The application of hyperspectral imaging for assessment of coating – evaluation of thickness and degradation level

Protective coatings protects structures by isolating materials from environmental factors. Coatings degradation leads to costly maintenance and potential safety risks. Current paint condition assessment techniques are impractical for large surfaces. This study developed a comprehensive framework combines hyperspectral imaging (HSI) and machine learning (ML) to assess paint condition and predict its thickness under accelerated aging. Aluminum plates painted with polyurethane-based paint were prepared with different thicknesses and aging states. The HYSPEX SWIR 384 camera was used for data collection and the QUV was used for samples aging. Utilizing ML and deep neural network (DNN) models to predict paint thickness from 43 to 218 µm resulted in an RMSE of 11.6 µm and R² of 0.91. For the degradation assessment, the DNN model delivered an R² of 0.896 and RMSE of 160 hours in predicting aging hours. The obtained results highlight the effectiveness of HSI for paint condition assessment.