IIIT-H’s Vahan Eye Targets Truck Plates

Hyderabad: When standard license plate recognition systems failed to read India’s hand-painted truck license plates, researchers at the International Institute of Information Technology Hyderabad (IIIT-H) developed a solution specifically designed for Indian highways.

Leveraging previous research from the Center for Visual Information Technology, the institute’s iHub-Data has deployed an AI-based system called Vahan Eye to track sand transport vehicles for the Telangana Mine Development Corporation. The aim is to track authorized trucks and prevent illegal sand mining.

iHub-Data CEO and IIIT-H Adjunct Faculty Dr. “Typical license plates are actually easy to detect,” said Veera Ganesh Yalla. “But in India, especially on trucks, number plates are often hand-painted, inconsistent and highly variable. The design and style are unique from vehicle to vehicle.”

Commercial automatic license plate recognition systems designed for uniform license plates have proven to be ineffective and costly. The IIIT-H team adapted a laboratory prototype, recreated the handwritten character recognition module, and integrated it into an open-source platform as a plug-in. “If anyone wants to plug our plate technology into their own platform, they can do so without rewriting everything,” Dr Yalla said.

The system was piloted at Chityal on the Vijayawada-Hyderabad highway. It tracks incoming trucks and cross-checks them against a whitelist of around 40,000 approved vehicles. The system, which has been operating since September, has continued to improve with live data despite low light conditions and visual obstructions such as decorative wreaths.

Developed by a small team of less than five engineers using deep learning models such as YOLO and RF-Detr, Vahan Eye is now customized for traffic violation detection. “Our intellectual property is that we figured out how to solve this hand-written license plate problem,” Yalla said, adding that the goal is to make this type of public-interest technology affordable and scalable for government use.