Combining GANs And Fuzzy Logic For Real-Time Image Enhancement In Autonomous Vehicles

Abstract

In this study, we present a novel approach for real-time image enhancement in autonomous vehicles by integrating Generative Adversarial Networks (GANs) with Fuzzy Logic. The increasing reliance on visual data for autonomous navigation necessitates high-quality image enhancement, particularly under challenging conditions such as low light or adverse weather. While GANs have shown great promise in generating high-resolution images, their deterministic nature and computational demands pose challenges for real-time applications. To address these issues, we propose a system that combines the image generation capabilities of GANs with the adaptability of Fuzzy Logic, allowing for context aware refinement of images based on environmental conditions. Experimental results demonstrate that our method significantly improves image quality, with a Peak Signal-to-Noise Ratio (PSNR) of 36.5 dB and a Structural Similarity Index (SSIM) of 0.93, outperforming traditional filtering methods and CNN-based enhancements. Despite a slight increase in processing time, the proposed system achieves a favourable balance between image quality and real-time performance, making it a robust solution for enhancing visual data in autonomous vehicles.