Abstract: Methods and systems for detecting defects on a specimen are provided. One system includes an inspection subsystem configured to generate images of a specimen and one or more computer subsystems configured for detecting defect candidates on the specimen. Detecting the defect candidates includes identifying a patterned feature in a test image included in the images generated of the specimen. Detecting the defect candidates also includes, for at least one pixel in the test image located within the patterned feature, determining a difference between a characteristic of the at least one pixel and the characteristic of other pixels in the test image located within a predetermined window of the at least one pixel. In addition, detecting the defect candidates includes detecting a defect candidate at the at least one pixel based on the determined difference.

Abstract: Method and system for predicting a label for an input sample. A first label is predicted for the input sample using a first machine learning (ML) model that has been trained to map samples to a first set of labels; If the first label satisfies prediction accuracy criteria it is outputted as the predicted label for the input sample; if the first label does not satisfy the prediction accuracy criteria, a second label is predicted for the input sample using a second ML model that has been trained to map samples to a second set of labels that includes the first set of labels and a set of additional labels, and the second label is outputted as the predicted label for the input sample.

Abstract: A prodrugged antibody has a blocking moiety attached to a Cys on its heavy or light chain via a linker having a cleavable moiety. The blocking moiety inhibits binding of the antibody to its antigen. Cleavage of the cleavable moiety releases the blocking moiety and restores ability of the antibody to bind to its antigen.

Abstract: Methods and systems for generating embeddings from molecular graphs, which may be used for classification of candidate molecules. A physical model is used to generate a set of task-relevant feature vectors, representing local physical features of the molecular graph. A trained embedding generator is used to generate a set of task-relevant structural embeddings representing connectivity among the set of vertices and task-relevant features of the set of vertices. The task-relevant feature vectors are combined with the task-relevant structural embeddings and provided as input to a trained classifier. The trained classifier generates a predicted class label representing a classification of the candidate molecule.

Abstract: The invention provides a method, apparatus, device and storage medium for predicting a protein binding site. The method comprises the steps of: receiving a protein sequence to be predicted, dividing the protein sequence by using a preset sliding window and sliding step to obtain a plurality of amino acid sub-sequences, building word vectors for the protein sequence according to the amino acid sub-sequences, extracting document features from word elements, building document feature vectors for the protein sequence according to the extracted document features, extracting protein chain biological features from the amino acid sub-sequences, building biological feature vectors for the protein sequence according to the extracted biological features, classifying the amino acid sub-sequences expressed with the document feature vectors and the biological feature vectors by using a preset amino acid residue classification model to obtain amino acid residue types for the protein sequence.

Abstract: A compact silicon waveguide mode converter, a dielectric meta-surface structure based on periodical oblique subwavelength perturbations, including a top silicon structure with oblique subwavelength perturbations etched in certain periods with period length of ?, a duty cycle and an oblique angle ? on the SOI substrate. The invention adopts an all-dielectric meta-surface structure with oblique subwavelength perturbation, which can achieve a compact mode conversion from fundamental mode to arbitrary high-order mode of silicon waveguide, and can improve the optical communication capacity greatly.

Abstract: A device for preparation of composite for on-site pipeline reinforcement includes: a temperature control stirring unit, an infiltration unit, and a vacuum unit communicated in sequence, the infiltration unit includes a spindle, reinforced fiber cloth, a flow-guiding net, and a vacuum bag film sleeved outside them, the spindle is stopped by two baffles, an adhesive feeding joint and an adhesive discharging joint are disposed at two ends of the spindle, respectively, the adhesive feeding joint and the adhesive discharging joint each includes an inner joint and an outer joint, sealing discs disposed at a junction between the inner joint and the outer joint, an outer wall of the inner joint and an outer side of the baffle are covered by a flow-leading net, and the flow-guiding net covered on the outer side of the baffle extends from an edge of the baffle into the adhesive storing compartment of the baffle.

Abstract: The present disclosure relates to the technical field of image processing, in particular to an ultrasound imaging device and system and a breast ultrasound apparatus.

Abstract: This application relates to a face image processing method, apparatus, computer device, and storage medium. The method includes acquiring a first face image and a second face image, the first face image and the second face image being images of real faces; generating a first updated face image with non-real face image characteristics based on the first face image; adjusting color distribution of the first updated face image according to color distribution of the second face image to obtain a first adjusted face image; acquiring a target face mask of the first face image, the target face mask being generated by randomly deforming a face region of the first face image; and blending the first adjusted face image and the second face image according to the target face mask to obtain a target face image. Accordingly, a diversity of target face images can be generated.

Abstract: Servers, methods and systems are disclosed for fair and secure vertical federated learning. Fair and secure vertical federated learning (FSVFL) systems are disclosed that achieve one or more of the following properties: model fairness, high security, high accuracy, high efficiency, and/or high generality. Private data is retained on local computing systems, which share only their model outputs, and a server or a trusted computing system shares only model gradients and randomly partitioned sets of data sample identifiers with untrusted computing systems. A fairness constraint protects protected classes of data samples against model training resulting in bias on the basis of the protected classes.

Abstract: The present disclosure provides a light-adjusting glass and a smart vehicle window. The light-adjusting glass has a transmittance adjustment region and an encapsulation region; the light-adjusting glass includes: a first and second substrates opposite to each other, and a dye liquid crystal layer between the first and second substrates in the transmittance adjustment region, and a frame sealant in the encapsulation region; the first substrate includes a first base and a first electrode layer on a side of the first base proximal to the dye liquid crystal layer; the second substrate includes a second base and a second electrode layer on a side of the second base proximal to the dye liquid crystal layer; a conductive structure is provided in the frame sealant; a first and second voltage transmission structures electrically insulated from each other are on the first base.

Abstract: A highly linear time amplifier with power supply rejection. In a reset stage, the threshold value of an over-threshold detector is used for resetting an output node of an amplifier, to eliminate the impact of power supply voltage changes on the threshold value of the threshold detector. A node capacitor unit is charged under the control of an input clock signal. After completion of charging, the node capacitor unit is discharged under the control of a synchronous clock signal. The time amplification gain only depends on the proportion of the charge and discharge current, and the charging and discharging time are completely linear in principle, which eliminates the nonlinearity of the traditional time amplifier, and reduces the negative impact of threshold change on system performance.

Abstract: Methods and systems for generating a semantically augmented image are disclosed. An embedding is generated for each object label associated with a segmented input image. For each embedding associated with a respective object label, a similarity score is computed between the embedding associated with the object label and an embedding representing an object class in an object bank storing a plurality of object images. At least one object is selected, the selected object being associated with a respective object image in the object bank, the selected at least one object being from an identified object class that is identified as contextually relevant to at least one object label associated with the segmented input image, based at least on the similarity score. The selected object is added into the segmented input image to generate the augmented image.

Abstract: A pipelined analog-to-digital converter and an output calibration method for the same. The pipelined analog-to-digital converter introduces an error calibration mechanism on the basis of traditional pipelined analog-to-digital converters through a control module, an equivalent gain error extraction module, an error storage module and a coding reconstruction module to compensate for gain errors and setup errors caused by operational amplifiers in a pipelined conversion module, so that the analog-to-digital conversion accuracy is improved, and requirements for the gain and bandwidth of the operational amplifier are relaxed, which can effectively reduce the power consumption of the analog-to-digital converter and the complexity of the corresponding analog circuit; a curve fitting method is adopted to obtain an ideal output sequence and then calculate errors; meanwhile, extraction and calibration of equivalent gain errors are all done in digital ways, and therefore accuracy thereof is high.

Abstract: Disclosed herein are methods of producing metal nanoparticle-decorated carbon nanotubes. The methods include forming a reaction mixture by combining a first solution with a second solution, wherein the first solution comprises polymer-coated metal nanoparticles comprising metallic nanoparticles coated with a polymer, and wherein the second solution comprises carbon nanotubes. The methods also include heating the reaction mixture to a temperature greater than a glass transition temperature of the polymer for a time sufficient to cause the polymer-coated metal nanoparticles to bind to the carbon nanotubes forming the metal nanoparticle-decorated carbon nanotubes.

Abstract: A video recording method and a video recording device are provided. The method includes: obtaining video data to be recorded; dividing, based on the video data, each frame of the video data into a region of interest and a background region by using a preset neural network model; and encoding the region of interest of the video data based on a first encoding bit rate, and the background region based on a second bit rate, and storing the encoded video data into a storage device through a video buffer.

Abstract: In a method for embedding user information in webpages, user information of a user who logs into a target website is obtained. The user information is converted into color values. The color values are embedded in a target webpage of the target website. If a webpage is leaked, a user who leaks the webpage can be traced.

Abstract: An implantable light delivery device comprises: a core portion cladded in a cladding, the core portion comprising upconversion nano-particles (UCNPs) encapsulated in an encapsulation material. The implantable light delivery device may be used in treatments such as photodynamic therapy to provide an emission of light at a wavelength configured by the selection of UCNPs. The encapsulation material may comprise hydrogel and the cladding may comprise fluorinated ethylene propylene.