IntelliVision’s new AI-Deep Learning-based LPR engine achieving near-perfect accuracy


One of the market leaders and pioneers in video analytics software, IntelliVision, is using the latest in computing technologies to achieve near-100% automatic license plate recognition (ALPR) reads under real-world conditions.

Headquartered in San Jose, California, IntelliVision provides one of the largest suites of advanced video analytic products currently available, with its software providing multiple big data analysis and search functions. The company’s technology has been recognized as the ‘Brains Behind the Eyes’ for many applications deploying and using cameras to analyze video content, extract metadata, send out real-time alerts, and provide intelligence to other home, business and security systems, used by leading companies and US government agencies.

IntelliVision has now announced that the latest version of its license plate recognition (LPR) and detection software, which uses a combination of artificial intelligence (AI), CNN (convolutional neural network) and deep learning, has achieved accuracy numbers as high as 99% in real-world customer environments.

The company’s License Plate Recognizer (LPR) has performed and matured over many years of application, taking into account multiple variables, such as movement and high speeds, which are natural to the environment of vehicle monitoring. Using still images or video feeds, the product can recognize and capture information located on license plates and automatically log this information for future inspection. Real-time searching can also be performed on each plate detected, comparing the information gathered with a stored database of license plates.

IntelliVision’s LPR engine is part of a portfolio of products for smart city, smart auto and intelligent transportation systems (ITS). The company’s products are in over 3 million cameras worldwide, including 15,000 cameras at 3,000 intersections in the USA, where they take the place of costly and often unreliable sub-pavement electromagnetic induction loops. The software is systems agnostic and works with various camera types and brands, as well as archived video.

“We have been continuously improving our LPR/ANPR technology for a number of years, as part of our AI and deep learning-based video analytics software,” said Vaidhi Nathan, IntelliVision’s CEO. “We are now achieving accuracy of up to 99% in countries as diverse as the USA, Mexico, Germany and India, and not just in the lab.”

Share this story:

About Author


Tom has edited Traffic Technology International (TTi) magazine and its Traffic Technology Today website since May 2014. During his time at the title, he has interviewed some of the top transportation chiefs at public agencies around the world as well as CEOs of leading multinationals and ground-breaking start-ups. Tom's earlier career saw him working on some the UK's leading consumer magazine titles. He has a law degree from the London School of Economics (LSE).