Self-sustained and AI-driven Railway Intrusion Object Detection System

I&T Solution Self-sustained and AI-driven Railway Intrusion Object Detection System
(REF: S-0973)
Trial Project
Solution Feature
  • Standalone wire-free visual sensor with built-in a 4G LTE network powered by the solar panel and rechargeable battery
  • An edge AI computing device powered by a portable standalone solar-battery system
  • A computer vision AI software for automated real-time railway intrusion object detection
  • A self-adaptive federated learning framework for cloud-edge updates of AI models in the software
  • A cloud hosted analytical system for generating alarms of railway intrusion object detection
Trial Application and Expected Outcome
  • Develop and setup the standalone visual sensor and AI computing device as well as make sure they are able to self-sustain for 24/7
  • Test and make-sure the data (captured video of the targeted railway scene) streaming from the visual sensor to the local AI computing device via wireless transmission
  • Implement the computer vision based railway intrusion object detection software and report the performance evaluation
  • Implement the cloud-edge coordinated analytics and alarming system as well as provide the final project demo
Additional Solution Information Project Details.pptx
Info on I&T Solution Provider
Solution Provider:Hong Kong Institute for Data Science
Address:P7311, Yeung Kin Man Academic Building, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
Contact Person:Zijun Zhang
Position:Associate Professor
Tel:34425328
Email: zijzhang@cityu.edu.hk

For details of the above I&T solution, please contact the I&T solution provider.