Diffractive Deep Neural Networks at Visible Wavelengths
نویسندگان
چکیده
منابع مشابه
Selectively Deep Neural Networks at Runtime
Deep networks are powerful tools for computer vision tasks, but they incur significant computational costs, even at test time. In this work, we present an approach for optimizing existing deep networks to enable faster running times with minimal loss of accuracy. In contrast with approaches such as model compression, we do not create a new network, but instead perform minimal modifications to t...
متن کاملDeep inelastic diffractive scattering at HERA
It is well known that soft hadron-hadron scattering can be described by Regge phenomenology, which models the total, elastic and single and double diffractive dissociation cross sections in terms of reggeon and pomeron exchange [1]. At high energies, it is the pomeron exchange mechanism that dominates the total cross section [2]. However, until recently, our understanding of the pomeron in term...
متن کاملA polychromatic approach to far-field superlensing at visible wavelengths
Breaking the diffraction barrier in the visible part of the electromagnetic spectrum is of fundamental importance. Far-field subwavelength focusing of light could, for instance, drastically broaden the possibilities available in nanolithography, light-matter interactions and sensing at the nanoscale. Similarly, imaging with a nanometric resolution could result in incredible breakthroughs in sof...
متن کاملPrecise Measurement of Lunar Spectral Irradiance at Visible Wavelengths
We report a measurement of lunar spectral irradiance with an uncertainty below 1 % from 420 nm to 1000 nm. This measurement uncertainty meets the stability requirement for many climate data records derived from satellite images, including those for vegetation, aerosols, and snow and ice albedo. It therefore opens the possibility of using the Moon as a calibration standard to bridge gaps in sate...
متن کاملTwo-dimensional imaging by far-field superlens at visible wavelengths.
We report that two-dimensional (2D) sub-diffraction-limited images can be theoretically reconstructed by a new metamaterial far-field superlens. The metamaterial far-field superlens, composed of a metal-dielectric multilayer and a one-dimensional (1D) subwavelength grating, can work over a broad range of visible wavelengths intrinsically. The imaging principle and the reconstruction process are...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Engineering
سال: 2021
ISSN: 2095-8099
DOI: 10.1016/j.eng.2020.07.032