Anomaly Detection in Television Digital Channel

Detecting anomalies in industrial processes is a field in constant advancement. However, automating this task presents significant challenges due to the complexity of the problem. In this paper, Deep Learning techniques were employed to detect anomalies in video footage during digital channel testing of televisions on a production line. A 3D Convolutional Neural Network was trained on a dataset containing two classes of videos: those with simulated defects and those without defects. The resulting model achieved an accuracy of 98,45% with a processing speed of 648 FPS.