They rank with rain, moisture and dust – spider webs are an irritating problem for any user of video surveillance cameras. Engineers at the software platform Videoloft are developing their next neural network – which is learning to detect when a spider web is covering a camera, causing false positive motion alerts. Hence the company is calling for volunteers to help train their neural network – they’ve asked their partners to give them access to cameras which are outdoors and covered in webs. The more spider webs it sees, the quicker their neural network will learn. It will the developers say be able to alert users who have spider webs, and in future versions Videoloft will be able to remove the false positives entirely.
Videoloft says its newly released analytics features are based on neural networks, a machine learning technique. Videoloft aim to make neural network technologies accessible to the security industry through their cloud CCTV platform.
Briefly; neural networks are based on the structures of the human brain that deal with vision and interpret images. By mimicking the way a human brain works, they can read information, learn from it and then predict the outputs of other similar information.
James West, Videoloft CEO and co-founder, said that neural networks have the ability to go above and beyond more common and traditional AI tools. He said: “The most recent dramatic AI advancements have been based on neural networks, yet we haven’t seen much adoption within the security industry.”
The company says that it can add neural network modules via remote software updates; no need to buy expensive hardware to get functionality, the firm adds. Visit www.videoloft.com.
