Spectral imaging method and system

    公开(公告)号:US10101206B2

    公开(公告)日:2018-10-16

    申请号:US15524827

    申请日:2015-11-05

    摘要: An imaging system and method are presents for use in reconstructing spectral data of an object. The imaging system comprises: an optical unit; a pixel array of a detector; and a data processor for receiving and processing image data indicative of light detected by the pixel array and generating reconstructed spectral data of the object being imaged. The optical unit is configured and operable for applying a predetermined coding to an input light field while creating an optical image thereof on a detection plane defined by the pixel array. Therefore, the image data is a function of the predetermined coding and a spectrum of the object to be determined.

    HIGH FREQUENCY SENSITIVE NEURAL NETWORK
    7.
    发明公开

    公开(公告)号:US20240281642A1

    公开(公告)日:2024-08-22

    申请号:US18563388

    申请日:2022-05-26

    IPC分类号: G06N3/0464

    CPC分类号: G06N3/0464

    摘要: A computer-implemented method of extracting high-frequency features from data, including receiving a first dataset; in a training phase, applying frequency-based guidance to learnable high-frequency filters in a neural network, the frequency based-guidance including promoting high eigenvalues associated with eigenvectors comprising the learnable high-frequency filters; extracting from the first dataset high-frequency features associated with the eigenvectors; and using the trained high-frequency filters to extract high-frequency features from a second dataset.