Abstract: An optical measurement system and apparatus for carrying out cataract diagnostics in an eye of a patient includes a Corneal Topography Subsystem, a wavefront aberrometer subsystem, and an eye structure imaging subsystem, wherein the subsystems have a shared optical axis, and each subsystem is operatively coupled to the others via a controller. The eye structure imaging subsystem is preferably a fourierdomain optical coherence tomographer, and more preferably, a swept source OCT.

Abstract: An image processing method is performed by a computer device, which includes: converting initial image data into super-resolution image data using a trained image processing network, a resolution of the super-resolution image data being greater than or equal to a target resolution; performing image quality enhancement processing on the super-resolution image data using the trained image processing network, to obtain first enhanced image data; when there is a face image in the first enhanced image data, performing face enhancement on the face image in the first enhanced image data using the trained image processing network to obtain second enhanced image data; and performing image sharpening processing on the second enhanced image data using the trained image processing network to obtain sharpened image data.

Abstract: Target audio data and frequency spectrum information of the target audio data is acquired. The target audio data includes an audio missing segment and context audio segments of the audio missing segment. The frequency spectrum information includes frequency spectrum features of the context audio segments. Feature compensation is performed on the frequency spectrum information of the target audio data based on the frequency spectrum features of the context audio segments to obtain compensated frequency spectrum information corresponding to the target audio data. The compensated frequency spectrum information indicates upsampled frequency spectrum information of the target audio data. Audio prediction is performed based on the compensated frequency spectrum information to obtain predicted audio data.

Abstract: A wide-field imaging and hyperspectral collaborative early warning system and method are provided. The system includes: a primary imaging system, a digital micromirror array, a first collimation system, a second collimation system, a first reflector, a second reflector, a wide-field imaging optical filter, a narrow-band optical filter, a wide-field imager and a hyperspectral interferometer. The system and method have the advantages that the wide-field imaging general survey and fine spectrum detection can be realized synchronously, the functions are multiple, and the false alarm rate is low; the target spatial resolution of the fine spectrum identification is adjustable, the micro-control unit flipping solution is simple to operate, and quick adjustment is realized; the adaptability is strong, the flipping time of the corresponding micro-control unit is adaptively adjusted for the measured target with too low or too strong radiation intensity, and the dynamic range of the system is effectively improved.

Abstract: Embodiments of this application disclose an optical cable connection box and an adapter, and belong to the field of optical cable connection technologies. The optical cable connection box includes a box body, at least one pair of adapters, and at least one optical fiber, and each adapter includes a first interface and a second interface that are opposite to each other; the at least one pair of adapters is located on the box body, the first interface of each adapter is located outside the box body, and the second interface of each adapter is located inside the box body; and the at least one optical fiber is located inside the box body and is connected between the second interfaces of each pair of adapters.

Abstract: This application discloses a prefabricated cable, and relates to the field of communication technologies. In one example, the prefabricated cable includes an optical cable, an optical fiber connector, a traction member, and a branch cable fastener. The optical cable sequentially includes an insertion section, an optical cable connection section, and an optical cable extension section. The optical fiber connector is connected to an end of the insertion section. The traction member includes a traction section and a traction member connection section. The traction member connection section is connected to the optical cable connection section through the branch cable fastener. The insertion section and the traction section extend in a same direction.

Abstract: A two-stage driver supplies current to an LED load. A first stage of the driver generates a bulk voltage. A second stage has a flyback transformer with a primary winding and a secondary winding. The second stage generates an output voltage to cause LED load current. The primary winding is turned on and off by a gating signal. Control logic within the second stage is responsive to initially turning on the driver to perform a startup short circuit test of the output circuit by applying a gating signal with a short on-time and a low switching frequency. If the output circuit is not shorted, the control logic increases the on-time and the switching frequency to detect if an output current is excessive. If the output current is not excessive, the control logic adjusts the on-time and the frequency to provide sufficient current to illuminate the LED load.

Abstract: An optical measurement system and apparatus for carrying out cataract diagnostics in an eye of a patient includes a Corneal Topography Subsystem, a wavefront aberrometer subsystem, and an eye structure imaging subsystem, wherein the subsystems have a shared optical axis, and each subsystem is operatively coupled to the others via a controller. The eye structure imaging subsystem is preferably a fourierdomain optical coherence tomographer, and more preferably, a swept source OCT.

Abstract: A fiber connector plug, a fiber adapter, a connector assembly, and a communications device, where the fiber connector plug includes a ferrule fastened to a fiber, and a main housing and a front cover that encircle the fiber. The front cover is fastened to the main housing and encircles the ferrule. A front end face of the ferrule is flush with a front end face of the front cover, or in an axial direction, a front end face of the ferrule is located between a front end face of the front cover and a rear end face of the front cover. A slot is formed between the front cover and the ferrule, and the slot is used to accommodate an end face of a ferrule sleeve of a fiber adapter. The ferrule is used to be inserted into the ferrule sleeve.

Abstract: A two-stage driver supplies current to a light emitting diode (LED) load. The driver includes a first stage and a second stage. The first stage has a first flyback converter. The first stage is configured to receive a non-regulated voltage input and to generate a substantially constant bulk voltage across a first-stage output filter capacitor. The second stage has a second flyback converter. The second stage is configured to receive the bulk voltage from the first stage. The second stage is further configured to generate a desired current through the LED load. The second stage is electrically isolated from the first stage such that the LED load does not share a common ground reference with the non-regulated voltage input to the first stage. The driver further includes an auxiliary power supply coupled to an auxiliary winding of a transformer of the first flyback converter to generate an auxiliary voltage.

Abstract: Aspects of the disclosure provide an image processing method, an image processing terminal, an image processing apparatus, and a non-transitory computer-readable storage medium. The method can include performing face detection on a target image to detect a face region in the target image. The method includes determining an expression class of the detected face region based on a trained expression recognition model and determining whether the determined expression class of the detected face region is a target expression class. If the determined expression class is the target expression class, head detection is performed to identify a head region in the target image and the identified head region is edited. If the determined expression class is not the target expression class, the head detection is not performed on the target image.

Abstract: A fiber connector plug includes a main housing sleeved over a periphery of a fiber and a first locking structure disposed on an outer surface of the main housing, where the first locking structure is used to fit a second locking structure on a fiber adapter, the first locking structure includes a sliding member and a locking part, the locking part is fastened to the main housing, and the sliding member is slidably connected between a first position and a second position to the main housing; the locking part is located between the sliding member and a ferrule; and when the sliding member is located in the first position, the sliding member fits the locking part to jointly lock the second locking structure.

Abstract: The present invention discloses a method and system for recognizing a geometric regularity image of a honeycomb structure. The method includes the steps of image acquisition, image processing, vertex extraction, cell reconstruction, and quality evaluation, wherein a step of binaryzation is set between the step of image processing and the step of vertex extraction, and is to set a pixel value of a background in the image to be 0 and set a pixel value of a honeycomb skeleton in the image to be 1 to form a binary image, and the step of quality evaluation is to calculate angular deviation values of all the cells and an average thereof as well as linear deviation values and an average thereof based on the reconstructed cell image, and determine whether the honeycomb structure is qualified or not by comparing with a set tolerance zone.

Abstract: An optical splitting device is provided, which includes a housing, at least one first optical splitter which is disposed in the housing, a multi-core input optical interface, a multi-core output optical interface, and at least one single-core output optical interface. The multi-core input optical interface, the multi-core output optical interface, and the at least one single-core output optical interface are disposed on an outer wall of the housing, and each first optical splitter includes an input end, a first output end, and at least one second output end. The multi-core input optical interface is connected to an input end of the at least one first optical splitter, the first output end of each first optical splitter is connected to the multi-core output optical interface, and the second output end of each first optical splitter is connected to the single-core output optical interface in a one-to-one correspondence.

Abstract: A method for facilitating a remote conference includes receiving a digital video and a computer-readable audio signal. A face recognition machine is operated to recognize a face of a first conference participant in the digital video, and a speech recognition machine is operated to translate the computer-readable audio signal into a first text. An attribution machine attributes the text to the first conference participant. A second computer-readable audio signal is processed similarly, to obtain a second text attributed to a second conference participant. A transcription machine automatically creates a transcript including the first text attributed to the first conference participant and the second text attributed to the second conference participant.

Abstract: A node mesh contains an originating node and several node groups, each node group consisting of one or more nodes with interfaces connected to other nodes of the node group. Each node of a node group has an associated route table with an association between an applied DC voltage and an output interface to couple the input signal to. When the originating node outputs a DC voltage accompanied by differential signaling, each node in turn directs the DC voltage and differential signaling to an output interface as directed by the node local route table to a local termination in a node, which may be coupled to a training processor of inference processor for machine learning applications.

Abstract: A Bias Unit Element (UE) has a digital input, a sign input, and a chop clock. The sign input is exclusive ORed with the chop clock to generate a signed chop clock. Each Bias UE comprises a positive Bias UE and a negative Bias UE, each comprising groups of NAND gates generating an output and a complementary output, each of which are coupled to differential charge transfer lines through binary weighted charge transfer capacitors to a differential charge transfer bus comprising a positive charge transfer line and a negative charge transfer line. The chopped sign input enables the positive Bias UE when the sign bit is positive and enables the negative Bias UE when the sign bit is negative.

Abstract: An architecture for a multiplier-accumulator (MAC) uses a common Unit Element (UE) for each aspect of operation, the MAC formed as a plurality of MAC UEs, a plurality of Bias UEs, and a plurality of Analog to Digital Conversion (ADC) UEs which collectively perform a scalable MAC operation and generate a binary result. Each MAC UE, BIAS UE and ADC UE comprises groups of NAND gates with complementary outputs arranged in NAND-groups, each NAND gate coupled to a differential charge transfer bus through a binary weighted charge transfer capacitor to form an analog multiplication product as a charge applied to the differential charge transfer bus. The analog charge transfer bus is coupled to groups of ADC UEs with an ADC controller which enables and disables the ADC UEs using successive approximation to determine the accumulated multiplication result.

Abstract: This disclosure relates to method and system for visual inspection of rotating components. The method includes representing rotation cycles of a rotating component as spatial features based on video or image frames, ascertaining and/or evolving Hidden Markov Model (HMM) chains for the cycles, ascertaining a count of the rotating component in the frames and/or labelling the frames with ascertained states of the HMM chains.

Abstract: An analog machine learning architecture uses modular analog multiplier-accumulator (AMAC) elements of fixed size to form a machine learning (ML) system with increasing feature map size. A single 3 × 3 × 64 AMAC array is arranged to provide a three layer ML architecture with first layer 3×3×64, second layer 3×3×128, and third layer 3×3×256 using arrangements of single 3×3×64 AMACs arranged in parallel, where the bias of each AMAC is separately established in a unique interval of time.