Infrared thermography technology based nondestructive testing in electrical equipment defect detection: the state-of-the-art by bibliography analysis

To ensure a safe operation and maintenance of electrical equipment, Infrared Non-Destructive Testing (IRNDT) has emerged as a preferred solution. This method is highly valued for its efficiency, accuracy, safety, and non-invasive detection of defects in electrical systems, thereby enhancing power system reliability. This paper outlines the principles, classifications, and factors affecting infrared imaging performance of IRNDT. It also discusses image processing techniques including noise reduction, enhancement, and segmentation. The article compares and analyzes traditional methods with machine learning and deep learning approaches in infrared image processing. Finally, it examines current limitations of IRNDT in electrical equipment inspection and proposes future research and development trends.