公开(公告)号：KR1020110070516A

公开(公告)日：2011-06-24

申请号：KR1020090127371

申请日：2009-12-18

Applicant: 한국전자통신연구원

Inventor： 장민수 ,  김재홍 ,  손주찬 ,  이재연 ,  박찬규 ,  박천수 ,  이대하 ,  조현성 ,  정연구 ,  강상승

IPC: G06K9/46 ,  G06K9/78 ,  G06T7/00

CPC classification number: G06K9/46 ,  G06K9/78 ,  G06T7/00

Abstract: PURPOSE: Motion based identification system and method are provided to perform personal identification resistant to an environmental noise using the performance characteristic of users without the request for the coordination of users. CONSTITUTION: An operation signal acquisition device(200) extracts the operating signal from the operation of user. A personal identification apparatus creates the personal identification information by processing the operating signal. A feature information extraction unit(110) extracts the feature information from the operating signals. A motion feature information storage unit(150) stores the motion feature information of users. A personal identification information generator(130) creates the personal identification information of users by comparing the feature information with the motion feature information.

Abstract translation: 目的：提供基于运动的识别系统和方法，使用用户的性能特征来执行抵抗环境噪声的个人识别，而无需用户协调。 构成：操作信号获取装置（200）从用户的操作中提取操作信号。 个人识别装置通过处理操作信号来创建个人识别信息。 特征信息提取单元（110）从操作信号中提取特征信息。 运动特征信息存储单元（150）存储用户的运动特征信息。 个人识别信息生成器（130）通过将特征信息与运动特征信息进行比较来创建用户的个人识别信息。

公开(公告)号：KR1020110070506A

公开(公告)日：2011-06-24

申请号：KR1020090127349

申请日：2009-12-18

Applicant: 한국전자통신연구원

Inventor： 장민수 ,  김재홍 ,  강상승 ,  손주찬 ,  박찬규 ,  박천수 ,  조현성 ,  이대하 ,  이재연 ,  정연구

IPC: G06F3/0346 ,  G06F3/03

CPC classification number: G06F3/0346 ,  G06F3/0304

Abstract: PURPOSE: Integrated infrared-acceleration pointing method and device are provided to obtain a precise pointing function in a noise environment by realizing a pointing function in a light source tracking system grasping the location of the light source. CONSTITUTION: An acceleration sensor(114) senses the acceleration for the movement of a light source(110). An optical sensor(120) senses the light radiated from the light source. A signal processing host(130) distinguishes the light source on a basis of the optical point position information of the light source received from the light sensor and the operation information on the light source received from the acceleration sensor.

Abstract translation: 目的：提供集成的红外加速指示方法和装置，通过在获取光源位置的光源跟踪系统中实现指向功能，在噪声环境中获得精确的指向功能。 构成：加速度传感器（114）感测光源（110）的运动的加速度。 光学传感器（120）感测从光源照射的光。 信号处理主机130基于从光传感器接收的光源的光点位置信息和从加速度传感器接收到的光源的操作信息来区分光源。

公开(公告)号：KR1020100065074A

公开(公告)日：2010-06-15

申请号：KR1020090096410

申请日：2009-10-09

Applicant: 한국전자통신연구원

Inventor： 강상승 ,  김재홍 ,  손주찬 ,  정연구

IPC: B25J13/08 ,  B25J21/02

Abstract: PURPOSE: A robot control system based on motion recognition and a method thereof are provided to ensure natural interaction between a user and a robot through control command input based on the hand motion of the user. CONSTITUTION: A robot control system based on motion recognition comprises a glove-type input device(100) and a motion recognition and control unit(110). The glove-type input device comprises a hand motion detection sensor(101) and a communication module(102). The motion recognition and control unit recognizes the motion information received from the glove-type input device and transmits a control command corresponding to the motion information to a controller(121) of a robot. The motion recognition and control unit comprises a motion analysis unit(111) and a control command unit(112). The motion analysis unit recognizes a motion pattern by analyzing the motion information and creates motion information data. The control command unit compares the motion information data with a stored control command and transmits a control command.

Abstract translation: 目的：提供基于运动识别的机器人控制系统及其方法，以通过基于用户的手动的控制命令输入来确保用户与机器人之间的自然交互。 构成：基于运动识别的机器人控制系统包括手套式输入装置（100）和运动识别和控制单元（110）。 手套型输入装置包括手动检测传感器（101）和通信模块（102）。 运动识别和控制单元识别从手套式输入装置接收的运动信息，并将与运动信息相对应的控制命令发送到机器人的控制器（121）。 运动识别和控制单元包括运动分析单元（111）和控制命令单元（112）。 运动分析单元通过分析运动信息来识别运动模式并创建运动信息数据。 控制命令单元将运动信息数据与存储的控制命令进行比较，并发送控制命令。

公开(公告)号：KR1020100065070A

公开(公告)日：2010-06-15

申请号：KR1020090094577

申请日：2009-10-06

Applicant: 한국전자통신연구원

Inventor： 장민수 ,  김재홍 ,  손주찬

IPC: G06F3/03 ,  G06F3/00

Abstract: PURPOSE: A motion recognition device and an operation recognition method using the same are provided to easily separate a motion trajectory, thereby removing a problem of inaccuracy of recognizing a user operation. CONSTITUTION: An input device(110) includes a light source(111) and an inertia sensor(113). A location change of the light source is detected during a user motion section determined from a sensing signal of the inertia sensor, a motion recognizing unit(120) extracts a user motion trajectory. A motion determining unit(121) calculates an energy variation amount from the sensing signal. The motion determining unit determines the user motion section according to an operation state of the user by comparing the calculated energy variation amount with at least one reference level.

Abstract translation: 目的：提供运动识别装置和使用其的操作识别方法以容易地分离运动轨迹，从而消除识别用户操作的不精确性的问题。 构成：输入装置（110）包括光源（111）和惯性传感器（113）。 在根据惯性传感器的感测信号确定的用户运动部分期间检测光源的位置变化，运动识别单元（120）提取用户运动轨迹。 运动确定单元（121）根据感测信号计算能量变化量。 运动确定单元通过将计算的能量变化量与至少一个参考水平进行比较，根据用户的操作状态来确定用户运动区间。

公开(公告)号：KR1020090065884A

公开(公告)日：2009-06-23

申请号：KR1020070133421

申请日：2007-12-18

Applicant: 한국전자통신연구원

Inventor： 김재홍 ,  강상승 ,  손주찬 ,  조현규 ,  조영조

IPC: G06F3/01 ,  G06F3/0346 ,  B25J13/00

CPC classification number: G06F3/011 ,  G06F3/0414 ,  B25J13/00 ,  G06F3/017

Abstract: A touching behavior recognition system and a method thereof are provided to recognize various touch behaviors through a sensor board and a signal processing board without attaching a sensor to a robot at an initial time. A touch behavior recognition system comprises a sensor board(100), a signal processing board(200) and a robot(300). The sensor board detects a sensor value corresponding to a touch behavior through an inertia sensor attached to a body portion. The signal processing board recognizes the touch behavior from the sensor value detected by the sensor board. The signal processing board transfers the touch behavior recognition result to an object.

Abstract translation: 提供了一种触摸行为识别系统及其方法，用于通过传感器板和信号处理板来识别在初始时间不将传感器附着到机器人的各种触摸行为。 触摸行为识别系统包括传感器板（100），信号处理板（200）和机器人（300）。 传感器板通过附接到主体部分的惯性传感器来检测与触摸行为相对应的传感器值。 信号处理板从传感器板检测到的传感器值识别触摸行为。 信号处理板将触摸行为识别结果传送到对象。

公开(公告)号：KR100893758B1

公开(公告)日：2009-04-20

申请号：KR1020070104133

申请日：2007-10-16

Applicant: 한국전자통신연구원

Inventor： 강상승 ,  김재홍 ,  손주찬 ,  조현규 ,  조영조

IPC: B25J9/16 ,  G06F19/00

CPC classification number: G06N3/008

Abstract: An emotion expressing system of robots and a method thereof are provided to maximize interaction between a robot and a human by providing natural emotion behavior description according to emotion change. An emotion expressing system of robots comprises: an emotion engine(200) producing the emotion information of a robot according to the information inputted from a plurality of sensors; an emotion expression and driving controller(300) controlling an object driving unit of the robot by determining the behavior description of the robot according to the emotion information; a hardware controller(400) controlling movement of the object driving unit according to the control method of the emotion expression and driving controller; and an exterior management operating part(100) including a tool editing the motion and action and a tool setting up environment or managing resources.

Abstract translation: 提供一种机器人表情系统及其方法，以通过根据情绪变化提供自然情绪行为描述来最大化机器人与人之间的相互作用。 机器人情感表达系统包括：根据从多个传感器输入的信息产生机器人的情感信息的情感引擎（200）; 通过根据情绪信息确定机器人的行为描述来控制机器人的对象驱动单元的情绪表达和驱动控制器（300）; 硬件控制器（400），其根据情绪表达式的控制方法和驱动控制器来控制对象驱动单元的移动; 以及包括编辑运动和动作的工具以及工具建立环境或管理资源的外部管理操作部分（100）。

公开(公告)号：KR1020090024898A

公开(公告)日：2009-03-10

申请号：KR1020070089841

申请日：2007-09-05

Applicant: 한국전자통신연구원

Inventor： 박찬규 ,  손주찬 ,  조현규 ,  조영조

IPC: G06T7/00 ,  H04N19/20

CPC classification number: G06T7/174 ,  G06T7/12 ,  G06T7/194 ,  G06T2207/10016

Abstract: An image object extracting device and a method thereof are provided to separate and extract a frontal-viewing object image from an inputting image by using multi-border line information. A border information detector(108a) performs a preprocessing operation that obtains border information and border line information about a reference background object image and an inputting image, and a background separator(108b) separates a background object image from the inputting image by using a border line difference image based on multi-border line information. A post-processor(108c) removes the background object image and an image noise from the inputting image through threshold value conversion and scale conversion to extract a frontal object image from the inputting image.

Abstract translation: 提供一种图像对象提取装置及其方法，用于通过使用多边界线信息从输入图像分离和提取正面观看对象图像。 边界信息检测器（108a）执行获取关于参考背景物体图像和输入图像的边界信息和边界线信息的预处理操作，并且背景分离器（108b）通过使用边界从背景对象图像与输入图像分离 基于多边界线信息的线差图像。 后处理器（108c）通过阈值转换和缩放转换从输入图像中去除背景对象图像和图像噪声，以从输入图像提取正面对象图像。

公开(公告)号：KR1020080050976A

公开(公告)日：2008-06-10

申请号：KR1020070077985

申请日：2007-08-03

Applicant: 한국전자통신연구원

Inventor： 고영철 ,  손주찬 ,  조영조

IPC: G06F15/00 ,  G06F9/44 ,  G06F17/00

CPC classification number: G06F17/30687 ,  G06F8/35 ,  G06F8/38 ,  G06F9/445 ,  G06F17/2765 ,  G06F17/30734

Abstract: A method for modeling a user intention based on human desire and a system thereof are provided to offer efficient, active and intelligent services in a ubiquitous environment by supporting provision of services where various user's intentions are reflected. A system for modeling a user intention based on human desire includes plural user intention model servers(410), plural ontology servers(420), a service providing device(430), a status information collector(440), and a remote machine(460). Each of user intention model servers is composed of a Bayesian network, includes a user intention model(412) for expressing a user's intention and a Bayesian network inference engine software(414) for analyzing the user intention model. Each of ontology servers includes an ontology(422) for describing the user's intention model, and an ontology inference engine software for inferring the user's intention by using the ontology and external input information. The user's intention can be used via the service providing device like a robot(432) or an agent(434). The remote machine is used for additionally extending the user's intention at a remote place.

Abstract translation: 提供了一种基于人类愿望建模用户意图的方法及其系统，以通过支持提供反映各种用户意图的服务来提供无所不在的环境中的有效，主动和智能的服务。 一种用于基于人类欲望建模用户意图的系统包括多个用户意图模型服务器（410），多个本体服务器（420），服务提供设备（430），状态信息收集器（440）和远程机器 ）。 每个用户意图模型服务器由贝叶斯网络组成，包括用于表达用户意图的用户意图模型（412）和用于分析用户意图模型的贝叶斯网络推理引擎软件（414）。 每个本体服务器包括用于描述用户意图模型的本体（422）和用于通过使用本体和外部输入信息来推断用户意图的本体推理引擎软件。 用户的意图可以通过诸如机器人（432）或代理（434）的服务提供设备来使用。 远程机器用于在远程的地方额外扩展用户的意图。

公开(公告)号：KR1020070122043A

公开(公告)日：2007-12-28

申请号：KR1020060057004

申请日：2006-06-23

Applicant: 한국전자통신연구원

Inventor： 하영국 ,  손주찬 ,  유원필 ,  조영조

IPC: H04B7/24 ,  H04L12/12 ,  G08B1/08

CPC classification number: H04B10/1149 ,  H04W84/18 ,  Y02D70/144 ,  Y02D70/162 ,  Y02D70/22 ,  Y02D70/46

Abstract: A method for radio sensor network communication by using infrared rays and RF communication and an apparatus therefor are provided to perform selectively infrared communication and RF communication for low power consumption, thereby realizing the RF sensor network communication with low power than performing communication with only an RF communication function. A first communication unit wakes up a neighbor node in a standby status through an infrared signal and transceives sensor data with the waked node. A second communication unit transceives the sensor data with the waked neighbor node through an RF(Radio Frequency) signal. A radio node processing unit(110) wakes up from the standby status if the wake-up signal of the infrared signal is received through the first communication unit, wakes up the neighbor node in the standby status through the first communication unit in order to transmit the received sensor data to a target node if the sensor data is received through one of the first and second communication units, and transmits the sensor data with the minimum power consumption through one of the first and second communication units.

Abstract translation: 提供了一种通过使用红外线和RF通信的无线电传感器网络通信方法及其设备，用于执行选择性的红外通信和RF通信以实现低功耗，从而实现低功率的RF传感器网络通信，而不仅仅是执行与RF 通讯功能。 第一个通信单元通过红外信号唤醒处于待机状态的邻居节点，并用唤醒节点收发传感器数据。 第二个通信单元通过射频（射频）信号与被唤醒的邻居节点收发传感器数据。 如果通过第一通信单元接收到红外信号的唤醒信号，则无线电节点处理单元（110）从待机状态唤醒，通过第一通信单元唤醒待机状态的邻居节点，以便发送 如果通过第一和第二通信单元之一接收到传感器数据，则将接收到的传感器数据传送到目标节点，并通过第一和第二通信单元中的一个传送具有最小功耗的传感器数据。

公开(公告)号：KR1020070061212A

公开(公告)日：2007-06-13

申请号：KR1020060065474

申请日：2006-07-12

Applicant: 한국전자통신연구원

Inventor： 박천수 ,  손주찬 ,  조영조

IPC: G06F17/00 ,  G06F17/21

CPC classification number: G06Q50/10 ,  G06F17/22 ,  G06F17/30005 ,  G06Q10/067

Abstract: A system and a method for generating BPEL4WS using OWL-S and a web service are provided to offer various services and apply to a robot platform having different service interface by mapping a process model and I/O(Input/Output) message information defined in the OWL-S to the process model defined in the BPEL4WS, and enabling WSDL(Web Service Description Language) to generate the BPEL4WS from the outside of the robot platform. A service plan analyzer(110) extracts a WSDL document describing a web service by receiving an OWL-S base service plan instance and analyzes an OWL-S base service process structure. A WSDL analyzer(120) analyzes the extracted WSDL document. A process model mapper(130) maps the process model of the BPEL4WS and the OWL-S by receiving an analysis result from the service plan analyzer and the WSDL analyzer. A process model generator(140) generates a BPEL4WS model on a memory by using a mapping rule output from the process model mapper. A file converter(150) converts and outputs the BPEL4WS model in a file form.

Abstract translation: 提供了使用OWL-S和Web服务生成BPEL4WS的系统和方法，以提供各种服务，并通过映射过程模型和I / O（输入/输出）消息信息中定义的应用于具有不同服务接口的机器人平台 OWL-S到BPEL4WS中定义的流程模型，并启用WSDL（Web服务描述语言）从机器人平台的外部生成BPEL4WS。 服务计划分析器（110）通过接收OWL-S基本服务计划实例来提取描述Web服务的WSDL文档，并分析OWL-S基本服务进程结构。 WSDL分析器（120）分析提取的WSDL文档。 过程模型映射器（130）通过从服务计划分析器和WSDL分析器接收分析结果来映射BPEL4WS和OWL-S的过程模型。 过程模型生成器（140）通过使用来自过程模型映射器的映射规则输出在存储器上生成BPEL4WS模型。 文件转换器（150）以文件形式转换并输出BPEL4WS模型。