ALPR is currently one of the most effective ways to identify a vehicle. It is a valuable tool for parking operators seeking to automate parking lots, whether they want to cut costs, increase customer satisfaction, or cut emissions and energy consumption. The dilemma lies in how ALPR is implemented to meet user demands, as well as keeping costs affordable.
ALPR is currently one of the most effective ways to identify a vehicle. It is a valuable tool for parking operators seeking to automate parking lots, whether they want to cut costs, increase customer satisfaction, or cut emissions and energy consumption. The dilemma lies in how ALPR is implemented to meet user demands, as well as keeping costs affordable.
ALPR processing can be done on the camera, in a nearby trunk or at the back end. ALPR equipment generally consists of a camera with integral illumination and a processor that hosts the ALPR software, said Jim Kennedy, President of Inex/Zamir. “The processor can handle multiple lanes for efficiency and cost savings. The camera/illuminator needs to be in the lane, and the processor can either be in the lane (a weatherized unit) or located elsewhere such as an electrical cabinet, entry or exit kiosk or rack-mounted in an administration building.”
A camera that has a built-in processor designed to interpret plates can run at its full potential regardless of the number of cameras in the system, said Chris Yigit, AutoVu Senior Product Manager, Genetec. “This would not be the case if you had a trunk unit that's shared by more than one camera.”
Another key to superior reading is image quality, Yigit said. “By processing on the edge, we can access the full, uncompressed image that needs to be interpreted. If we were to send this to the back office instead of processing on the edge, we would need to send 30 fps of uncompressed video per camera over the network; the bandwidth requirement would be obscene. The alternative would be to compress the video, which would technically work, but the read performance would drop. Altering just a few pixels in the compression affects the highly sophisticated algorithms.”
ALPR algorithms can be built into the camera, but a big consideration is cost per lane, Kennedy said. “If you run multiple cameras into one processor, which can be located near the cameras, you still have the ability to reduce bandwidth traffic going back to a central site (perhaps even more so), and you are only investing in a single processor for multiple lanes.”
MOVING FORWARD
As the number of vehicles increases, so will the frustration of finding a parking space. Be it for parking lot operators or building managers, efficient parking can increase customer satisfaction and revenue. The environmental impact of emission and energy consumption can also be relieved through smarter parking. “Most of these measures lower gasoline consumption and the associated emissions by helping people to burn less gasoline searching for parking. This adds up over many vehicles over time,” Martin said. Looking forward, there are a number of areas where technology revolutionizes the parking experience.