How Machine Learning Training Algorithms Perform

1min Snip

00:00

Play full episode

Summary

Transcript

Episode notes

Our standard thermal cameras do not have the capability to extract optimal information from the thermal radiation field. So we are baking in some of the physical understanding of how heat emanates and heat propagates in the environment that is involved in the machine learning training algorithms. And they're able to sort of take all that information and spit out an image that is more useful and more like daytime vision, I guess. Exactly,. exactly. How well would you say this system performs?

In this episode:

00:46 How to see in the dark like it’s daytime

There are many methods for better night-vision, but often these rely on enhancing light, which may not be present, or using devices which can interfere with one another. One alternative solution is to use heat, but such infrared sensors struggle to distinguish between different objects. To overcome this, researchers have now combined such sensors with machine learning algorithms to make a system that grants day-like night-vision. They hope it will be useful in technologies such as self-driving cars.

Research article: Bao et al.

News and Views: Heat-assisted imaging enables day-like visibility at night

09:27 Research Highlights

Benjamin Franklin’s anti-counterfeiting money printing techniques, and how much snow is on top of Mount Everest really?

Research Highlight: Ben Franklin: founding father of anti-counterfeiting techniques

Research Highlight: How much snow is on Mount Everest? Scientists climbed it to find out

11:47 Briefing Chat

We discuss some highlights from the Nature Briefing. This time, the cost to scientists of English not being their native language, and the mysterious link between COVID-19 and type 1 diabetes.

Nature News: The true cost of science’s language barrier for non-native English speakers

Nature News: As COVID-19 cases rose, so did diabetes — no one knows why

Subscribe to Nature Briefing, an unmissable daily round-up of science news, opinion and analysis free in your inbox every weekday.

Hosted on Acast. See acast.com/privacy for more information.