Analysis Of Srgan To Upscaling Cctv Image
DOI:
https://doi.org/10.34818/eoe.v9i3.17985Abstract
Abstract Detailed information from high-resolution images is needed when analyzing the content contained in the image. However, sometimes the image has a low resolution so that the image is difficult to interpret. Based on these problems, a research was carried out on improving the quality of image resolution especially in forensic field. CCTV video results that have lowquality images and degraded with many noises, bad illumination, distortions, and blurs. So, we can remove that using Super-Resolution (SR) method. This can help during to identification, image interpretation, and analysis process clearly. Using Super-Resolution methods, a high-resolution image is obtained from a set of low-resolution images. The research was conducted based on the Super-Resolution Generative Adversarial Network (SRGAN). SRGAN is a generative model method that can generate data (images) with good quality. From the experiments that have been carried out, the system that has been built is proven to be able to produce images with good quality with the highest PSNR value is 24,873 and the highest SSIM value is 0.831. Keywords: CCTV, Super-Resolution, Generative Adversarial Network, PSNR, SSIMReferences
[1] B. C. Tom, N. P. Galatsanos, and A. K. Katsaggelos, “Reconstruction of a High Resolution Image from Multiple Low Resolution Images,” Super-Resolution Imaging, pp. 73–105, 2005, doi: 10.1007/0-306-47004-7_4.
[2] J. Zhu, C. Zhou, D. Fan, and J. Zhou, “A new method for superresolution image reconstruction based on surveying adjustment,” J. Nanomater., vol. 2014, 2014, doi: 10.1155/2014/931616.
[3] Q. He and R. Schultz, “Super-Resolution Reconstruction by Image Fusion and Application to Surveillance Videos Captured by Small Unmanned Aircraft Systems,” Sens. Fusion its Appl., 2010, doi: 10.5772/9978.
[4] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 6, pp. 1137–1149, 2017, doi: 10.1109/TPAMI.2016.2577031.
[5] K. He, G. Gkioxari, P. Dollar, and R. Girshick, “Mask R-CNN,” Proc. IEEE Int. Conf. Comput. Vis., vol. 2017-Octob, pp. 2980–2988, 2017, doi: 10.1109/ICCV.2017.322.
[6] J. Yang, J. Wright, T. S. Huang, and Y. Ma, “Image super-resolution via sparse representation,” IEEE Trans. Image Process., vol. 19, no. 11, pp. 2861–2873, 2010, doi: 10.1109/TIP.2010.2050625.
[7] R. Timofte, V. De, and L. Van Gool, “Anchored neighborhood regression for fast example-based super-resolution,” Proc. IEEE Int. Conf. Comput. Vis., pp. 1920–1927, 2013, doi: 10.1109/ICCV.2013.241.
[8] C. Dong, C. C. Loy, K. He, and X. Tang, “Image Super-Resolution Using Deep Convolutional Networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 38, no. 2, pp. 295–307, 2016, doi: 10.1109/TPAMI.2015.2439281.
[9] B. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Cnn实际训练的,” Commun. ACM, vol. 60, no. 6, pp. 84–90, 2012.
[10] C. Dong, C. C. Loy, and X. Tang, “Accelerating the super-resolution convolutional neural network,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 9906 LNCS, pp. 391–407, 2016, doi: 10.1007/978-3-319-46475-6_25.
[11] J. Kim, J. K. Lee, and K. M. Lee, “Deeply-recursive convolutional network for image super-resolution,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 1637–1645, 2016, doi: 10.1109/CVPR.2016.181.
[12] S. Sharma, K. Shanmugasundaram, and S. K. Ramasamy, “FAREC - CNN based efficient face recognition technique using Dlib,” Proc. 2016 Int. Conf. Adv. Commun. Control Comput. Technol. ICACCCT 2016, no. 978, pp. 192–195, 2017, doi: 10.1109/ICACCCT.2016.7831628.
[13] A. Toshev and C. Szegedy, “DeepPose: Human pose estimation via deep neural networks,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 1653–1660, 2014, doi: 10.1109/CVPR.2014.214.
[14] I. Goodfellow et al., “Generative adversarial networks,” Commun. ACM, vol. 63, no. 11, pp. 139–144, 2020, doi: 10.1145/3422622.
[15] C. Ledig et al., “href{https://ieeexplore.ieee.org/abstract/document/8099502}{Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network},” Cvpr, vol. 2, no. 3, p. 4, 2017, [Online]. Available: http://openaccess.thecvf.com/content_cvpr_2017/papers/Ledig_Photo-Realistic_Single_Image_CVPR_2017_paper.pdf.
[16] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, “Inception-v4, inception-ResNet and the impact of residual connections on learning,” 31st AAAI Conf. Artif. Intell. AAAI 2017, pp. 4278–4284, 2017.
[17] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., pp. 1–14, 2015.
