-
公开(公告)号：KR1020090099622A
公开(公告)日：2009-09-23
申请号：KR1020080024715
申请日：2008-03-18
Applicant: 한국전자통신연구원
IPC: G06T7/00 , H04N21/232
CPC classification number: G06K9/00711 , G06K9/4642
Abstract: A feature extraction apparatus of a video signal and an extraction method thereof, a video recognition system and an identification method for the video recognition system are provided to extract the features of a video signal regardless of the conversion of the frame and compression rates. A feature extraction apparatus of a video signal includes a frame rate conversion unit(102a), a black/white scale conversion unit(102b), a frame size normalization unit(102c) and a block feature extraction unit(102d). The frame rate conversion unit performs the conversion according to a preset frame rate when a video signal is inputted, and the black/white scale conversion unit converts the converted video signal into a black/white signal. The frame size normalization unit normalizes the black/white signal into the size of a preset image. The block feature extraction unit splits the normalized signal into the blocks of certain size, then extracts a feature value depending on the image brightness for each block and outputs the feature vector corresponding to the feature value.
Abstract translation: 提供视频信号的特征提取装置及其提取方法，视频识别系统和用于视频识别系统的识别方法，以提取视频信号的特征，而不管帧的转换和压缩率如何。 视频信号的特征提取装置包括帧速率转换单元（102a），黑/白比例转换单元（102b），帧大小归一化单元（102c）和块特征提取单元（102d）。 当输入视频信号时，帧速率转换单元根据预设的帧速率执行转换，并且黑/白比例转换单元将转换的视频信号转换为黑/白信号。 帧大小归一化单元将黑/白信号归一化为预设图像的大小。 块特征提取单元将归一化信号分解成一定大小的块，然后根据每个块的图像亮度提取特征值，并输出与特征值相对应的特征向量。
-
公开(公告)号：KR1020090065099A
公开(公告)日：2009-06-22
申请号：KR1020070132543
申请日：2007-12-17
Applicant: 한국전자통신연구원
CPC classification number: G06K9/46 , G06F17/30247
Abstract: A system for managing digital image features and a method thereof are provided to extract an intrinsic feature from a digital still image to efficiently mange a large sized image, thereby enabling a user to efficiently a necessary still image. A still image feature extractor(300) extracts feature information of a digital image. A still image feature classifier(400) classifies the extracted feature information and feature information stored in a still image feature database(500). A still image searcher(600) outputs images larger than a specific threshold value as a search result by measuring degree of similarity between feature information read from the still image feature database and the extracted feature information.
Abstract translation: 提供了一种用于管理数字图像特征的系统及其方法，用于从数字静止图像中提取固有特征以有效地管理大尺寸图像，从而使得用户能够有效地获得必要的静止图像。 静止图像特征提取器（300）提取数字图像的特征信息。 静止图像特征分类器（400）将所提取的特征信息和存储在静止图像特征数据库（500）中的特征信息进行分类。 静止图像搜索器（600）通过测量从静止图像特征数据库读取的特征信息与所提取的特征信息之间的相似度，输出大于特定阈值的图像作为搜索结果。
-
公开(公告)号：KR100862616B1
公开(公告)日：2008-10-09
申请号：KR1020070037399
申请日：2007-04-17
Applicant: 한국전자통신연구원
CPC classification number: G11B27/11 , G06F17/30743 , G06F17/30778
Abstract: An audio fingerprint searching system and method using index information are provided to reduce searching time by generating index based on statistical characteristics of audio fingerprint of an audio file. An audio fingerprint searching system using index information comprises an index process part(3), a DB group(2), and an audio fingerprint searching device(1). The index process part extracts an audio fingerprint from an audio file and transmits the audio fingerprint and information about music to a corresponding DB. The DB group generates index based on statistical characteristics of the audio fingerprint, and stores position information for matching the index, audio fingerprint, and music information continuously. The audio fingerprint searching device creates new index based on statistical characteristics of an audio fingerprint of newly input audio file, and searches information about corresponding music for the new audio file by searching the new index from the DB group.
Abstract translation: 提供使用索引信息的音频指纹搜索系统和方法，以通过基于音频文件的音频指纹的统计特性生成索引来减少搜索时间。 使用索引信息的音频指纹搜索系统包括索引处理部分（3），DB组（2）和音频指纹搜索装置（1）。 索引处理部分从音频文件中提取音频指纹，并将音频指纹和关于音乐的信息发送到对应的DB。 数据库组基于音频指纹的统计特性生成索引，并且连续地存储用于匹配索引，音频指纹和音乐信息的位置信息。 音频指纹搜索装置根据新输入的音频文件的音频指纹的统计特性创建新索引，并通过从DB组中搜索新的索引来搜索关于新音频文件的相应音乐的信息。
-
公开(公告)号：KR1020080010028A
公开(公告)日：2008-01-30
申请号：KR1020060069970
申请日：2006-07-25
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30864
Abstract: A system and a method for searching illegal contents automatically in the Internet by using a connection/search application are provided to search illegal contents automatically in a charged P2P(Peer-to-Peer)/webhard by editing an illegal contents searching/downloading script file based on configuration information of the connection/search application in a website, and controlling the application with the script file to search/download the illegal contents automatically. A keyword input part(13) receives a keyword related to the contents from a searcher and manages the keyword with a keyword group. A script file editor(14) edits the illegal contents searching/downloading script file by using a Windows class ID, which controls the connection/search application(11), extracted from the configuration information of the application. A script file automating part(15) controls the application to search/download the illegal contents related to the keyword group automatically according to the script file. An illegal contents information storing part(16) stores illegal contents information searched/downloaded by the application. An information manager(12) manages the configuration/connection information of the application.
Abstract translation: 提供了一种通过使用连接/搜索应用程序在互联网中自动搜索非法内容的系统和方法，用于通过编辑非法内容搜索/下载脚本文件来自动搜索在点对点（Peer-to-Peer）/ webhard中的非法内容 基于网站中的连接/搜索应用的配置信息，并且利用脚本文件控制应用程序来自动搜索/下载非法内容。 关键词输入部（13）从搜索者接收与内容有关的关键词，并用关键字组来管理该关键词。 脚本文件编辑器（14）通过使用从应用程序的配置信息中提取的控制连接/搜索应用程序（11）的Windows类ID来编辑非法内容搜索/下载脚本文件。 脚本文件自动化部（15）控制应用程序根据脚本文件自动搜索/下载与关键字组相关的非法内容。 非法内容信息存储部（16）存储由应用搜索/下载的非法内容信息。 信息管理器（12）管理应用的配置/连接信息。
-
公开(公告)号：KR1020070061285A
公开(公告)日：2007-06-13
申请号：KR1020060083242
申请日：2006-08-31
Applicant: 한국전자통신연구원
IPC: G10L19/018
Abstract: A method for watermarking an audio signal is provided to embed a watermark in a digital audio signal by using hybrid transform to maximize embedding strength while minimizing deterioration of quality of the audio signal after the audio signal is watermarked. A method for watermarking a digital audio signal includes a step of grouping 1024 samples of original audio signals into one frame and performing DCT(Discrete Cosine Transform) on the signals, a step of dividing the audio signals into a low-frequency coefficient and a high-frequency coefficient by using DCT coefficients, a step of embedding a watermark in the low-frequency coefficient by using a watermark embedder and performing IDCT(Inverse Discrete Cosine Transform) on the low-frequency coefficient to generate a tonal signal, a step of embedding a watermark in the high-frequency coefficient to generate a nontonal signal, and a step of summing up the tonal signal and the nontonal signal to generate a watermarked signal.
Abstract translation: 提供一种用于对音频信号进行水印处理的方法，通过使用混合变换将水印嵌入到数字音频信号中，以最大化嵌入强度，同时使音频信号被水印后的音频信号的质量降低最小化。 一种用于对数字音频信号进行水印处理的方法包括将原始音频信号的1024个样本分组成一帧并对该信号执行DCT（离散余弦变换）的步骤，将音频信号划分为低频系数和高频 通过使用DCT系数的频率系数，通过使用水印嵌入器在低频系数中嵌入水印并对低频系数执行IDCT（逆离散余弦变换）以生成音调信号的步骤，嵌入步骤 高频系数中的水印以产生非波顿信号，以及将色调信号和非波形信号相加以产生加水印信号的步骤。
-
Patent Agency Ranking
公开(公告)号：KR101527370B1
公开(公告)日：2015-06-10
申请号：KR1020110125457
申请日：2011-11-28
Applicant: 한국전자통신연구원
IPC: G06F17/30
CPC classification number: G06F17/30743
Abstract: 본발명은음악의감정정보를이용한추천음악검색방법및 장치에관한것으로, 더욱상세하게는, 사용자로부터소정의검색조건이입력되면, 상기입력된검색조건에서밸런스(Valence) 값및 어로우절(Arousal) 값을포함하는감정수치를추출하고, 상기감정수치에대응되는복합감정에대한정보를포함하는감정모델을이용하여, 상기검색된감정수치정보에대응되는감정순위조합을추출하고, 음악감정데이터베이스에서상기추출된감정순위조합에대응되는음악정보를검색하여추천음악목록을출력함으로써, 복합감정에대한음악검색의수행이가능하도록하여검색결과에대한사용자만족도를향상시키는발명이다.
-
37.发明公开이미지 기반 데이터베이스 구축 방법, 이미지 기반 데이터베이스를 이용한 거리 검색 서비스 제공 방법 및 장치 审中-实审基于图像生成数据库的方法，用于提供道路搜索服务的方法和用于执行方法的装置
公开(公告)号：KR1020140111106A
公开(公告)日：2014-09-18
申请号：KR1020130024061
申请日：2013-03-06
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30241 , G01C21/34 , G06T7/10
Abstract: A method for building a database based on images, which effectively builds database and shortens a road searching time is disclosed. The method comprises steps of securing an image of a crossroad; extracting information on features from the secured image of the crossroad; storing the information on features by matching the information on features with information on a map of the crossroad. Therefore, the resource and time consumed for building the database can be reduced and easy road and street searching service is provided to users.
Abstract translation: 公开了一种基于图像构建数据库的方法，该方法有效地构建了数据库并缩短了道路搜索时间。 该方法包括固定十字路口图像的步骤; 从十字路口的安全图像中提取特征信息; 通过将特征信息与十字路口的地图上的信息进行匹配来存储特征信息。 因此，可以减少建立数据库所需的资源和时间，为用户提供便利的道路和街道搜索服务。
-
公开(公告)号：KR1020130125706A
公开(公告)日：2013-11-19
申请号：KR1020130006058
申请日：2013-01-18
Applicant: 한국전자통신연구원
CPC classification number: H04N19/85 , H04N21/2393 , H04N21/44222 , H04N21/83
Abstract: The present invention relates to forensic marking for real-time content service, which can reduce the load of a content providing server and increase storage efficiency by deciding a method for forensic processing based on the frequency of request by content, etc. and performing forensic marking through the combination of pre-processed content files for content according to the decided forensic processing method; or by performing forensic marking in real time for an uncompressed original content file. [Reference numerals] (AA) Start;(BB) End;(S300) Receive a user request;(S302) Check a content;(S304) Requested frequency > T?;(S306) Load pre-processed content files;(S308) Combine the pre-processed content files;(S310) Perform forensic marking;(S312) Load a bar-compressed original content;(S314) Perform forensic marking at a file or buffer unit;(S318) Download/streaming;(S361) Encoding
Abstract translation: 本发明涉及用于实时内容服务的取证标记，其可以通过基于内容请求的频率等来确定用于取证处理的方法来减少内容提供服务器的负载并提高存储效率，并执行取证标记 通过预处理内容文件的组合，根据决定的取证处理方式对内容进行组合; 或通过对未压缩的原始内容文件实时执行取证标记。 （S30）请求频率> T（S306）加载预处理的内容文件;（S308）请求的频率> T 3;（S308） ）组合预处理的内容文件;（S310）执行取证标记;（S312）加载条形压缩的原始内容;（S314）在文件或缓冲单元执行取证;（S318）下载/流式传输;（S361） 编码
-
公开(公告)号：KR1020120076664A
公开(公告)日：2012-07-09
申请号：KR1020100125862
申请日：2010-12-09
Applicant: 한국전자통신연구원
IPC: G06Q50/10
Abstract: PURPOSE: A music information customizing method and apparatus are provided to improve user convenience by offering a customized service reflecting user's feeling. CONSTITUTION: A receiving unit(101) receives musical inclination evaluation of a user to the currently reproduced music from the user. A tracking unit(102) traces the expression coordinates of the user's musical inclination using the musical inclination evaluation and ages the expression coordinates to trace the final expression coordinates of the user's musical inclination. A searching unit(104) searches a playlist corresponding to the user's musical inclination.
Abstract translation: 目的：提供音乐信息定制方法和设备，通过提供反映用户感觉的定制服务来提高用户便利性。 构成：接收单元（101）从用户接收对当前再现的音乐的用户的音乐倾向评估。 跟踪单元（102）使用音乐倾向评估跟踪用户的音乐倾向的表达式坐标，并使表情坐标老化以跟踪用户音乐倾向的最终表情坐标。 搜索单元（104）搜索与用户的音乐倾向相对应的播放列表。
-
公开(公告)号：KR1020120062359A
公开(公告)日：2012-06-14
申请号：KR1020100123584
申请日：2010-12-06
Applicant: 한국전자통신연구원
IPC: G06F21/16 , H04N21/8358 , G06F17/30
CPC classification number: G06F21/10 , G06F2221/0713 , G06F2221/0733 , G06F21/16 , G06F17/30943 , H04N21/8358
Abstract: PURPOSE: A digital contents forensic marking apparatus and a method thereof are provided to remove delay time due to insertion of forensic mark and to rapidly supply a service. CONSTITUTION: A forensic marking unit(130) generates first and second content which insert first and second binary forensic marks. If a content service request is generated, the forensic marking unit combines first and second contents based on user information which requests the content service. Content database(150) stores the first and the second content.
Abstract translation: 目的：提供一种数字内容取证标记装置及其方法，以消除由于插入法医标记引起的延迟时间并快速提供服务。 构成：法医标记单元（130）生成插入第一和第二二进制取证标记的第一和第二内容。 如果生成内容服务请求，则取证标记单元基于请求内容服务的用户信息来组合第一和第二内容。 内容数据库（150）存储第一和第二内容。
-
-
-
-
-
-
-
-
-
Search Results
78
records
(took 0.133 seconds)
