Abstract
In Japan, there are 1.2milion kilometers or more of the road, of which 900, 000 km of road has been paved. Paved road has been painted compartment line to separate many vehicles and driver-pedestrian. Compartment line is important for drivers and pedestrians because they can recognize for they are aware of the width of the roadway and the sidewalk by the compartment line. Therefore, municipality should inspect the ongoing compartment line but they are inspected by people so the burden for road management is big. The purpose of this research would be to automatically detect the damage of the compartment line. The image obtained from cameras, applying random forest. In this experiment, we used more than about 50, 000 of images by capturing a Walking camera and a vehicle-mounted camera. Baseline and precision comparison result, the detection of damage to compartment line, it was confirmed that the proposed method is effective.
Original language | English |
---|---|
Title of host publication | Adjunct Proceedings of the 13th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2016 |
Publisher | Association for Computing Machinery |
Pages | 53-58 |
Number of pages | 6 |
Volume | 28-November-2016 |
ISBN (Electronic) | 9781450347594 |
DOIs | |
Publication status | Published - 2016 Nov 28 |
Event | 13th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2016 - Hiroshima, Japan Duration: 2016 Nov 28 → 2016 Dec 1 |
Other
Other | 13th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2016 |
---|---|
Country | Japan |
City | Hiroshima |
Period | 16/11/28 → 16/12/1 |
Keywords
- Camera
- City management
- Image processing
- Machine learning
- Smart sensing
- Vehicle
ASJC Scopus subject areas
- Human-Computer Interaction
- Computer Networks and Communications
- Computer Vision and Pattern Recognition
- Software