Long Distance Video Analytics for Crowdedness Detection

I&T Wish Long Distance Video Analytics for Crowdedness Detection
(REF: W-0491)
Matched I&T Solution
Trial Project
Summary and Challenges
  • In this project, we aim to provide an effective method of crowdedness detection for large scale events. A video analytics system is to be developed for determine the crowdedness level in a number of outdoor venues from a long-distance location.
  • The System shall interface with the CCTV system, which are installed in a high position (mounting on the light pole(s)), to retrieve live CCTV video streams for detection. The system would be deployed to the cloud platform to further analyze. In addition, the analytic results (crowdedness level) would be shown in a web portal and published in government website via API.
  • Project Challenges:
    1. For environmental factors, different angles and heights of devices, variations of illumination, weather, background complexity shall be considered;
    2. For detection, camera quality and photography skills (such as partly occlusion on human body or human head etc.) shall be considered;
    3. Provision of handshaking mechanism with cloud service;
    4. Only 7-10 days would be provided for tunning of CCTVs view angles.
Expected Outcome
  • A video analytics system that provides:
    1. Generic AI module for crowdedness analysis;
    2. System interface for video streaming from cameras installed by another contractor;
    3. System interface for crowdedness status publishing on other government website via API;
    4. Web portal - Dashboard UI about visualization of crowdedness. - Management UI for venue’s coordinator to manually adjust/override the crowdedness status. - Account management module. - Report function showing the status change and the people count.
    5. An overall of 90% or above accuracy is preferred;
    6. The AI model shall be scalable and adaptable that can be applied new data for training and analysing via Application Programming Interface (API).
Expected Trial Duration 6-month
Contact Information
I&T Wish Proposer:Electrical and Mechanical Services Department (EMSD)
Contact Person:YEUNG Po Ching, Anthea
Position:Electronics Engineer/Information Technology Development/5
Tel:3741 8879
Email: pcyeung@emsd.gov.hk
Upload Date 2023-08-09
Closing Date 2023-08-16
Propose I&T Solution Closed