Environmental contamination and pollution under specific weather conditions can cause discharges on the surface of insulators in high-voltage lines and stations, leading to possible service disruptions. This paper presents the first attempt at a new approach in surface discharge detection, based on artificial intelligence and computer vision techniques. The research is carried out as part of a collaboration with TERNA (the Italian TSO). Videos, collected by the LANPRIS test station camera for the purpose of monitoring the aging of insulators, were used to explore the potential of an object detection approach using an artificial neural network to recognize insulators in the videos and to distinguish insulators with normal behavior from those subject to surface discharge phenomena. The input data are described in the context of the LANPRIS experiment; the model chosen for object detection and the pipeline, created to analyze the video files, are presented in this paper; the tools used for this work are explained in detail; and the results of the research are discussed. This study is the first part of a larger work aimed at experimenting with artificial intelligence and computer vision techniques in systems that monitor very important components of the electrical system, such as the insulators of high-voltage lines.