LAUSR

Abstract Internal curing has been widely used in high-strength concrete because it can replenish water for cement hydration and reduce self-shrinkage. The internal curing effects mostly depend on the self-shrinkage changes and mechanical parameters obtained through a destructive test. This research proposes a fast and nondestructive detection method based on image recognition analysis to determine the effects of internal curing material content on the performance of high-strength concrete. The superabsorbent polymer (SAP) was chosen as the internal curing material. Laboratory uniaxial compression tests, scanning electron microscopy (SEM), and energy dispersive spectroscopy were used to analyze the mechanical properties and phase characteristics of the concrete. Based on the SEM images, the gray level co-occurrence matrix was used to extract texture features of the concrete, and effective SEM image samples were screened out to verify the correlation between texture features and mechanical strength. The results showed that: (1) the four characteristic values of GLCM have a significant linear correlation with SAP content; (2) with the increase in SAP content, self-shrinkage decreases and reaches the smallest point when the SAP content is 0.24 % of the cementing material content; and (3) the relative dynamic modulus of elasticity and maximum shrinkage are negatively correlated with the energy and correlation and positively correlated with the contrast and entropy. Such results provide a simple method of determining the influence of internal curing content on concrete performance.