专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2008014831A3 METHOD FOR OBSERVATION OF A PERSON IN AN INDUSTRIAL ENVIRONMENT 审中-公开
标题翻译：在工业环境中观察人的方法
公开(公告)号：WO2008014831A3
公开(公告)日：2008-04-03
申请号：PCT/EP2007003037
申请日：2007-04-04
申请人： DAIMLER AG , KRESSEL ULRICH , KRUEGER LARS , PROGSCHA WERNER , WOEHLER CHRISTIAN , KUMMERT FRANZ , SCHMIDT JOACHIM , OTT RAINER
发明人： KRESSEL ULRICH , KRUEGER LARS , PROGSCHA WERNER , WOEHLER CHRISTIAN , KUMMERT FRANZ , SCHMIDT JOACHIM , OTT RAINER
IPC分类号： B25J9/16 , F16P3/14 , G06K9/00
CPC分类号： F16P3/142 , G05B2219/40202 , G06K9/00201 , G06K9/00342
摘要： Manufacturing processes profit from close collaboration between man and machine to a major extent. While people behave flexibly and adaptably they have a tendency to make errors, however, when carrying out repetitive work processes, while machines operate quickly and exactly, but in this case are static and not very flexible. In order to allow better danger protection in the context of a safe man-machine interaction, monitored spatial areas are assigned a position-dependent and time-dependent variable hazard degree as a function of the current position and the current movement state of a machine or of a machine element, and of the person being observed. For this purpose, a multiple camera system is used to record image data about the person in order to obtain information that is as detailed as possible about the position and orientation of a person, in particular also with respect to that person's body parts in an industrial environment. This image data is then examined for the image of a person so that, once a person has been detected in the image data, this person hypothesis is matched to an articulate, virtual 3D model of the human body. This virtual body model is then continuously matched to the movement behaviour of the person detected in the image data. A hazard potential is determined from the knowledge of the position and the movement behaviour of the virtual body model in space. The hazard potential determined in this way is subjected to a threshold-value comparison process in order to intervene in the movement control for the machine or the machine parts if this threshold value is exceeded.
摘要翻译：制造流程从人与机器的密切合作中受益匪浅。尽管人们的行为灵活且适应性强，但重复性工作过程容易出错，机器工作迅速准确，但是静态且不灵活。为了能够在安全的人机交互的上下文中改进的风险的保护，监控空间区域，根据当前位置和当前运动的机器或机械元件，以及分配的地点的危险的时间相关的变量度的被观察者的状态。为了这个目的，通过一个多相机系统的装置中，检测人的图像数据，以获得详细信息地了解一个人的位置和方向，特别是在工业环境中的相对于它们的身体部位。然后检查该图像数据以对人进行成像，以便当在图像数据中检测到人时，该人假设适应于人体的铰接的虚拟3D模型。随后，该虚拟身体模型不断适应于在图像数据中检测到的人的运动行为。了解虚拟人体模型在空间中的位置和运动行为，确定潜在危险。以这种方式确定的潜在危险进行阈值比较，以便当超过该阈值时影响机器或机器部件的运动控制。

2. 发明申请

WO03025843B1 MODEL-BASED OBJECT CLASSIFICATION AND TARGET RECOGNITION 审中-公开
标题翻译：基于模型的目标分类和目标探测
公开(公告)号：WO03025843B1
公开(公告)日：2003-10-16
申请号：PCT/DE0203423
申请日：2002-09-16
申请人： EADS DEUTSCHLAND GMBH , BAUR HENRIK , EGGERS HELMUTH , KRUEGER LARS , OTT RAINER
发明人： BAUR HENRIK , EGGERS HELMUTH , KRUEGER LARS , OTT RAINER
IPC分类号： G06K9/32 , G06K9/68
CPC分类号： G06K9/325
摘要： The invention relates to a method for the model-based classification and/or target recognition of an object, which comprises the following steps: recording an image of an object, defining a characteristic feature represented by an object part, defining at least one condition, which is linked to said characteristic feature and which indicates the applicability thereof and carrying out the classification and/or target recognition of the objet by recording said characteristic feature, if the condition indicates the applicability of the characteristic feature.
摘要翻译：本发明涉及一种用于基于模型的分类和/或目标识别对象的方法，包括以下步骤：记录一个对象的图像; 确定一个特性是所述物体的一部分; 确定至少一个条件是vernüpft与特征和指示特征的适用性; 并进行分类和/或通过观察特性目标识别对象的，如果条件指示所述特征的适用性。

3. 发明申请

WO03025843A3 MODEL-BASED OBJECT CLASSIFICATION AND TARGET RECOGNITION 审中-公开
标题翻译：基于模型的对象分类和目标识别
公开(公告)号：WO03025843A3
公开(公告)日：2003-08-07
申请号：PCT/DE0203423
申请日：2002-09-16
申请人： EADS DEUTSCHLAND GMBH , BAUR HENRIK , EGGERS HELMUTH , KRUEGER LARS , OTT RAINER
发明人： BAUR HENRIK , EGGERS HELMUTH , KRUEGER LARS , OTT RAINER
IPC分类号： G06K9/32 , G06K9/68
CPC分类号： G06K9/325
摘要： The invention relates to a method for the model-based classification and/or target recognition of an object, which comprises the following steps: recording an image of an object, defining a characteristic feature represented by an object part, defining at least one condition, which is linked to said characteristic feature and which indicates the applicability thereof and carrying out the classification and/or target recognition of the objet by recording said characteristic feature, if the condition indicates the applicability of the characteristic feature.
摘要翻译：本发明涉及一种用于对象的基于模型的分类和/或目标识别的方法，包括以下步骤：记录对象的图像; 确定代表对象一部分的特征; 确定与该特征相关联的至少一个条件并指示该特征的适用性; 以及如果条件指示特征的适用性，则通过检测特征来执行对象的分类和/或目标识别。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式