Ukrainians created a neural network to identify quarantine violators


Using a neural network, you can control the employees of companies where the use of masks is determined

Ukrainians have created a neural network that detects quarantine violators: how it works and where they will use it

Out of curiosity, the Ukrainian company Fulcrum has developed a neural network that can find people in crowded medical masks. The project is not commercial, according to the developer, the venture was driven by a net interest in how likely the venture is.

“Our goal was to verify whether people wearing masks on the streets could only be identified using webcams. It is not a commercial project, but it has interested us. In just two weeks, we were able to answer – yes, ”the developers say.

It took only two weeks to train the neural network. The developers described the process of its creation on the blog. First, networks were taught to find masks on images, and then proceed to video processing.

Webcams typically record short videos of 10–15 minutes. These videos can be sent to the server, where such a program will process them. This can be useful if a company or organization, for example, wants to ensure that all its employees wear masks.

“As for the roads, it is still science fiction and difficult to accomplish,” the company said.

Summary PAYSPACE magazine

Remember that the popularity of dressing and barrier systems has provoked difficulties with technologies such as Face ID, which are evaluated as a serious security threat in some countries. Especially in China.

Facial recognition systems are a common technique for PRC, here they are everywhere: at airports, on the streets, in shops and hotels. It is virtually an element of local infrastructure. And now, due to the proliferation of Kovid-19, biometric detection difficulties are encountered – as masks cover more than half of the face.

Read: Filters against COVID-19: Ford created a unique respirator using seat ventilation technology.

By content: AIN.UA


Source link

Leave a Reply