Anomaly detection method and apparatus for multi-type data

    公开(公告)号:US11423260B1

    公开(公告)日:2022-08-23

    申请号:US17589888

    申请日:2022-01-31

    Abstract: The present disclosure provides an anomaly detection method and apparatus for multi-type data. According to the anomaly detection method for multi-type data, an adversarial learning network is trained, so that a generator in the adversarial learning network fits a distribution of a normal training sample and learns a potential mode of the normal training sample, to obtain an updated adversarial learning network, an anomaly evaluation function in the updated adversarial learning network is constructed according to a reconstruction error generated during training, and the updated adversarial learning network is constructed into an anomaly detection model, to perform anomaly detection on inputted detection data by the anomaly detection model, to obtain an anomaly detection result. A mode classifier is introduced to effectively resolve difficult anomaly detection when a distribution of detected data is similar to that of normal data, further improving the accuracy of anomaly detection.

    Dark-field confocal microscopy measurement apparatus and method based on time-varying fractional-order vortex demodulation

    公开(公告)号:US12204081B1

    公开(公告)日:2025-01-21

    申请号:US18812680

    申请日:2024-08-22

    Abstract: This application relates to the technical field of confocal microscopy measurement and provides a dark-field confocal microscopy measurement apparatus and method based on time-varying fractional-order vortex demodulation. The apparatus includes a time-varying modulated illumination module, an optical scanning module, a signal collection and demodulation module, a function generator, and a sample platform. The function generator is separately connected with. The time-varying modulated illumination module is configured to emit fractional-order vortex light to the optical scanning module. The optical scanning module is configured to transmit the fractional-order vortex light to the to-be-measured sample on the sample platform and transmit a reflected light signal to the signal collection and demodulation module. The signal collection and demodulation module is configured to collect the reflected light signal, and perform dark-field confocal detection on the reflected light signal based on a reference signal, to obtain measurement information of the to-be-measured sample.

    METHOD AND SYSTEM FOR RANDOMLY GENERATING POROUS MEDIUM MODEL

    公开(公告)号:US20240403518A1

    公开(公告)日:2024-12-05

    申请号:US18642930

    申请日:2024-04-23

    Abstract: Provided are a method and a system for randomly generating a porous medium model. The method includes following steps: setting a porosity, resolution and a size of a pre-generated porous medium model; initializing the porous medium model and generating position information of particles; extracting particle profile edges; obtaining filled particles; carrying out a collision detection on the filled particles and preset particles, and determining effectiveness of a particle generation position; presetting a cyclic pop-up condition, and if a judgment result meets the cyclic pop-up condition, continuing; otherwise, updating Fourier parameters; adding a particle configuration meeting the cyclic pop-up condition to a model generation area, and storing parameters; determining whether the generated model meets preset generation requirements, and if so, outputting a porous medium model.

Patent Agency Ranking