Image-Based Quantification of Aggregate Segregation and Orientation in Hardened Concrete

The segregation and orientation of coarse aggregates significantly influence the mechanical behaviour and durability of cement concrete. This study investigates these characteristics using digital image analysis techniques through ImageJ and MATLAB software. Concrete samples were prepared following ASTM C33-01 standards, using two aggregate sizes (1½” and 1”), and produced under both controlled (laboratory) and uncontrolled (field-simulated) conditions to examine the effects of boundary confinement on aggregate distribution. Additionally, the influence of vertical reinforcement (simulated using three #4 steel bars) on aggregate behaviour was also assessed. Concrete cylinders and slabs were cast, and core samples were extracted and sectioned at bottom, middle, and top levels. Surface images were acquired using a flatbed scanner, yielding 48 high-resolution digital cross-sections. ImageJ macros and MATLAB algorithms were employed to process and quantify aggregate segregation and orientation. Results indicate that image-based analysis provides a reliable and effective method for assessing aggregate distribution in hardened concrete. Findings reveal that segregation varies significantly with casting conditions and height within the specimens. Uncontrolled samples exhibited greater peripheral segregation compared to centrally confined ones. Reinforced samples also showed altered distribution patterns due to the presence of steel bars. These insights contribute to a better understanding of mix performance and can guide improvements in construction quality control and concrete durability.