A method for data-mining of defects on building façade elements

Abstract

It has been recognized that the value of buildings gradually decreases over time due to aesthetic, functional, and structural deteriorations caused by various defects. In response to these deteriorations, the value of buildings can be restored by various types of maintenance operations. However, minimizing depreciation is better than maintenance. In this regard, the defects that cause depreciation should be reduced through data analysis on existing buildings. This study suggests the Pixelization Method (PM) for data mining on building façade elements to collect data on façade defects. PM is a technique that supports the creation of a database by simplifying the analysis of building façades as well as the input data. The database, developed with PM, is expected to be used in various areas such as analysis of the relationship between the defects and the facade elements or the prediction of defects. © 2016 SERSC.