Building Hights and Floor Estimation Using 3D Maps, Central Part of Kucukcekmece, Istanbul, Turkey


ERENER A., SARP G., Karaca M. I.

1st Springer Conference of the Arabian-Journal-of-Geosciences (CAJG), Hammamet, Tunisia, 12 - 15 November 2018, pp.159-162 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1007/978-3-030-01440-7_37
  • City: Hammamet
  • Country: Tunisia
  • Page Numbers: pp.159-162
  • Keywords: Lidar technique, Building floor estimation, 3D maps, Urban areas

Abstract

The estimation of accurate, fast and up-to-date building and building floor data is inevitable for 3D urban map projects. These maps mostly required for tracking building construction speed, monitoring horizontal and vertical urban growth and illegal constructions, updating building inventories, preparing feasible urban plans, assessment of hazard and risk and creating infrastructure plans. Up-To-Date the passive remote sensing technologies are to detect and map urban features and to obtain land use and land cover maps. They can record large and continuous land cover information in urban areas. The buildings can be perceived as different from each other in urban areas by using high-resolution remote sensing images. For this reason, high-resolution satellite imagery is very useful for obtaining areas and locations of buildings that are difficult to identify compared to medium resolution satellite imagery. However, these systems have several constraints in creating 3D maps of urban features and detecting urban building heights. Active sensors can overcome some of these constraints when used together with passive systems. These systems create highly accurate 3D height maps for buildings, therefore, can be used to estimate an accurate floor value for each building in urban areas. 3D building detection studies have shown that the Lidar technique is promising and suitable for 3D object detection. In this study, a combination of aerial photograph and Lidar data were used to produce individual building heights and then estimate building floor in the urban central part of Kucukcekmece, Istanbul. The accuracy of the proposed algorithm was evaluated for each building floor using ground truth data and has proved an overall accuracy of 79% and a kappa equal to 0.74 which is a promising result.