rob F (2011) Blind deconvolution using a normalized sparsity measure. In: International conference on computer vision and pattern recogintion (CVPR)Google Scholar
  8. 8.
    Gong D, Yang J, Liu L, Zhang Y, Reid ID, Shen C, van den Hengel A, Shi Q (2017) From motion blur to motion flow: a deep learning solution for removing heterogeneous motion blur. In: International conference on computer vision and pattern recogintion (CVPR), pp 3806–3815Google Scholar
  9. 9.
    Isola P, Zhu J, Zhou T, Efros AA (2017) Image-to-image translation with conditional adversarial networks. In: International conference on computer vision and pattern recogintion (CVPR), pp 5967–5976Google Scholar
  10. 10.
    Jian S, Wenfei C, Zongben X, Jean P (2015) Learning a convolutional neural network for non-uniform motion blur removal. In: International conference on computer vision and pattern recogintion (CVPR)Google Scholar
  11. 11.
    Jinshan P, Zhe H, Zhixun S, Ming-Hsuan Y (2014) Deblurring text images via l0-regularized intensity and gradient prior. In: International conference on computer vision and pattern recogintion (CVPR)Google Scholar
  12. 12.
    Kaiming H, Xiangyu Z, Shaoqing R, Jian S (2016) Deep residual learning for image recognition. In: International conference on computer vision and pattern recogintion (CVPR)Google Scholar
  13. 13.
    Kupyn O, Budzan V, Mykhailych M, Mishkin D, Matas J (2017) Deblurgan: blind motion deblurring using conditional adversarial networks. arXiv:1711.07064
  14. 14.
    Lee C, Xie S, Gallagher PW, Zhang Z, Tu Z (2015) Deeply-supervised nets. In: Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, AISTATS 2015, San Diego, California, USA, May 9-12, 2015Google Scholar
  15. 15.
    Li X, Jiaya J (2010) Two-phase kernel estimation for robust motion deblurring. In: Europeon conference on computer vision (ECCV)Google Scholar
  16. 16.
    Li X, SJ RJ, Ce L, Jiaya J (2014) Deep convolutional neural network for image deconvolution. In: Conference and workshop on neural information processing systems (NIPS)Google Scholar
  17. 17.
    Michal H, Jan K, Pavel Z, Filip Š (2015) Convolutional neural networks for direct text deblurring. In: British machine vision conference (BMVC)Google Scholar
  18. 18.
    Nah S, Kim TH, Lee KM (2017) Deep multi-scale convolutional neural network for dynamic scene deblurring. In: International conference on computer vision and pattern recogintion (CVPR), pp 257–265Google Scholar
  19. 19.
    Nie L, Wang M, Zha Z, Chua T (2012) Oracle in image search: A content-based approach to performance prediction. ACM Trans Inf Syst 30(2):13:1–13:23CrossRefGoogle Scholar
  20. 20.
    Nie L, Wang X, Zhang J, He X, Zhang H, Hong R, Tian Q (2017) Enhancing micro-video understanding by harnessing external sounds. In: Proceedings of the 2017 ACM on multimedia conference, MM 2017, Mountain View, CA, USA, October 23-27, 2017, pp 1192–1200Google Scholar
  21. 21.
    Nie L, Yan S, Wang M, Hong R, Chua T (2012) Harvesting visual concepts for image search with complex queries. In: Proceedings of the 20th ACM multimedia conference, MM ’12, Nara, Japan, October 29 - November 02, 2012, pp 59–68Google Scholar
  22. 22.
    Noroozi M, Chandramouli P, Favaro P (2017) Motion deblurring in the wild. In: Pattern recognition - 39th german conference (GCPR), pp 65–77Google Scholar
  23. 23.
    Qi S, Jiaya J, Aseem A (2008) High-quality motion deblurring from a single image. In: ACM transactions on graphics (TOG)Google Scholar
  24. 24.
    Ramakrishnan S, Pachori S, Gangopadhyay A, Raman S (2017) Deep generative filter for motion deblurring. In: International conference on computer vision (ICCV), pp 2993–3000Google Scholar
  25. 25.
    Rob F, Barun S, Aaron H, Sam TR, William TF (2006) Removing camera shake from a single photograph. In: ACM Transactions on graphics (TOG)Google Scholar
  26. 26.
    Ruomei Y, Ling S (2013) Image blur classification and parameter identification using two-stage deep belief networks. In: British machine vision conference (BMVC)Google Scholar
  27. 27.
    Sergey I, Christian S (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167
  28. 28.
    Yangqing J, Evan S, Jeff D, Sergey K, Jonathan L, Ross G, Sergio G, Trevor D (2014) Caffe: convolutional architecture for fast feature embedding. arXiv:1408.5093

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Electrical and Electronics EngineeringNanyang Technological UniversitySingaporeSingapore
  2. 2.The Hong Kong University of Science and TechnologyHong KongChina
  3. 3.Fachbereich InformatikTechnische Universität DarmstadtDarmstadtGermany
  4. 4.Shenzhen Institutes of Advanced TechnologyChinese Academy of SciencesShenzhenChina

Personalised recommendations