Security comes in layers, both physical and informational. Both of those vectors can be enhanced with AI license plate reading (LPR) technology. In a school or campus setting, as cars approach, license plates can be read and matched to authorized vehicles. Such technology can create automatic alerts based on parameters such as whenever an unauthorized car/truck approaches the building or enters the parking lot. Artificial intelligence (AI) can monitor hundreds of camera feeds simultaneously, making it suitable for any size campus.
In addition to license plate recognition to identify staff member cars or school stakeholders, AI LPR technology can recognize simple car shapes, such as distinguishing between a car or bus in a no parking zone or bus zone.
Once an unauthorized vehicle is identified, security alerts can be passed directly to on-campus security and first responders, with a contextualized alert (e.g. “WARNING: UNKNOWN TRUCK APPROACHING SOUTH DOCK”) and a photo screenshot of the suspicious vehicle.
As AI data sets continue to grow, larger data sets can now account for the explosion of various vanity plate formats offered in every state. Smarter AI algorithms are being trained to operate and read plates in sub-optimal conditions such as rain, snow, poor angles, and partially obscured visual ranges.
With a well-trained LPR AI, simple 720p cameras with video feeds at 30fps are sufficient to supply the AI with a recognizable plate. In addition, the AI required for competent LPR can be containerized and run directly on a GPU edge server, thus avoiding the addition of a remote cloud hosted AI, where bandwidth or network issues might defeat the data for recognition going out or for the alerts coming back.
