Abstract | ||
---|---|---|
Adequate illumination of city streets during night hours is essential to ensure road safety. However, even for developed cities, monitoring streetlights still remain a tedious task that relies on manual inspection reports. Existing systems mostly rely on vehicle-mounted camera or sensors fitted at every light post that is not cost-effective and scalable. In contrary, in this paper, we develop a novel cost-effective system LiSense to monitor illumination levels of street lights and detect as well as localize malfunctioning light posts. The system utilizes ambient light and GPS sensors and uses crowdsourcing. Sensor trails collected by our App from 2-wheeler covering 160 km suburban city road detects all malfunctioning street lights more than 96% in accuracy with a mean localization error of 6 meters. To the best of our knowledge, this is the first of its kind approach to monitoring street light condition which is cost-effective, scalable and suitable for developing regions. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/ICDMW.2018.00092 | ICDM Workshops |
Keywords | Field | DocType |
Lighting,Roads,Urban areas,Monitoring,Sensor systems,Global Positioning System | Data mining,Computer science,City street,Crowdsourcing,Real-time computing,Global Positioning System,Developing regions,Street light,Scalability | Conference |
ISSN | ISBN | Citations |
2375-9232 | 978-1-5386-9288-2 | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Munshi Yusuf Alam | 1 | 0 | 0.34 |
Shahrukh Imam | 2 | 0 | 0.34 |
Harshit Anurag | 3 | 0 | 0.34 |
Sujoy Saha | 4 | 47 | 13.03 |
Subrata Nandi | 5 | 71 | 21.37 |
Mousumi Saha | 6 | 2 | 4.13 |