Video Identification Service for Parking
The role of video identification in parking and access control systems is growing. In many countries video comes standard in parking facilities. Automatic License Plate Recognition (ALPR) video identification service is an excellent way to increase throughput on entry and exit, customer loyalty and improve overall parking experience. Typical applications seen:
- Automatic access gates for registered plates of known residents or prepaid slots.
- Gateless parking garages where payments is handled in a subscription based on detected vehicles.
- Special rates for returning visitors.
Basic principles of identification and billing per registration number may seem straightforward. Nonetheless, ALPR requirements in parking are much higher when it shift from a nice-to-have feature to only means of identification. One would require extreme accuracy with low error rates in detection and identification when, for example, automatic billing is required.
Removing concerns about recognition quality in parking transactions enables parking solutions integrators to focus on their key area of expertise. Delegating responsibility in an SLA is a good way to achieve this. Experienced suppliers of video identification services should accept performance requirements as part of their contract and reward structure.
Changing quality and process
In current practice license plate recognition in parking uses cameras. These cameras capture vehicles entering, leaving, and moving through the parking facility. An ALPR engine inside the camera or deployed on a local PC reads the captured video. It will report the detected registration number to the parking back office for further processing. At this stage registration numbers are still independent from the full vehicle flow. In fact detected registration numbers may contain read errors. Or worse: license plates may not have been identified automatically altogether. At this point a back office will ensure quality of the identification. Back office can combine entries and exits of the same vehicle into trustworthy billable transactions.
Countless facilities adopted the described setup. They successfully operate video based parking management process in a similar fashion. And it has become easy to do so. Dozens of parking management applications are on the market today. Each filling in one or more logical sub components of the parking process. A key concern in this setup remains dealing with challenges of video processing and ALPR specifically. Turning vehicles passages into trustworthy billable transactions is a minute process. Reliable transactions require smart processing workflows for automatic and manual verification. At the same time the need for human reviewers introduces new risks to availability and scalability.
Concerns will disappear when video processing is sourced as an external video identification service. The parking operator can leverage the know-how, platform and infrastructure of the video identification service provider. Typically these services improve availability and reliability as well as improve ALPR quality. We already see the first steps towards externally sourced services in parking. Several parking facilities use lane cameras only to push transactions with images directly to a web service. These services are hosted locally or in the cloud by the video identification service provider.
A good video identification service combines all detections to single transactions. In it, workflows, business logic, redundant ALPR engines, and manual reviews guarantee the quality of the identification. The parking back office just needs to fetch passages as trustworthy transactions from the web service. Apart from savings on cost and risk, predefined accuracy enables new video-only based possibilities like ticketless free-flow parking.
Our video identification service
Q-Free Intrada® Synergy Server (ISS) is a versatile cloud-based vehicle identification service for video based parking and access control systems using our Intrada ALPR. ISS can receive transactions including images with cars and license plates directly from entrance/exit cameras. Following that, ISS logic will push registration numbers and exceptions towards the parking back office as trustworthy transactions. With the possibility to enable manual reviews ISS can ensure 100% identification on all humanly readable images. After that output is direct control of operational processes like payments, invoicing or enforcement.
ISS is available as hosted cloud service and can also be installed on-premises.