-
公开(公告)号：KR1020050040638A
公开(公告)日：2005-05-03
申请号：KR1020030076099
申请日：2003-10-29
Applicant: 한국전자통신연구원
Abstract: 본 발명은 음성지원 브라우저의 웹 문서 본문 추출방법에 관한 것으로, 컨텐츠 변환기를 구비하는 음성지원 브라우저의 HTML 문서에서 핵심이 되는 부분의 문장을 자동으로 추출하는 본문 추출부에 의해 추출된 문장만을 음성으로 출력하여 사용자에게 제공하는 것이다.
본 발명의 HTML DOM 문서에서 핵심이 되는 본문 문단을 추출하는 방법은, 상기 HTML DOM 문서의 최상위 노드에서 자식 노드가 존재하는지를 판단하여, 상기 자식 노드가 존재하는 경우, 상기 자식 노드의 WIDTH 속성 값, 모든 문자수의 합(TN), 문자열, 평균 문자수(MT) 및 상기 자식 노드의 문자열에서 제목과 일치하는 단어의 개수(TM)를 구한 후, 상기 구해진 WIDTH, TN, MT, TM을 이용하여 가중치 값을 계산하고, 상기 구해진 가중치 값이 가장 큰 노드를 선택하여, 후보 노드에 추가하고, 상기 후보 노드에 자식 노드가 있는 경우, 상기한 단계를 반복 실행한다.
한편, 상기 후보 노드나 최상위 노드에 자식노드가 없는 경우에는, 최적의 노드를 선택하여, 하위 문장을 추출한 후, VoiceXML 파일을 생성한다.-
公开(公告)号：KR1020050035784A
公开(公告)日：2005-04-19
申请号：KR1020030071567
申请日：2003-10-14
Applicant: 한국전자통신연구원
Abstract: 본 발명은 HTML 컨텐츠 변환을 통한 음성지원 웹 브라우저 및 음성지원 방법에 관한 것으로, 사용자가 휴대단말 등에서 웹 문서를 검색할 때 시각뿐만 아니라 음성 등 다양한 사용자 인터페이스가 가능하도록 함으로써 보행 중이나 이동 중에도 화면에 구애받지 않고 웹 검색을 할 수 있다.
특히, 본 발명의 웹 브라우저는 문서를 다운로드 하는 페이지 로더, 상기 페이지 로더에서 다운로드 된 문서가 HTML(Hyper Text Makeup Language) 문서일 경우, 상기 문서를 파싱하여 HTML DOM(Document Object Model)을 생성하는 HTML 파서, 상기 DOM으로부터 음성지원이 필요한 부분을 추출하고, 문서의 상관관계에 따른 음성인식 그래머를 생성하여 VoiceXML 문서를 생성하는 컨텐츠 변환기, 상기 VoiceXML(Voice eXtensible Markup Language) 문서에서 문서를 파싱하여 상기 문서의 내용을 TTS(Text to Speech)로 음성출력하고, 음성인식에 필요한 그래머(Grammar)와 사용자로부터 입력된 음성 명령어를 음성인식기에 전달하고, 상기 음성인식기로부터 인식된 단어를 넘겨받아서 처리하는 VoiceXML 해석기 및 상기 VoiceXML 해석기에서 넘겨받은 문장을 음성으로 출력하는 TTS(Text to Speech)와 상기 그래머에 의해 인식된 단어를 상기 VoiceXML 해석기로 전달하는 음성인식기를 포함하여 구성되는 것을 특징으로 한다.-
公开(公告)号：KR100463838B1
公开(公告)日：2004-12-29
申请号：KR1020020079175
申请日：2002-12-12
Applicant: 한국전자통신연구원
IPC: H04L12/16
Abstract: PURPOSE: A method for downloading a program of a bootloader for a UIA(Universal Information Appliance) system and supporting NFS(Network File System) development environment is provided to offer the better download function than a serial port by using network. CONSTITUTION: A UIA bootloader initializes a system(S10) and detects whether a CS89001 chip is provided and operated in a development board of a UIA system development board(S11). The UIA bootloader sets an MAC(Media Access Control) address of the CS8900A of the development board of the UIA system, a name of a program to be downloaded and a server directory by using an environment setup command(S12). The UIA bootloader performs a command for driving a BOOTP to be allocated an IP address from a DHCP(Dynamic Host configuration Protocol) server(S13). The UIA bootloader downloads a user's desired program to a memory of the UIA system through network from the server(S14) and stores it in a designated flash memory region(S15).
Abstract translation: 目的：提供一种下载UIA（通用信息设备）系统的引导加载程序并支持NFS（网络文件系统）开发环境的方法，以提供比使用网络的串行端口更好的下载功能。 构成：UIA引导程序初始化系统（S10）并检测是否在UIA系统开发板（S11）的开发板上提供并操作了CS89001芯片。 UIA引导加载程序通过使用环境设置命令（S12）设置UIA系统的开发板的CS8900A的MAC（媒体访问控制）地址，要下载的程序的名称和服务器目录。 UIA引导装载程序执行用于从DHCP（动态主机配置协议）服务器驱动BOOTP以分配IP地址的命令（S13）。 UIA自举程序从服务器通过网络将用户期望的程序下载到UIA系统的存储器（S14），并将其存储在指定的闪存区域中（S15）。
-
公开(公告)号：KR1020040056700A
公开(公告)日：2004-07-01
申请号：KR1020020083245
申请日：2002-12-24
Applicant: 한국전자통신연구원
IPC: G06F3/02
Abstract: PURPOSE: A software keyboard and a software key-in method are provided to discriminate frequently used keys from non-frequently used keys, to always display the frequently used keys, and to display the remaining keys whenever necessary according to their purpose. CONSTITUTION: The keyboard comprises a keyword selection group(110), an indispensable key selection group(120), a number selection group(130), and two trays(140, 150). The keyword selection group(110) includes a Hangul basis selection key(111), an alphabet basis key(112), a Hangul/alphabet key(113) and a user definition selection key(114). The keyboard can be combined according to a user inclination, and the combined keyboard is displayed with the two trays(140, 150). The indispensable key selection group(120) includes a tab(121), a back space(122), a space(123), an enter(124), a page up(125), a page down(126) and an escape(127).
Abstract translation: 目的：提供软件键盘和软件键入方法，用于将经常使用的键与非常用的键进行区分，以便始终显示常用的键，并根据其目的在必要时显示剩余的键。 构成：键盘包括关键字选择组（110），不可缺少的键选择组（120），数字选择组（130）和两个托盘（140,150）。 关键词选择组（110）包括韩文基础选择键（111），字母表键（112），韩文/字母键（113）和用户定义选择键（114）。 键盘可以根据用户的倾向进行组合，组合键盘与两个托盘（140,150）一起显示。 不可或缺的按键选择组（120）包括标签（121），后部空间（122），空间（123），输入（124），页面向上（125），下翻页（126）和转义 （127）。
-
公开(公告)号：KR1020040051276A
公开(公告)日：2004-06-18
申请号：KR1020020079175
申请日：2002-12-12
Applicant: 한국전자통신연구원
IPC: H04L12/16
CPC classification number: H04L67/1097 , G06F9/4401
Abstract: PURPOSE: A method for downloading a program of a bootloader for a UIA(Universal Information Appliance) system and supporting NFS(Network File System) development environment is provided to offer the better download function than a serial port by using network. CONSTITUTION: A UIA bootloader initializes a system(S10) and detects whether a CS89001 chip is provided and operated in a development board of a UIA system development board(S11). The UIA bootloader sets an MAC(Media Access Control) address of the CS8900A of the development board of the UIA system, a name of a program to be downloaded and a server directory by using an environment setup command(S12). The UIA bootloader performs a command for driving a BOOTP to be allocated an IP address from a DHCP(Dynamic Host configuration Protocol) server(S13). The UIA bootloader downloads a user's desired program to a memory of the UIA system through network from the server(S14) and stores it in a designated flash memory region(S15).
Abstract translation: 目的：为UIA（通用信息设备）系统下载程序，支持NFS（网络文件系统）开发环境，提供了一种比串行端口更好的下载功能的方法。 规定：UIA引导加载程序初始化系统（S10），并检测在UIA系统开发板的开发板中是否提供和操作CS89001芯片（S11）。 UIA引导加载器通过使用环境设置命令来设置UIA系统的开发板的CS8900A的MAC（媒体访问控制）地址，要下载的程序的名称和服务器目录（S12）。 UIA引导加载程序执行用于驱动从DHCP（动态主机配置协议）服务器分配IP地址的BOOTP的命令（S13）。 UIA引导装载机通过来自服务器的网络将用户期望的程序下载到UIA系统的存储器（S14），并将其存储在指定的闪存区域（S15）中。
-
Patent Agency Ranking
公开(公告)号：KR1020040046171A
公开(公告)日：2004-06-05
申请号：KR1020020074009
申请日：2002-11-26
Applicant: 한국전자통신연구원
IPC: G06F17/27
CPC classification number: G06F17/272 , G06F17/2247 , G06F17/30905 , G06F17/3092
Abstract: PURPOSE: A system and a method for parsing an element based web document are provided to integrate/browse the contents written in various markup languages offered from the wire/wireless web sites with no restriction in a specification of a portable terminal. CONSTITUTION: A vocabulary parser(310) separates tokens based on the markup and the non-markup by referring to a token table(311) for all markup data required for a document type to be supported. A syntax parser(320) generates a tree-base object based on a GUI(Graphic User Interface) of the terminal after parsing each syntax by parsing a contents model(321) based on a DTD(Document Type Definition) of each document. The vocabulary parser includes a comment parser processing a comment or a blank, a markup start tag parser generating the token by recognizing a markup start tag, an attribute parser generating the token by parsing an attribute, and a general user data parser generating the token by parsing general user data.
Abstract translation: 目的：提供用于解析基于元素的web文档的系统和方法，以便在便携式终端的规格中没有限制地整合/浏览由有线/无线网站提供的各种标记语言编写的内容。 构成：词汇解析器（310）通过参考用于要支持的文档类型所需的所有标记数据的令牌表（311），基于标记和非标记来分离令牌。 语法解析器（320）基于每个文档的DTD（文档类型定义）解析内容模型（321）来解析每个语法之后，基于终端的GUI（图形用户界面）生成树基对象。 词汇解析器包括处理注释或空白的注释解析器，通过识别标记开始标签生成令牌的标记开始标签解析器，通过解析属性生成令牌的属性解析器以及生成令牌的通用用户数据解析器 解析一般用户数据。
-
公开(公告)号：KR100424349B1
公开(公告)日：2004-03-24
申请号：KR1020010080722
申请日：2001-12-18
Applicant: 한국전자통신연구원
IPC: H04B7/26
Abstract: The invention relates to a call control method for handoff guarantee in a mobile terminal for minimizing handoff failure rate which continuously guarantees the same quality of service is during handoff of an existing call from the current cell into a new cell in a high-speed radio communication network which is constituted of micro-cells to support multimedia traffic. The bandwidth reservation is performed in the adjacent cells, call approval control is performed by temporarily storing a handoff request into a queue when an available bandwidth is insufficient, and a radio bandwidth is dynamically managed so as to efficiently use communication network resources. Further, new call block rate and handoff failure rate are periodically monitored for adaptation to dynamic variation of the communication network traffic state, by which the amount of a reserved bandwidth is dynamically adjusted according to the state so as to efficiently use communication network resources.
Abstract translation: 本发明涉及一种用于移动终端中的切换保证的呼叫控制方法，用于最小化切换失败率，该切换失败率连续地保证在将来自当前小区的现有呼叫切换到高速无线电通信中的新小区期间相同的服务质量 网络由微蜂窝构成以支持多媒体业务。 在相邻小区中执行带宽预留，当可用带宽不足时，通过将切换请求临时存储到队列中来执行呼叫批准控制，并且动态管理无线电带宽以有效地使用通信网络资源。 此外，为了适应通信网络业务状态的动态变化，周期性地监视新的呼叫阻塞率和切换失败率，通过该变化，根据状态动态地调整预留带宽的量，以便有效地使用通信网络资源。
-
公开(公告)号：KR1020020019200A
公开(公告)日：2002-03-12
申请号：KR1020000052348
申请日：2000-09-05
Applicant: 한국전자통신연구원
IPC: G06F15/02
Abstract: PURPOSE: An agent and method for automatically installing PDA(Personal Digital Agent) application software is provided to install an instant update agent, executing simple and repeated work instead of a user, at an individual user terminal. CONSTITUTION: If a power switch is turned on(300), an agent installed at a user PDA, the functions of a WIM(Web Information Manager), a CM(Communication Manager). An UPM(User Preference Manager), and a SIM(Software Installation Manager) are executed respectively in parallel(310). The CM tests a message confirmed state(331). If information that new information has been posted on the web arrives from the WIM, the CM loads option information stored in a user terminal(333). In case that it is information corresponding to an alarm function, the CM informs the user of it(335). If an installation instruction is given by the user(337), the CM gives a downloading and installation instruction to the SIM(336).
Abstract translation: 目的：提供用于自动安装PDA（个人数字代理）应用软件的代理和方法，以安装即时更新代理，在单个用户终端上执行简单而重复的工作而不是用户。 规定：如果电源开关打开（300），安装在用户PDA上的代理，WIM（Web信息管理器），CM（通信管理器）的功能。 分别执行UPM（用户偏好管理器）和SIM（软件安装管理器）（310）。 CM测试消息确认状态（331）。 如果网络上发布新信息的信息从WIM到达，则CM加载存储在用户终端中的选项信息（333）。 在与报警功能对应的信息的情况下，CM通知用户（335）。 如果用户给出安装说明（337），则CM向SIM（336）提供下载和安装指令。
-
公开(公告)号：KR100296704B1
公开(公告)日：2001-08-07
申请号：KR1019970065675
申请日：1997-12-03
Applicant: 한국전자통신연구원
IPC: G06F17/30
Abstract: PURPOSE: A multimedia computer based on a network and a method for controlling a variable structure using the same are provided to transmit, store, and control multimedia data by connecting individual multi media hardware devices to a large-sized capacity storage as a hard disk on a high-speed network and to enhance a processing performance of a system by connecting additional hardware devices to the network at any time. CONSTITUTION: A main module(10) comprises the Strong ARM RISC processor of the digital company which is a low-power and high performance central processor(10.1), a high speed network connection device(10.2), and a memory(10.3). A video module(11) comprises the ARM 7500 RISK processor which is a low-power central processor(11.1), a high speed network connection device(11.2), a video controller(11.3), and a camera(11.4). A graphic module(13) comprises a low-power central processor(12.1), a high speed network connection device(12.2), a graphic controller(12.3), and a monitor(12.4). An audio module(13) comprises a low-power central processor(13.1), a high speed network connection device(13.2), an audio controller(13.3), a mike(13.4), and a speaker(13.5). Each module uses a low-power central processor and a high speed network connection device being operated in a power of 3.3 volt commonly. Each module is connected to a high speed network through a built-in high speed network connection device.
-
公开(公告)号：KR1020000026691A
公开(公告)日：2000-05-15
申请号：KR1019980044349
申请日：1998-10-22
Applicant: 한국전자통신연구원
IPC: H04N5/30
Abstract: PURPOSE: An intelligent CCD(charge coupled device) camera is to capture only a required moment and compact an image of the moment and transmits the image to a central controlling computer, thereby facilitating a maintenance thereof. CONSTITUTION: An intelligent CCD camera for monitoring and recording comprises a plurality of image capture units which are disposed at a place to obtain an image of a correspondent monitoring area at a constant time interval and compares the images having the constant time interval and sorts out a difference between the images, and a central controlling part which is connected with the image obtaining parts to remote-control the image obtaining parts and receives/stores images only selected by the image obtaining parts. The image obtaining part comprises a CCD camera(211) for monitoring an area at a constant time interval, a signal processor(204) for controlling an image selecting process, an image processing unit for converting an analog image from the CCD camera into a digital image and transmitting the digital image to the signal processor.
Abstract translation: 目的：智能CCD（电荷耦合器件）摄像机仅捕获所需时刻并压缩瞬间的图像，并将图像传输到中央控制计算机，从而便于维护。 构成：用于监视和记录的智能CCD摄像机包括多个图像捕获单元，其被设置在一个位置以获得恒定时间间隔的相应监视区域的图像，并且比较具有恒定时间间隔的图像并对其进行排序 图像之间的差异，以及与图像获取部件连接以远程控制图像获取部分并且接收/存储仅由图像获取部件选择的图像的中央控制部件。 图像获取部分包括用于以恒定时间间隔监视区域的CCD照相机（211），用于控制图像选择处理的信号处理器（204），用于将来自CCD照相机的模拟图像转换为数字图像的图像处理单元 图像并将数字图像发送到信号处理器。
-
-
-
-
-
-
-
-
-
Search Results
194
records
(took 0.026 seconds)
