AI IMPLEMENTATION ON NVIDIA JETSON NANO FOR EDGE COMPUTING ON CROWD MANAGEMENT SYSTEM
Nowadays in Indonesia, many cities have implemented smart cities where many CCTVs are installed in various corners of the city and then monitored through the command center. However, with the increasing number of video data taken from CCTV, the process of detecting crowded areas will be difficult to...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/62339 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Nowadays in Indonesia, many cities have implemented smart cities where many CCTVs are installed in various corners of the city and then monitored through the command center. However, with the increasing number of video data taken from CCTV, the process of detecting crowded areas will be difficult to do manually. Therefore we need a device that can monitor and detect crowded areas automatically and digitally using a device called a Crowd Management System (CMS). CMS will perform face detection to determine the level of crowd in an area monitored by CCTV. In general, cloud computing-based CMS will face the problem of bandwidth requirements where a large internet connection is needed between CCTV cameras and data processors to transfer video data. One of solution for this problem is to process data directly on CCTV which is known as edge computing. The data processing uses Artificial Intelligent (AI), so an embedded device capable of running AI is needed. Therefore, this study will design an edge computing-based CMS using SSD-Mobilnetv2 on a Nvidia Jetson Nano. |
|---|