-
公开(公告)号：KR1020120008211A
公开(公告)日：2012-01-30
申请号：KR1020100068944
申请日：2010-07-16
Applicant: 성균관대학교산학협력단
Abstract: PURPOSE: An object recognition and posture assumption method of a robot is provided to efficiently discover the postures possible to occur around three dimensional lines by using three dimensional lines based on direction and relationship. CONSTITUTION: The object recognition and posture assuming method of robot is as follows. A prior first interpretation is propagated to a state space by using information about robot motions(S210). A probability distribution function of each posture candidate of an object is obtained based on the first evidence selected among obtained evidences(S220). The probability value [probability value] toward each posture candidate of the object material calculates based on 'the second evidence among the obtained evidence' (S230). The probability value for each posture candidates of the object is computed based on the second evidence among the obtained evidences(S230). The second interpretation is created based on probability distribution function and probability value (S240). Current first interpretation is created based on the propagated first and second interpretations(S250) and returns to the very stage for propagating to sate space of the first interpretation.
Abstract translation: 目的：提供机器人的物体识别和姿态假设方法，通过使用基于方向和关系的三维线，有效发现三维线周围可能出现的姿势。 构成：机器人的物体识别和姿态假设方法如下。 先前的解释通过使用关于机器人运动的信息传播到状态空间（S210）。 基于从获得的证据中选择的第一证据，获得对象的每个姿势候选者的概率分布函数（S220）。 基于“获得的证据中的第二个证据”，对象物的每个姿势候选者的概率值[概率值]计算（S230）。 根据获得的证据中的第二个证据来计算对象的每个姿势候选者的概率值（S230）。 第二个解释是基于概率分布函数和概率值创建的（S240）。 基于传播的第一和第二解释创建当前的第一解释（S250），并且返回到传播到第一解释的状态空间的阶段。
-
92.发明公开구조 광 패턴을 이용한 영상의 경계를 결정하는 방법, 이러한 방법이 기록된 기록매체 및 구조 광 패턴을 이용한 영상의 경계 인식 시스템 有权使用结构化光线决定图像边界的方法，使用结构光照的图像边界识别系统的记录介质
公开(公告)号：KR1020110124902A
公开(公告)日：2011-11-18
申请号：KR1020100044352
申请日：2010-05-12
Applicant: 성균관대학교산학협력단
IPC: G06T7/00
CPC classification number: G06T7/13 , G06T7/514 , G06T11/003
Abstract: PURPOSE: A method for determining the boundary of an image using a structure optical pattern is provided to improve the accuracy of the boundary detection. CONSTITUTION: An image processing unit detects the boundary by using crossing of two signals in a graph if the reflectivity of the surface of object is constant(220,230). If the reflectivity is not constant, the image processing unit calculates a reflection index(240). The image processing unit calculates an incoming light(250). The image processing unit finds a cross point between the graph of an incoming light intensity corresponding to a captured signal pattern image and the graph of the incoming light intensity corresponding to the captured reverse signal pattern image. The image processing unit detects the boundary by the cross point(260).
Abstract translation: 目的：提供一种使用结构光学图案确定图像边界的方法，以提高边界检测的准确性。 构成：如果物体表面的反射率恒定，则图像处理单元通过使用图中的两个信号的交叉来检测边界（220,230）。 如果反射率不恒定，则图像处理单元计算反射指数（240）。 图像处理单元计算入射光（250）。 图像处理单元在与捕获的信号图案图像相对应的入射光强度的图表与对应于所捕获的反向信号图案图像的入射光强度的曲线图之间找到交叉点。 图像处理单元通过交叉点检测边界（260）。
-
公开(公告)号：KR101067839B1
公开(公告)日：2011-09-27
申请号：KR1020090003086
申请日：2009-01-14
Applicant: 성균관대학교산학협력단
Abstract: 정전기장 유도 방식을 이용하는 멀티 노즐 잉크젯 프린트 헤드가 개시된다. 본 발명의 일 실시예에 따른 잉크젯 프린트 헤드는, 상기 노즐 마다에 형성되는 제1 전극; 및 상기 노즐과 이격된 측면에 위치하여 이웃한 어느 2개의 노즐 사이 마다를 가로막는 형태로 배치되며, 상기 제1 전극과의 사이에서 전위차를 형성함으로써 상기 노즐로부터 잉크를 토출시키기 위한 정전기장을 유도하는 제2 전극을 포함할 수 있다. 본 발명의 실시예에 의하면, 인접 노즐 간 전계 간섭을 차폐함으로써 크로스토크 현상을 방지할 수 있는 효과가 있다.
잉크젯 프린트 헤드, 정전기장 유도 방식, 멀티 노즐, 크로스토크.-
公开(公告)号：KR1020110077937A
公开(公告)日：2011-07-07
申请号：KR1020090134625
申请日：2009-12-30
Applicant: 성균관대학교산학협력단
CPC classification number: G06T7/246 , G06K9/4671
Abstract: PURPOSE: A pose real-time estimation method of an object is provided to estimate a pose according to an image and track the object. CONSTITUTION: An object pose real-time estimation system receives an image that an object is photographed(S210). The system extracts a corner point from the inputted image(S220). In case the corner point is newly offered, the system corrects the corner point as time required to the obtainment of the corner point(S231,S232). According to the corrected corner point, the system estimates the pose of the object and tracks the object(S234).
Abstract translation: 目的：提供对象的姿态实时估计方法，以根据图像估计姿势并跟踪对象。 构成：物体姿态实时估计系统接收被拍摄物体的图像（S210）。 系统从输入图像中提取角点（S220）。 在新提供角点的情况下，系统将角点修正为获得角点所需的时间（S231，S232）。 根据修正的角点，系统估计对象的姿态并跟踪对象（S234）。
-
公开(公告)号：KR100948954B1
公开(公告)日：2010-03-23
申请号：KR1020080008100
申请日：2008-01-25
Applicant: 성균관대학교산학협력단
CPC classification number: B41J2/16 , B41J2/1628 , B41J2/1629 , B41J2/1631
Abstract: 본 발명은 정전기력을 이용한 잉크분사장치, 그 제조방법 및 잉크 공급방법에 관한 것으로, 제1기판 상면에 구비되는 노즐과 상기 노즐 내부에 위치하는 하부전극 및, 제1기판 하면에 구비되는 잉크 유입로가 일체로 형성된 하부 전극부; 제2기판의 상면에 형성되는 상부전극 및 제2기판의 하면으로부터 상기 상부전극까지 관통되어 형성된 잉크 토출공을 구비하는 상부 전극부; 및 상기 노즐과 잉크 토출공이 수직으로 정렬되도록 상기 하부 전극부와 상부 전극부를 본딩하는 접착층을 구비하여, 상기 잉크 유입로를 통해 노즐로 공급되는 잉크에 상기 하부전극과 상부전극 사이의 전위차에 의해 발생한 정전기력이 가해질 때 노즐 끝단에 메니스커스가 형성되어, 메니스커스 끝단에서 상부 전극부의 잉크 토출공을 통해 미소 액적 크기의 잉크가 토출되도록 이루어지는 것을 특징으로 한다.
이와 같은 구성을 구비함으로써, 잉크분사장치에서의 잉크 선정이 용이하고 정전기력의 집중 효율이 높으며 제조공정을 단순화할 수 있는 효과가 있다.
정전기력, 잉크분사, 잉크젯, 프린터헤드-
Patent Agency Ranking
公开(公告)号：KR100940283B1
公开(公告)日：2010-02-05
申请号：KR1020080010298
申请日：2008-01-31
Applicant: 성균관대학교산학협력단
IPC: G06F17/30
CPC classification number: G06F17/10
Abstract: 동일해상도의 옥트리 구조에서의 주소검색방법이 개시된다. 주소 검색방법은 옥트리 구조 내의 각 셀에 옥트리 깊이 레벨에 따라 값이 증가되는 고유한 주소값을 부여하여, 옥트리 구조 내의 이웃하는 셀들간의 주소 차이값이 일정한 수열적 규칙성을 나타내도록 옥트리 셀들의 주소 인코딩을 실시하는 단계; 주소 인코딩된 각 옥트리 셀들의 주소 차이값이 갖는 수열적 규칙성을 이용하여, 선택된 옥트리 셀에 면으로 인접한 이웃셀들의 주소값을 찾아내는 단계를 포함한다. 본 발명에 따르면, 옥트리 셀에 이웃한 이웃셀들의 주소값을 효율적으로 검색할 수 있게 한다..
동일해상도, 옥트리 구조, 이웃 셀, 주소검색Abstract translation: 公开了一种在具有相同分辨率的八叉树结构中挖掘地址的方法。 搜索具有相同分辨率的八叉树结构中的相邻小区的地址值的方法可以包括通过给出根据八叉树的深度级别增加的固有地址值来执行八叉树小区的地址编码，所述固有地址值在八叉树结构中的每个小区如 八叉树结构中相邻小区的寻址差值具有顺序规则; 以及通过使用每个八叉树单元的地址编码地址差值的顺序规则来搜索与所选择的八叉树单元接触的相邻小区的地址值。 根据本发明的实施例，可以有效地搜索与八叉树单元接触的相邻小区的地址值。
-
公开(公告)号：KR1020090092486A
公开(公告)日：2009-09-01
申请号：KR1020080017750
申请日：2008-02-27
Applicant: 성균관대학교산학협력단
IPC: G06T17/00
CPC classification number: G06T17/205 , G06T3/4053 , G06T2210/56
Abstract: A self modeling method of a 3D rotating symmetrical object and an apparatus thereof are provided to grasp the whole shape of an object by estimating the geometric shape of the object from partial data. The main axis of a 3D point cloud is estimated(S10). The 3D point cloud is divided into 3D points belonging to each plane. The center points of each plane are obtained. A center point cloud is obtained(S20). An intrinsic vector having the biggest eigenvalue is selected as a new main axis(S30). When the previous main axis and new main axis exist within a threshold value, the new main axis is selected as the direction vector of the center axis of the 3D object(S42).
Abstract translation: 提供3D旋转对称物体的自建模方法及其装置，以通过从部分数据估计物体的几何形状来掌握物体的整体形状。 估计3D点云的主轴（S10）。 3D点云被分为属于每个平面的3D点。 获得每个平面的中心点。 获得中心点云（S20）。 选择具有最大特征值的固有矢量作为新的主轴（S30）。 当前一主轴和新主轴存在于阈值内时，选择新的主轴作为3D对象的中心轴的方向矢量（S42）。
-
公开(公告)号：KR1020090087667A
公开(公告)日：2009-08-18
申请号：KR1020080013057
申请日：2008-02-13
Applicant: 성균관대학교산학협력단
CPC classification number: G06T17/005
Abstract: A 3D object or environment representation method based on a multi resolution octree structure are provided to prevent waste of an unnecessary octree cell and prevent increase of octree generation time. A size of an area that one point occupies in a real space is calculated by using a camera parameter(S102). Point density inside an octree cell is calculated by using the calculated size of the area and the number of points(S103). The calculated point density is compared with a predetermined threshold value(S104). When the point density is larger than the threshold value, an average value and a variance value of coordinates of points inside a corresponding octree cell are calculated(S105). Point distribution inside the octree cell is checked by using the calculated average value and variance value(S106). Whether to disassemble the octree cell is determined according to the distribution(S108).
Abstract translation: 提供了一种基于多分辨率八叉树结构的3D对象或环境表示方法，以防止浪费不必要的八叉树单元并防止增加八叉树生成时间。 通过使用相机参数来计算一点占据实际空间的区域的大小（S102）。 通过使用计算出的面积大小和点数来计算八叉树单元内的点密度（S103）。 将计算出的点密度与预定阈值进行比较（S104）。 当点密度大于阈值时，计算相应八叉树单元内的点的坐标的平均值和方差值（S105）。 通过使用计算的平均值和方差值来检查八叉树单元内的点分布（S106）。 根据分布确定是否拆分八叉树单元（S108）。
-
99.发明公开
公开(公告)号：KR1020080089148A
公开(公告)日：2008-10-06
申请号：KR1020080005137
申请日：2008-01-17
Applicant: 성균관대학교산학협력단
Abstract: A central information processing system and a method for a service robot having layered indoor information according to a recognition and reasoning level are provided to integrally manage information about an indoor space and objects therein in order to enable plural of service robots to share the indoor information by providing necessary indoor information to the service robots and receiving modified information from the service robots. A central information processing system and a method for a service robot includes a database(20), a broker unit, a spatial information manager(33), an object information manager(34), and a space & object reasoning unit(35). The broker unit includes at least one of interfaces(31) corresponding to each of the service robots. Each of the interfaces analyzes an information request of a corresponding robot, transforms a format of the requested information to a predetermined robot-readable format, and transmits the transformed information to the robot. The interface analyzes an information request from a corresponding service robot and transforms a format of the received information request to a predetermined format that can be processed in the central information processing system. On the contrary, the interface receives a result of the information request from the central information processing system and transforms a format of the result to a predetermined format analyzable by the service robot. The spatial information manager and the object information manager hierarchically compose information about an indoor space and objects therein according to a recognition/reasoning level and stores the layered information.
Abstract translation: 提供一种用于根据识别和推理级别分层的室内信息的服务机器人的中央信息处理系统和方法，以一体地管理关于室内空间及其中的对象的信息，以使多台服务机器人能够通过 向服务机器人提供必要的室内信息，并从服务机器人接收修改的信息。 中央信息处理系统和服务机器人的方法包括数据库（20），代理单元，空间信息管理器（33），对象信息管理器（34）和空间与对象推理单元（35）。 代理单元包括对应于每个服务机器人的接口（31）中的至少一个。 每个接口分析相应机器人的信息请求，将请求的信息的格式转换成预定的机器人可读格式，并将变换后的信息发送给机器人。 接口分析来自对应的服务机器人的信息请求，并将接收到的信息请求的格式转换成可在中央信息处理系统中处理的预定格式。 相反，接口从中央信息处理系统接收到信息请求的结果，并将结果的格式转换成由服务机器人可分析的预定格式。 空间信息管理器和对象信息管理器根据识别/推理级别分层地构成关于室内空间和对象的信息，并存储分层信息。
-
公开(公告)号：KR1020080079969A
公开(公告)日：2008-09-02
申请号：KR1020070026811
申请日：2007-03-19
Applicant: 성균관대학교산학협력단
CPC classification number: G01C3/00 , G01C11/14 , G01C11/30 , G01N21/88 , G01N2021/9516
Abstract: A method and a system for structural light based depth imaging using a geometric condition are provided to correct distorted information by removing or correcting a corresponding point error between projectors and imaging devices. A method for structural light based depth imaging using a geometric condition comprises a step of detecting, removing, or correcting a corresponding point error between plural projectors and imaging devices, which is caused by reflection and blur, by using a geometric condition between the projectors and the imaging devices.
Abstract translation: 提供了使用几何条件的基于结构光的深度成像的方法和系统，以通过移除或校正投影仪和成像装置之间的对应点误差来校正失真的信息。 使用几何条件的基于结构光的深度成像的方法包括通过使用投影仪和投影仪之间的几何条件来检测，去除或校正由反射和模糊引起的多个投影仪和成像装置之间的对应点误差的步骤 成像装置。
-
-
-
-
-
-
-
-
-
Search Results
101
records
(took 0.034 seconds)
