Evaluation of speckle-interferometry descriptors to measuring drying-of-coatings

In the industrial process of painting, paint-drying is an important stage because of its high impact in the final result. Its study is of relevance to improve the properties of the resulting coating. Amalvy's experiments to measure the speed of drying on surfaces, based on techniques of speckle interferometry, have been used as a starting point in the evaluation of other methods, which allow to measure the process with greater accuracy. Haralick's descriptors have been studied in depth, then filters based on mathematical morphology techniques, a natural complexity measure and, finally, local binary patterns. Measures of speed of drying based on gravimetrical information were obtained and used as a gold standard. The comparison of different techniques was based on their ability to predict its values through a linear regression model. Morphological descriptors showed a low dependance with the sampling time, a desired property. Permutation entropy and local binary patterns evinced similar drying curves, showing a remarkable inflection point, coincident with the passage on the constant drying area to a later state, defined by a slower diffusion of the solvent through the dry coat of the surface. More precise descriptors of drying phenomena have been identified in this study.

► We compared different techniques to measure speed of paint drying trough speckle images. ► We obtained more precise descriptors of drying phenomena in this study. ► Morphological descriptors showed outstanding results, providing accuracies above 99% in all cases. ► Permutation entropy and local binary patterns showed an inflection point, coincident with a notable change of drying rate.