Abstract:
Backlit image enhancement is a crucial task in improving the quality and visibility of the underexposed regions in an image caused by the difference in illumination betwe...Show MoreMetadata
Abstract:
Backlit image enhancement is a crucial task in improving the quality and visibility of the underexposed regions in an image caused by the difference in illumination between the background and the foreground. Traditional methods struggle to effectively handle the dynamic range compression required for backlit image enhancement and fail to properly balance the exposure between the background and foreground areas. In this article, we propose a novel backlit image enhancement architecture using a modified U-Net and a unique 1\,\, {}\times {}1 discriminator-based conditional generative adversarial network (cGAN). Our approach incorporates custom hyperparameters, a tailored loss function, and bilateral-guided upsampling (BGU) for efficient enhancement of large images. We evaluated our model on the backlit image dataset (BAID) and achieved superior results in various image evaluation metrics, such as peak signal-to-noise ratio (PSNR), natural image quality evaluator (NIQE), and \Delta E ^{\circ} , demonstrating its effectiveness in enhancing both backlit and low-light (LOL) images. We have also compared the computational efficiency and resources used by our method with other state-of-the-art methods. Extending our work, we have incorporated our model into a pipeline for enhancing backlit traffic images and videos, which is then used to detect the license plates of vehicles with improved accuracy.
Published in: IEEE Transactions on Instrumentation and Measurement ( Volume: 72)