Abstract: Embodiments of the present disclosure provide a datum selection method for minimizing hole position errors in group hole machining of large components, comprising: 1) determining a type of a computer numerical control (CNC) machine tool and establishing a topological structure of the CNC machine tool; 2) establishing a theoretical postural model of a tool center point during a motion; 3) establishing a hole position error model; 4) establishing an average error model of hole positions in group hole machining; and 5) obtaining a machining datum for group holes of corresponding components. For the skeleton and skinned group hole machining of aircraft components, different principles of datum selection are provided respectively, which can effectively improve the positional accuracy of the skeleton or skinned group hole machining, at the same time provides a more scientific and reasonable approach for the datum selection in group hole machining of large components.

Abstract: According to a method for pushing a notification provided in some embodiments of the present disclosure, a to-be-notified event is obtained; a first notification condition and a second notification condition are determined; a first notification page is displayed on the display when the first notification condition is met, to display, to a user, first association information when the first notification condition is met, for example, event status information, to-be-executed behavior information, and operation association information; and a second notification page is displayed on the display when the second notification condition is met after the first notification page is displayed, to display, to the user, second association information when the second notification condition is met. A plurality of notification pages are pushed for a to-be-notified event, and each notification page is pushed when a notification condition is met as the event develops.

Abstract: A method for controlling an application instance to be online or offline includes: acquiring target indication information, the target indication information including an identification of a target application instance; according to the target indication information and stored target instances, determining an online or offline control instruction of the target application instance, wherein the target instances including application instances that are started and are in an offline state; and according to the online or offline control instruction of the target application instance, controlling the target application instance to be online or offline.

Abstract: A few-shot defect detection method based on metric learning, including: (S1) performing data enhancement on a to-be-detected few-shot defect data set through a G2-Generative adversarial network (G2-GAN); (S2) extracting features of a defect data set similar to the to-be-detected few-shot defect data set based on an adaptive convolution kernel-based convolutional neural network (SKM-CNN) to generate a pre-training model; and (S3) transferring the pre-training model to a few-shot defect detection network (S2D2N) based on metric learning; and performing target feature extraction and metric learning in sequence to realize rapid identification and location of defects.

Abstract: An object of the present invention is to provide a shock absorber that can use a single bracket regardless of difference in thickness of a knuckle bracket. A shock absorber according to the present invention includes: a shock absorber main body having an outer shell and a rod movably inserted into the outer shell; a knuckle bracket attached to an outer circumference on a bottom end side of the outer shell; and a bracket including a curved piece curved along a circumferential direction of an outer circumference of the outer shell and an attachment piece extending from the curved piece and to which a mounting component is attached, the curved piece including a bracket attached to the outer shell by being welded to the outer circumference of the outer shell and within a range from a top end to a bottom end of the knuckle bracket in an axial direction of the outer shell.

Abstract: A vernier permanent magnet motor with a stator having coded auxiliary teeth which belongs to the technical field of permanent magnet motors. The motor includes a stator and a rotor that are coaxially sleeved, and an air gap is formed between the stator and the rotor. Coded auxiliary tooth structures are provided on a side surface in the stator close to the air gap, and first permanent magnets are provided on part of the coded auxiliary tooth structures. A plurality of pairs of second permanent magnets and ferromagnets are arranged around a side surface in the rotor close to the air gap. The first permanent magnets and the ferromagnets generate a first magnetic field, the second permanent magnets and the coded auxiliary tooth structures generate a second magnetic field, and the magnetic fields are superimposed to form working magnetic fields.

Abstract: A subtle defect detection method based on coarse-to-fine strategy, including: (S1) acquiring data of an image to be detected via a charge-coupled device (CCD) camera; (S2) constructing a defect area location network and preprocessing the image to be detected to initially determine a defect position; (S3) constructing a defect point detection network; and training the defect point detection network by using a defect segmentation loss function; and (S4) subjecting subtle defects in the image to be detected to quantitative extraction and segmentation via the defect point detection network.

Abstract: The present disclosure disclose a method for feature detection of complex defects based on multimodal data, including feature extraction of multimodal data, multimodal feature cross-guided learning, multimodal feature fusion, and defect classification and regression. Feature extraction networks for multimodal two-dimensional data are constructed first, and a defect data set is sent to the networks for training; during training, cross-guided learning is implemented by using a multimodal feature cross-guidance network; then feature fusion is performed by using a weight adaptive method; and finally a defect detection task is implemented by using a classification subnetwork and a regression subnetwork.

Abstract: Methods, devices, and computer-readable media for multi-task based lifelong learning. A method for lifelong learning includes identifying a new task for a machine learning model to perform. The machine learning model trained to perform an existing task. The method includes adaptively training a network architecture of the machine learning model to generate an adapted machine learning model based on incorporating inherent correlations between the new task and the existing task. The method further includes using the adapted machine learning model to perform both the existing task and the new task.

Abstract: A cross-scale defect detection method based on deep learning, including: (S1) building a vision data acquisition system to acquire a surface image of a part to be processed; and building a defect dataset; (S2) building a deep learning-based cross-scale defect detection model; and inputting the defect dataset obtained in the step (S1) into the deep learning-based cross-scale defect detection model for model training; and (S3) building a defect detection system according to the deep learning-based cross-scale defect detection model and the vision data acquisition system; and detecting a defect of the surface image of the part to be processed.

Abstract: The present disclosure provides a method for analyzing a minor defect based on a progressive segmentation network, including: acquiring an original image for a surface of a component, and cropping the original image into a plurality of patches; inputting each of the patches to a minor defect feature extraction network to extract an image feature; classifying the patch into a defective image or a non-defective background image according to an extracted image feature; inputting an extracted image feature of the defective image to a defect segmentation network to obtain a segmentation mask image of a corresponding defect; and quantitatively analyzing the defect according to the segmentation mask image to obtain information such as an area, a length and a width of the defect.

Abstract: Systems and methods for presenting and sorting summaries of alerts triggered by search queries in data aggregation and analysis systems. An example method may comprise: causing, by one or more processing devices, one or more alert summaries to be displayed, each alert summary corresponding to an alert and representing one or more instances of the alert, the alert defined by a search query and a triggering condition; wherein an instance of the alert corresponds to a particular dataset that (i) is generated by executing the search query over time-series data falling within a particular time range in a set of time ranges over which the search query has been instructed to search, and (ii) satisfies the triggering condition for the alert; wherein an alert summary includes an indication of at least one of: a total count of alert instances generated by the alert, or a count of alert instances generated by the alert that have not been viewed by a user.

Abstract: A few-shot defect detection method based on metric learning, including: (S1) performing data enhancement on a to-be-detected few-shot defect data set through a G2-Generative adversarial network (G2-GAN); (S2) extracting features of a defect data set similar to the to-be-detected few-shot defect data set based on an adaptive convolution kernel-based convolutional neural network (SKM-CNN) to generate a pre-training model; and (S3) transferring the pre-training model to a few-shot defect detection network (S2D2N) based on metric learning; and performing target feature extraction and metric learning in sequence to realize rapid identification and location of defects.

Abstract: A subtle defect detection method based on coarse-to-fine strategy, including: (S1) acquiring data of an image to be detected via a charge-coupled device (CCD) camera; (S2) constructing a defect area location network and preprocessing the image to be detected to initially determine a defect position; (S3) constructing a defect point detection network; and training the defect point detection network by using a defect segmentation loss function; and (S4) subjecting subtle defects in the image to be detected to quantitative extraction and segmentation via the defect point detection network.

Abstract: The present disclosure provides a point cloud denoising method based on deep learning for an aircraft part, in which different degrees of Gaussian noise are added based on a theoretical data model of the aircraft part, a heightmap for each point in the theoretical data model is generated, and a deep learning training set is constructed. A deep learning network is trained based on the constructed deep learning training set, to obtain a deep learning network model. A real aircraft part is scanned via a laser scanner to obtain measured point cloud data. The normal information of the measured point cloud is predicted based on the trained deep learning network model. Based on the predicted normal information, a position of each point in the measured point cloud data is further updated, thereby completing denoising of the measured point cloud data.

Abstract: An online geometric/thermal error measurement and compensation system for computer numerically controlled (CNC) machine tools belonging to the technical field of error testing and compensation of CNC machine tools. The online CNC machine tool geometric/thermal error measurement and compensation system includes two parts: the hardware platform and the measurement and compensation software. The hardware platform includes a unidirectional acceleration sensor, a precision integrated circuit (IC) temperature sensor, a multi-channel temperature data collector, and a geometric/thermal error measurement and compensation host. The error measurement and compensation software runs in the geometric/thermal error measurement and compensation host and realizes testing and compensation of geometric and thermal errors in machine tools, which are communicated to the FANUC CNC system.

Abstract: An ophthalmological imaging device includes illumination optics having an illumination light source that outputs an illumination light, a scanner that redirects the illumination light toward a portion of an object to be imaged, and an optical image capture device including a camera that receives backscattered light that is scattered from the illumination light by the object and captures first and second images of the backscattered light. The device also includes a control processor that controls the scanner and the optical image capture device to cause the optical image capture device to capture the first and second images of the object. The first and second images are captured by the camera at different times and extracted from different portions of the backscattered light. The device also includes an image processor that generates a stereoscopic image from the first and second images of the object.

Abstract: An APP remote control method and related devices. wherein method includes: a service device receiving a first operation instruction, where the first operation instruction is generated by a terminal device based on a first input on a start of an application icon of a first application program, and/or on a user interface of the first application program, the first application program runs on the service device, and the application icon and the user interface of the first application program are displayed on the terminal device; the service device responding to the first operation instruction through the first application program, and generating a first response result; and the service device extracting a first interface drawing instruction from the first response result, and sending it to the terminal device, where the first interface drawing instruction is used to draw an interface.

Abstract: A method includes training, using at least one processor, a specialized teacher model to perform visual object instance segmentation in order to segment and classify objects in first training images. The first training images contain foreground objects without backgrounds. The method also includes training, using the at least one processor, a student model to perform visual object instance segmentation in order to segment and classify objects in second training images. The second training images contain the foreground objects and the backgrounds. Training the student model includes using selected outputs of the specialized teacher model. The method further includes deploying the trained student model to perform visual object instance segmentation in an external device.

Abstract: A method for measuring a distance between trace lines of a 3D braided material, including: (S1) establishing a vision data acquisition system using a vision sensor; (S2) acquiring, by the vision data acquisition system, a training data of the trace lines of the 3D braided material; (S3) constructing a deep learning model for recognizing the trace lines of the 3D braided material; and inputting the training data acquired in step (S2) to the deep learning model to obtain a trained deep learning model; and (S4) positioning a location of the trace lines of the 3D braided material in batch images according to the trained deep learning model obtained in step (S3); and measuring a distance between adjacent trace lines.