| Download | - View author's version: AI-AR for remote visual bridge inspection by drone (PDF, 840 KiB)
|
|---|
| Author | Search for: Lapointe, Jean-François1; Search for: Sekkati, Hicham1; Search for: Allili, Mohand Saïd; Search for: Hebbache, Loucif; Search for: Amirkhani, Dariush; Search for: Hammouche, Nadir |
|---|
| Affiliation | - National Research Council of Canada. Digital Technologies
|
|---|
| Format | Text, Article |
|---|
| Conference | 11th International Conference on Structural Health Monitoring of Intelligent Infrastructure (SHMII-11), August 8-12, 2022, Montreal, Quebec, Canada |
|---|
| Subject | SHMII-11; visual inspection; remote inspection; bridge inspection; drone; artificial intelligence; augmented reality |
|---|
| Abstract | Regular inspections of bridges and overpasses are required to ensure the safety of those infrastructures. This paper presents a system developed to do the remote visual inspection of concrete bridges and overpasses by combining drone, artificial intelligence and augmented reality technologies with the goal to assist bridge inspectors in conducting a remote visual inspection. It also reports on the results of a first field trial of this complex human-machine system. Deep learning networks can be trained to detect defects on independent images but the user interface must be adapted to track defects in a stream of images at a speed consistent with human vision to allow inspection. This first field trial allows for an initial estimate of a speed up factor of ten (10x) to conduct visual inspections. The developed system could lead to significant cost reduction to inspect the bridges and overpasses as well as significant reduction of downtimes for these transportation infrastructures. |
|---|
| Publication date | 2022-08-12 |
|---|
| Publisher | Concordia University
International Society for Health Monitoring of Intelligent Infrastructure (ISHMII) |
|---|
| In | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | 9504cb21-624a-4f6a-be13-6dd06003f0bd |
|---|
| Record created | 2024-01-24 |
|---|
| Record modified | 2024-11-22 |
|---|
