This institute’s AI solution can detect motorcyclists without helmets

Researchers at IIT-Hyderabad have developed an artificial intelligence-based solution to catch motorcyclists driving without helmets, a report in The Economic Times stated.

The AI is designed to automatically identify errant motorcyclists in surveillance videos, for which the institute has signed a memorandum of understanding (MoU) with the Hyderabad City Police to access the feed. The technology will soon be launched and the institute has filed its patents for it, the report added.

The research team behind the project at IIT-Hyderabad comprising C Krishna Mohan, Dinesh Singh and C Vishnu plan to bring on board an industry partner to push the technology into the market.

According to Mohan, the technology has been tested and experimented on sparse traffic at the IIT-Hyderabad campus as well as in dense traffic in Hyderabad city.

How it works

CCTV cameras and the servers of the central police control room will both have the tech embedded in them. The software, which is inserted on a card attached to the CCTV cameras, picks out errant motorcyclists by alerting the central database, the ET report stated.

“It will be fully automatic along with a web interface to verify the alerts by the operators (traffic police, etc.). From there, it will be connected to the existing RTO website to generate challans (fines) and send a notification to the riders through SMS,” Singh was quoted as telling ET.

The technology comprises convolutional neural network, which can scan images mimicking the human brain. Besides catching riders without helmets, the tech can be used to detect other kinds of traffic problems, the report said.