Overcoming Blurred Vision: A YOLO-NAS and Cycle GAN-based Approach for Accurate ANPR in Challenging Images

Author(s)	Dr. Nitin Manik Gaikwad, Dr. Soojey Ramchandra Deshpande
Country	India
Abstract	This paper introduces a novel two-stage methodology to enhance the accuracy of Number Plate Recognition (NPR) systems, particularly in challenging scenarios characterized by blurred images. The proposed approach integrates the You Only Look One Network Architecture (YOLO-NAS) for precise vehicle detection and Generative Adversarial Networks (CYCLE GANs) for effective de-blurring of license plates. In the first stage, YOLO-NAS ensures accurate identification of vehicles even in the presence of image blur. The second stage employs a specially crafted CYCLE GAN architecture to de-blur license plates, preserving critical details for accurate character recognition. This integrated method not only overcomes the limitations of traditional NPR systems but also sets a new standard for accuracy and reliability in challenging conditions. The proposed solution holds significant potential for diverse applications, including toll roads, parking areas, and restricted zones, marking a transformative leap in addressing blurred vision challenges and enhancing the efficiency of NPR systems in real-world scenarios.
Keywords	CYCLE GAN, NPR, CNN, YOLO-NAS and OCR
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-27
DOI	https://doi.org/10.36948/ijfmr.2026.v08i03.79722

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research