The International Institute of Information Technology Hyderabad (IIIT-H) has developed an artificial intelligence (AI)-based traffic violation detection system that uses dashcams mounted on police patrol vehicles to identify traffic offences and assist in generating e-challans.
Called VIOLA, the system has been developed by IIIT-H's Applied AI Research Centre (INAI). It analyses video captured by dashcams to detect violations such as riding without a helmet, triple riding and wrong-side driving. The system also uses automatic number plate recognition (ANPR) to identify vehicles and forwards verified cases to the Telangana Police's existing e-challan system.
The project builds on earlier AI research by Prof. Ravi Kiran S. of IIIT-H's Centre for Visual Information Technology (CVIT), which focused on helmet detection. INAI expanded the research into a deployable traffic enforcement system.
Instead of relying solely on fixed CCTV cameras, the system uses four dashcams installed in police patrol vehicles. According to the institute, the mobile cameras expand enforcement coverage without requiring additional roadside infrastructure.
Police personnel continue to verify each violation before a challan is issued, with the AI system functioning as a decision-support tool rather than replacing human oversight.
According to IIIT-H, the system has helped generate several traffic challans during its pilot phase. The institute said around 96% of the violations flagged by the AI have been approved by police after verification.
The developers are working on adding capabilities such as speed detection and identifying road infrastructure issues, including potholes, damaged roads and missing lane markings, using the same dashcam footage.
The project was developed through IIIT-H's Applied AI Research Centre with support from iHub-Data, the institute's technology innovation hub. The team said it is in discussions with police departments outside Telangana to explore wider deployment of the system.