Bangalore Police says AI-powered system helped stop firecracker violations during Deepavali

Bengaluru City Police claimed to have alleviated the problem of firecrackers going off during the Deepavali festival night, claiming that around 41% of complaints reported after the 10pm deadline were suppressed through proactive AI-based detection methods.

Joint Commissioner of Police (Administration) Kuldeep Kumar Jain said enforcement cameras installed at strategic locations in the city are designed to detect firecrackers and smoke in public places and immediately inform the command center and dispatch the competent Hoysala forces. He said the system has significantly reduced air and noise pollution.

This technique was first used in Bangalore to effectively suppress firecrackers exploding especially at night.

“The AI-powered Firecracker Detection and Response Initiative has successfully validated the Bangalore Safe City Project’s ability to integrate advanced AI tools for real-time civil enforcement.Through collaboration with the Bangalore City Police and IT companies, the project has “We have demonstrated measurable improvements in awareness, enforcement accuracy and public safety. This initiative has established a strong precedent for leveraging Safe City’s infrastructure to deploy AI-based governance systems across major urban centers in India.” Jainism.

The effort leveraged the city’s extensive Safe City surveillance network, which includes more than 7,500 cameras and Integrated Command and Control Center (ICCC) infrastructure. Awiros Appstack’s Firecracker Detection Video AI app, powered by Awiros Video Intelligence OS, analyzed live feeds from over 200 existing cameras installed in high-density residential and commercial areas such as Srirampuram, KR Market, HSR Layout, Haralur, and Marathahalli.

The AI ​​algorithm is programmed to detect firecracker flashes, smoke, and excessive crowd activity in real-time. Upon detection, the system automatically generated a warning packet containing location details, timestamps, and visual evidence. These alerts were immediately sent via WhatsApp to the ICCC and field forces, allowing control room operators to confirm the incident and coordinate an immediate police response.