Abstract
Removing motion blur has been an important issue in computer vision literature. Motion blur is caused by the relative motion between the camera and the photographed object. However, in recent years, some achievements have been made in the research of image deblurring by using deep learning algorithms. In this paper, an enhanced adversarial network model is proposed. The proposed model can use the weight of feature channel to generate sharp image and eliminate draughtboard artefacts. In addition, the mixed loss function enables the network to output high-quality image. The proposed approach is tested using GOPRO datasets and Lai datasets. In the GOPRO datasets, the peak signal-to-noise ratio of the proposed approach is up to 28.674, and DeblurGAN is 27.454. And the structural similarity measure can be achieved up to 0.969, and DeblurGAN is 0.939. Furthermore, the images were obtained from China’s Chang’e 3 Lander to test the new algorithm. Due to the elimination of the chessboard effect, the deblurred image has a better visual appearance. The proposed method achieved higher performance and efficiency in qualitative and quantitative aspects using the benchmark dataset experiments. The results also provided various insights into the design and development of the camera pointing system, which was mounted on the Lander for capturing images of the moon and rover for Chang’e space mission.
Similar content being viewed by others
References
Polyu.edu.hk (2015) Life on Mars? Award-wining PolyU device could dig up the answer [online]. http://www.polyu.edu.hk/openingminds/en/story.php?sid=3. Accessed 10 Nov 2017
Polyu.edu.hk (2017) PolyU 70th Anniversary [online]. https://www.polyu.edu.hk/cpa/70thanniversary/memories_achievement10.htm. Accessed 3 Nov 2017
Gupta A, Joshi N, Zitnick CL, Cohen M, Curless B (2010) Single image deblurring using motion density functions. In: Proceedings of the ECCV, pp 171–184
Harmeling S, Hirsch M, Schölkopf B (2010) Space-variant single-image blind deconvolution for removing camera shake. In: Proceedings of the NIPS, Vancouver, British Columbia, Canada, pp 829–837
Hirsch M, Schuler CJ, Harmeling S, Scholkopf B (2011) Fast removal of non-uniform camera shake. In: Proceedings of the ICCV, pp 463–470
Whyte O, Sivic J, Zisserman A, Ponce J (2010) Non-uniform deblurring for shaken images. In: Proceedings of the CVPR, pp 491–498
Li W (2011) Parameter estimation and algorithm research of motion blur image restoration. M.S. dissertation, Department of Computer Application, Anhui University, Anhui, China
Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S et al (2014) Generative adversarial nets. In: Proceedings of the ICNOIP, pp 2672–2680
Sun J, Cao W, Xu Z, Ponce J (2015) Learning a convolutional neural network for non-uniform motion blur removal. In: Proceedings of the CVPR, pp 769–777
Nah S, Kim TH, Lee KM (2018) Deep multi-scale convolutional neural network for dynamic scene deblurring. arXiv preprint arXiv:1612.02177v2
Ramakrishnan S, Pachori S, Gangopadhyay A, et al (2017) Deep generative filter for motion deblurring. In: Proceedings of the ICCV, pp 2993–3000
Kupyn O, Budzan V, Mykhailych M, Mishkin D, Matas J (2017) Deblurgan: blind motion deblurring using conditional adversarial networks. arXiv preprint arXiv:1711.07064
Arjovsky M, Chintala S, Bottou L (2017) Wasserstein generative adversarial networks. In: Proceedings of the ICML, pp 214–223
Johnson J, Alahi A, Li FF (2016) Perceptual losses for real-time style transfer and super-resolution. In Proceedings of the ECCV, pp 694–711
Su S, Delbracio M, Wang J, Sapiro G, Heidrich W, Wang O (2017) Deep video deblurring for hand-held cameras. In: Proceedings of the CVPR, pp 1279–1288
Zhang K, Luo W, Zhong Y, Mia L, Liu W (2018) Adversarial Spatio-Temporal Learning for Video Deblurring. In: Proceedings of the TIP, pp 281–301
Chen H, Gu J, Gallo O, Liu M, Veeraraghavan A, Kautz J (2018) Reblur2Deblur: deblurring videos via self-supervised learning. In: ICCP
Mirza M, Osindero S (2014) Conditional generative adversarial nets. Comput Sci 63:2672–2680
Arjovsky M, Chintala S, Bottou L (2017) Wasserstein gan. arXiv preprint arXiv:1701.07875
Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Proceedings of the ICML
Shi W, Caballero J, Huszár F, et al (2016) Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 1874–1883
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 7132–7141
Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D et al (2015) Going deeper with convolutions. In: Proceedings of the CVPR, pp 1–9
Szegedy C, Vanhoucke V, Ioffe S et al (2016) Rethinking the inception architecture for computer vision. In: Proceedings of the CVPR
Szegedy C, Ioffe S, Vanhoucke V (2017) Inception-v4, inception-resnet and the impact of residual connections on learning. arXiv:1602.07261
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the CVPR, pp 770–778
Odena A, Dumoulin V, Olah C (2016) Deconvolution and checkerboard artifacts. http://distill.pub/2016/deconvcheckerboard/
Liand C, Wand M (2016) Precomputed real-time texture synthesis with Markovian generative adversarial networks. In: Proceedings of the ECCV, pp 702–716
Isola P, Zhu J, Zhou T, Efros A (2017) Image-to-image translation with conditional adversarial networks. In: Proceedings of the CVPR, pp 1125–1134
Maas AL, Hannun AY, Ng AY (2013) Rectifier nonlinearities improve neural network acoustic models. Proc. ICML 30(1):3
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. Comput Sci 362:1140
Deng J, Dong W, Socher R, Li LJ, Li K, Li FF (2009) ImageNet: a large-scale hierarchical image database. Computer Vision and Pattern Recognition. In: Proceedings of the CVPR, pp 248–255
Jiao J, Tu WC, He S, Lau RWH (2017) FormResNet: formatted residual learning for image restoration. In Proceedings of the CVPRW, pp 1034–1042
Kingmaand DP, Ba J (2017) Adam: a method for stochastic optimization. arXiv: arXiv:1412.6980v9
Chinese Academy of Sciences, China National Space Administration, The Science and Application Center for Moon and Deepspace Exploration. Chang’e 3data:Rover P anoramic Camera. http://planetary.s3.amazonaws.com/data/change3/pcam.html. 14 May 2018
Polyu.edu.hk (2017) The Hong Kong Polytechnic University-The Space Exploration Journey [online]. https://www.polyu.edu.hk/web/filemanager/en/content_155/1960/Appendix_Milestones_SpaceProjects_.pdf. Accessed 24 Nov 2017
Hore A, Ziou D (2010) Image quality metrics: PSNR vs. SSIM. In: Proceedings of the ICRP, pp 2366–2369
Acknowledgments
This work is supported by the Natural Science Foundation of Guangdong Province (2015A030310172). It is also partly supported by a Grant from the Department of Industrial and Systems Engineering of the Hong Kong Polytechnic University (H-ZG3K) and a Grant from Shenzhen Technology University (2018010802008).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Zhang, Y., Ma, S.Y., Zhang, X. et al. EDGAN: motion deblurring algorithm based on enhanced generative adversarial networks. J Supercomput 76, 8922–8937 (2020). https://doi.org/10.1007/s11227-020-03189-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-020-03189-y