Real-time Detection of Illegal Parking Offences using Video Analytics

Project Reference :


Institution :

Nanyang Technological University (NTU)

Principal Investigator :

Professor Huang Guangbin

Technology Readiness :

7 ( Prototype in real environment)

Technology Categories :

AI- Computer Vision

Background/Problem Statement

The global market for Smart Cities is projected to reach a revised size of US$2.5 Trillion by 2026 with Smart Governance & Education projected to record a 24.4% CAGR and reach US$915.5 Billion by the end of the analysis period. The ongoing pandemic has provided an opportunity for expediting the use of digital technologies for creating smart cities and improving the lives of its citizens. Among the different initiatives that are being explored is the development of smart traffic control and parking management systems.

Illegal parking causes many problems such as traffic jams and safety risks to other road users. Hence, checking for illegally-parked vehicles has become an important task of modern urban management. The current enforcement process of checking illegally-parked vehicles highly depends on enforcement officers on foot patrol. This can be labour-intensive and time-consuming.


An AI-based solution has been developed to automate the processing of illegal parking enforcement, provide real-time illegal parking analysis and reduce the involvement of human effort in on-spot checking of illegal parking offences.

An in-vehicle camera mounted on a mobile enforcement vehicle uses real-time video to capture, detect and identify the type of parking offence and offending vehicle.

The trained AI model analyses the contents of live video streams to learn information about roads, vehicles, parking lots, restriction signs, etc., followed by a multi-channel synchronised decision system to identify whether vehicles are parked properly or legally.

If an offending event is detected, the type of infringement will be reported online and in real-time to a cloud-based server.


Cost and time savings due to a reduction in the involvement of manpower in manually checking and recording the details of each illegal parking event.

Potential Application(s)

Automatic detection of parking offences without manual effort

We welcome interest from the industry for collaboration/ co-development / customisation of the technology into a new product or service. If you have any enquiries or are keen to collaborate, please contact us.