Researchers at the University of California Irvine are developing an AI-powered night vision camera that is capable of shooting full color images in complete darkness. Put simply, humans are able to see light in the visible spectrum (400-700 nm), and although some night vision systems use infrared light that is not perceptible to humans, their images rendered are transposed to a digital display presenting a monochromatic image in the visible spectrum. Read more for two videos and additional information.
So, the researchers developed an imaging algorithm powered by optimized deep learning architectures whereby infrared spectral illumination of a scene could be used to predict a visible spectrum rendering of the scene as if it were being viewed by a human withing the visible light spectrum. This means that it would be possible to digitally render a visible spectrum scene to humans when they are otherwise in complete “darkness” and only illuminated with infrared light.
- Fine Detail Performance: 16 megapixel Micro Four Thirds sensor with no low pass filter resulting in a near 10 percent boost in fine detail resolving...
- Mirrorless Interchangeable Lens Camera Kit: With 2 super compact lenses 12 32mm F3.5 5.6 and 45150mm F4.0 5.6, shoot lighter and faster with the...
- Class Leading Dual Image Stabilization: In body 5 axis dual image stabilization works in both photo and motion picture recording—including 4K...
To achieve this goal, we used a monochromatic camera sensitive to visible and near infrared light to acquire an image dataset of printed images of faces under multispectral illumination spanning standard visible red (604 nm), green (529 nm) and blue (447 nm) as well as infrared wavelengths (718, 777, and 807 nm). We then optimized a convolutional neural network with a U-Net-like architecture to predict visible spectrum images from only near-infrared images,” said the researchers.
