A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Anomaly detection in urban drainage with stereovision
Abstract This work introduces RADIUS, a framework for anomaly detection in sewer pipes using stereovision. The framework employs three-dimensional geometry reconstruction from stereo vision, followed by statistical modeling of the geometry with a generic pipe model. The framework is designed to be compatible with existing workflows for sewer pipe defect detection, as well as to provide opportunities for machine learning implementations in the future. We test the framework on 48 image sets of 26 sewer pipes in different conditions collected in the lab. Of these 48 image sets, 5 could not be properly reconstructed in three dimensions due to insufficient stereo matching. The surface fitting and anomaly detection performed well: a human-graded defect severity score had a moderate, positive Pearson correlation of 0.65 with our calculated anomaly scores, making this a promising approach to automated defect detection in urban drainage.
Highlights 3D reconstruction of sewer pipe surface (including defects) with stereovision. Generic pipe surface model that is robust to different pipe shapes. RANSAC-based algorithm fits point cloud data of regular pipe surface and anomalies. Global anomaly score quantifies the deviation from the pipe model per image pair.
Anomaly detection in urban drainage with stereovision
Abstract This work introduces RADIUS, a framework for anomaly detection in sewer pipes using stereovision. The framework employs three-dimensional geometry reconstruction from stereo vision, followed by statistical modeling of the geometry with a generic pipe model. The framework is designed to be compatible with existing workflows for sewer pipe defect detection, as well as to provide opportunities for machine learning implementations in the future. We test the framework on 48 image sets of 26 sewer pipes in different conditions collected in the lab. Of these 48 image sets, 5 could not be properly reconstructed in three dimensions due to insufficient stereo matching. The surface fitting and anomaly detection performed well: a human-graded defect severity score had a moderate, positive Pearson correlation of 0.65 with our calculated anomaly scores, making this a promising approach to automated defect detection in urban drainage.
Highlights 3D reconstruction of sewer pipe surface (including defects) with stereovision. Generic pipe surface model that is robust to different pipe shapes. RANSAC-based algorithm fits point cloud data of regular pipe surface and anomalies. Global anomaly score quantifies the deviation from the pipe model per image pair.
Anomaly detection in urban drainage with stereovision
Meijer, Dirk (author) / Luimes, Rianne (author) / Knobbe, Arno (author) / Bäck, Thomas (author)
2022-04-21
Article (Journal)
Electronic Resource
English
Obstacle detection in urban traffic using stereovision
| IEEE | 2005
Obstacle Detection in Urban Traffic Using Stereovision
| British Library Conference Proceedings | 2005
3D lane detection system based on stereovision
| IEEE | 2004
Reconstruction de visages par stereovision active
| British Library Conference Proceedings | 1994
A stereovision-based crack width detection approach for concrete surface assessment
| Online Contents | 2016