Facial recognition is a classic computer vision task. State-of-the-art facial recognition systems can accurately identify and differentiate between faces on standard datasets. However, the impact of varying light intensities poses a significant challenge to visual identification systems. Despite widespread acknowledgement and research into the issue of inadequate lighting, its nuanced effects, and tailored strategies for mitigation in specific scenarios remain largely unexplored. To facilitate facial recognition under low light conditions, image enhancement is therefore a very necessary step. In our study, we developed and tested a variety of facial recognition pipelines that incorporate different low light image enhancement methods with a state-of-the-art face recognition model in order to test the efficacy of using these techniques in real-time facial recognition systems. Furthermore, we have curated a unique facial recognition dataset, which is derived from a subset of the DigiFace dataset, for the purpose of testing the face recognition pipelines on both well-lit and low light versions of the same image. Ultimately, our investigation involved a comparative analysis of the Face Recognition System performance across three scenarios: normal illuminated images, low light versions of the images, and post-enhancement versions of the images, for all possible pipelines.