Efficient-VDSR Network for Super-Resolution
DOI:
https://doi.org/10.54097/hset.v9i.1865Keywords:
Super-resolution; Deep Learning; Residual Learning; VDSR Network; Bicubic interpolation; Quantitative Comparison.Abstract
Super-resolution image processing is an important research topic in the field of image processing. The exploration in this paper comes from the use of VDSR network, which can improve the learning rate by only detecting and learning the residuals since the difference between low-resolution and high-resolution images lies more in the high-frequency detail part. Increasing the depth of the network can significantly improve the learning rate and further affect the final rendering of the image, but the increase of weight layers also increases the number of iterations and training time. This paper intends to reduce the number of iterations by decreasing the number of weight layers and correspondingly shorten the training time of the model. Then, we create low-resolution images and improve the resolution of the images by using bicubic interpolation method and the trained network respectively, and then obtain the evaluation data of the image quality by final visualization and quantitative comparison. The results show that reducing the number of weight layers still leads to a high accuracy of the images and will reduce the number of iterations and time accordingly.
Downloads
References
W. T. Freeman, E. C. Pasztor, and O. T. Carmichael. Learning low-level vision. International journal of computer vision, 40(1):25–50, 2000. 1
M. Irani and S. Peleg. Improving resolution by image registration. CVGIP: Graphical models and image processing, 53(3):231–239, 1991. 1
J. S. Isaac and R. Kulkarni, “Super-resolution techniques for medical image processing,” in Proc. Int. Conf. Technologies Sustain. Develop., 2015, pp. 1–6.
M. W. Thornton, P. M. Atkinson, and D. a. Holland. Sub-pixel mapping of rural land cover objects from fine spatial resolution satellite sensor imagery using super-resolution pixel-swapping. International Journal of Remote Sensing, 27(3):473–490, 2006. 1
B. K. Gunturk, A. U. Batur, Y. Altunbasak, M. H. Hayes, and R. M. Mersereau. Eigenface-domain super-resolution for face recognition. IEEE Transactions on Image Processing, 12(5):597–610, 2003. 1
Kim, J., J. K. Lee, and K. M. Lee."Accurate Image Super-resolution Using Very Deep Convolutional Networks."Proceedings of the IEEE®Conference on Computer Vision and Pattern Recognition.2016, pp. 1646-1660.
C. Dong, C. C. Loy, K. He, and X. Tang, “Learning a deep convolutional network for image super-resolution,” in Proc. Eur. Conf. Comput. Vis., 2014, pp. 184–199.
I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2014, pp. 2672–2680.
Zhihao Wang , Jian Chen , and Steven C. H. Hoi ,” Deep Learning for Image Super-resolution: A Survey”, IEEE Transactions on Pattern Analysis and Machine Intellgence,VoL.43,No.10,October 2021,pp.3369-3377
C. Dong, C. C. Loy, K. He, and X. Tang. Image super-resolution using deep convolutional networks. TPAMI, 2015. 1-4
He K, Zhang X, Ren s, et al. Deep residual learning for image recognition[C]. IEEE Conference on Computer Vision,2015,115(3):211-252.
Grubinger, M., P. Clough, H. Müller, and T. Deselaers."The IAPR TC-12 Benchmark:A New Evaluation Resource for Visual Information Systems."Proceedings of the OntoImage 2006 Language Resources For Content-Based Image Retrieval.Genoa, Italy.Vol. 5, May 2006, p. 10.
Y. Bengio, P. Simard, and P. Frasconi. Learning long-term dependencies with gradient descent is difficult. Neural Networks, IEEE Transactions on, 5(2):157–166, 1994. 3, 4
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
