利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

10 条记录 (耗时 0.014 秒)

    VERFAHREN ZUM AUTOMATISCHEN ERSTELLEN VON ZWEI- ODER DREIDIMENSIONALEN GEBÄUDEMODELLEN
    2.
    发明公开
    VERFAHREN ZUM AUTOMATISCHEN ERSTELLEN VON ZWEI- ODER DREIDIMENSIONALEN GEBÄUDEMODELLEN 有权
    VERFAHREN ZUM AUTOMATISCHEN ERSTELLEN VON ZWEI- ODER DREIDIMENSIONALENGEBÄUDEMODELLEN

    公开(公告)号：EP2923333A1

    公开(公告)日：2015-09-30

    申请号：EP12799096.8

    申请日：2012-11-20

    申请人： Siemens Aktiengesellschaft

    发明人： BIRCHBAUER, Josef, Alois ,  KLUCKNER, Stefan ,  WINDISCH, Claudia

    IPC分类号： G06T7/00 ,  G06T17/05

    CPC分类号： G06T17/05 ,  G06K9/00637 ,  G06T7/11 ,  G06T7/194

    摘要： The invention relates to a method for the automatic creation of two- or three-dimensional building models, consisting of at least one digital aerial image, and for automatic provision with semantic building information, and comprises,
    inter alia , the following steps: generating training data on the basis of cadastral data and/or geoinformation data as a two-dimensional first vector model comprising semantic information, and thereby training a classifier; generating a second two- or three-dimensional vector model from the segmented and classified aerial image; and transferring semantic information into the second vector model from the first vector model by comparison with the latter.

    摘要翻译： 本发明涉及一种用于自动创建由至少一个数字航空图像组成的二维或三维建筑模型以及用于具有语义建筑信息的自动提供的方法，并且包括以下步骤：产生训练 基于地籍数据和/或地理信息数据的数据作为包括语义信息的二维第一矢量模型，从而训练分类器; 从分割和分类的空间图像生成第二二维或三维矢量模型; 并且通过与后者进行比较，将语义信息从第一向量模型转移到第二向量模型中。

    METHOD AND SYSTEM FOR SEMANTIC SEGMENTATION IN LAPAROSCOPIC AND ENDOSCOPIC 2D/2.5D IMAGE DATA
    9.
    发明公开
    METHOD AND SYSTEM FOR SEMANTIC SEGMENTATION IN LAPAROSCOPIC AND ENDOSCOPIC 2D/2.5D IMAGE DATA 审中-公开
    用于腹腔镜和内窥镜2D / 2.5D图像数据的语义分割的方法和系统

    公开(公告)号：EP3289562A1

    公开(公告)日：2018-03-07

    申请号：EP15722833.9

    申请日：2015-04-29

    申请人： Siemens Aktiengesellschaft

    发明人： KLUCKNER, Stefan ,  KAMEN, Ali ,  CHEN, Terrence

    IPC分类号： G06T7/00

    CPC分类号： G06T7/11 ,  G06K9/3233 ,  G06T7/13 ,  G06T7/344 ,  G06T2207/10028 ,  G06T2207/10068 ,  G06T2207/30004

    摘要： A method and system for semantic segmentation laparoscopic and endoscopic 2D/2.5D image data is disclosed. Statistical image features that integrate a 2D image channel and a 2.5D depth channel of a 2D/2.5 laparoscopic or endoscopic image are extracted for each pixel in the image. Semantic segmentation of the laparoscopic or endoscopic image is then performed using a trained classifier to classify each pixel in the image with respect to a semantic object class of a target organ based on the extracted statistical image features. Segmented image masks resulting from the semantic segmentation of multiple frames of a laparoscopic or endoscopic image sequence can be used to guide organ specific 3D stitching of the frames to generate a 3D model of the target organ.

    VERFAHREN ZUR BILDBASIERTEN VERÄNDERUNGSERKENNUNG
    10.
    发明公开
    VERFAHREN ZUR BILDBASIERTEN VERÄNDERUNGSERKENNUNG 审中-公开
    VERFAHREN ZUR BILDBASIERTENVERÄNDERUNGSERKENNUNG

    公开(公告)号：EP2859531A1

    公开(公告)日：2015-04-15

    申请号：EP12727816.6

    申请日：2012-06-06

    申请人： Siemens Aktiengesellschaft

    发明人： BIRCHBAUER, Josef Alois ,  KLUCKNER, Stefan ,  WINDISCH, Claudia

    IPC分类号： G06T7/20

    CPC分类号： G01C11/00 ,  G06K9/00208 ,  G06T7/20 ,  G06T2200/04 ,  G06T2207/10024 ,  G06T2207/10028 ,  G06T2207/10032 ,  G06T2207/30181

    摘要： The invention relates to a method for image-based alteration recognition for three-dimensional objects and/or surface sections. In this case, a three-dimensional model for an object is generated (2) from a plurality of photographs for a particular time, and a reference model or comparison model for the object is also provided. A visual focus on the three-dimensional model of the object is then defined and a 2.5D view of the object is derived (3) for this visual focus - in the form of a Cartesian data record. In this data record, at least besides two-dimensional coordinate values, a colour value derived from the three-dimensional model and a depth value derived from the three-dimensional model are also recorded (4) as data attributes for the respective data point. The respective data attributes of the respective data point are then compared (5) with the corresponding data from the reference model in order to establish discrepancies and/or changes. In this way, the method according to the invention very easily reduces complexity for recognition or ascertainment of alterations and/or differences between three-dimensional object photographs that are available in the form of what are known as point clouds. The data records, which comprise both geometric and colour information for the respective data points, allow very easy recognition of alterations in a three-dimensional object.

    摘要翻译： 本发明涉及一种用于三维物体和/或表面部分的基于图像的变更识别方法。 在这种情况下，从特定时间的多张照片中产生（2）对象的三维模型，并且还提供对象的参考模型或比较模型。 然后定义对象的三维模型的视觉焦点，并且以笛卡尔数据记录的形式导出（3）对于该视觉焦点的对象的2.5D视图。 在该数据记录中，至少除了二维坐标值之外，还记录从三维模型导出的颜色值和从三维模型导出的深度值（4）作为各个数据点的数据属性。 然后将相应数据点的相应数据属性与来自参考模型的相应数据进行比较（5），以建立差异和/或变化。 以这种方式，根据本发明的方法非常容易地降低了用于识别或确定以所谓的点云的形式可用的三维物体照片之间的变化和/或差异的复杂性。 包括各个数据点的几何和颜色信息的数据记录允许非常容易地识别三维对象中的变化。