Practical test: first ML experiences with Google Coral and TensorFlow Lite
Some Raspberry Pi projects use machine learning to recognize objects, words or gestures, or to process data using neural networks. If you want to run “AI applications” in the form of deep or machine learning (DL, ML) in the field of IoT, for example on an old laptop, Raspberry Pi or other “computers single boards ”(SBC), you come across but quickly to the limits of computing power. Face or object recognition via a camera, for example, runs in slow motion on the Raspi. This may be sufficient for experiments in this area and a “proof of concept”, but not for practical use.
But there is a remedy: Raspi, PC & Co. can be extended with special chips for ML applications. The Google Coral USB3 key used in the suite (from € 69.90) speeds up object recognition on a Raspberry Pi with a camera with a lameness of 3 to 5 at more than 30 images per second. In this article, we take a look at the inner technical workings of the stick and use a sample project to determine what the stick can do.
- Access to all heise + content
- proprietary testing, advice and history: independent, well-founded
- Read c’t, iX, MIT Technology Review, Mac & i, Make, c’t photography directly in your browser
- register once – read on all devices – can be canceled monthly
- first month free, then € 12.95 per month
- Weekly newsletter with personal reading recommendations from the editor-in-chief
Start the month FREE
Start your FREE month now
Already subscribed to heise +?
Sign up and read
Register now and read the articles right away
More information about heise +