Bike Flow Survey Focuses on Cycling’s Future
Hong Kong is one of the most densely populated cities on earth with congested roads and public transportation systems that can be at capacity during rush hour.
Prof. Kuo Yong-Hong
But there is an increasing interest in a simpler form of transport with cycling growing as a hobby, for sport and for commuting.
In the Hong Kong Smart City Blueprint, smart mobility is one of the main initiatives outlined by the government. “Bicycle-friendly” new towns are explicitly addressed in the blueprint for their strategic importance in enhancing the mobility of the city.
To create a “bicycle-friendly” environment, understanding cycling activities in the city is key. With no available government or industry data on the travel behaviour and demands of cyclists an interdisciplinary team from HKU is conducting a comprehensive study of bike usage at the Hong Kong Science Park near Shatin.
Prof. Kuo Yong-Hong, who is running the Computer Vision-based Bike Flow Estimation project, said they have installed thermal cameras along a five-kilometre section of the Shatin to Tai Po cycle track to capture cycling activity.
“Our project aims to leverage advanced technologies, including thermal cameras, traffic engineering techniques and data science tools to infer travel behaviour and demands.” Prof. Kuo said. “At the lab at HKU, we have our computer vision models to offer insights into the city’s cycling mobility.”
Prof. Kuo said the system is able to measure the transport demands induced by pedestrians, people on bikes or on scooters or other modes of transport, and he added that the thermal cameras are unable to capture personal details.
Prof. Kuo, from the Department of Data and Systems Engineering, is working with transportation researchers from the Department of Civil Engineering as well as traffic management systems company Yunex Traffic and LocoBike, Hong Kong's largest dockless bike sharing platform.
The team is conducting bike flow analysis and safety evaluations of thousands of hours of observations by utilising AI integrated analytics models to generate real-time data on cyclists’ movements.
Professor Kuo said the aim of the project is to estimate the travel demands within the cycling community that uses the tracks at Hong Kong Science Park and then to translate the research into impactful real world applications benefiting society.
Yunex Traffic engineer Martin Cheung said the goal through this collaboration with HKU is to develop a cost effective, intelligent and reliable solution for bicycle traffic management which supports a smarter and a more sustainable urban future.