Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System
The maintenance and inspection of sewer pipes are essential to urban infrastructure but remain predominantly manual, resource-intensive, and prone to human error. Advancements in artificial intelligence (AI) and computer vision offer significant potential to automate sewer inspections, improving rel...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/23/7786 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849220388317298688 |
|---|---|
| author | Jan Thomas Jung Alexander Reiterer |
| author_facet | Jan Thomas Jung Alexander Reiterer |
| author_sort | Jan Thomas Jung |
| collection | DOAJ |
| description | The maintenance and inspection of sewer pipes are essential to urban infrastructure but remain predominantly manual, resource-intensive, and prone to human error. Advancements in artificial intelligence (AI) and computer vision offer significant potential to automate sewer inspections, improving reliability and reducing costs. However, the existing vision-based inspection robots fail to provide data quality sufficient for training reliable deep learning (DL) models. To address these limitations, we propose a novel multi-sensor robotic system coupled with a DL integration concept. Following a comprehensive review of the current 2D (image) and 3D (point cloud) sewage pipe inspection methods, we identify key limitations and propose a system incorporating a camera array, front camera, and LiDAR sensor to optimise surface capture and enhance data quality. Damage types are assigned to the sensor best suited for their detection and quantification, while tailored DL models are proposed for each sensor type to maximise performance. This approach enables the optimal detection and processing of relevant damage types, achieving higher accuracy for each compared to single-sensor systems. |
| format | Article |
| id | doaj-art-89bc9ada3c4b441f9ef4b1aad439195e |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-89bc9ada3c4b441f9ef4b1aad439195e2024-12-13T16:32:46ZengMDPI AGSensors1424-82202024-12-012423778610.3390/s24237786Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor SystemJan Thomas Jung0Alexander Reiterer1Department of Sustainable Systems Engineering, University of Freiburg, Georges-Köhler-Allee 10, 79110 Freiburg im Breisgau, GermanyDepartment of Sustainable Systems Engineering, University of Freiburg, Georges-Köhler-Allee 10, 79110 Freiburg im Breisgau, GermanyThe maintenance and inspection of sewer pipes are essential to urban infrastructure but remain predominantly manual, resource-intensive, and prone to human error. Advancements in artificial intelligence (AI) and computer vision offer significant potential to automate sewer inspections, improving reliability and reducing costs. However, the existing vision-based inspection robots fail to provide data quality sufficient for training reliable deep learning (DL) models. To address these limitations, we propose a novel multi-sensor robotic system coupled with a DL integration concept. Following a comprehensive review of the current 2D (image) and 3D (point cloud) sewage pipe inspection methods, we identify key limitations and propose a system incorporating a camera array, front camera, and LiDAR sensor to optimise surface capture and enhance data quality. Damage types are assigned to the sensor best suited for their detection and quantification, while tailored DL models are proposed for each sensor type to maximise performance. This approach enables the optimal detection and processing of relevant damage types, achieving higher accuracy for each compared to single-sensor systems.https://www.mdpi.com/1424-8220/24/23/7786automated inspectiondamage detectionsewer pipesartificial intelligencerobotic inspectioncomputer vision |
| spellingShingle | Jan Thomas Jung Alexander Reiterer Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System Sensors automated inspection damage detection sewer pipes artificial intelligence robotic inspection computer vision |
| title | Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System |
| title_full | Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System |
| title_fullStr | Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System |
| title_full_unstemmed | Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System |
| title_short | Improving Sewer Damage Inspection: Development of a Deep Learning Integration Concept for a Multi-Sensor System |
| title_sort | improving sewer damage inspection development of a deep learning integration concept for a multi sensor system |
| topic | automated inspection damage detection sewer pipes artificial intelligence robotic inspection computer vision |
| url | https://www.mdpi.com/1424-8220/24/23/7786 |
| work_keys_str_mv | AT janthomasjung improvingsewerdamageinspectiondevelopmentofadeeplearningintegrationconceptforamultisensorsystem AT alexanderreiterer improvingsewerdamageinspectiondevelopmentofadeeplearningintegrationconceptforamultisensorsystem |