Real-time image recognition compute module uses NVidia GPU
This high-performance embedded computer (HPEC) uses the Jetson TX2 to provide twice the performance of its predecessor, or run at more than twice the power efficiency, while drawing less than 7.5W. It features a fanless, conduction-cooled design and offers high performance in a small form factor.
“With the increased power to performance ratio that the Jetson TX2 infuses into our next generation A176 GPGPU, we believe the use of deep learning in embedded computing is poised to grow exponentially,” said Dan Mor, product line manager, Aitech Defense Systems. “The industry hasn’t seen this type of processing in a compact form factor to date – it’s truly redefining what is possible.”
In addition to incorporating the new NVIDIA Jetson TX2 module, the unit now supports more hardware I/O and software options (1553, ARINC 429, Camera Link Frame Grabber), allowing faster integration. Internal microSD storage enables more design flexibility to scale the supercomputer to more complex compute-intensive applications. This includes data convolutions and transpositions, image and data manipulation, application of digital filters, image and frame object edge detection, and image recognition and data analysis.
Aitech; https://rugged.com/a176-cyclone-gpgpu-fanless-small-ff-redibuilt-supercomputer
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
