-
公开(公告)号：KR1020120045744A
公开(公告)日：2012-05-09
申请号：KR1020100107502
申请日：2010-11-01
Applicant: 한국전자통신연구원
CPC classification number: G09B7/00 , G06Q50/2057 , G06Q50/10 , G06T17/00
Abstract: 본발명은체험형학습콘텐츠의저작장치및 방법에있어서, 사용자의영상이투영되어학습을수행할수 있도록 3D 화면이나화면내에서의연결동작을등을정의하는체험형학습을위한 3D 콘텐츠를저작할수 있는저작장치를제공하고, 저작장치에서 3D 화면내학습을위한시나리오에따라스테이트, 액션존등을정의하고, 스테이트와액션존등이정의된지하철, 박물관등의다양한공간에사용자의 3D 화면을투영시켜사용자가가상으로해당공간에서의동작을체험할수 있도록함으로써실제공간에서와같은체험을경험할수 있어학습효과를높이도록한다.
Abstract translation: 目的：提供一种经验学习内容创作设备及其方法，以使用户能够通过创作三维内容来通过用户的运动投影的图像来学习运动。 规定：当要求经验学习内容创作时，显示创作内容的创作画面（S202）。 通过根据在创作画面中的内容的场景定制的输入信息来加载三维对象和二维对象来构建虚拟世界（S204）。 在虚拟世界中定义已建立运动被激活的动作区（S206）。 国家根据虚拟语言中的教育情景来定义（S208）。
-
公开(公告)号：KR1020100073186A
公开(公告)日：2010-07-01
申请号：KR1020080131780
申请日：2008-12-22
Applicant: 한국전자통신연구원
CPC classification number: G06K9/18 , G06K9/00288 , G06K9/00402 , G06T5/001 , G06T5/50
Abstract: PURPOSE: A maker recognition apparatus and a method thereof are provided to maximize the recognition rate of a marker by performing power estimation for a maker of a teaching material. CONSTITUTION: A marker recognition apparatus includes a marker detection unit(10), a pattern recognition unit(20), a pose estimation unit(30) and a marker restoration unit(40). The marker detection unit detects a marker in a photographed image of a teaching material, and a pattern recognition unit recognizes a pattern of a detected marker. The pose estimation unit estimates the pose of the material including a recognized marker as a pose converting matrix. The pose conversion matrix is obtained from a camera conversion vector.
Abstract translation: 目的：提供制造商识别装置及其方法，以通过对教材的制造者进行功率估计来最大化标记的识别率。 构成：标记识别装置包括标记检测单元（10），模式识别单元（20），姿势估计单元（30）和标记恢复单元（40）。 标记检测单元检测教材的拍摄图像中的标记，并且模式识别单元识别检测到的标记的图案。 姿态估计单元将包括识别标记的材料的姿态估计为姿势转换矩阵。 姿态转换矩阵是从相机转换矢量获得的。
-
公开(公告)号：KR1020100057984A
公开(公告)日：2010-06-03
申请号：KR1020080116626
申请日：2008-11-24
Applicant: 한국전자통신연구원
CPC classification number: G06K9/00241 , G06K9/00255 , G06T7/507 , G08B13/194
Abstract: PURPOSE: A face image verifying device and a method thereof are provided to verify whether a face image is real by using a characteristic capable of being created with the different characteristics in a 2D fake face image and 3D real face image. CONSTITUTION: A face detection part(302) detects a face range from an input image. A lighting part(110) examines illumination to the face range of a user. An illumination information recognition part(312) stores location information of the illumination and a gray scale value range of shadows capable of being created with the illumination. A shadow detection unit(308) is detects the shadows by the illumination in the face range image. A real image decision device(310) analyzes the detected shadows and calculates a gray scale value. In case the gray scale value is included in the gray scale value range, the real image decision device decides the face image as the real face image.
Abstract translation: 目的：提供一种面部图像验证装置及其方法，以通过使用能够在2D假面部图像和3D真实面部图像中以不同特征创建的特征来验证脸部图像是否是真实的。 构成：面部检测部（302）从输入图像检测面部范围。 照明部分（110）检查照明到用户的脸部范围。 照明信息识别部分（312）存储照明的位置信息和能够用照明产生的阴影的灰度值范围。 阴影检测单元（308）通过面部范围图像中的照明来检测阴影。 真实图像判定装置（310）分析所检测的阴影并计算灰度值。 如果灰度值包括在灰度值范围内，则真实图像判定装置将面部图像判定为真实的脸部图像。
-
公开(公告)号：KR1020080025610A
公开(公告)日：2008-03-21
申请号：KR1020060090310
申请日：2006-09-18
Applicant: 한국전자통신연구원
CPC classification number: G06K9/0061
Abstract: An iris recognition method and an apparatus for the iris recognition method are provided to detect the iris accurately and efficiently even though an eye is covered by eyebrows or an eyelid in an image of the eye. An iris recognition method includes a step of receiving an image of an eye(S100) and filtering the image with a threshold value to obtain a binary image(S105), a step of searching candidate central search regions for the pupil center using histogram information of the binary image(S110), a step of considering points belonging into the candidate central search regions in the binary image as candidate pupil center points and performing a masking operation on concentric circles obtained by varying the radius of a circle having each of the candidate pupil center points as a center point using eight pupil boundary mask templates to detect the boundary and the center point of the pupil(S115), and a step of performing a masking operation on concentric circles obtained by varying the radius of a circle having the detected center point of the pupil as a center point with six iris boundary mask templates to detect the boundary of the iris(S120).
Abstract translation: 提供了一种虹膜识别方法和用于虹膜识别方法的装置，即使眼睛被眼睛的图像中的眉毛或眼睑覆盖，也能够准确有效地检测虹膜。 虹膜识别方法包括接收眼睛图像的步骤（S100）并用阈值对图像进行滤波以获得二进制图像（S105），使用直方图信息搜索瞳孔中心的候选中心搜索区域的步骤 二进制图像（S110），考虑属于二进制图像中的候选中央搜索区域的点作为候选瞳孔中心点，并对通过改变具有每个候选瞳孔的圆的半径获得的同心圆执行掩蔽操作的步骤 中心点作为中心点，使用八个瞳孔边界掩模模板来检测瞳孔的边界和中心点（S115），以及对通过改变具有检测到的中心的圆的半径获得的同心圆进行掩蔽操作的步骤 瞳孔的点作为具有六个虹膜边界掩模模板的中心点，以检测虹膜的边界（S120）。
-
公开(公告)号：KR1020030055702A
公开(公告)日：2003-07-04
申请号：KR1020010085757
申请日：2001-12-27
Applicant: 한국전자통신연구원
IPC: G06F9/06
CPC classification number: G06F21/6218 , G06F2221/2107 , G06F2221/2113
Abstract: PURPOSE: A file security system using a security level and a method for managing an encryption key are provided to encrypt or decrypt a file having the security level by using the encryption key of each security level, and to offer the file to a user, or save the file in a disk. CONSTITUTION: The disk(130) stores the encryption key matched with the security level set by an access control module(120), a key file storing the encryption key, and the file encrypted by the encryption key. A kernel memory(140) stacks the encryption key stored in the disk(130) according to the driving of the encryption file system(110). The encryption file system(110) draws out the encryption key matched with the security level of the file to read or stored by the user from the kernel memory(140), and transmits the file decrypted or encrypted by the encryption key to the user, or stores the file in the disk.
Abstract translation: 目的：提供使用安全级别的文件安全系统和用于管理加密密钥的方法，以通过使用每个安全级别的加密密钥对具有安全级别的文件进行加密或解密，并将文件提供给用户，或 将文件保存在磁盘中。 构成：磁盘（130）存储与由访问控制模块（120）设置的安全等级相匹配的加密密钥，存储加密密钥的密钥文件和通过加密密钥加密的文件。 内核存储器（140）根据加密文件系统（110）的驱动堆叠存储在盘（130）中的加密密钥。 加密文件系统（110）从内核存储器（140）中提取与用户要读取或存储的文件的安全级别匹配的加密密钥，并将用加密密钥解密或加密的文件发送给用户， 或将文件存储在磁盘中。
-
Patent Agency Ranking
公开(公告)号：KR1020150081797A
公开(公告)日：2015-07-15
申请号：KR1020140001649
申请日：2014-01-07
Applicant: 한국전자통신연구원
CPC classification number: H04N5/23254 , G06K9/00369 , H04N7/18
Abstract: 본발명의객체추적장치및 방법에관한것으로, 본발명에따른장치는, 입력되는영상으로부터객체를포함하는영역을검출하는객체검출부, 입력되는영상의매 프레임마다상기객체에대응하는영역으로부터복수의특징점들을검출하고, 현재프레임및 이전프레임으로부터검출된특징점들을비교하여상기객체의위치를추적하는객체추적부, 및특징점들의평균치및 최대유사도값중 어느하나에근거하여산출된상태변수추정치를이용하여상기객체의중첩상태를판별하는중첩판별부를포함하며, 상기객체가중첩상태가아닌경우만해당객체의정보를업데이트하도록한다.
Abstract translation: 本发明涉及一种被摄体跟踪方法和装置。 本发明的装置包括：被摄体检测单元，用于从图像输入中检测包含被摄体的区域; 对象跟踪单元，用于从与图像输入的每个帧相对应的对象区域中检测多个特征点，并通过比较从当前帧和先前帧检测到的特征点来跟踪对象的位置; 以及重叠确定单元，用于通过使用基于特征点的平均值和最大相似度值之一计算的估计状态变量来确定对象的重叠状态，其中在被摄体不在 重叠状态。
-
公开(公告)号：KR101530634B1
公开(公告)日：2015-06-23
申请号：KR1020100107502
申请日：2010-11-01
Applicant: 한국전자통신연구원
CPC classification number: G09B7/00
Abstract: 본발명은체험형학습콘텐츠의저작장치및 방법에있어서, 사용자의영상이투영되어학습을수행할수 있도록 3D 화면이나화면내에서의연결동작을등을정의하는체험형학습을위한 3D 콘텐츠를저작할수 있는저작장치를제공하고, 저작장치에서 3D 화면내학습을위한시나리오에따라스테이트, 액션존등을정의하고, 스테이트와액션존등이정의된지하철, 박물관등의다양한공간에사용자의 3D 화면을투영시켜사용자가가상으로해당공간에서의동작을체험할수 있도록함으로써실제공간에서와같은체험을경험할수 있어학습효과를높이도록한다.
-
公开(公告)号：KR101500711B1
公开(公告)日：2015-03-10
申请号：KR1020120006393
申请日：2012-01-19
Applicant: 한국전자통신연구원
IPC: G06T7/20
CPC classification number: G06T7/2033 , G06T7/246 , G06T7/277 , G06T2207/10016 , G06T2207/10024 , G06T2207/20076 , G06T2207/30196 , G06T2207/30232
Abstract: 본 발명은 컬러 히스토그램을 이용한 사람 추적 방법에 관한 것으로서, 파티클 필터를 기반으로 사람을 추적할 때 사람 한 명에 대해 하나의 타겟 컬러 히스토그램만을 가지고 추적을 수행하지 않고 사람의 포즈에 따라 적응적으로 다른 타겟 컬러 히스토그램을 통해 추적을 수행하여 사람 추적의 정확도를 향상시킬 수 있다.
Abstract translation: 公开了使用颜色直方图的人物跟踪方法。 使用根据本发明的颜色直方图的人类跟踪方法可以根据人的姿态更自适应地执行使用不同目标颜色直方图的人物跟踪，而不是仅将一个目标颜色直方图应用于一个人的跟踪过程，使得精度 的人追踪可以增加。 人体跟踪方法包括执行输入视频数据的色彩空间转换; 基于色空间转换数据计算粒子的状态方程; 计算状态方程，计算人姿态自适应观测可能性; 使用观察可能性重新采样粒子，并估计人的状态值; 并更新目标颜色直方图。
-
公开(公告)号：KR101381594B1
公开(公告)日：2014-04-10
申请号：KR1020080131661
申请日：2008-12-22
Applicant: 한국전자통신연구원
IPC: G06Q50/20
Abstract: 본 발명은 교실의 학생 및 원격지의 강사가 등장하는 가상공간을 구현하여 학생이 학습 콘텐츠에 실제로 들어간 것처럼 느끼면서 가상현실 콘텐츠와 상호작용하는 학습을 지원하는 교육 장치 및 방법에 관한 것이다. 본 발명은 학생을 촬영한 영상에서 학생의 영상을 추출하는 영상 컴퓨터와, 원격지 강사를 촬영한 영상에서 원격지 강사의 영상을 추출하는 원격지 강사 컴퓨터와, 가상공간이 구현된 학습 콘텐츠를 실행하며 상기 원격지 강사 컴퓨터와 실시간 통신을 수행하는 교실 강사 컴퓨터, 및 상기 학습 콘텐츠의 가상공간에 상기 학생의 영상 및 상기 원격지 강사의 영상을 합성하는 가상현실 컴퓨터를 포함하는 가상현실을 이용한 교육 장치를 제공한다. 본 발명에 의하면, 공간적 제약을 극복하면서도 현실감 있는 체험 기반의 교육을 제공할 수 있게 되는 효과가 있다.
가상현실, 스테레오 카메라, 영상합성, 원격교육, 콘텐츠-
公开(公告)号：KR1020140039888A
公开(公告)日：2014-04-02
申请号：KR1020120106685
申请日：2012-09-25
Applicant: 한국전자통신연구원
CPC classification number: G06N99/005 , G06K9/6231 , G06K9/6235 , G06K9/6256 , G06K9/6282 , G06K2009/6236
Abstract: Disclosed is a learning method using extracted data characteristics for simplifying learning process and improving the accuracy of the estimation. The learning method using the extracted data characteristics comprises the steps of: dividing input learning data into two groups on the criteria set in advance; extracting the data characteristics that can distinguish the divided groups; and performing a learning by using the extracted characteristics. The present invention can simplify the learning process and improve the accuracy of the estimation because the learning data is classified in stages for extracting the characteristics that can easily distinguish each group. [Reference numerals] (AA) Start; (BB) Yes; (CC) No; (DD) End; (S110) Dividing input learning data into two groups; (S120) Extracting the data characteristics that can distinguish the divided groups; (S130) Division of sub-groups is necessary?; (S140) Diving to the sub-groups; (S150) Learning by using the extracted characteristics
Abstract translation: 公开了一种使用提取的数据特征来简化学习过程并提高估计精度的学习方法。 使用提取的数据特征的学习方法包括以下步骤：根据预先设定的准则将输入学习数据分成两组; 提取可以区分分组的数据特征; 并通过使用所提取的特征来执行学习。 本发明可以简化学习过程并提高估计的准确性，因为学习数据被分阶段分类以提取可以容易地区分每个组的特性。 （附图标记）（AA）开始; （BB）是的; （CC）否; （DD）结束; （S110）将输入学习数据分为两组; （S120）提取可以区分分组的数据特征; （S130）分组是必要的吗？ （S140）潜水到小组; （S150）通过使用提取的特征进行学习
-
-
-
-
-
-
-
-
-
Search Results
63
records
(took 0.022 seconds)
