India, home to the world's second-largest road network, is leveraging NVIDIA's accelerated computing technology to tackle congestion at its numerous toll booths, most of which are currently operated manually.
Calsoft, an Indian-American technology company and NVIDIA Metropolis partner, has implemented a sophisticated vision AI application to automate toll collection across India's highways. This initiative aims to reduce traffic delays and improve commute times on the country's nearly 4 million miles of roads.
"Particularly challenging was night-time detection," said Vipin Shankar, senior vice president of technology at Calsoft. "Another challenge was model accuracy improvement on pixel distortions due to environmental impacts like fog, heavy rains, reflections due to bright sunshine, dusty winds and more."
The solution, which has been deployed in several leading metropolitan cities as part of a pilot programme, uses automatic number plate recognition (ANPR) to read passing vehicle plates and charge the associated driver's unified payments interface (UPI) account. Despite the challenge of India's non-standardised licence plates, the system has achieved approximately 95% accuracy in plate reading.
This technological advancement represents a significant step towards addressing India's traffic congestion issues. As the system proves successful, it could pave the way for widespread automation of toll collection across the country, potentially transforming the experience of millions of drivers on India's vast highway network.