Enhancing License Plate Super-Resolution: A Layout-Aware and Character-Driven Approach
This addresses a real-world challenge in traffic surveillance for law enforcement and security applications, but it is incremental as it builds on existing super-resolution and LPR techniques.
The paper tackles the problem of license plate recognition from low-resolution surveillance images by introducing a novel loss function and GAN-based training, resulting in significant improvements in character reconstruction quality that outperform state-of-the-art methods.
Despite significant advancements in License Plate Recognition (LPR) through deep learning, most improvements rely on high-resolution images with clear characters. This scenario does not reflect real-world conditions where traffic surveillance often captures low-resolution and blurry images. Under these conditions, characters tend to blend with the background or neighboring characters, making accurate LPR challenging. To address this issue, we introduce a novel loss function, Layout and Character Oriented Focal Loss (LCOFL), which considers factors such as resolution, texture, and structural details, as well as the performance of the LPR task itself. We enhance character feature learning using deformable convolutions and shared weights in an attention module and employ a GAN-based training approach with an Optical Character Recognition (OCR) model as the discriminator to guide the super-resolution process. Our experimental results show significant improvements in character reconstruction quality, outperforming two state-of-the-art methods in both quantitative and qualitative measures. Our code is publicly available at https://github.com/valfride/lpsr-lacd