Multi-type change detection of building models by integrating spatial and spectral information

Liang Chien Chen*, Chih Yuan Huang, Tee-Ann Teo

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

12 Scopus citations

Abstract

Detecting building changes followed by updates is preferable for efficient revisions to building models. Additionally, more change types can be detected with spatial information provided by building models for reducing land surveying work. Therefore, for efficient building of model revision and land surveys, this work applies a new multi-type change detection scheme with new light detection and ranging (LIDAR) point clouds, new aerial images and existing building models. By integrating the spatial information from LIDAR data and image-based spectral information, this work identifies changes to existing buildings and identifies newly built and changed buildings. To provide an initial value for further revisions, new building regions are generated from change detection results. Experimental results demonstrate that the proposed scheme has high accuracy for both change type determination and building region generation. To provide comprehensive observations, experimental results deemed unreliable are scrutinized.

Original languageEnglish
Pages (from-to)1655-1681
Number of pages27
JournalInternational Journal of Remote Sensing
Volume33
Issue number6
DOIs
StatePublished - 1 Jan 2012

Fingerprint

Dive into the research topics of 'Multi-type change detection of building models by integrating spatial and spectral information'. Together they form a unique fingerprint.

Cite this