Stochastic modelling of chloride-induced pitting corrosion of reinforcement bars in concrete.

Chloride induced corrosion, caused primarily by salt spray in marine environments, airborne salts and de-icing salts, is one of the most common deterioration processes in reinforced concrete structures. At present, most of the models found in literature describe uniform corrosion and those that do address loca-lized corrosion focus on a simplified definition of the reduced cross-sectional area of corroded rebars without due attention to physical characteristics and spatial variability. This may be attributed to the limitations of cur-rent manual methods used in evaluating the corrosion characteristics on the surface of reinforcement. In this paper, an automated procedure for the acquisition of corrosion depth data on rebars based on 3D laser scan-ning is investigated. Moreover, the first results of an analysis process based on image analysis using wavelet theory are presented. These results show a promising way of improving the classification of corrosion depths. This can be useful for the relation between spatial distribution of corrosion and mechanical properties of the corroded element.