Attending ISC West? Get Your Free Ticket and Meet Us There!
Get a Free Ticket to ISC West
License plate recognition (LPR) has traditionally depended on fixed-angle cameras, optimized for a single lane or detection point. But as urban environments demand more adaptable infrastructure, combining pan-tilt-zoom (PTZ) capability with reliable on-camera recognition unlocks a new set of possibilities.
The Axis Q6355-LE and Q6358-LE, both based on the ARTPEC-9 system-on-chip, demonstrate that PTZ and LPR are no longer mutually exclusive. Running CAMMRA AI entirely on the edge, these units enable multi-spot license plate and vehicle recognition, with no dependency on external servers or cloud compute.
During testing, CAMMRA AI maintained accurate LPR and make/model recognition as the camera cycled through preset positions. While PTZ cameras observe one area at a time, this level of real-time adaptability allows a single unit to efficiently monitor multiple zones in sequence. It is especially valuable in scenarios where continuous coverage of all areas is not required, but flexible, on-demand recognition across different points is essential.
The integration with ARTPEC-9’s deep learning processing unit (DLPU) enables both recognition and semantic post-processing to be handled entirely within the camera. This brings a new level of autonomy for edge-based analytics, removing the need for external compute resources.
Beyond just enabling inference, ARTPEC-9 represents a major leap in architectural design. Earlier PTZ models lacked the capability to support real-time deep learning at the edge, but this new generation includes a dedicated DLPU built specifically for AI workloads. To harness its full potential, FF Group restructured the internal architecture of CAMMRA AI, aligning it with ARTPEC-9’s parallel processing framework. This was more than just a software update, it was a foundational reengineering of how the system processes video data.
Initial performance tests show a 20 to 30 percent improvement in speed and recognition consistency while the camera cycles through presets. This makes the solution well-suited for use cases where LPR coverage can be rotated between zones, rather than maintained continuously at each point. It is not a replacement for multiple fixed LPR cameras unless used in combination with external sensors or trigger mechanisms.
The system still provides advantages in streamlined deployment, requiring only one IP address and one VMS license, ideal for simpler setups or bandwidth-limited environments.
From a hardware perspective, both models feature 1/2-inch sensors, 31x optical zoom, and laser focus. This combination enables sharp, high-quality imaging and reliable recognition, even in low light or adverse weather conditions. Whether installed on rooftops or mast-mounted trailers, the cameras support effective long-distance capture when properly configured for license plate recognition.
For optimal LPR accuracy, the camera operates with increased shutter speed to reduce motion blur. This setup naturally lowers the effective IR illumination range compared to default surveillance settings, with real-world performance typically optimized for distances up to 150 meters. Laser-assisted focus ensures the camera quickly locks onto vehicles when switching views, minimizing out-of-focus time and preserving situational awareness.
Thanks to laser-assisted focus, the camera instantly locks onto targets as it moves between presets. There is virtually no lag during zoom or repositioning, which helps maintain continuous situational awareness. Every vehicle is captured clearly without gaps or delay.
One of the more practical outcomes of this approach is for ANPR cameras mounted on mobile surveillance trailers. Traditionally fixed focus ANPR cameras would be mounted on trailers and would need to be manually adjusted for optimum accuracy which has a labour and transport cost associated especially in more remote deployments. The advent of ANPR on PTZ now allows for faster deployment and remote adjustment capabilities offering trailer operators a lower total cost of ownership especially for those who provide Video Surveillance as a service. Most mobile surveillance trailer operators only deploy PTZ cameras for these reasons.
The solution lends itself particularly well to temporary installations, mobile deployments, and scenarios requiring dynamic coverage. It also suits use cases where LPR is not continuous but needs to be available on demand, with minimal setup and high positional flexibility.
In short, pairing CAMMRA AI with Axis’s ARTPEC-9-based PTZ platforms introduces a more scalable, cost-efficient approach to vehicle recognition. The ability to move, refocus, and accurately recognize in real time all within a single device, offers clear benefits for integrators and operators designing next-generation mobility infrastructure.
.svg)