首页> 外国专利> Method of automatic processing of a signal scans of the type comprising a succession of stages is slow variations studied and representatives of the phases a rapid variations, and a device for implementing it.

Method of automatic processing of a signal scans of the type comprising a succession of stages is slow variations studied and representatives of the phases a rapid variations, and a device for implementing it.

机译:自动处理包括连续阶段的类型的信号扫描的方法是研究缓慢的变化,并且代表阶段的是快速变化,以及用于实现该方法的装置。

摘要

Method for automatically processing a digitised signal of the type comprising a succession of slowly varying phases representing the phenomenon under study, and rapidly varying phases, consisting in rejecting the said rapid phases and in extrapolating the slowly varying phases, and in carrying out at least the following steps: - the said signal is sampled and digitised, the sampling frequency being greater than the Shannon frequency, - the digitised signal is differentiated with a digital differentiator filter, - a global function for membership of the slow phase for each of the sampled values is constructed in the course of a first sweep of the derivative of the sampled digital signal, - the global form of the signal is detected by fitting a curve to the derivative of the signal using the least squares criterion, - a fine function for membership of the slow phase for the sampled values is constructed in the course of a second sweep, - the sampled values are classified according to their membership of the slow phases or of the rapid phases, - the rapid phases are eliminated, - a slow-phase cumulated signal is constructed. The invention relates in particular to the processing of images, or the processing of nystagmic signals or automatic classification. The invention also relates to a device for implementing this method.
机译:自动处理类型的数字化信号的方法,该方法包括代表研究中的现象的一系列缓慢变化的相位和快速变化的相位,包括拒绝所述快速相位和推断缓慢变化的相位,以及至少执行步骤如下:-对所述信号进行采样和数字化,采样频率大于香农频率;-通过数字微分滤波器对数字化信号进行微分;-全局函数,用于每个采样值的慢相成员在采样的数字信号的导数的第一波扫描过程中构造,-通过使用最小二乘准则将曲线拟合到信号的导数来检测信号的整体形式,-隶属函数在第二次扫描过程中构造了采样值的慢相,-根据对于慢相或快速相的成员,-消除快速相,-构建慢相累积信号。本发明尤其涉及图像的处理,眼震信号的处理或自动分类。本发明还涉及用于实施该方法的设备。

著录项

  • 公开/公告号FR2610123B1

    专利类型

  • 公开/公告日1989-06-09

    原文格式PDF

  • 申请/专利权人 INSTITUT NAL SANTE RECHERC MEDIC;

    申请/专利号FR19870001001

  • 发明设计人 MOHAMMAD ARZI;

    申请日1987-01-28

  • 分类号G06F15/42;G06F15/66;A61B3/00;A61B5/00;

  • 国家 FR

  • 入库时间 2022-08-22 06:30:14

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号