-
公开(公告)号：KR101429962B1
公开(公告)日：2014-08-14
申请号：KR1020100116115
申请日：2010-11-22
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30029
Abstract: 본 발명은 로봇을 이용하여 사용자의 기억을 보조하는 시스템 및 방법에 관한 것이다. 본 발명은 사용자의 요청에 따라 주변 상황과 관련된 영상 데이터를 획득하고, 사용자의 피드백에 따라 획득되거나 외부로부터 수집된 데이터에 대해 고유 ID를 부여하고 사용자 정보, 위치, 시간, 인물 정보 등을 고려하여 분류 관리하며, 미리 정해진 질의들로부터 사용자에 의해 선택된 질의에 따라 관리 중인 데이터를 선별하여 사용자의 기억 보조를 위한 것으로써 제공된다. 본 발명에 따르면, 사용자 중심의 시점에서 일상 활동을 기록할 수 있으며, SNS와 연계하여 로봇의 DB에 저장되어 있지 않은 사진도 쉽게 접할 수가 있다. 또한, 사용자 동의 하에 영상을 촬영하기 때문에 프라이버시 문제도 해결할 수 있다.
-
公开(公告)号：KR1020140054763A
公开(公告)日：2014-05-09
申请号：KR1020120120657
申请日：2012-10-29
Applicant: 한국전자통신연구원
CPC classification number: G06N99/005 , G06T17/05 , G06T7/00 , G06T17/00 , G06T17/005 , G06T17/30
Abstract: Disclosed are a device and a method for mapping probability distribution based on object and system properties. The device for mapping the probability distribution includes: a sensor information collection unit collecting sensor information from multiple sensors; an object recognition unit recognizing an object by combining and deducting the sensor information, and obtaining object information, which is information of the object; and a probability distribution generation unit determining whether an object property model, which includes at least one of kinematic properties, physical properties, and probability properties, is applied to correspond to the object information, and generating the probability distribution based on object properties by obtaining the object properties corresponding to the object information.
Abstract translation: 公开了一种基于对象和系统属性映射概率分布的设备和方法。 用于映射概率分布的装置包括：从多个传感器收集传感器信息的传感器信息收集单元; 对象识别单元，通过组合和减去传感器信息来识别对象，获得作为对象的信息的对象信息; 以及概率分布生成单元，确定是否应用包含运动特性，物理特性和概率属性中的至少一个的对象属性模型来对应于所述对象信息，并且基于对象属性生成所述概率分布， 对象属性对应于对象信息。
-
公开(公告)号：KR1020130108784A
公开(公告)日：2013-10-07
申请号：KR1020120030508
申请日：2012-03-26
Applicant: 한국전자통신연구원
Abstract: PURPOSE: Automotive headlight control device and method are provided to automatically control the brightness of an automotive headlight by using the front image information photographed in a front camera. CONSTITUTION: A pedestrian detection unit (230) detects the entire body and the entire face of a pedestrian from image information photographing the front of a car. A judgment configuration unit (270) judges existence of pedestrian in the front of a car based on the entire body and the entire face of the detected pedestrian and a current driving speed of the car, and configures a brightness value of the headlight with a tail lamp if a pedestrian exists. A control unit (290) controls the brightness of the headlight according to the configured brightness value. [Reference numerals] (210) Image input unit; (230) Pedestrian detection unit; (250) Driving information input unit; (270) Judgment configuration unit; (290) Control unit; (400) Forward camera; (500) Headlight
Abstract translation: 目的：提供汽车前照灯控制装置和方法，通过使用在前置照相机中拍摄的前置图像信息来自动控制汽车前灯的亮度。 构成：步行者检测单元（230）从拍摄汽车前方的图像信息中检测行人的整个身体和整个脸部。 判断构成单元（270）基于检测到的行人的整体和整个面部以及轿厢的当前行驶速度来判断轿厢前方的行人的存在，并且利用尾部构成前灯的亮度值 如果存在行人灯， 控制单元（290）根据配置的亮度值控制前灯的亮度。 （附图标记）（210）图像输入单元; （230）行人检测单位; （250）驾驶信息输入单元; （270）判断配置单位; （290）控制单元; （400）前照相机; （500）大灯
-
公开(公告)号：KR101295855B1
公开(公告)日：2013-08-12
申请号：KR1020090104132
申请日：2009-10-30
Applicant: 한국전자통신연구원
IPC: G06F3/0346 , G06F3/042 , G06F3/03
Abstract: 사용자의 동작을 인식하여 전자기기를 제어하는 적외선 송신 장치와, 적외선 송신 장치를 기반으로 한 동작 인식으로 제어되는 전자기기는, 적외선 신호를 송출하는 적외선 신호 발생 장치와, 적외선 신호 발생 장치의 온/오프를 제어하는 스위치를 포함하는 장갑형 적외선 송신 장치를 포함한다. 또한 스위치의 온/오프 사이에 송출된 적외선 신호를 수신하는 통신 모듈과, 적외선 신호의 위치 변화에 근거한 적외선 송신 장치의 동작을 샘플링하여 데이터화 하고, 적외선 송신 장치의 동작 데이터에 대응하는 제어 명령을 전자기기 제어부에 송신하는 동작 관리부를 포함하는 전자기기를 포함한다.
-
公开(公告)号：KR1020130082223A
公开(公告)日：2013-07-19
申请号：KR1020120003295
申请日：2012-01-11
Applicant: 한국전자통신연구원
CPC classification number: G06K9/00865 , G06K9/3258 , G06K9/348 , G06K2209/01
Abstract: PURPOSE: A method and an apparatus for ordering of off-line characters from a natural scene are provided to suggest angular normalization, structural feature position determination, and structural feature numeral string generation methods, thereby highly recognizing off-line characters of various fonts regardless of angles or sizes of the characters. CONSTITUTION: A pre-processor (100) extracts off-line characters from an image, performs binarization of the off-line characters through a predetermined threshold, and performs thinning of the binarized off-line characters to have a fixed thickness. A normalizer (200) normalizes the pre-processed characters into characters corresponding to a predetermined standard. A code allocator (300) allocates chain codes that indicate a progressive direction of each stroke and intersection points that indicate strokes are crossed each other to points constituting the normalized characters. An ordering unit (400) transforms the characters with allocated codes or points into numeralized data using structural features containing end points or intersection points of the characters. [Reference numerals] (100) Pre-processor; (210) End point detecting unit; (220) Text information rotating unit; (300) Code allocator; (410) Structural feature code generating unit; (420) Ordering unit; (500) Modeling unit; (600) Database; (AA) Text information
Abstract translation: 目的：提供一种从自然场景排序离线字符的方法和装置，以提出角度归一化，结构特征位置确定和结构特征数字串生成方法，从而高度识别各种字体的离线字符，而不管 角度或大小的字符。 规定：预处理器（100）从图像中提取离线字符，通过预定阈值执行离线字符的二值化，并且执行二值化离线字符的稀疏化以具有固定的厚度。 归一化器（200）将预处理的字符归一化为与预定标准对应的字符。 代码分配器（300）分配指示每个笔画的逐行方向的链代码，指示笔划的交点与构成归一化字符的点交叉。 排序单元（400）使用包含字符的端点或交点的结构特征将具有分配的代码或点的字符变换成数字化数据。 （附图标记）（100）预处理器; （210）端点检测单元; （220）文字信息旋转单元; （300）代码分配器; （410）结构特征码生成单元; （420）订购单位; （500）建模单位; （600）数据库; （AA）文本信息
-
Patent Agency Ranking
公开(公告)号：KR1020130081063A
公开(公告)日：2013-07-16
申请号：KR1020120002030
申请日：2012-01-06
Applicant: 한국전자통신연구원
CPC classification number: H04W4/027 , G08B21/0423 , H04W4/90
Abstract: PURPOSE: A behavior pattern collecting apparatus, a behavior pattern analyzing system using the same, and a method thereof can minimize the feeling of tightness or the inconvenience of a wearer. CONSTITUTION: A behavior pattern collecting apparatus (100) comprises an acceleration sensor (110), a behavior pattern extractor (120), a radio communications unit (140), a position update unit (170), and a data collector (130). The acceleration sensor senses the movement of a wearer and generates an acceleration signal of 3-axial direction. The behavior pattern extractor classifies and stores a behavior pattern corresponding to the acceleration signal, and corresponds it to a generated acceleration signal. The radio communication unit connects a relay extracting a behavior pattern with wireless telecommunication. The position update unit updates position estimate information including a relay identification code of a relay currently telecommunicating on a real time basis. The data collector periodically generates collection information including equipment identification number, position estimate information, or an extracted behavior pattern. [Reference numerals] (110) Acceleration sensor; (120) Behavior pattern extractor; (121) Behavior categorization unit; (122) Fall detection unit; (130) Data collector; (140) Radio communications unit; (150) Emergency call unit; (155) Emergency call button; (160) Infrared light sensor; (170) Position update unit; (180) Temperature sensor; (182) Illumination sensor
Abstract translation: 目的：行为模式收集装置，使用该行为模式分析系统的行为模式分析系统及其方法可以使佩戴者的紧张感或不便感最小化。 构成：行为模式收集装置（100）包括加速度传感器（110），行为模式提取器（120），无线电通信单元（140），位置更新单元（170）和数据收集器（130）。 加速度传感器感测佩戴者的移动并产生3轴方向的加速度信号。 行为模式提取器对与加速度信号相对应的行为模式进行分类并存储，并将其对应于生成的加速度信号。 无线电通信单元将提取行为模式的中继与无线电信连接起来。 位置更新单元更新包括当前正在进行电信的继电器的继电器识别码的位置估计信息。 数据收集器周期性地产生包括设备识别号，位置估计信息或提取的行为模式的收集信息。 （附图标记）（110）加速度传感器; （120）行为模式提取器; （121）行为分类单位; （122）跌落检测单元; （130）数据采集器; （140）无线电通信单元; （150）紧急呼叫单位; （155）紧急呼叫按钮; （160）红外光传感器; （170）位置更新单位; （180）温度传感器; （182）照明传感器
-
公开(公告)号：KR1020130072063A
公开(公告)日：2013-07-01
申请号：KR1020110139610
申请日：2011-12-21
Applicant: 한국전자통신연구원
IPC: G06T7/00
CPC classification number: G06K9/00 , G06K2209/19 , G06T7/12 , G06T7/74 , G06T2207/30164
Abstract: PURPOSE: A device and method for recognizing product components are provided to increase a mounting rate of components by estimating a pose angle and kinds using feature vectors. CONSTITUTION: An image pre-processing unit of a learning unit (100) detects a component area using a component edge by extracting each component edge from inputted component images. A feature extractor of the learning unit extracts component features in the component area, and generates the feature vectors using the component features. A recognition unit (200) inputs the feature vectors in a nerve network in advance in order to recognize the type of the component. [Reference numerals] (110) Image input unit; (120) Image pre-processing unit; (130) Feature extractor; (140) Nerve network learning unit
Abstract translation: 目的：提供用于识别产品部件的装置和方法，以通过使用特征向量估计姿态角度和种类来增加部件的安装速度。 构成：学习单元（100）的图像预处理单元通过从输入的分量图像中提取每个分量边缘，使用分量边缘来检测分量区域。 学习单元的特征提取器提取组件区域中的组件特征，并使用组件特征生成特征向量。 识别单元（200）预先输入神经网络中的特征向量以识别该组件的类型。 （附图标记）（110）图像输入单元; （120）图像预处理单元; （130）特征提取器; （140）神经网络学习单元
-
公开(公告)号：KR1020130072038A
公开(公告)日：2013-07-01
申请号：KR1020110139582
申请日：2011-12-21
Applicant: 한국전자통신연구원
IPC: H04N21/45 , H04N21/441
CPC classification number: H04N21/4383 , H04H60/45 , H04N21/4415 , H04N21/466
Abstract: PURPOSE: A device and a method for setting a TV start channel are provided to automatically set a desired channel when users turn on the TV. CONSTITUTION: A camera (130) films the front side of TV when the TV is turned on. A user management unit (140) recognizes a user by images filmed by the camera. A preference channel determiner (150) determines a preference channel on the basis of history information on channels that are frequently set by the user. A TV controller (120) sets the channels as a TV start channel. [Reference numerals] (110) User input unit; (120) TV controller; (130) Camera; (140) User management unit; (150) Preference channel determiner; (160) State information memory; (170) Channel history management unit; (180) Saving unit (user information, channel history information); (AA) Current user; (BB) Current time; (CC) Current setting channel; (DD) Channel setting time
Abstract translation: 目的：提供一种用于设置电视启动频道的设备和方法，以便在用户打开电视机时自动设置所需频道。 构成：当电视机打开时，相机（130）拍摄电视机的正面。 用户管理单元（140）通过由照相机拍摄的图像识别用户。 偏好信道确定器（150）基于用户频繁设置的信道上的历史信息来确定偏好信道。 电视控制器（120）将频道设置为电视启动频道。 （附图标记）（110）用户输入单元; （120）电视控制器; （130）相机; （140）用户管理单元; （150）优先通道确定器; （160）国家信息记忆; （170）频道历史管理单元; （180）保存单元（用户信息，通道历史信息）; （AA）当前用户; （BB）当前时间; （CC）当前设定通道; （DD）通道设定时间
-
公开(公告)号：KR1020130063261A
公开(公告)日：2013-06-14
申请号：KR1020110129686
申请日：2011-12-06
Applicant: 한국전자통신연구원
IPC: G05D1/02
CPC classification number: G05D1/0246 , G06K9/00221
Abstract: PURPOSE: A method for closely moving a robot to a user is provided to move the robot to a specific user closely by finding the specific user effectively. CONSTITUTION: A face detector detects at least one face in a photographed image(325). A face cluster generator generates at least one face cluster by clustering the detected faces according to a face size(330). An approximate distance calculator obtains an average face area which is the average of faces included in a random face cluster among the generated face clusters(340). The approximate distance calculator obtains an approximate distance between the robot and the average face area using the size of the average face area(345). A movement controller moves the robot toward the average face area(350). The movement control part performs face recognition for each face included in the face cluster, and moves the robot toward the user if a specific user is found(365). [Reference numerals] (310) Photograph an image at a certain angle; (315) Photographed at all angles?; (320) Rotate a robot; (325) Detect a face in each angle image; (330) Generate clusters by face size using the detected faces; (335) Select a face cluster; (340) Obtain an average face area; (345) Calculate the approximate distance and direction of the average face area; (350) Move the robot toward the average face area; (355) Perform facial recognition for each face included in the face clusters; (360) The wanted user is included?; (365) Approach to the found user; (AA) Start; (BB,DD) No; (CC,EE) Yes; (FF) End
Abstract translation: 目的：提供一种将机器人紧密移动到用户的方法，通过有效地查找特定用户，将机器人紧密地移动到特定用户。 构成：脸部检测器检测拍摄图像中的至少一个脸部（325）。 面部群集生成器根据面部大小聚类检测到的面部来生成至少一个面部群集（330）。 近似距离计算器获得平均面部面积，其是生成的面部群集（340）中包括在随机面部群集中的面部的平均值。 近似距离计算器使用平均面部面积的大小来获得机器人与平均面部区域之间的近似距离（345）。 移动控制器使机器人朝向平均脸部区域（350）移动。 移动控制部对于面部群集中的每个脸部进行面部识别，如果发现特定的用户，则向机器人移动（365）。 （附图标记）（310）以一定角度拍摄图像; （315）在各个角度拍摄？ （320）旋转机器人; （325）检测每个角度图像中的脸部; （330）使用检测到的脸部通过面部大小生成聚类; （335）选择一个面团; （340）获取平均面积; （345）计算平均面积的近似距离和方向; （350）将机器人朝平均脸部区域移动; （355）对面群中包含的每个面进行面部识别; （360）被包含的用户？ （365）找到用户的方法; （AA）开始; （BB，DD）否; （CC，EE）是; （FF）结束
-
公开(公告)号：KR101259560B1
公开(公告)日：2013-04-30
申请号：KR1020100028249
申请日：2010-03-30
Applicant: 한국전자통신연구원
Abstract: 본 발명은 노인이나 거동이 불편한 보행자 등을 대상으로 실생활에서 자주 발생하는 낙상사고를 실시간으로 감지하여 낙상으로 인한 2차 사고에 적극적으로 대비하기 위한 낙상 감지 장치 및 그 방법에 관한 것으로, 본 발명의 일 실시예인 낙상 감지 장치는 사용자의 가속도 및 기울기 정보를 감지하는 가속도 센서, 가속도 센서에서 추출되는 가속도 및 기울기 정보를 이용하여 낙상여부를 결정하는 판단 모듈 및 판단 모듈에 의해 낙상이 결정되면 낙상 발생 정보를 전송하는 무선 통신 모듈로 구성된다.
-
-
-
-
-
-
-
-
-
Search Results
191
records
(took 0.031 seconds)
