The official source code for the project "An Intelligent System For 24/7 Real-Time Traffic Surveillance on Edge Devices".
The primary goal of this project is to develop a state-of-the-art traffic surveillance system optimized for edge devices. This system is designed to extract real-time information under various weather conditions, from the clear, sunny days to the most challenging foggy or rainy scenarios. By leveraging the latest advancements in edge computing, the project aims to provide immediate, accurate traffic data analysis directly where it's most needed – at the edge of the network, closest to the traffic itself.
Traffic conditions can change rapidly, influenced by factors such as time of day, accidents, events, and especially weather conditions. Traditional traffic surveillance systems often struggle to adapt to these changes swiftly due to their reliance on central processing. Our project addresses these challenges by:
-
Enhancing Real-Time Processing: Utilizing the power of edge computing to process and analyze data in real time, significantly reducing the latency often found in centralized systems.
-
Adapting to Weather Variabilities: Implementing advanced algorithms capable of adjusting to different weather conditions, ensuring reliable data collection and analysis regardless of environmental changes.
The Edge Traffic Surveillance System Project is more than a traffic management solution; it's a step towards smarter cities. By enhancing our ability to manage traffic efficiently, we aim to reduce congestion, minimize accident response times, lower pollution levels, and improve the overall quality of urban living.
TBA ...
Current modules of project:
- ETSS-02-VehicleDetTrack
- ETSS-03-Lane
- ETSS-04-SpeedEst
- .
- .
- ETSS-07-ImageEnhance
- ETSS-08-Data
- Realistic-Traffic-Surveillance_GeneratedSample
This project is supported by the Institute for Information & Communications
Technology Promotion (IITP) grant funded by the Korea government (MSIP) (2021-0-01364).
TBA ...