A diffractive optical neural network device includes a plurality of diffractive substrate layers arranged in an optical path. The substrate layers are formed with physical features across surfaces thereof that collectively define a trained mapping function between an optical input and an optical output. A plurality of groups of optical sensors are configured to sense and detect the optical output, wherein each group of optical sensors has at least one optical sensor configured to capture a positive signal from the optical output and at least one optical sensor configured to capture a negative signal from the optical output. Circuitry and/or computer software receives signals or data from the optical sensors and identifies a group of optical sensors in which a normalized differential signal calculated from the positive and negative optical sensors within each group is the largest or the smallest of among all the groups.

A system includes a diffractive optical element configured to receive a first beam and a second beam interfering with one another to generate a first interference pattern. The diffractive optical element is also configured to forwardly diffract the first beam and the second beam to output a third beam and a fourth beam. The third beam and the fourth beam interfere with one another to generate a second interference pattern. The system also includes a detector configured to detect the second interference pattern.

Display methods and apparatus are described. In some embodiments, to generate an image, light is selectively emitted from one or more light-emitting elements (such as a μLEDs) in a light-emitting layer. The emitted light from each element is collimated using, for example, an array of microlenses having small apertures. Each beam of collimated light is split by a first diffractive grating into a first generation of child beams, and the first generation of child beams is split by a second diffractive grating into a second generation of child beams. Beams in the second generation of child beams that are not parallel to the original beam of collimated light may be blocked by a spatial light modulator (e.g. an LCD panel). The un-blocked beams operate in some respects as if they had been generated using optics with an aperture larger than the apertures of the microlenses.

The optical component includes a first substrate, a first diffractive layer formed on the first substrate, a second substrate, a second diffractive layer formed on the second substrate, and a bonding material disposed between the first substrate and the second substrate and connecting the first substrate and the second substrate. The second diffractive layer is disposed opposite to the first diffractive layer, and both the first diffractive layer and the second diffractive layer are located between the first substrate and the second substrate. A gap is formed between the first diffractive layer and the second diffractive layer.

An apparatus for sorting orbital angular momentum eigenstates of one or more photons includes at least one transformation PBOE configured to sort orbital angular momentum eigenstates of the one or more photons, at least one phase correction PBOE configured to sort spin angular momentum eigenstates of the one or more photons. A method for sorting orbital angular momentum eigenstates of one or more photons includes using at least one transformation PBOE to sort orbital angular momentum eigenstates of the one or more photons, and using at least one phase correction PBOE to sort spin angular momentum eigenstates of the one or more photons.

An all-optical Diffractive Deep Neural Network (D.sup.2NN) architecture learns to implement various functions or tasks after deep learning-based design of the passive diffractive or reflective substrate layers that work collectively to perform the desired function or task. This architecture was successfully confirmed experimentally by creating 3D-printed D.sup.2NNs that learned to implement handwritten classifications and lens function at the terahertz spectrum. This all-optical deep learning framework can perform, at the speed of light, various complex functions and tasks that computer-based neural networks can implement, and will find applications in all-optical image analysis, feature detection and object classification, also enabling new camera designs and optical components that can learn to perform unique tasks using D.sup.2NNs. In alternative embodiments, the all-optical D.sup.2NN is used as a front-end in conjunction with a trained, digital neural network back-end.

A lighting apparatus for vehicles with a number of semiconductor-based light sources and a projection device for generating the specified light distribution with a cut-off line. The projection device features a correction device with at least two lenses. The surface of at least one of the lenses is designed as a diffractive lens surface for achromatization in a visible wavelength range. The two lenses are made from different lens materials. The surfaces of at least two lenses are designed as refractive lens surfaces that have their optical power calculated based on a temperature range and/or expansion coefficient of the lens material of at least two lenses such that adding the optical power of the lenses yields a predefined total optical power of the correction device.

Systems, methods, and apparatus are disclosed for attenuating an incident polarized light beam using a plurality of LCPGs and one or more switchable liquid crystal layers. When four LCPGs are used, a spacing between first and second LCPGs can be equal to a spacing between third and fourth LCPGs. Pi and FCL cells can also be used in place of more traditional LC switches. Switching of the LC switch can be imparted via an AC bias.

One embodiment of the invention provides a pattern generation device includes a light source, a first HPDLC cell, and a second HPDLC cell. The first HPDLC cell is disposed downstream of a light path of the light source and contains a first phase modulation pattern. The second HPDLC cell is disposed downstream of the light path of the first HPDLC cell and contains a diffraction grating pattern.

An optical guide has at least two diffraction gratings in serial in front of a light source in order to diffract a light beam from the light source twice. The first diffraction grating could split the light beam into several parallel light beams along a first axis, and the second diffraction grating could split the light beams into several points of light along a second axis, and so on and so forth. By rotating the diffraction gratings relative to one another and by adjusting the distance between the diffraction gratings, a user of the optical guide could adjust the angle of the axis points and adjust a relative distance of the points of light relative to one another. These light beams could provide convenient guides for users in a variety of applications.