Fusion of multi source data for atmospheric corrosion evaluation using sensors and image recognition

Detecting corrosion in Q235 steel is crucial for ensuring the safety and maintenance of industrial facilities. In this study, we present a novel approach that combines image recognition techniques with corrosion sensor data to improve both the accuracy and real-time capabilities of corrosion monitoring. Our findings show that integrating image recognition significantly enhances the predictive power of atmospheric corrosion models. This improvement is attributed to the strong correlation between image texture features, contrast, and material corrosion. By integrating diverse data sources, we have developed a rapid atmospheric corrosion evaluation model, Qcorr, for efficient assessment of outdoor Q235 steel corrosion. We believe the Qcorr model will be a valuable tool for practical corrosion detection.