Extraction and measurement of recessive and conventional cracks in underground engineering concrete structures

Abstract

Crack damage is a crucial aspect of monitoring the health of concrete structures. This study introduces the concept of recessive cracks and proposes novel methods for their visualization. These methods enable the early detection and potential prevention of damage in underground concrete structures before cracks fully develop. Parameters such as infiltration time, evaporation rate, infiltration shadow width, and damage depth were measured during the visualization of recessive cracks, thereby achieving the primary objective of assessing the extent of damage caused by these cracks. Additionally, a novel algorithm, the “Similarity–Difference Algorithm,” was developed to track the development and evolution of recessive cracks at different stress stages. Furthermore, this study addresses the extraction, segmentation, and measurement of conventional cracks with spiderweb-like interlacing. Finally, outdoor experiments were conducted to validate the reliability and suitability of the proposed method for real-world applications.