Using geometric features and history information to detect features such as car exhaust in point maps
    1.
    发明授权
    Using geometric features and history information to detect features such as car exhaust in point maps 有权
    使用几何特征和历史信息来检测点图中的汽车排气等功能

    公开(公告)号:US08818609B1

    公开(公告)日:2014-08-26

    申请号:US13678240

    申请日:2012-11-15

    Applicant: Google Inc.

    Abstract: Disclosed herein are systems and methods for classifying regions of a scanning zone of a light detection and ranging (LIDAR) scan. One example includes a method for receiving information indicative of a light detection and ranging (LIDAR) scan of a vehicle. The method further involves generating a point map of the scanning zone. The method further involves generating a density profile of the point map that is indicative of density of one or more regions of the scanning zone characterized at one or more distances. The method further involves generating an elevation profile of the point map that is indicative of elevation of one or more reflected features in the scanning zone characterized at one or more distances. The method further involves classifying regions of the scanning zone based on the density profile and the elevation profile.

    Abstract translation: 本文公开了用于对光检测和测距(LIDAR)扫描的扫描区域的区域进行分类的系统和方法。 一个示例包括用于接收指示车辆的光检测和测距(LIDAR)扫描的信息的方法。 该方法还涉及生成扫描区域的点图。 该方法还包括产生指示在一个或多个距离处表征的扫描区域的一个或多个区域的密度的点图的密度分布。 该方法还包括产生指示在一个或多个距离上表征的扫描区域中的一个或多个反射特征的高程的点图的仰角分布。 该方法还包括基于密度分布和仰角分布对扫描区域的区域进行分类。

Patent Agency Ranking