Comprehensive property retrieval and measurement of concrete spalling using machine vision for post-earthquake safety assessments

Abstract

Current procedures in post-earthquake safety and structural assessment are performed by a skilled triage team of structural engineers/certified inspectors. These procedures, in particular the physical measurement of the damage properties, are time-consuming and qualitative in nature. Spalling has been accepted as an important indicator of significant damage to structural elements during an earthquake, and thus provides a sound springboard for a model in machine vision automated assessment procedures as is proposed in this research. Thus, a novel method that automatically detects regions of spalling on reinforced concrete columns and measures their properties in image data is the specific focus of this work. According to this method, the region of spalling is first isolated by way of a local entropy-based thresholding algorithm. Following this, the properties of the spalled region are depicted by way of classification of the extent of spalling on the column. The region of spalling is sorted into one of three categories by way of a novel global entropy-based adaptive thresholding algorithm in conjunction with well-established image processing methods in template matching and morphological operations. These three categories are specified as the following: (1) No spalling; (2) Spalling of cover concrete; and (3) Spalling of the core concrete (exposing reinforcement). In addition, the extent of the spalling along the length of the column is quantified. The method was tested on a database of damaged RC column images collected after the 2010 Haiti Earthquake, and comparison of the results with manual measurements indicate the validity of the method.