Nature Podcast November 18, 2020 Scientists have grave concerns over ethical and societal impacts of facial-recognition technology. Cities across the globe are installing thousands of surveillance cameras equipped with facial recognition technology. Although marketed to reduce crime, researchers worry that these systems are ripe for exploitation and are calling for strict regulations on their deployment. Despite concerns surrounding consent and use, researchers are still working on facial recognition technology. Nature surveyed 480 researchers who have published papers on facial recognition, AI, and computer science. The results revealed that many researchers think there is a problem. Podcast
Category Archives: Imaging technology
New technology allows cameras to capture colors invisible to the human eye
Phys.org November 5, 2020 In upconversion imaging phase matching severely limits the spectral bandwidth, therefore requires serial acquisitions to cover a large spectrum. Researchers in Israel have designed an upconversion imaging scheme covering the mid‐IR based on adiabatic frequency conversion. They presented mid‐IR multicolor imaging and demonstrated simultaneous imaging on a CMOS camera of radiation spanning a spectrum from 2 to 4 µm. This approach being coherent and ultrafast in essence, spectrally resolved spatiotemporal imaging is further demonstrated that allows spatially distinguishing the temporal evolution of spectral components. The findings has applications in a variety of fields from computer gaming […]
Physicists circumvent centuries-old theory to cancel magnetic fields
Phys.org October 28, 2020 Controlling magnetism, essential for a wide range of technologies, is impaired by the impossibility of generating a maximum of magnetic field in free space. An international team of researchers (Spain, Italy, UK) circumvented the limits to shape magnetic fields by creating a device comprised of a careful arrangement of electrical wires. This creates additional fields that counter act the effects of the unwanted magnetic field. While a similar effect has been achieved at much higher frequencies, this team has achieved the same at low frequencies and static fields—such as biological frequencies—which will unlock a host of useful […]
Painting with light: Novel nanopillars precisely control intensity of transmitted light
EurekAlert September 4, 2020 By shining white light on a glass slide stippled with millions of tiny titanium dioxide pillars, an international team of researchers (China, USA – NIST, University of Maryland) has reproduced luminous hues and subtle shadings of a painting. By adding or dropping a particular color, or wavelength, of light traveling in an optical fiber, scientists can control the amount of information carried by the fiber. By altering the intensity, researchers can maintain the brightness of the light signal as it travels long distances in the fiber. The approach has potential applications in improving optical communications and […]
Quantum light squeezes the noise out of microscopy signals
Phys.org September 8, 2020 A team of researchers in the US (Tulane University, MIT, University of Colorado) demonstrated the first practical application of nonlinear interferometry by measuring the displacement of an atomic force microscope microcantilever with quantum noise reduction of up to 3 dB below the standard quantum limit, corresponding to a quantum-enhanced measurement of beam displacement. They minimized photon backaction noise while taking advantage of quantum noise reduction by transducing the cantilever displacement signal with a weak squeezed state while using dual homodyne detection with a higher power local oscillator. This approach may enable quantum-enhanced broadband, high-speed scanning probe […]
Seeing objects through clouds and fog
EurekAlert September 9, 2020 The presence of scattering places fundamental limits on our ability to image through fog, rain, dust, or the atmosphere. Conventional approaches for imaging through scattering media operate at microscopic scales or require a priori knowledge of the target location for 3D imaging. Researchers at Stanford University have introduced a technique that co-designs single-photon avalanche diodes, ultra-fast pulsed lasers, and a new inverse method to capture 3D shape through scattering media. They demonstrated acquisition of shape and position for objects hidden behind a thick diffuser at macroscopic scales. The technique complements other vision systems that can see […]
Researchers develop flat lens a thousand times thinner than a human hair
Nanowerk August 24, 2020 In a conventional lens an increase in refraction index increases the field of view in proportion to the flatness of the lens. To image with a single element over a wide field of view an international team of researchers (Brazil, UK, China) used photolithography to design a lens consisting of a single nanometric layer of silicon on arrays of nanoposts that interact with light. The resulting metalens mimics a totally flat lens with an infinite refraction index, which could not be obtained with a conventional lens. It has a field of view, which ideally can reach […]
Deep learning and metamaterials make the invisible visible
Nanowerk August 11, 2020 Due to the diffraction limit seeing and recognizing an object whose size is much smaller than the illumination wavelength is a challenging task for an observer placed in the far field. Researchers in Switzerland have demonstrated that combining deep learning with lossy metalenses allows recognizing and imaging largely subwavelength features directly from the far field. Their acoustic learning experiment shows that, despite being 30 times smaller than the wavelength of sound, the fine details of images can be successfully reconstructed and recognized in the far field, which is crucially favored by the presence of absorption. They […]
Recognising fake images using frequency analysis
EurekAlert July 16, 2020 To date, deep-fake images have been analysed using complex statistical methods. Researchers in Germany converted the images into the frequency domain using the discrete cosine transform and express the image as the sum of many different cosine functions. Natural images consist mainly of low-frequency functions. The analysis has shown that images generated by GANs (Generative Adversarial Networks) exhibit artefacts in the high-frequency range. The researchers’ experiments showed that these artefacts do not only occur in GAN generated images. They are a structural problem of all deep learning algorithms. Frequency analysis is therefore an effective way to […]
Detecting hidden nanostructures by converting light into sound
Nanowerk July 8, 2020 Using laser-induced, extremely high-frequency ultrasound researchers in the Netherlands detected diffraction gratings buried below a stack of tens of 18-nm-thick SiO2 and Si3N4 layers and an optically opaque metal layer. The shape and amplitude of a buried metal grating were encoded on the spatial phase of the reflected acoustic wave. They showed that the complex shape of the diffracted signal as a function of time can be reproduced using a comprehensive numerical model that includes the generation, propagation, and optical detection of the acoustic waves. The results show that laser-induced ultrasound is a promising technique for […]