Autonomous vehicles perceive information about their surrounding environment using camera input. However, the reduced visibility imposed by challenging weather conditions limits the camera’s ability to capture clear images. This increases the risk of accidents in cars, as vehicle control and navigation systems are severely compromised. Therefore, this paper evaluates the performance of different image enhancement techniques, ranging from conventional model-based methods to advanced neural networks, to improve the quality of images under foggy conditions. The methods are evaluated on benchmark foggy datasets, Dense-Haze and DAWN, using reference-based and non-reference-based metrics, respectively. The findings of this study serve as a reference for future research in developing weather-robust autonomous driving systems.