Abstract: A method for identifying a type of choroidal neovascularization in a retina of a human eye is disclosed, the method comprising receiving, in a processor, optical coherence tomography data of the eye; generating, in the processor, volume segments of the eye using the optical coherence tomography data and a neural network; and identifying, in the processor, the type of choroidal neovascularization in the eye, using the volume segments, the type of choroidal neovascularization comprising one or more of the following: classic choroidal neovascularization and occult choroidal neovascularization.

Abstract: The present disclosure provides a driving method and a display device. The driving method is applied to a display panel, including a plurality of rows of scanning lines, a plurality of columns of data lines, and a plurality of rows and a plurality of columns of pixel circuits, the pixel circuits being respectively electrically connected to corresponding rows of scanning lines and corresponding columns of data lines. The driving method includes: controlling a first charging time to be greater than 0.5 times a second charging time, and controlling the first charging time to be less than the second charging time; wherein the first charging time is a time for charging an Ath-row Mth-column pixel circuit via an Ath data voltage on the Mth-column data line; the second charging time is a time for charging the Bth-row Mth-column pixel circuit via the Ath data voltage and a Bth data voltage.

Abstract: A micro light emitting diode structure includes a temporary substrate, a plurality of micro light emitting elements, a plurality of light blocking structures, and a connection layer. The micro light emitting elements and the light blocking structures are disposed on the temporary substrate and arranged alternately. Each of the light blocking structures includes a light blocking layer, and a light shielding layer disposed thereon. The micro light emitting elements and the light blocking structures are fixed to the temporary substrate by the connection layer. The connection layer is a part of a plurality of fixing structures. A reflectivity of the light blocking layer is greater than a reflectivity of the connection layer, and a Young's modulus of the light blocking layer is greater than a Young's modulus of the fixing structures, and the Young's modulus of the fixing structures is greater than a Young's modulus of the light shielding layer.

Abstract: A ribbon speaker includes a housing, first and second magnet assemblies, a ribbon diaphragm, a coil, and a flexible substrate. The first and second magnet assemblies is arranged in the housing. The same poles of the first magnet assembly and the second magnet assembly are arranged opposite to each other. The diaphragm is positioned between these magnet assemblies. The coil, placed on this diaphragm, includes two first connection portions. The flexible substrate includes a body, two connection portions, and two soldering portions. The body's first part is inside the housing, while the second part arranged outside. The second connections on the body's first part electrically connected to the coil's first connections. The externally positioned soldering portions facilitate easier assembly, significantly reducing soldering complexity. The soldering portions are respectively and electrically connected to the second connection portions.

Abstract: A method of processing video data includes receiving a picture; and filtering a current block of the picture, through a neural network and based on local correlations of proximate samples and distant, non-local correlations of non-proximate samples relative to the current block, to generate a filtered current block. The neural network comprises one or more backbone blocks and one or more transformer blocks. Each of the one or more transformer blocks is associated with a backbone block of the one or more backbone blocks. At least one of the backbone blocks is configured to capture the local correlations, relative to the current block and the proximate samples of the current block, and at least one of the transformer blocks is configured to generate features, based on applying an attention mechanism, that capture the distant, non-local correlations, relative to the current block and the non-proximate samples, in the picture for processing.

Abstract: Methods and devices for decoding video data are described. An example method includes in-loop filtering a current block of the video data using a neural network-based in-loop filter to generate an in-loop filtered current block, wherein the neural network-based in-loop filter is trained using an architecture comprising a U-Net architecture comprising one or more residual blocks and one or more transform blocks; and outputting the in-loop filtered current block.

Abstract: A cerium-zirconium-aluminum-based composite material, a cGPF catalyst and a preparation method thereof are provided. The cerium-zirconium-aluminum-based composite material adopts a stepwise precipitation method, firstly preparing an aluminum-based pre-treated material, then coprecipitating the aluminum-based pre-treated material with zirconium and cerium sol, and finally roasting at high temperature to obtain the cerium-zirconium-aluminum-based composite material. The cerium-zirconium-aluminum-based composite material has better compactness and higher density, and when it is used in cGPF catalyst, it occupies a smaller volume of pores on the catalyst carrier, such that cGPF catalyst has lower back pressure and better ash accumulation resistance, which is beneficial to large-scale application of cGPF catalyst.

Abstract: The present invention provides a federated large model adaptive learning system. Based on the combination of multiobjective optimization and incremental learning, multiple optimization indexes are constructed, and adaptive mini model incremental learning is designed. A gradient scaling method of mini models is proposed for data privacy protection under federated learning, to make full use of gradient information. A correlation between the generalization ability and sampling data is revealed to propose a generalization ability evaluation function. With respect to the real problems of performance degradation and fault faced by industrial equipment during operation, multiple optimization objectives are designed in combination with the generalization ability evaluation function, and the models are updated and repaired adaptively through multiobjective evolutionary learning, to improve the usability of large models in real industrial scenarios.

Abstract: A pixel circuit includes a current generator, a switch, and a time controller. The pixel circuit is configured in a display panel. The current generator provides a driving current, and controls the driving current according to a pulse amplitude modulation mechanism. The switch is coupled in series with the current generator and a light-emitting component, and is turned on or off according to a control signal, wherein the control signal is a pulse width modulation signal. The time controller generates the control signal, receives a set voltage, and adjusts a pulse width of the control signal according to a voltage value of the set voltage.

Abstract: A micro light emitting diode display panel including a plurality of display units and a control element is provided. The plurality of display units are arranged in an array and each of the plurality of display units includes a first sub-pixel. The first sub-pixel includes a first micro light-emitting diode and a second micro light-emitting diode. The control element is configured to control light emission of the first micro light-emitting diode and the second micro light-emitting diode and determines operating currents of the first micro light-emitting diode and the second micro light-emitting diode. In a same display image, the operating current of the first micro light-emitting diode increases as an operating temperature of the first sub-pixel increases.

Abstract: A device for decoding video data is configured to determine first neural network (NN) weights for a layer of a convolutional neural network (CNN) filter; derive a quantization value for the layer of the CNN filter based on values of the first NN weights; convert the first NN weights to second NN weights based on the quantization value; and filter a block of the video data using the second NN weights.

Abstract: A multi-layer display module includes a first display panel, and a second display panel. Dimension of long side of the first display panel is D1, and the first display panel has first pixel resolution P1. The second display panel is located on one side of the first display panel and overlapped with the first display panel. There is a space d between the first display panel and the second display panel. Dimension of the long side of the second display panel is D2, and the second display panel has the second pixel resolution P2. Transmittance of the second display panel is T2. The multi-layer display module complies with T2>40%, P1?P2, and D ? 2 * T ? 2 ? d ? D ? 2 ? "\[LeftBracketingBar]" P ? 1 - P ? 2 ? "\[RightBracketingBar]" * P ? 2 .

Abstract: A method includes epitaxially growing a Ge1-xSnx channel layer over a substrate. The Ge1-xSnx channel layer is in a metastable state. A Ge1-ySny barrier layer is epitaxially grown over the Ge1-xSnx channel layer to form a two-dimensional hole gas in the Ge1-xSnx channel layer. The Ge1-xSnx channel layer and the Ge1-ySny barrier layer are etched to form a first opening and a second opening in the Ge1-xSnx channel layer and the Ge1-ySny barrier layer. A first source/drain electrode and a second source/drain electrode are deposited in the first opening and the second opening, respectively. A gate electrode is formed over the Ge1-ySny barrier layer.

Abstract: An electrode structure, a display panel, an electronic device are provided, the electrode structure includes a first electrode portion, a second electrode portion and a conductive connection portion, the first electrode portion includes a first connection bar having a first side and a second side and a plurality of first electrode strips, ends of adjacent first electrode strips away from the first connection bar are open; the second electrode portion includes a second connection bar at a position of the first side away from the second side and a plurality of second electrode strips, the second connection bar includes a third side and a fourth side; the second electrode strips are connected with the second connection bar, ends of adjacent second electrode strips away from the second connection bar are open; ends of the conductive connection portion are connected with the first connection bar and the second connection bar.

Abstract: Introduced are a method, a software apparatus, an electronic device, and a storage medium for simulating inerter using finite element analysis. The method involves: acquiring structural parameters of an inerter; simulating the mechanical properties of racks on rigid rods and flywheels within the inerter using the structural parameters and a finite element platform; formulating constraint equations and transformation formula to simulate translational-rotational conversion and inertia amplification mechanism of the inerter; integrating the force and constraint information to achieve the finite element simulation of the inerter. The finite element simulation method for the inerter addresses the challenge of lacking “inerter units” and the inability to simulate inertia components in general finite element software platforms. It also overcomes the technical bottleneck of real-time dynamic coupling simulation between inerter and complex engineering structures.

Abstract: A system comprising a first image processing arrangement and a second image processing arrangement, wherein the first image processing arrangement comprises a controller configured to: a) receive an image; b) select a task and a task identifier associated with task data; c) compress the image based on the task data; and d) transmit the compressed image to the second image processing arrangement for processing, and wherein the second image processing arrangement comprises a controller configured to: e) receive the compressed image and task identifier; f) retrieve task parameters associated with the task identifier; g) process the compressed image based on the task parameters; h) determine results and i) transmit the at least indications of the determined results to the first image processing arrangement, and the controller of the first image processing arrangement is further configured to: j) receive at least indications of a result of the processing from the second image processing arrangement; and k) indicat

Abstract: An electronic device includes a substrate, a hyperbolic magnet, a pair of depletion gates, a pair of barrier gates and a accumulation gate. The hyperbolic magnet is over the substrate and has a first magnet portion and a second magnet portion separated from each other. The first magnet portion and the second magnet portion have a first convex surface and a second convex surface facing the first convex surface, respectively. The depletion gates are separated from each other and between the first convex surface and the second convex surface over the substrate. The barrier gates are between the depletion gates. The accumulation gate is over the depletion gates and between the barrier gates.

Abstract: Aspects of the disclosure are directed to information processing. In accordance with one aspect, information processing includes a databus; a memory system coupled to the databus; and a graphics processing unit (GPU) coupled to the memory system and the databus, wherein the GPU is configured to do the following: retrieve a first plurality of atomic operations containing a first plurality of data values for a shared memory location; compute a first aggregate data value based on the first plurality of data values; and generate a first aggregate atomic operation containing the first aggregate data value.

Abstract: The current disclosure describes a new vertical tunnel field-effect transistor (TFET). The TFET includes a source layer over a substrate. A first channel layer is formed over the source layer. A drain layer is stacked over the first channel layer with a second channel layer stacked therebetween. The drain layer and the second channel layer overlap a first surface portion of the first channel layer. A gate structure is positioned over the channel layer by a second surface portion of the channel layer and contacts a sidewall of the second channel layer.

Abstract: A method of making non-grain oriented (NGO) electrical steel is disclosed. The method includes tapping liquid steel out of a primary steelmaking furnace, deoxidizing the liquid steel before or after transferring the deoxidized liquid steel to a ladle metallurgy furnace, removing sulfur at the ladle metallurgy furnace (LMF), adding fluxes and deoxidizer to the ladle slag and/or skimming off ladle slag to prevent sulfur reversion, transferring the deoxidized liquid steel from the ladle metallurgy furnace to an RH degasser for carbon removal by blowing oxygen, and adding fluxes at the RH degasser before oxygen blowing to fortify the bottom layer of the ladle slag to prevent sulfur reversion. The removal of oxygen and sulfur prior to transferring the liquid steel to the RH degasser facilitates nitrogen removal and prevents carbon pick up during the step of adding fluxes and arcing for sulfur removal. Oxygen blowing at the RH also lowers titanium pickup.