Image processing applied to the analysis of segregation in lightweight aggregate concretes

Lightweight concretes are susceptible to present aggregate segregation due to density differences between its components. In fact, during concrete vibration, lightweight aggregate tends to float, which may affect the concrete’s mechanical properties. This fact justifies the needs for the identification and quantification of this phenomenon. The present work presents a means of evaluating the segregation via an image processing based technique. Moreover, an index for classifying concretes according to the aggregates uniformity distribution is proposed. The examples and validation tests indicate that this uniformity analysis may be efficiently applied to the evaluation of aggregate distribution on concrete samples.

► We propose an image based technique to analyse aggregates segregation in concrete.
► Concretes are classified via a uniformity scale based on image processing data.
► Synthetic and actual images of concrete were classified into 4 uniformity levels.
► It was detected a relation between aggregate fraction and the mixture uniformity.
► Results show good agreement with observed experimental features.