利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

4 条记录 (耗时 0.009 秒)

    A SYSTEM AND METHOD FOR IDENTIFYING, ANALYSING AND COUNTING REPETITIONS OF PHYSICAL EXERCISES BY AN INDIVIDUAL, IN PARTICULAR FOR USE IN REHABILITATION PROGRAMS
    3.
    发明公开
    A SYSTEM AND METHOD FOR IDENTIFYING, ANALYSING AND COUNTING REPETITIONS OF PHYSICAL EXERCISES BY AN INDIVIDUAL, IN PARTICULAR FOR USE IN REHABILITATION PROGRAMS 审中-实审

    公开(公告)号：EP4407628A1

    公开(公告)日：2024-07-31

    申请号：EP23153803.4

    申请日：2023-01-27

    申请人： Henesis S.r.l.

    发明人： LAI, Stefano ,  MANTOAN, Alice ,  SERNISSI, Francesca ,  ASCARI, Luca ,  SYMONS, Frederick

    IPC分类号： G16H20/30 ,  G16H40/67 ,  G16H50/20

    CPC分类号： G16H20/30 ,  G16H40/67 ,  G16H50/20

    摘要： A system (1) for identifying, analysing, and counting repetitions of physical exercises by an individual, the system (1) comprising:
    - at least one inertial sensor (2) that is wearable by the individual and is configured to sample corresponding signals representative of biometric and/or medical parameters of the individual so as to obtain sampled data;
    - a backbone neural network (6) configured to extract features arranged in a spatial vector, said backbone neural network (6) having been pre-trained by a Simple Framework for Contrastive Learning of Visual Representation, SimCLR, technique;
    - a first neural network (7) configured to detect exercise events in response to the receipt of the extracted features coming from the backbone neural network (6);
    - a second neural network (8) configured to identify the exercise event out of a pool of K different exercises the individual may be assigned;
    - a third neural network (9) configured to count repetitions of said identified exercise;
    - a version of the backbone neural network (6) that has been adapted to the tasks of binary exercise classification, multiclass exercise classification and regression of the phase indicator function, by fine-tuning the backbone neural network (6) combined with the three neural networks (7), (8) and (9) on labelled data to achieve the aforementioned tasks.

    PANTOGRAPH MONITORING SYSTEM AND METHOD
    4.
    发明公开
    PANTOGRAPH MONITORING SYSTEM AND METHOD 审中-公开
    系统和方法用于监控的受电弓

    公开(公告)号：EP2619727A1

    公开(公告)日：2013-07-31

    申请号：EP11761570.8

    申请日：2011-09-21

    申请人： Henesis S.r.l.

    发明人： SACCHI, Matteo ,  PIAZZI, Aurelio ,  CAGNONI, Stefano ,  ASCARI, Luca

    IPC分类号： G06T7/00

    CPC分类号： G06T7/00 ,  G06T7/0004 ,  G06T2207/30136

    摘要： A method for automatic diagnostics of images related to pantographs, comprising the steps of: capturing an image that shows a pantograph of a locomotive, the image being taken from an aerial view during the travel of the locomotive, the image comprising the gliding area of a plurality of slippers of the pantograph; identifying, by means of a module for classifying the pantograph model, the model of the pantograph within a plurality of pantograph models, on the basis of the image captured; determining, by means of a module for classifying materials, a material of which the slippers are composed among a plurality of materials, on the basis of the pantograph model identified; and determining a value related to the state of wear for each one of the plurality of slippers, on the basis of the type of material determined.