Introduction

Cities and communities around the world are turning in large numbers towards using AI edge computing applications as a solution to many of their needs. From Smart City and Smart Community initiatives to Intelligent Building management and Smart Security, there are many AI edge based applications to help improve the quality of life and wellbeing of people who live in these cities, communities and buildings.

One way in which AI is being deployed is through the use of facial recognition and behavioral analysis to help improve safety and security in these communities. Facial recognition can help easily identify and track persons of interest, while behavioral analysis improves security with the use of virtual fences and identifying suspicious behavior. These technologies can even be used to track and locate missing and lost items. The benefits of these AI Edge technologies help create a higher quality of life through citizens feeling safer in their communities and homes.

One company in China is leading the deployment of AI Edge and Smart City solutions and has begun development of a new edge computing platform to power advanced facial recognition applications. This facial recognition system is designed to be deployed in a range of AI applications such as person of interest tracking for Smart Cities, intelligent security and access control in Smart Buildings, and identifying residents and non-residents in Smart Community roles.

Challenges

With a powerful new AI Edge system with unique requirements, the company required an embedded hardware solution which could not only meet the demands of being an embedded AI edge platform, but also the specific needs and support required by their AI system. These challenges and requirements needed to be met in order to realize the full potential of their AI edge software.

Multiple Camera Input

To provide better coverage and tracking for edge applications, the facial recognition system requires multiple camera inputs. A single system supporting multiple camera inputs also helps to reduce the overall cost of deployment, making intelligent security solutions more attractive to cities and communities on a tighter budget.

Powerful AI Processor

With a powerful facial recognition software meant to be deployed in edge applications, the AI system needs to be powerful enough to run the software effectively.

Embedded System

To allow effective and flexible deployment of the company’s AI solution, the system needs to be deployed in a range of embedded configurations for cities, communities and buildings. This requires a system that is not only compact to fit in a range of spaces, but also rugged to deal with a range of environments and weather conditions, as well as being easy to install.

Proprietary Software Protection

As a large developer of AI technology, the company needed a system that helped secure their software. To ensure their proprietary software was safe from attack, the company required specialized encryption protection be put in place to prevent the software from being copied or stolen.

Custom OS

The specialization of the company’s AI software required the development of a custom operating system and system functions. The company devised a custom Linux build, but needed a partner to help test and refine the OS to ensure smooth operation.

Solution

AAEON is an award winning manufacturer of embedded AI Edge solutions. The company approached AAEON to not only provide a hardware solution for their AI software, but to provide the support needed to provide testing and preparation needed to bring their AI Edge system to market. Working closely with the company, AAEON determined the best solution to be the BOXER-8120AI embedded AI system with the NVIDIA Jetson TX2.

Excellent I/O Features

The BOXER-8120AI features a powerful compliment of I/O features, including four Gigabit Ethernet ports, perfect for creating an IP camera network. The company also request the BOXER-8120AI be configured with both COM and CANbus ports, allowing the system to connect with even more sensors or edge nodes, providing deeper integration into Smart City and Smart Community frameworks.

Powered by NVIDIA Jetson TX2

The BOXER-8120AI is powered by the NVIDIA Jetson TX2. The Jetson TX2 features a hexacore processor, created by combining the dual-core Denver 2 and quad-core ARM Cortex-A57 processors. It also features 8 GB of LPDDR4 memory and up to 32 GB of onboard eMMC storage. With support of up to 256 CUDA cores of processing, the BOXER-8120AI with NVIDIA Jetson TX2 provides the performance necessary for the AI system.

Compact, Rugged Design

The BOXER-8120AI is built compact and rugged. At only 30mm thick, it can fit into almost any tight space. With a fan-less design and all aluminum chassis, it can operate in various conditions, including a wide temperature range of -20°C to 50°C. Additionally, the company was impressed by the aesthetic design of BOXER-8120AI.

Software Security

AAEON works with our clients to provide embedded systems which help keep their software safe and prevent illicit copying and theft of intellectual property. The BOXER-8120AI comes standard with several key protections in place. Working together with the company, we configured the BOXER-8120AI with a high-end encryption protection system to address their specific needs and concerns.

OEM/ODM Services

AAEON provided support through our OEM/ODM Services, from adding an additional CANBus port to providing custom software protections. Working closely with the company, AAEON helped test and refine the custom Linux build the company chose to use. Furthermore, AAEON customized the BIOS, eliminating unneeded drivers and hardware controllers, making the system efficient and lean. This improved end-user experience and also helped accelerate testing times and reduce time to market.

Impact

In working closely with the company, AAEON was able to help in providing not only the hardware solution, but also OEM/ODM support to help them bring their facial recognition AI solution to market quickly.

With the deployment of the company’s facial recognition edge AI system, cities can expand their Smart City efforts in traffic and accident monitoring, Smart Communities can provide better safety for their residents, and buildings with Intelligent Security can ensure the people and property within are secure. As AI Edge applications continue to grow in number and ability to provide intelligent solutions, cities and communities around the world will be able to enjoy a higher quality of life for their residents.

BOXER-8120AI Embedded AI System

The BOXER-8120AI is designed for edge AI computing. Part of the BOXER-8100AI family of embedded AI solutions, it is equipped with the NVIDIA Jetson TX2, supporting 256 CUDA cores and a range of AI frameworks including Tensorflow, Caffe2, and Mxnet. Users can also install their own AI inference software on the BOXER-8120AI.

The BOXER-8120AI has four Intel LAN (i211) ports to support IP cameras, and it also boasts a series of rugged design features such as anti-dust specifications, a wide operating temperature range, a wide range power input, and an aluminum chassis. The BOXER-8120AI’s compact, cost-effective design and powerful processing abilities mean you’ll be able to deploy your AI platform anywhere.

AAEON Manufacturer Services provide configuration, customization and OEM/ODM support to ensure your AI application can be delivered quickly and effectively. Whether it’s changing an I/O port, or reconfiguring the BIOS, AAEON is a committed hardware solutions partner, and ensures the specific needs of your project are met.