Damage assessment on concrete surfaces using multi-spectral image analysis

To monitor the state of conservation of structures is extremely important to plan maintenance interventions, thus allowing minimizing associated costs. The aim of this paper is to investigate if Very High Spatial Resolution (VHSR) multispectral images can be used to detect, analyse and measure areas with biological colonization, stains and/or repair materials on exposed concrete surfaces. With this aim, the authors present a method, SurfCrete - Multi-spectral image analysis of concrete surfaces, which was tested to classify surface images of exposed concrete buildings where biological colonization, exposed aggregates, repairing mortars and cracks were previously detected on a visual inspection. This technique was performed using four multi-spectral bands (blue, green, red and near infrared). A vegetation index was also considered, as additional band information, to evaluate its usefulness in this kind of applications. SurfCrete allowed to achieve a global classification accuracy of 94%, which indicates that the identification of different materials, as well as damage, can be obtained from VHSR multispectral images using the developed methodology with a high accuracy. Surface mapping and statistical results are automatically obtained and stored in digital format which can be valuable when applied to large exposed concrete areas. Furthermore, results demonstrate that most accurate outputs can be derived using the near infrared band. (C) 2012 Elsevier Ltd. All rights reserved.