مجله ماشین بینایی و پردازش تصویر

مجله ماشین بینایی و پردازش تصویر

ارائه یک شبکه مولد ترکیبی مبتنی بر توجه جهت افزایش وضوح تصاویر

نوع مقاله : مقاله پژوهشی

نویسندگان
1 دانشجوی کارشناسی ارشد هوش مصنوعی و رباتیکز، دانشکده فنی و مهندسی،گروه مهندسی کامپیوتر، دانشگاه اراک
2 گروه کامپیوتر/دانشکده فنی و مهندسی / دانشگاه اراک / شهر اراک/استان مرکزی/ایران
چکیده
بهبود کیفیت تصاویر در حوزه بینایی کامپیوتری به عنوان یکی از چالش‌های کلیدی مطرح شده است. در این مقاله، یک مدل جدید شبکه عصبی مولد متخاصم وضوح فوق‌العاده مبتنی بر توجه معرفی شده است که با ترکیب مدل‌های شبکه بهبود وضوح تصویر واقعی و شبکه توجه کانال‌محور رسوبی برای بهبود بازسازی تصاویر با وضوح بالا و کاهش اعوجاج‌ها و نویزهای پیچیده توسعه داده شده است. اعوجاج‌های تصویر شامل تحریفات هندسی، تاری و از بین رفتن جزئیات هستند که در تصاویر با وضوح پایین یا فشرده‌شده معمولاً مشاهده می‌شوند.در این مدل، قابلیت‌های شبکه بهبود وضوح تصویر واقعی برای کاهش نویز و افزایش وضوح به همراه توانایی شبکه توجه کانال‌محور رسوبی در حفظ جزئیات دقیق به کار گرفته شده‌اند. نتایج تجربی بر روی چند نمونه از مجموعه داده‌های شناخته‌شده در این حوزه نشان داده‌ است که مدل پیشنهادی بهبود عملکردی حدود 5 درصد به‌طور متوسط در معیار PIQEو ... را نسبت به روش‌های موجود مانند شبکه بهبود وضوح تصویر با تقویت توجه و شبکه بهبود وضوح تصویر واقعی در معیارهای شناخته‌شده بازسازی کور تصویر ارائه داده است. این بهبودها، مدل را قادر ساخته‌اند تا تصاویر را با وضوح بالا و حداقل اعوجاج و نویز بازسازی کند.
کلیدواژه‌ها

   [1]      M. Impieri, "Methods for Blind Super-Resolution of satellite images," Politecnico di Torino, 2023.
[2]        Z. Zhu, Y. Lei, Y. Qin, C. Zhu, and Y. Zhu, "IRE: improved image super-resolution based on real-ESRGAN," IEEE Access, vol. 11, pp. 45334-45348, 2023.
[3]        R. Li and W. Zhou, "Image Super-Resolution Reconstruction of Landslide Based on Real-ESRGAN," in Proceedings of the 2024 8th International Conference on Control Engineering and Artificial Intelligence, 2024, pp. 203-208.
[4]        Z. Huang, "Research on Super-Resolution Processing of Improved RCAN on SMT Blurred Images," in 2024 7th International Conference on Artificial Intelligence and Big Data (ICAIBD), 2024: IEEE, pp. 478-483.
[5]        Z. Lin et al., "Revisiting rcan: Improved training for image super-resolution," arXiv preprint arXiv:2201.11279, 2022.
[6]        Z. Wei, Y. Huang, Y. Chen, C. Zheng, and J. Gao, "A-ESRGAN: Training real-world blind super-resolution with attention U-Net Discriminators," in Pacific Rim International Conference on Artificial Intelligence, 2023: Springer, pp. 16-27.
[7]        S. Higashiyama et al., "Usefulness of a No-Reference Metric for Evaluation of Images in Nuclear Medicine-A Comparative Study with Visual Assessment," 2021.
[8]        J. Lin et al., "Suppressing Uncertainties in Degradation Estimation for Blind Super-Resolution," arXiv preprint arXiv:2406.16459, 2024.
[9]        A. H. Khan, C. Micheloni, and N. Martinel, "IDENet: Implicit degradation estimation network for efficient blind super resolution," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 6065-6075.
[10]      X. Wang, L. Sun, A. Chehri, and Y. Song, "A review of GAN-based super-resolution reconstruction for optical remote sensing images," Remote Sensing, vol. 15, no. 20, p. 5062, 2023.
[11]      ع. م. اسمعیلی زینی, ع. م. لطیف, و ق. برید لقمانی, "بزرگ‌نمایی تصویر با استفاده از تقریب بهترین صفحه," محاسبات نرم, vol. 5, no. 1, pp. 26-35, 2021. [Online]. Available: https://scj.kashanu.ac.ir/article_111402_3f35fb9a923d66814a83fd00efb75db9.pdf.
[12]      ع. دلشادی, و. مهرداد, و م. دولتشاهی, "بهینه‌سازی ساختاری شبکه عصبی عمیق سلسله مراتبی خودتوجه و رمزگذاری پویای تک متغیره برای بزرگنمایی تصاویر دیجیتال," مجله بینایی ماشین و پردازش تصویر, vol. 10, no. 3, pp. 47-63, 2023. [Online]. Available: https://jmvip.sinaweb.net/article_172083_cf349b830a5c8eac52b386fd954fff9c.pdf.
[13]      C. M. Ward, J. Harguess, B. Crabb, and S. Parameswaran, "Image quality assessment for determining efficacy and limitations of Super-Resolution Convolutional Neural Network (SRCNN)," in Applications of Digital Image Processing XL, 2017, vol. 10396: SPIE, pp. 19-30.
[14]      D. Vint, G. Di Caterina, J. Soraghan, R. Lamb, and D. Humphreys, "Evaluation of performance of VDSR super resolution on real and synthetic images," in 2019 Sensor Signal Processing for Defence Conference (SSPD), 2019: IEEE, pp. 1-5.
[15]      B. M. Kuriakose, "EDSR: Empowering super-resolution algorithms with high-quality DIV2K images," Intelligent Decision Technologies, vol. 17, no. 4, pp. 1249-1263, 2023.
[16]      C. Fiscone et al., "Generalizing the Enhanced-Deep-Super-Resolution neural network to brain MR images: a retrospective study on the Cam-CAN dataset," Eneuro, vol. 11, no. 5, 2024.
[17]      H. Wang, "MSRDN: A Super-Resolution Network for Human Body," in 2024 3rd International Conference on Innovations and Development of Information Technologies and Robotics (IDITR), 2024: IEEE, pp. 178-182.
[18]      D. Kong, L. Gu, X. Li, and F. Gao, "Multi-Scale Residual Dense Network for the Super-Resolution of Remote Sensing Images," IEEE Transactions on Geoscience and Remote Sensing, 2024.
[19]      S. Anwar and N. Barnes, "Densely residual laplacian super-resolution," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 3, pp. 1192-1204, 2020.
[20]      I. Sayal, "Image transformation using super resolution," 2024.
[21]      I. J. Goodfellow, "On distinguishability criteria for estimating generative models," arXiv preprint arXiv:1412.6515, 2014.
[22]      I. Goodfellow et al., "Generative adversarial networks," Communications of the ACM, vol. 63, no. 11, pp. 139-144, 2020.
[23]      X. Wang et al., "Esrgan: Enhanced super-resolution generative adversarial networks," in Proceedings of the European conference on computer vision (ECCV) workshops, 2018, pp. 0-0.
[24]      X. Li and S. He, "Blind image quality evaluation method based on cyclic generative adversarial network," IEEE Access, 2024.
[25]      L. Zhang, L. Wang, Z. Yan, Z. Jia, H. Wang, and X. Tang, "Star Generative Adversarial VGG Network-Based Sample Augmentation for Insulator Defect Detection," International Journal of Computational Intelligence Systems, vol. 17, no. 1, pp. 1-16, 2024.
[26]      J. Lin et al., "Improving image restoration through removing degradations in textual representations," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 2866-2878.
[27]      Z. Zhang et al., "Rethink arbitrary style transfer with transformer and contrastive learning," Computer Vision and Image Understanding, vol. 241, p. 103951, 2024.
[28]      E. Goceri, "GAN based augmentation using a hybrid loss function for dermoscopy images," Artificial Intelligence Review, vol. 57, no. 9, p. 234, 2024.
[29]      Y. Li, F. Xiao, W. Liang, and L. Gui, "Multiply complementary priors for image compressive sensing reconstruction in impulsive noise," ACM Transactions on Multimedia Computing, Communications and Applications, vol. 20, no. 6, pp. 1-22, 2024.
[30]      G. George, "Pretrained U-Net: in-depth analysis of binary image segmentation in underwater marine environment," Machine Learning for Computational Science and Engineering, vol. 1, no. 1, p. 4, 2025.
[31]      K. Huang, X. Yin, T. Gu, and W. Gao, "Perceptual-Centric Image Super-Resolution using Heterogeneous Processors on Mobile Devices," 2024.
[32]      S. Wu and J. Zhang, "TargetSR: Towards Semantic location Real-World Image Super-Resolution with Diffusion Prior," in 2024 International Joint Conference on Neural Networks (IJCNN), 2024: IEEE, pp. 1-8.
[34]      J. Greffier et al., "Comparison of two deep-learning image reconstruction algorithms on cardiac CT images: A phantom study," Diagnostic and Interventional Imaging, vol. 105, no. 3, pp. 110-117, 2024.
[35]      G. Omarova et al., "No-reference quality assessment of medical images using contrast enhancement," Journal of Theoretical and Applied Information Technology, vol. 101, no. 1, pp. 267-281, 2023.
[36]      X. Li, J. Pan, J. Shang, A. Souri, and M. Gao, "An improved blind/referenceless image spatial quality evaluator algorithm for image quality assessment," International Journal of Computational Science and Engineering, vol. 27, no. 1, pp. 48-56, 2024.
[37]      A. Gavrovska, A. Samčović, and D. Dujković, "No-Reference Image Quality Assessment Based on Machine Learning and Outlier Entropy Samples," Pattern Recognition and Image Analysis, vol. 34, no. 2, pp. 275-287, 2024.
[38] H. F. Tolie and M. R. Faraji, "Screen content image quality assessment using distortion-based directional edge and gradient similarity maps," SignalProcessing: Image Communication, vol. 101, p. 116562, 2022