[ad_1]
Edge devices are typically defined as any hardware element that controls the flow of data between the boundaries of two networks. These include not only routers, switches and gateways, but also a range of IoT gadgets such as Ring Cameras, Industrial Robots, Smart Medical Devices or Self-Guided Camera UAVs. Intel's NCS2 is essentially a stand-alone neural network connected to a USB key. It should make it faster and easier to develop these devices by offloading much of the processing power needed to train them at its integrated Movidius Myriad X vision processing unit (VPU).
<img alt = "NCS2a" data-credit-link-back = "" data-dam-provider = "" data-local-id = "local-1- 530272-1542144248804 "data-media-id =" b67ebe4d-a3d1-3ed1-aa1d-434b06fced0e "data-original-url =" https://s.yimg.com/os/creatr-uploaded-images/2018-11/ 6cc05fd0-e78a-11e8 -bedf-400c7f6e494f "data-title =" NCS2a "src =" https://o.aolcdn.com/images/dims?crop=1600%2C1199%2C0&quality=85&format=jpg&resize=1600%2C1199&image_uri= https% 3% 2F% 2Fs.yimg.com% 2Fos% 2Fcreatr-Upload-Images% 2F2018-11% 2F6cc05fd0-e78a-11e8-bedf-400c7f6e494f
"We can do incredible things with artificial intelligence, and one of the fastest growing types of data, as we all know, is video data," said Steen Graham, director General Channels and Ecosystem in the group Internet of Things of Intel, to reporters. press call Friday. "The camera is the ultimate sensor."
Basically, it allows you to develop artificial intelligence algorithms and artificial vision systems locally (that is to say on your laptop) without the need for a cloud connection or even to the Internet. Users simply need to run it on a Linux-based operating system with a USB 3.0 port. In addition, the processing cores added to the NCS2 would allow it to run eight times faster than its predecessor, released last year.
<img alt = "ncs2b" data-credit-link-back = "" data-dam-provider = "" data-local-id = "local-2- 3734618-1542144282839 "data-media-id =" 6ff08a32-b5b8-3474-9f73-bb38affbf0e1 "data-original-url =" https://s.yimg.com/os/creatr-uploaded-images/2018-11/ 8036ac90-e78a-11e8 -bbf7-7b0122794370 "data-title =" ncs2b "src =" https://o.aolcdn.com/images/dims?crop=1600%2C1199%2C0&quality=85&format=jpg&resize=1600%2C1199&image_uri= https% 2F% 2Fs.yimg.com% 2Fos% 2Fcreatr-the-images-sent% 2F2018-11% 2F8036ac90-e78a-11e8-bbf7-7b0122794370 & customer = a1acac3e1b3290917d92 & signature = 8b40be9bc29nm & hell
Intel has already pre-released the NCS2 to a handful of developers who have used it to develop a variety of useful applications, including Clean Water AI, which combines artificial vision with a microscope to detect harmful bacteria in the water; BlueScan AI, which scans the skin for signs of melanoma; and the ASL classification, which translates American sign language into text, all in real time.
The NSC2 costs 100 USD and is available via Intel. There are no restrictions on who can buy one. Whether you are a researcher in artificial intelligence or an amateur, if you have a Benjamin, you can get an NCS2. This could be exciting because, with a broader developer base, the rest of us will likely see more and better improvements to our existing IoT devices and devices, as well as new gadgets and features that we do not have. We have not even considered it yet.
Source link
