Automatic mapping of moisture affectation in exposed concrete structures by fusing different wavelength remote sensors. Issue 6 (8th November 2015)

• Record Type: Journal Article
• Authors: Conde, B.; Del Pozo, S.; Riveiro, B.; González‐Aguilera, D.
• Is Part Of: Structural control and health monitoring. Volume 23:Issue 6(2016)
• Page Start: 923
• Page End: 937
• Publication Date: 2015-11-08
• Subjects: moisture detection -- concrete structures -- multispectral analysis -- unsupervised classification; Structural engineering -- Periodicals; Structural control (Engineering) -- Periodicals; Automatic data collection systems -- Periodicals; Detectors -- Periodicals; 624.17
• DOI: 10.1002/stc.1814
• Languages: English
• ISSNs: 1545-2255

Abstract

Summary: Water content is a critical parameter for the early detection of moisture degradation in exposed concrete structures. Traditionally, visual inspection is the most extended procedure to detect superficial pathologies caused by moisture in concrete constructions, principally when access is limited. For such cases, remote sensing is a valuable tool to recover radiometric information useful for detecting and quantifying different degrees of affectation caused by water. This paper presents an approach to identifying and evaluating the water content in a real concrete structure by fusing several sensors recording data in different wavelengths. In particular, a procedure to integrate three‐dimensional intensity data collected by two terrestrial laser scanners (Riegl‐Z390i and FARO Focus 3D) with two‐dimensional radiometric data provided by a six‐band multispectral camera and a commercial digital camera (MCA6 Tetracam and Canon EOS 5D) is developed. After data fusion in a two‐dimensional space, a multiband image was created for further spectral analysis. Finally, an unsupervised classification using clustering algorithms was performed to identify the degrees of affectation and the most suitable remote sensor for moisture mapping. Comparisons between the sensors used in this survey reveal that intensity imagery from both laser scanners has high potential for the recognition and characterisation of the degree of moisture in this type of structure. Copyright © 2015 John Wiley & Sons, Ltd.