Toradex worked in collaboration with Amazon Web Services (AWS) and NXP® Semiconductors to create the AI Embedded Vision Starter Kit, for developing cloud-connected computer vision and machine learning (ML) designs.
The kit functions as a reference implementation and was built using best in class cloud, industrial-grade edge software and hardware. It simplifies the creation of products in industries such as industrial automation, agriculture, medical equipment, and many more.
The AI Vision Starter Kit includes:
• Toradex Apalis System on Module (SoM) featuring NXP’s high performance i.MX 8QuadMax applications processor - optimized for safety and reliability
• Toradex Ixora Carrier Board
• Allied Vision Alvium 1500 industrial-grade MIPI CSI-2 camera
• All required cables and a power supply to get started
• Full software stack, including source code for running the device as well as for cloud deployment
• Extensive documentation
• 50 USD AWS credit
Today’s smart connected devices must meet a wide range of requirements:
• Secure connectivity for integration with business tools, remote monitoring and updates
• Maximum uptime and reliability, even with intermittent connectivity
• Small, rugged and cost-optimized computing hardware
• Optimized computer vision and machine learning technologies
• Short time-to-market, even with limited development resources
While there are solutions that solve each of these challenges, solving all of them together is not trivial. NXP, AWS and Toradex created a starter kit which does exactly that – it features connectivity, high reliability, a reduced time to market, computer vision and machine learning optimizations all in a small form factor.
“This AI Vision Starter Kit allows all customers to more easily unlock the embedded vision and machine learning potential of the i.MX 8QuadMax applications processor,” says Alex Dopplinger, industrial marketing manager at NXP. “The step-by-step instructions deploy a comprehensive reference software stack with just a few clicks, and combined with AWS and Toradex tools support, extends the reference design to many other uses.”
On the device side the kit comes with Torizon™, an easy-to-use industrial Linux® platform offered by Toradex. AWS IoT Greengrass is used to manage secure communications with the cloud as well as deliver offline capabilities. A local UI via HDMI is provided utilizing Web technologies. The cloud stack is deployed directly from the device in just a few steps via AWS CloudFormation as a dedicated AWS instance. AWS enabled the NXP i.MX 8 applications processor as a target for Amazon SageMaker Neo. Amazon SageMaker Neo allows developers to optimize a ML model for a certain device. Finally, to get everything up and running even faster, a trained and optimized model is included in the kit as well.
The source code for the software is available on Github and supported by extensive documentation in the Toradex Developer Center, making it easy to modify the implementation for your own needs.
To learn more visit: https://www.toradex.com/imx8-embedded-vision-starter-kit
To experience the solution and register for a chance to win a complementary starter kit, join the webinar Jumpstart Cloud-connected Computer Vision and Machine Learning Designs on February 6th visit: https://www.toradex.com/webinars/jumpstart-cloud-connected-computer-vision-and-machine-learning-designs
Toradex is a global company headquartered in Switzerland. It is focused on offerings to make high-reliable embedded computing easy.
Toradex offers Arm®-based System on Modules (SoMs) and customized SBCs. Powered by NXP® i.MX 6, i.MX 7, i.MX 8 and Vybrid; and NVIDIA® Tegra 2, 3 and TK1 SoCs, these pin-compatible SoMs offer scalability and come with free premium support, long-term product availability, and industrial grade software such as Torizon.
Toradex relies on exceptional engineering, modern infrastructure and advanced automation to consistently and rapidly deliver the latest and greatest hardware and software to its customers.
For more information, please visit https://www.toradex.com/.