Electronics Weekly Electronics Design & Components Tech News
Developed with Cloudtop to run on Xmos’ xcore.ai branded silicon, the AI-based software needs no cloud connection – all processing is local – and is aimed at car parks, optimised to read slow-moving number plates at a between three and five metres range using low-cost cameras.
“Down-scaling machine learning models so that they can run on mass-producible silicon like xcore.ai affords developers the funding and design flexibility to experiment,” said Cloudtop co-founder Professor Zhang.
“For smart parking, cloud connectivity and huge processing power is simply overkill,” said Xmos marketing v-p Aneet Chopra. “It makes automatic licence plate recognition networks far more expensive than they need to be, makes maintenance more complex, and comes rife with privacy concerns inherent to the cloud.”
Xmos and Cloudtop are showing the software at the tinyML Summit in San Francisco (28-30th March).
The automatic number plate recognition sortware product page is here
