University of Twente Student Theses

Login

Amplifying the Analyst: Machine Learning Approaches for Buried Utility Characterization

Versloot, C.W.A. (2019) Amplifying the Analyst: Machine Learning Approaches for Buried Utility Characterization.

[img] PDF
3MB
Abstract:The Netherlands has a complex underground infrastructure comprising various types of cables and pipelines. Given inadequate registration procedures from the past, large deviations occur between the estimated and actual positions of such utilities. Annually, this results in many utility strikes during excavation work. By consequence, a business model has emerged for utility mapping companies that identify the actual positions of cables and pipelines for their customers. The analyst's job, which is essentially a classification problem with respect to utility material type, is however repetitive and labor-intensive. Fortunately, today, human intellect can be amplified by machines through Machine Learning. In this work, three classes of Convolutional Neural Networks are designed and trained for this purpose harnessing Ground Penetrating Radar data, with promising results. It subsequently presents a prototypical web application that embeds the models, allowing them to be used in practice. Early validation activities confirm its business value.
Item Type:Essay (Master)
Clients:
TerraCarta B.V., Hoogeveen, the Netherlands
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:38 earth sciences, 54 computer science, 56 civil engineering
Programme:Business Information Technology MSc (60025)
Link to this item:https://purl.utwente.nl/essays/79045
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page