Connect with us

Hi, what are you looking for?

Tech & Science

MIT’s new camera sensor could erradicate photo overexposure

MIT has developed a better way that uses a new “unbounded HDR” technique to allow camera sensors to absorb more light than they normally could. Ordinary sensors become saturated with light after receiving a certain amount, an effect that the new Modulo (modular arithmetic) sensor avoids.
The Next Web explains that the camera can reset its sensor capacitors at the point at which it would normally become saturated with light. It can then absorb light again and use an inverse modulo algorithm to calculate how much has been received in total.
The details were presented to the International Conference on Computational Photography. The technique has been likened to a 24-hour clock that resets itself every 12 hours instead of counting all the way through to 24 hours, becoming less accurate than if it had been set against another source halfway through.
Washed out skies and grey-shaded blacks are the classic symptoms of overexposure. Caused by letting too much light into an image, the effect can occur when the camera measures the light from the main subject and then gets overwhelmed in the other areas of the photo.
Traditional techniques to remedy it include darkening the affected areas in editing software or enabling high dynamic range (HDR) on devices that support it. HDR blends a regular shot with a deliberately underexposed and overexposed one to achieve stronger shadows and bright areas in one photo.
HDR can produce some stunning photos but also frequently falls short of expectations. Images can be ruined by blur and ghosting if the camera user’s hand moves because the sensor has to take several photos and blend them together. It would be much better if overexposure could be reduced without having to take more photos, something that Modulo could achieve.
The technology is a long way off being commercialised but could make its way into cameras in the future. MIT says that it could be useful in other environments besides photography. Robots and autonomous vehicles could also benefit from better light detection with less risk of dramatic over-exposure or blur introduced by HDR.

Written By

You may also like:

Life

Scam: “AI voice cloning” has been gaining traction on Google, with 23,000 monthly searches on average.

Life

If you receive a call from an unknown number without a caller ID and hear your friend or family member make an unusual request,...

World

The U.S. said it was terminating the legal status of hundreds of thousands of immigrants, giving them weeks to leave the country.

Business

Jaguar's ambition to seduce younger, richer drivers was on full display in Paris with a presentation of its newest prototype, the Type 00.