Laparoscopic videos are suffering from various distortions during the surgery which lead to loss of visual quality. These distortions have impact on a surgeon’s visibility and other related tasks such as segmentation and instrument tracking in robot-assisted surgery and image guided navigation systems. The distortions in a laparoscopic video are due to technical problems in the equipment or side-effects of the instruments such as smoke. To address these problems, most of the existing solutions rely on making some changes to the technical equipment using one of the many available troubleshooting options. The existing solutions are time-consuming and not robust enough. Therefore, automated video enhancement systems are required to avoid previous problems. Identification of distortion is the main and important component in the feedback loop to enhance the video quality in real time. This research work aims to address this problem by developing a deep learning model for distortion classification as the first step towards designing a robust enhancement system. This study aims to improve the performance metrics of the model including accuracy, precision, and recall. Additionally, it aims to speed up the process of distortion classification to run the enhancement system in real time. The output of this project is a fast, robust, and accurate model that is able to classify various distortions in laparoscopic videos.