Edge AI system using a Thermal Camera for Industrial Anomaly Detection
Keywords: EdgeAi, TinyML, Predictive Maintenance, IoT, Industry 4.0
Authors: Vitor Oliveira, António H. J. Moreira
Preventive maintenance plays an important role in reducing long-term maintenance costs, unplanned downtime, and improving the lifetime of industrial machines. A common trait of machines is that they produce heat while working resulting in temperature patterns. Temperature can be a key parameter for monitoring the performance and condition of machines, further aiding the diagnostics of machine problems. This paper presents an Internet of Things (IoT) system that monitors and detects thermal anomalies in industrial machines using deep neural networks (DNNs). The proposed system uses edge Ai technology that enables the DNN to run and make predictions inside the embedded microcontroller reducing the amount of data needed to be transmitted to any external server. Furthermore, this system uses a platform that centralizes multiple sensors with the option to communicate with a server that runs two additional neural network that are specialized in highlighting zones of interest in the thermal image and monitoring the temperature behavior over time. Overall, the system performed well and identified most anomalies correctly. It also presented high adaptability to different industrial environments.