| I&T Solution |
Long Distance Video Analytics for Crowdedness Detection
(REF: S-1516) |
| Trial Project |
|
| Solution Feature |
- Real-time video stitching from multiple angle cameras, to achieve a panoramic view of the crowd
- Use of adaptive image preprocessing to handle variations in illumination, weather, and background,and data augmentation to simulate
- Ability to count the number of people and crowdedness level in the stitched videos.
- Ability to accurately locate the Region-Of-Interest (ROI) with a Pan-tilt-zoom (PTZ) camera and record more videos.
- uses a scalable and adaptable AI model that can be trained on new data
|
| Trial Application and Expected Outcome |
- Setup 1-2 PTZ camera (multi-angle camera with a coverage of 180 degree) with pre-trained data
- Counting of numbers of people in different environment and provide crowdedness score up to 90% accuracy with AI trainable model.
- Alert or Notification if there is over crowded which is configurable.
|
| Additional Solution Information |
Long Distance Video Analytics for Crowdedness Detection Proposal by Beyond Limits.pptx.pdf
|
| Info on I&T Solution Provider |
| Solution Provider | : | Master Concept (Hong Kong) Limited | | Address | : | 2/F, The Whitney, 183 Wai Yip Street, Kwun Tong, Kowloon, Hong Kong | | Contact Person | : | Rachel Lee |
| Position | : | Account Manager | | Tel | : | 51716684 | | Email | : |
rachel.lee@hkmci.com | | Webpage | : | https://hkmci.com/ |
|
