公开(公告)号：KR100900316B1

公开(公告)日：2009-06-02

申请号：KR1020070064972

申请日：2007-06-29

Applicant: 한국전자통신연구원

Inventor： 김병규 ,  김종호 ,  조창식

IPC: H04N19/147 ,  H04N19/103

Abstract: 본 발명은 외부로부터 수신되는 비디오 신호 매크로 블록의 율-왜곡 비용을 측정하는 단계, 수신된 매크로 블록의 율-왜곡 비용과 인접하는 매크로 블록의 율-왜곡 비용을 비교하여 검색 임계치를 결정하는 단계, 수신된 매크로 블록의 인터 모드(inter mode) 별 오차합을 측정하여 최소 오차합을 가지는 인터 모드를 결정하는 단계, 결정된 인터 모드의 최소 오차합과 검색 임계치에 상응하여 인트라 모드(intra mode) 예측 검색을 결정하는 단계를 포함하는 실시간 비디오 압축을 위한 인트라 모드 예측 검색 방법을 제공할 수 있다.
H.264, 인트라(Intra) 예측 검색, 인터(Inter) 예측 검색

公开(公告)号：KR1020090025691A

公开(公告)日：2009-03-11

申请号：KR1020070090732

申请日：2007-09-07

Applicant: 한국전자통신연구원 ,  삼성전자주식회사

Inventor： 김범호 ,  김선태 ,  전형국 ,  조창식 ,  여창욱 ,  설종철 ,  금준승 ,  이요섭 ,  김일섭

IPC: H04N7/12 ,  H04N7/08 ,  H04N7/015

CPC classification number: H04N21/42615 ,  H04N21/4341

Abstract: A media file de-multiplexing method and an apparatus thereof are provided to minimize the process of finding a file multiplexing format and CODEC and analyzing a media file by discovering the file multiplexing format by using file extender information. A media file is inputted to an input unit(401). An extender database(407) stores the kinds of file extenders for showing all demultiplexing modes existing in a demultiplexing database(409). An extender searcher(403) searches the same extender as the extender of the inputted media file. The demultiplexing database stores at least one demultiplexing mode capable of demultiplexing the inputted media file. A demultiplexing searcher(405) searches a demultiplexing mode corresponding to the inputted media file.

Abstract translation: 提供一种媒体文件解复用方法及其装置，以通过使用文件扩展器信息发现文件复用格式来最小化查找文件多路复用格式和CODEC的过程并分析媒体文件。 媒体文件被输入到输入单元（401）。 扩展器数据库（407）存储用于显示存在于解复用数据库（409）中的所有解复用模式的文件扩展器的种类。 扩展搜索器（403）搜索与输入的媒体文件的扩展器相同的扩展器。 解复用数据库存储能够解复用输入的媒体文件的至少一种解复用模式。 解复用搜索器（405）搜索与输入的媒体文件对应的解复用模式。

公开(公告)号：KR1020090017724A

公开(公告)日：2009-02-19

申请号：KR1020070082078

申请日：2007-08-16

Applicant: 한국전자통신연구원

Inventor： 석진욱 ,  이정우 ,  김경일 ,  조창식

IPC: H04N19/103 ,  H04N19/51

CPC classification number: H04N19/523 ,  H04N19/109 ,  H04N19/14 ,  H04N19/147 ,  H04N19/176 ,  H04N19/61

Abstract: A block mode determining method is provided to minimize deterioration of image quality as minimizing operations for block mode determination. By performing motion estimation with respect to an inputted video frame, a current macro block and a corresponding reference macro block are determined. An inter-block average and distribution value are calculated with respect to the determined current macro block and corresponding reference macro block. By using the inter-block average and distribution value, inter-block bit generation possibility prediction value is calculated. It is determined whether it is a block mode needing addition motion estimation or a block mode needing no additional motion estimation according to the calculated bit generation possibility prediction value.

Abstract translation: 提供块模式确定方法以使图像质量的劣化最小化，以最小化块模式确定的操作。 通过对输入的视频帧进行运动估计，确定当前宏块和对应的参考宏块。 相对于确定的当前宏块和对应的参考宏块计算块间平均和分布值。 通过使用块间平均和分布值，计算块间比特生成可能性预测值。 根据计算出的比特生成可能性预测值，确定是否需要加法运动估计的块模式或不需要额外的运动估计的块模式。

公开(公告)号：KR1020090007849A

公开(公告)日：2009-01-21

申请号：KR1020070071020

申请日：2007-07-16

Applicant: 한국전자통신연구원

Inventor： 배수영 ,  조창식

IPC: G06Q50/10

CPC classification number: G06Q30/02

Abstract: A device and a method for sharing contents of a terminal placed in the same service space are provided to enable a contents receiver to search and use the contents of a contents provider through a receiver terminal by permitting the contents receiver placed in the same service space to share various contents of the contents provider. A content classifier(112) classifies shared joint contents, which is shared among contents receivers, stored in a memory based on a predetermined rule. A shared contents list generator(113) generates a shared contents list storing information related to the classified shared contents. A contents receiver searcher(114) searches whether the contents receiver wanting to share the classified contents is searched among the contents receivers entering the network. A contents processor(115) checks whether the shared contents selected by the contents receiver is available in the contents receiver, and processes the shared contents depending on a checking result. A shared contents provider(116) presents the shared contents list to the contents receiver according to a search result.

Abstract translation: 提供一种用于共享放置在同一服务空间中的终端的内容的装置和方法，以使得内容接收器能够通过允许放置在相同服务空间中的内容接收器通过接收机终端来搜索和使用内容提供商的内容 分享内容提供商的各种内容。 内容分类器（112）基于预定规则对存储在存储器中的内容接收器之间共享的共享内容进行分类。 共享内容列表生成器（113）生成存储与所分类的共享内容相关的信息的共享内容列表。 内容接收机搜索器（114）搜索进入网络的内容接收机是否搜索想要共享分类内容的内容接收机。 内容处理器（115）检查由内容接收器选择的共享内容是否在内容接收器中可用，并且根据检查结果处理共享内容。 共享内容提供商（116）根据搜索结果将共享内容列表呈现给内容接收者。

公开(公告)号：KR1020080029333A

公开(公告)日：2008-04-03

申请号：KR1020060095244

申请日：2006-09-29

Applicant: 한국전자통신연구원

Inventor： 석진욱 ,  조창식 ,  이정우

IPC: H04N19/103 ,  H04N19/147

CPC classification number: H04N19/107 ,  H04N19/109 ,  H04N19/11 ,  H04N19/14 ,  H04N19/147 ,  H04N19/176 ,  H04N19/61

Abstract: A video compressing/encoding device based on a feedback structure for fast encoding and an optimal mode decision method are provided to enable a separate RD(Rate Distortion) estimator to estimate a RD value simply by operation, to which linear parameters are applied, and update the linear parameters in real time through a feedback process, thereby being unnecessary to perform an MC operation for block modes except the inter mode of a maximum division block and the intra mode of a minimum division block to reduce an operation amount necessary for optimal mode determination significantly. A B-slice checking unit(200) performs B-slice check for current frame data. A maximum inter-mode estimation bit value calculator(300) performs ME(Motion Estimation) and MC(Motion Compensation) for an inter mode using a maximum division block to obtain an estimation bit value. A minimum intra ME bit value calculator(500) performs ME and MC for an intra mode using a minimum division block to obtain an estimation bit value. A linear estimation bit value calculator(400) obtains estimation bit values for modes except the inter mode of the maximum division block and the intra mode of the minimum division block by using linear parameters and statistics. A mode decision unit(600) compares the estimation bit values obtained by the maximum inter-mode estimation bit value calculator and the minimum intra ME bit value calculator, and the linear estimation bit value calculator to decide a proper encoding mode. An encoding unit(100) encodes the current frame data in a mode decided by the mode decision unit.

Abstract translation: 提供了一种基于用于快速编码的反馈结构和最佳模式决定方法的视频压缩/编码装置，以使单独的RD（速率失真）估计器可以简单地通过应用了哪些线性参数的操作来估计RD值，并且更新 通过反馈处理实时地进行线性参数，从而不需要对除了最大分割块的帧间模式和最小分割块的帧内模式之外的块模式执行MC操作以减少最佳模式确定所需的操作量 显著。 B片检查单元（200）对当前帧数据执行B片检查。 最大模式间估计位值计算器（300）使用最大分割块对帧间模式执行ME（运动估计）和MC（运动补偿），以获得估计位值。 最小内部ME位值计算器（500）使用最小分割块对帧内模式执行ME和MC以获得估计位值。 线性估计比特值计算器（400）通过使用线性参数和统计来获得除了最大分割块的帧间模式和最小分割块的帧内模式之外的模式的估计位值。 模式判定单元（600）比较由最大模式间估计比特值计算器和最小内部ME位值计算器获得的估计比特值和线性估计比特值计算器来确定适当的编码模式。 编码单元（100）以由模式判定单元决定的模式对当前帧数据进行编码。

公开(公告)号：KR100582549B1

公开(公告)日：2006-05-22

申请号：KR1020040109598

申请日：2004-12-21

Applicant: 한국전자통신연구원

Inventor： 이동훈 ,  조창식 ,  배수영 ,  마평수

IPC: G06Q50/10 ,  G06F15/16

公开(公告)号：KR1020030025386A

公开(公告)日：2003-03-29

申请号：KR1020010058241

申请日：2001-09-20

Applicant: 한국전자통신연구원

Inventor： 조창식 ,  마평수

IPC: H04N21/2387

CPC classification number: H04N21/26616 ,  H04N7/17336 ,  H04N21/2385 ,  H04N21/4334 ,  H04N21/47202 ,  H04N21/6405 ,  H04N21/6587

Abstract: PURPOSE: A method for pausing video playback and releasing the pause in a video system is provided to improve efficiency of a VOD system by supporting pause/resume operation of the system by using the minimum number of channel. CONSTITUTION: A client(201) sends an SID(Session ID) and a present playback position to a video server(205). The video server judges whether the SID is the only main session of a regular channel(202). If so, the video server calculates a stable paused time. The video server stops transmission of the regular channel. The video server judges whether the transmitting stop time is actually included in the calculated stable paused time. If so, the transmission of the regular channel is resumed. In the case that the SID is not the only main session and the actual transmission stop time is non included in the calculated stable paused time, the video server judges whether only the regular channel exists. If so, the video server calculates a stable paused time. A minimum value between data received in the regular channel and data received in a patching channel(203) becomes the stable paused time. The video server transmits the stable paused time to the client.

Abstract translation: 目的：提供一种用于暂停视频播放和释放视频系统中的暂停的方法，以通过使用最小数量的频道来支持系统的暂停/恢复操作来提高VOD系统的效率。 构成：客户端（201）向视频服务器（205）发送SID（会话ID）和当前播放位置。 视频服务器判断SID是否是常规频道的唯一主会话（202）。 如果是这样，视频服务器会计算一个稳定的暂停时间。 视频服务器停止正常频道的传输。 视频服务器判断发送停止时间是否实际包含在计算出的稳定暂停时间内。 如果是这样，则恢复正常信道的传输。 在SID不是唯一的主会话并且实际的传输停止时间不包括在计算的稳定暂停时间的情况下，视频服务器判断是否仅存在常规信道。 如果是这样，视频服务器会计算一个稳定的暂停时间。 在常规信道中接收的数据与在修补信道（203）中接收的数据之间的最小值变为稳定的暂停时间。 视频服务器向客户端发送稳定的暂停时间。

公开(公告)号：KR1020010027716A

公开(公告)日：2001-04-06

申请号：KR1019990039598

申请日：1999-09-15

Applicant: 한국전자통신연구원

Inventor： 마평수 ,  조창식 ,  진윤숙 ,  신규상

IPC: H04L12/18

CPC classification number: H04N21/4334 ,  G11B20/10527 ,  H04N21/231

Abstract: PURPOSE: A video storage server for reducing an initial latency time and an operating method thereof are provided to transmit video data to a client system at a request while storing the video data, so as to shorten the initial latency time and maximize the number of users capable of support. CONSTITUTION: A data storage unit(311) sequentially alternatively stores video divided by read-unit amounts in two disk sets distributively, and stores one read-unit amount of video data in many disks configuring the disk sets. An admission control manager(301) decides whether to admit an initial disk read request, on the basis of the maximum number of clients capable of support. A request handler(305) processes the initial disk read request admitted in the admission control manager(301) within an admission latency time, and schedules disk read requests under processing to be processed at least once within corresponding rounds. And the request handler(305) reads the read-unit amount of video data from the data storage unit(311) according to scheduled orders. And a delivery manager(313) delivers a message of a client received from a network to the admission control manager(301). If the initial disk request is admitted from the admission control manager(301), the delivery manager(313) transmits read-unit amount of video data supplied from the request handler(305) to a corresponding client through a network.

Abstract translation: 目的：提供用于减少初始等待时间的视频存储服务器及其操作方法，以便在存储视频数据的同时将视频数据发送到客户端系统，以缩短初始等待时间并最大化用户数量 能够支持。 构成：数据存储单元（311）依次将按读取单位量划分的视频分开存储在两个磁盘组中，并将一个读取单位数量的视频数据存储在配置磁盘组的许多磁盘中。 许可控制管理器（301）基于能够支持的最大客户端数量来决定是否允许初始磁盘读取请求。 请求处理器（305）在接纳延迟时间内处理准入控制管理器（301）中允许的初始磁盘读取请求，并且在相应的回合内调度要处理至少一次的磁盘读取请求。 并且请求处理器（305）根据调度的顺序从数据存储单元（311）读取读取单元数量的视频数据。 并且传送管理器（313）将从网络接收的客户端的消息传递到接纳控制管理器（301）。 如果从准入控制管理器（301）允许初始盘请求，则传送管理器（313）通过网络将从请求处理器（305）提供的视频数据的读取单元数量发送到对应的客户端。

公开(公告)号：KR1020000015619A

公开(公告)日：2000-03-15

申请号：KR1019980035646

申请日：1998-08-31

Applicant: 한국전자통신연구원

Inventor： 조창식 ,  윤석진 ,  신규상 ,  마평수

IPC: G06F17/21

Abstract: PURPOSE: An apparatus and method is provided to prevent a mistake of a user previously detecting a connection of cursor and a use of erroneous icon. CONSTITUTION: The method for generating a web document comprises the steps of: analyzing icons comprehended in a document worked at an editor according to each class to check whether or not a use of the icon satisfies a spacial limit condition and a connection between icons is logical; dividing the icons to a lattice to calculate an icon location in a table by a location of the lattice; generating a HTML document by use of the location information of the icon; analyzing a document worked at a sight editor(204) to generate an SQL door for an icon of a quality when the icon of the quality is used; and generating a CGI code(102) for searching a database by the SQL door.

Abstract translation: 目的：提供一种装置和方法，以防止用户以前检测到光标的连接和使用错误图标的错误。 构成：生成Web文档的方法包括以下步骤：分析根据每个类在编辑器中工作的文档中理解的图标，以检查图标的使用是否满足空间限制条件，并且图标之间的连接是逻辑的 ; 将图标划分为格子，以通过格子的位置计算表中的图标位置; 通过使用图标的位置信息生成HTML文档; 分析在视觉编辑器（204）处工作的文档，以在使用质量的图标时生成质量图标的SQL门; 以及生成用于通过所述SQL门搜索数据库的CGI代码（102）。

公开(公告)号：KR1019990032414A

公开(公告)日：1999-05-15

申请号：KR1019970053455

申请日：1997-10-17

Applicant: 한국전자통신연구원

Inventor： 윤일지 ,  윤석진 ,  조창식

IPC: G06F3/14

Abstract: 본 발명은 에이취.티.엠.엘(HTML) 문서 작성을 위한 테이블 태그 자동 생성 방법에 관한 것으로서, 본 발명에서 제공하는 테이블 태그 자동 생성 방법은 인터넷 서비스의 일종인 웹 서비스 상에 보여주기 위한 형태로 사각형의 이미지 또는 텍스트 영역을 편집하는 제 1 단계와, 상기 제 1 단계에서 생성된 이미지 또는 텍스트 영역의 각 변을 따라 직선을 그어 테이블을 생성하는 제 2 단계와, 상기 제 2 단계에서 생성된 테이블을 좌에서 우로 한줄씩 스캐닝하면서 테이블 태그를 생성하는 제 3 단계로 구성되어, 화면 상에 어떤 식으로 위치한 HTML 요소들도 편집 시 화면 상에 나타난 모습과 일치된 결과가 나올 수 있도록 HTML 문서를 생성해 낼 수 있으므로, 보다 더 편집이 자유스러운 위지윅/ 드래그-앤-드롭 방식의 HTML 편집기를 만들어 낼 수 있으며, 본 발명 을 채택한 편집기를 이용하면 테이블 태그를 자동으로 생성해 줌으로써, 테이블 태그에 대해서 알지 못하는 사용자라도 쉽게 고급스러운 HTML 문서를 만들 수 있다는 장점이 있다.