Topic: Technology
Researchers at UCLA created a special mirror called a 'lying mirror' that can hide information in an image. This mirror uses light and a special surface to transform the input image into a misleading pattern. The team tested this concept with great success, even under different lighting conditions.
Mirrors normally show us what's in front of them. But researchers at the University of California, Los Angeles (UCLA) have created something new: a 'lying mirror' that hides information carried by an input image.
The lying mirror uses a combination of a reflective mirror and a special surface called a diffractive surface. This surface is designed to bend light in a specific way, creating a misleading pattern from the original image. Instead of using a computer to alter the image digitally, the lying mirror performs this transformation through light and matter interactions.
The UCLA team optimized the phase profile of the structured surface using deep learning. This means that many different input objects could be mapped to a predefined 'dummy' output image. The researchers tested their concept with great success, even when input objects were rotated, shifted, or scaled, and when noise was added to the images.
The team experimentally validated the concept using a programmable micro-mirror array in the visible spectrum. They successfully transformed randomly selected objects into a predefined dummy output image, which was a handwritten digit '8'. The researchers also developed a broadband lying mirror that operates across a continuous spectral range, extending the concept beyond discrete RGB illumination.
Why It Matters
This new technology has potential applications in security, defense, and anti-surveillance technologies. It could also be used in entertainment industries to create new visual effects.
Key Facts
- Researchers at UCLA created a 'lying mirror' that can hide information in an image.
- The lying mirror uses light and a special surface to transform the input image into a misleading pattern.
- The team optimized the phase profile of the structured surface using deep learning.
- The concept was experimentally validated using a programmable micro-mirror array in the visible spectrum.
- A broadband lying mirror that operates across a continuous spectral range was also developed.
Key Terms
- Diffractive Surface
- A surface designed to bend light in a specific way, creating a misleading pattern from the original image.
- Deep Learning
- A type of machine learning that uses complex algorithms to optimize the phase profile of the structured surface.
Implications
This new technology has potential applications in security, defense, and anti-surveillance technologies. It could also be used in entertainment industries to create new visual effects.
Leave a comment