Classification of detection states in construction progress monitoring

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The research conducted in this publication focusses on automated progress monitoring. The recording of the current as-built state of a construction site is achieved by photogrammetric methods (e.g. UAVs) and compared to an as-planned (4D) BIM model. To visualize the detected elements and evaluate their respective detection rate, a schema for the classification of each individual element is presented. It compares the as-built ground truth with the actually detected elements and thus facilitates a quick and easy interpretation of the current construction state. Temporal data from construction schedules is added to further complete the provided information through visualization. Additionally, the classification helps to identify possible lacks of detection algorithms. New parameters for detection algorithms can be applied and the results are immediately visible with an easily understandable color scheme.

Original languageEnglish
Title of host publicationeWork and eBusiness in Architecture, Engineering and Construction - Proceedings of the 11th European Conference on Product and Process Modelling, ECPPM 2016
EditorsSymeon E. Christodoulou, Raimar Scherer
PublisherCRC Press/Balkema
Pages473-480
Number of pages8
ISBN (Print)9781138032804
StatePublished - 2016
Event11th European Conference on Product and Process Modelling, ECPPM 2016 - Limassol, Cyprus
Duration: 7 Sep 20169 Sep 2016

Publication series

NameeWork and eBusiness in Architecture, Engineering and Construction - Proceedings of the 11th European Conference on Product and Process Modelling, ECPPM 2016

Conference

Conference11th European Conference on Product and Process Modelling, ECPPM 2016
Country/TerritoryCyprus
CityLimassol
Period7/09/169/09/16

Fingerprint

Dive into the research topics of 'Classification of detection states in construction progress monitoring'. Together they form a unique fingerprint.

Cite this