With recent advancement in deep learning areas, computer vision research has changed from hard coded features to end-to-end trained deep neural network. Super-resolution is one of such areas which is influenced by deep learning advancement. Super-resolution is the technique for reconstructing high-resolution images from a given set of images. It is very important to acquire better quality images in satellite images, medical images and surveillance monitors where analysis of low quality images is extremely difficult. In this paper a novel approach to solve the problem of super-resolution image is presented. Proposed method trained the network using feedforward convolutional neural network and combined with perceptual loss function which measure the semantic differences between images and helps in reducing the computational complexity of overall super-resolution images. The proposed method also uses the adversarial network which helps in achieving the finer details in images.

Super-Resolution, Convolutional Neural network, Sub-Pixel Convolutional Layer, Perceptual Loss

[1] C. Dong, C. C. Loy, K. He, and X. Tang, “Image super-resolution using deep convolutional networks” IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(2):295–307, 2016.
[2] Yang, C.Y, Huang,J.B., Yang, M.H,”Exploiting self-similarities for single frame super-resolution”, In: IEEE Asian Conference on Computer Vision, pp.497–510 (2010).
[3] D. Glasner, S. Bagon and M. Irani,”Super-Resolution from a Single Image”,Proc. Int. Conf. Computer Vision, Kyoto, Japan,2009.
[4] Shreyas Fadnavis Int, Image Interpolation Techniques in Digital Image Processing:”An Overview, Journal of Engineering Research and Applications” ISSN: 2248-9622, Vol. 4, Issue 10( Part -1), pp.70-73, October 2014.
[5] Amisha J Shah, Suryakant B.Gupta and Rujul Makwana, “Single Image Super-Resolution via Non Sub-sample Contourlet Transform based Learning and a Gabor Prior”, International Journal of Computer Applications (0975–8887)Volume 64–No.18, February 2013.
[6] M. Elad. “Sparse and Redundant Representations: From Theory to Applications in Signal and Image Processing”, Springer Publishing Company, Incorporated, 1st edition, 2010.
[7] Yang C.Y, Ma C, Yang M.H,“Single-image super resolution: a benchmark”, Springer, Computer Vision (ECCV) pages 372-386,2014.
[8] S. Schulter, C. Leister, and H. Bischof,“Fast and accurate image upscaling with super-resolution forest” IEEE, Conference on computer vision and pattern recognition, pages 3791-3799,2015.
[9] N.S.Lele,”Image Classification Using Convolutional Neural Network”, International Journal of Scientific Research in Computer Science and Engineering, Vol.6, Issue.3, pp.22-26 , 2018.
[10] W. Shi, J. Caballero, F. Huszar, J. Totz, A. P. Aitken, R. Bishop, D. Rueckert, and Z. Wang, “Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network”, In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1874–1883, 2016.
[11] Johnson, Justin, Alexandre Alahi, and Li Fei-Fei, “Perceptual losses for real-time style transfer and super-resolution”, European Conference on Computer Vision. Springer International Publishing, 2016.
[12] He, K., Zhang, X., Ren, S., Sun, J. “Deep residual learning for image recognition” arXiv preprint arXiv: 1512.03385 (2015).
[13] C. Ledig, L. Theis, F. Huszar, J. Caballero, A. Cunning-ham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang, and W. Shi. “Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network”, arXiv.org, Sept. 2016.
[14] Lin, T.Y, Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L: “Microsoft coco: Common objects in context. In: Computer Vision– ECCV 2014”, Springer (2014) 740–755.