Thermographic inspection using a microcontroller-based camera positioning system

Abstract

Infrared thermal cameras can be used to detect early-stage thermal anomalies in electrical or mechanical equipment and structures, before they lead to the interruption of their operation. In this work, we present a system to control the positioning and movement of a thermographic camera, mounted on a mechanical support with six degrees of freedom, employing the low cost microcontroller board Arduino. The proposed system moves the camera to positions set and configured a priori, and provides an automatic diagnosis of target elements on the camera's field of view, based on thermographic images by applying image processing and artificial neural network techniques. The surrounding region of a step-up transformer located in the outer area of a hydroelectric power plant was selected to validate the system. The system also allows custom remote inspections within the camera's field of view since it can be accessed online and be operated in manual mode.