Title : RECOGNIZING BIOLOGICAL CRUSTS IN CIVIL ENGINEERING STRUCTURES USING INTENSITY DATA FROM TERRESTRIAL LASER SCANNER
Authors : Raaj Ramsankaran and Sravan Chinthaparthi
Publication : 13th – 16th February 2013
Volume :
Pages : 10-Jan
Price : 250
Abstract : In this study an attempt has been made for extracting information on the presence of biological crusts on concrete structures using Terrestrial laser scanning (TLS) intensity data. TLS (static or mobile) is a non-destructive and non-invasive technique whose use has expanded greatly in recent years in the field of graphic and metric documentation of objects in which no direct contact is involve, and is becoming a standard for 3D recording and modeling of complex and remote engineering structures. The methodology adopted is divided into two major parts. First step is to create an ortho-image using intensity values obtained from the scanner. Subsequently, a detailed study comparing some areas with evidence of biological crusts with the data shown in RGB ortho-image was performed manually and considered to be the ground truth data. In second step, an unsupervised classification called ISODATA clustering was performed on the ortho-image obtained from the first step. The obtained results are then compared with the ground truth data and found to be very satisfactory, confirming the potential of this technology for assessing surface deteriorations on engineering structures at inaccessible locations.