Rebar recognition of underwater concrete via total focusing method enhanced with adaptive windowing and amplitude-phase coherence factor

Ground penetration radar is the primary method used for internal detection of concrete. However, water interference prevents the ground penetration radar from operating effectively underwater. To overcome this limitation, this study presents an ultrasonic-based method for accurately detecting inclusions in concrete. The process begins by applying an adaptive window technique to full matrix data acquired from an ultrasonic array, aiming to suppress interfacial wave components. In addition, an amplitude-phase coherence factor is incorporated into the total focusing method to mitigate the effects of heterogeneous scattering. A dedicated algorithm is further developed to estimate the location and diameter of rebars. The process is validated through simulations, laboratory experiments, and field tests. Rebar diameter errors are 1.51 mm, 1.31 mm, and 1.38 mm, while localization errors are 4.15 mm, 7.01 mm, and 6.66 mm, respectively. This work demonstrates the feasibility of ultrasonic imaging for visualizing internal features in underwater concrete, contributing to nondestructive testing and structural health monitoring of concrete infrastructure.