Computer vision algorithm based on fiber optic sensors and infrared thermal radiation images for fatigue detection under simulated operating conditions

Abstract

In modern industrial and engineering fields, fatigue detection of equipment and structures is an important link to ensure safety and extend service life. Traditional detection methods often rely on direct physical monitoring, which has certain limitations. In recent years, infrared thermal radiation imaging technology has attracted wide attention because of its non-contact and high sensitivity. This study aims to explore a new fatigue detection method based on infrared thermal radiation images by combining optical fiber sensor and computer vision algorithm, so as to improve the accuracy and real-time performance of fatigue diagnosis. In this study, a fiber optic sensor is used to monitor strain data in real time by applying periodic loads to different material and structural samples in an experimental environment. At the same time, infrared thermal imaging camera was used to obtain the temperature distribution information of the material surface. The infrared thermal radiation image is combined with the sensor data, and the deep learning algorithm is used to extract the feature and identify the fatigue state. The experimental results show that the infrared thermal radiation image can effectively reflect the temperature change of the material in the fatigue process, and complement the mechanical information provided by the optical fiber sensor. Through the constructed computer vision model, the classification accuracy of fatigue state is obviously better than the traditional detection means, which provides a new and effective method for fatigue detection, which can realize more efficient and accurate real-time monitoring, and has a wide application prospect.