ARBOR Embedded Board Works as an ANPR Camera Engine for an Expressway Speed Enforcement Project

Our client is a leading company in ANPR/LPR technology incorporated in Singapore. The company’s flagship product is their self-developed ANPR system which employs learning technique and neural network algorithms to run edge video analysis on IP cameras. With over 25 years professional experience in ANPR applications, their award-winning solution has been trusted for decades by local governments for border security and their immigration checkpoints. Other landmark installations include Singapore Parliament House, Hong Kong Jockey Club and Macau International Airport.

The project is a tender for implementing an Outdoor Video Camera System for speed enforcement purposes in South East Asia. The system comprises a series of components from the frontend imaging equipment to the backend management servers. With regard of the ANPR camera frontend system, it is to be installed at expressway roads as well as slip roads, and the camera needs to be able to capture the license plates at various speeds and during the adverse day and night lighting conditions. The camera should also have the Optical Character Recognition (OCR) capability to recognize license plates of different colors, font sizes, and types. And to facilitate 24/7 automated monitoring and alert the operator upon detection of anomalies, the system needs to enable intelligent analysis, both real-time and post-incident.
Instead of using just off-the-shelf IP cameras, where the risk of blur image may result in recognition difficulties, our client chose to customize their ANPR camera. This project-specific camera was done by integrating a high-end USB3.0 camera to an industrial high-performance x86 system to perform the computationally intensive image processing algorithms.

ARBOR’s EmCORE-i89M2 single board computer was selected by our client as the ANPR camera’s engine. The EmCORE-i89M2 features two USB 3.0 interfaces for USB 3.0 camera connection. With an Intel® Skylake Quad Core i7-6822EQ processor, the computer can provide the computing power necessary to process a vast amount of real-time data. Its Gigabit Ethernet interface enables high-speed streaming of videos and data to the remote viewing and management terminals.
The EmCORE-i89M2 was also selected because of its wide temperature capability from -40~85° C to ensure the operation in tropical areas. A heatsink is also attached to enhance thermal performance. Plus, conformal coating is applied to the board to avoid short circuits in highly humid environments.
Considering the space constraints in the camera enclosure, compact size and ease of wiring were two major concerns of selecting the computer. The EmCORE-i89M2’s compact size of 3.5” form factor and ARBOR’s support on cabling design makes it well suited. Also, the support of M.2 storage greatly saves space due to its ultra compact size and no need for extra cabling. Not to mention, it offers a higher transfer speed than the SATA interface.

With an in-depth understanding of industrial imaging technology, our client understands what matters to an ANPR camera. ARBOR’s wide range of off-the-shelf embedded offerings allows our customers to select one with the best performance and thus making an adequate ANPR camera engine for the project. The ANPR camera, coupled with edge computing capability and storage, and sophisticated video analytics capability from our client, form a robust edge analytics solution for the tender project. ARBOR’s offerings in conjunction with our client’s expertise together make the project a success.