IRD trusts in AI for traffic count and classification

IRD has announced its iTheia video-based traffic counting and classifying system that uses artificial intelligence (AI). Instead of classifying vehicles based solely on axle spacing or vehicle length parameters, iTheia classes vehicles based on visual input and a machine learning algorithm.
October 13, 2020
IRD1

 

The system is easy to install on existing poles at the roadside. iTheia easily handles three lanes' classification in a single direction, or four lanes' bidirectional at highway speeds. IRD’s testing indicates that iTheia outperforms road tube counters for count accuracy in bidirectional applications, despite being a non-intrusive solution.

IRD claims that iTheia can outperform radar systems when classifying vehicles into six size-based classes and that the system outperforms radar when counting traffic in highly congested or stop-and-go conditions.

While six-class schemes are common in non-intrusive applications, iTheia is capable of classifying vehicles based on the standard 13-category FHWA class definitions used in the United States.

Software is provided with iTheia for selecting the classification scheme, initiating counts and exporting data. A near real-time video stream makes it simple to verify correct camera placement and system operation.

Data is delivered in standard formats, suitable for import into most traffic data and analysis systems. Video is also saved for quality control purposes, with each vehicle separately captured and placed in order at the top of the video with the class and lane identified and running count/classification totals at the bottom of the video.

Top features of iTheia include no video uploading for third-party processing; bidirectional traffic data; safe and simple operation; and reliable performance day or night in all weather conditions.

Companies in this article