专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO1989012868A1 METHOD AND APPARATUS FOR READING CHARACTERS 审中-公开
标题翻译：读取字符的方法和装置
公开(公告)号：WO1989012868A1
公开(公告)日：1989-12-28
申请号：PCT/US1989002393
申请日：1989-06-01
申请人： NCR CORPORATION
发明人： NCR CORPORATION , HO, Benedict, Chuck-Min , FRANKLIN, Gene, Robert
IPC分类号： G06K09/18
CPC分类号： G06K9/183 , G06K7/0163 , G06K7/084
摘要： In a method and apparatus for use in reading magnetic characters formed by bars separated by short intervals or long intervals, for example CMC7 characters, the interval between adjacent bars of a character is obtained by calculating the zero crossing point between a peak of a first polarity and a peak of a second polarity associated with a bar in the associated scanning signal. The zero crossing point may be regarded as the midpoint of the associated bar. The intervals are then determined by obtaining the distance or interval between the zero crossing points of adjacent bars. Characters are then recognized on the basis of the thus determined intervals.
摘要翻译：在用于读取由以短间隔或长间隔分隔的条形成的磁性字符（例如CMC7字符）的方法和装置中，通过计算第一极性的峰值之间的过零点，获得字符的相邻条之间的间隔以及与相关联的扫描信号中的条相关联的第二极性的峰值。零交叉点可以被认为是相关联的条的中点。然后通过获得相邻条的过零点之间的距离或间隔来确定间隔。然后基于如此确定的间隔来识别字符。

2. 发明申请

WO1996039677A1 METHOD AND APPARATUS FOR CHARACTER RECOGNITION OF HAND-WRITTEN INPUT 审中-公开
标题翻译：用于手写输入字符识别的方法和装置
公开(公告)号：WO1996039677A1
公开(公告)日：1996-12-12
申请号：PCT/US1996005884
申请日：1996-04-26
申请人： MOTOROLA INC.
发明人： MOTOROLA INC. , PARTHASARATHY, Kannan
IPC分类号： G06K09/18
CPC分类号： G06K9/6857
摘要： A method and apparatus for recognition of hand-written input is disclosed where hand-written input composed of a sequence of (x, y, pen) points (125) is preprocessed into a sequence of strokes (122). A short list of candidate characters that are likely matches for the hand-written input is determined by finding a fast matching distance (600) between the input sequence of strokes and a sequence of strokes representing each candidate character of a large character set (160) where the sequence of strokes for each candidate character is derived from statistical analysis of empirical data. A final sorted list of candidate characters which are likely matches for the hand-written input (106) is determined by finding a detailed matching distance between the input sequence of strokes and the sequence of strokes for each candidate character of the short list (1000). A final selectable list of candidate characters is presented to a user.
摘要翻译：公开了一种用于识别手写输入的方法和装置，其中由（x，y，笔）序列（125）组成的手写输入被预处理成笔画序列（122）。通过在输入的笔画序列和表示大字符集（160）的每个候选字符的笔画序列之间找到快速匹配距离（600）来确定手写输入可能匹配的候选字符的简短列表，其中每个候选人物的笔画序列是从经验数据的统计分析中得出的。对于手写输入（106）可能匹配的候选字符的最终排序列表通过在短列表（1000）的每个候选字符之间找到笔画的输入序列和笔画序列之间的详细匹配距离来确定，。向用户呈现候选人物的最终可选列表。

3. 发明申请

WO1991006923A1 RATIO DETECTION CHARACTER RECOGNITION APPARATUS 审中-公开
标题翻译：比例检测字符识别装置
公开(公告)号：WO1991006923A1
公开(公告)日：1991-05-16
申请号：PCT/US1990006134
申请日：1990-10-23
申请人： UNISYS CORPORATION , CANU, Jean, Pierre , DESFORGES, Bernard , HEUTTE, Gérard
发明人： UNISYS CORPORATION
IPC分类号： G06K09/18
CPC分类号： G06K9/186
摘要： An apparatus for recognising characters (20) on a document (10) comprises a scanning head (24) and decoding circuits (78, 72, 86, 92, 80), feeding a controller (84). The controller (84) detects the identity of the characters (20A - 20N) by calculating the ratio found by dividing the time between a first positive pulse (32) and a second positive pulse (34) by the time between the first positive pulse (32) and a last negative pulse (36) and comparing it against a stored table of ratios.
摘要翻译：用于识别文档（10）上的字符（20）的装置包括扫描头（24）和解码电路（78,72,86,92,80），馈送控制器（84）。控制器（84）通过计算通过将第一正脉冲（32）和第二正脉冲（34）之间的时间除以第一正脉冲（32）和第二正脉冲（34）之间的时间而求出的比率来检测字符（20A-20N） 32）和最后一个负脉冲（36），并将其与存储的比率表进行比较。

4. 发明申请

WO1989006018A1 METHOD AND APPARATUS FOR READING CHARACTERS 审中-公开
标题翻译：读取字符的方法和装置
公开(公告)号：WO1989006018A1
公开(公告)日：1989-06-29
申请号：PCT/US1988004311
申请日：1988-12-05
申请人： NCR CORPORATION
发明人： NCR CORPORATION , MAZUMDER, Ali, Tazammul
IPC分类号： G06K09/18
CPC分类号： G06K9/186
摘要： In a method and apparatus for recognizing MICR characters having positive peak values, negative peak values, and substantially zero values occurring at predetermined locations along the associated waveform, a first count derived from the speed of the document moving past the associated reader (10) is used for deriving nominal rate clocking pulses for clocking the peak and zero values into storage registers (DR-1 to DR-8) associated with the apparatus. The occurrences of positive and negative peak values in the waveform are used to adjust the first count depending upon whether the MICR characters are moving faster or slower than a nominal rate, with the adjustment being used to provide an adjusted count which is used for determining subsequent clocking pulses.
摘要翻译：在用于识别具有正峰值，负峰值和基本上零值的MICR字符的方法和装置中，沿相关联的波形在预定位置出现，从文档移动通过相关阅读器（10）的速度导出的第一计数是用于导出标称速率时钟脉冲，用于将峰值和零值计时到与设备相关联的存储寄存器（DR-1至DR-8）中。波形中正和负峰值的出现用于根据MICR字符是否比标称速率更快或更慢地移动来调整第一计数，其中调整用于提供用于确定后续的调整计数时钟脉冲。

5. 发明申请

WO1986003314A1 PROCESS AND ARRANGEMENTS FOR THE IDENTIFICATION MARKING AND RECOGNITION OF OBJECTS 审中-公开
标题翻译：识别标识和识别对象的过程和安排
公开(公告)号：WO1986003314A1
公开(公告)日：1986-06-05
申请号：PCT/EP1985000648
申请日：1985-11-27
申请人： KAPPNER, Helmut, A.
IPC分类号： G06K09/18
CPC分类号： G06K7/10871 , G06K9/18
摘要： A process for recognizing objects to which a permanent coding (23), has been applied consisting of segmented symbols in plain characters (30-36), in which further location symbols (24, 25) are added to the object, which indicate a predetermined position of the coding. The coding (23) and the location symbols (24, 25) are picked up by an electronic camera, the output signals of which is fed to a picture processing computer (14). For this, scanning masks (30, 31) are produced which, inside the picture received, derive the anticipated position of the location symbols (24, 25) and whose surface projects over those of the location symbols (24, 25). Those symbols which are located in the masks (30, 31) are recognized as location symbols (24, 25), and their co-ordinates are computed. In this way the precise position of the coding (23) is determined and further masks (44) scan the coding (23).
摘要翻译：用于识别已经应用了包括平原字符（30-36）中的分段符号的永久编码（23）的对象的过程，其中另外的位置符号（24,25）被添加到对象，其指示预定的编码的位置。编码（23）和位置符号（24,25）由电子照相机拾取，电子照相机的输出信号被馈送到图像处理计算机（14）。为此，产生扫描掩模（30,31），其在所接收的图像内部导出位置符号（24,25）的预期位置，并且其表面突出于位置符号（24,25）的位置符号。位于掩模（30,31）中的那些符号被识别为位置符号（24,25），并且它们的坐标被计算。以这种方式，确定编码（23）的精确位置，并且进一步掩码（44）扫描编码（23）。

6. 发明申请

WO2004038650A1 ORIENTATION-INDICATING CYCLIC POSITION CODES 审中-公开
标题翻译：方位指示循环位置代码
公开(公告)号：WO2004038650A1
公开(公告)日：2004-05-06
申请号：PCT/AU2003/001343
申请日：2003-10-10
申请人： SILVERBROOK RESEARCH PTY LTD , LAPSTUN, Paul
发明人： LAPSTUN, Paul
IPC分类号： G06K09/18
CPC分类号： G06K19/06075 , G06K7/14 , G06K7/1421 , G06K7/1456 , G06K7/1473 , G06K19/06009 , G06K19/06037 , G06K19/06168
摘要： Machine-readable coded data disposed on or in a substrate in accordance with a layout, the layout having at least order n rotational symmetry, where n is at least two, the layout encoding an orientation codeword comprising a sequence of an integer multiple m of of n symbols, where m is one or more, each encoded symbol being distributed at n locations about a center of rotational symmetry of the layout such that decoding the symbols at each of the n orientations of the layout produces n representations of the orientation codeword, each representation comprising a different cyclic shift of the orientation codeword and being indicative of the degree of rotation of the layout, and wherein the orientation codeword is fault tolerant.
摘要翻译：根据布局布置在基板上或基板中的机器可读编码数据，布局具有至少n阶旋转对称性，其中n至少为2，编码定向码字的布局，其包括整数倍m的序列 n个符号，其中m是一个或多个，每个编码符号分布在围绕布局的旋转对称中心的n个位置处，使得对布局的n个取向中的每个n个取向处的符号的解码产生定向码字的n个表示，每个表示包括定向码字的不同循环移位并且指示布局的旋转程度，并且其中定向码字是容错的。

7. 发明申请

WO1991006922A1 CHARACTER RECOGNITION APPARATUS 审中-公开
标题翻译：字符识别装置
公开(公告)号：WO1991006922A1
公开(公告)日：1991-05-16
申请号：PCT/US1990006133
申请日：1990-10-23
申请人： UNISYS CORPORATION , CANU, Jeane, Pierre , DESFORGES, Bernard , HEUTTE, Gérard
发明人： UNISYS CORPORATION
IPC分类号： G06K09/18
CPC分类号： G06K9/186
摘要： A character recognition apparatus uses a magnetic head to read magnetic printed coded characters from a passing cheque (10). A head (24) using a circuit including a delay line (58) drives a controller (76) which, should a character not instantly be recognised by matching against templates, first varies the presumed speed by plus or minus 10% to see if a match can be found, then suppresses each detected edge in the character, in turn, to determine whether or not a match can be found. Each edge is used to create subsequent timing windows wherein a subsequent edge may be detected. The use of timing windows suppresses erroneous edge detection.

8. 发明申请

WO1984000631A1 COLOR CODED SYMBOLIC ALPHANUMERIC SYSTEM 审中-公开
标题翻译：颜色编码符号铝质系统
公开(公告)号：WO1984000631A1
公开(公告)日：1984-02-16
申请号：PCT/US1982001011
申请日：1982-07-22
申请人： KREMPEL, Ralf
IPC分类号： G06K09/18
CPC分类号： G09B17/00
摘要： The KREMPEL CODE is a VISUAL COMMUNICATION SYSTEM, utilizing in a prearranged sequence various markings in different colors, to form an informational code similar to a written language.
摘要翻译： KREMPEL CODE是一种VISUAL COMMUNICATION SYSTEM，以预先安排的顺序，以不同颜色的各种标记使用，以形成类似于书面语言的信息代码。

9. 发明申请

WO1993004441A1 A PREMAGNETIZING HEAD 审中-公开
标题翻译：预备头
公开(公告)号：WO1993004441A1
公开(公告)日：1993-03-04
申请号：PCT/SE1992000549
申请日：1992-08-14
申请人： ICL DATA AB , STIERNSPETZ, Allan
发明人： ICL DATA AB
IPC分类号： G06K09/18
CPC分类号： G06K9/186 , H01F13/00
摘要： The present invention relates to a premagnetizing head for charging the magnetizable pigment of characters printed on a document. The head includes a permanent magnet (10) whose one pole is connected to a magnetically conductive metal plate (12) which extends away from the magnet (10) in a direction generally transversal to the north-south pole axis of the magnet and up to the document (D) whose characters are to be magnetized and read. The permanent magnet (10) is oriented so that its northpole-southpole axis will extend generally parallel with the plane of the document (D) to be read.
摘要翻译：本发明涉及一种用于对打印在文件上的字符的可磁化颜料进行充电的预磁化头。头部包括一个永磁体（10），其一极连接到导电金属板（12），导电金属板（12）沿着大体横向于磁体的南北极轴线的方向远离磁体（10）延伸，直到文件（D）的字符将被磁化和读取。永磁体（10）被定向成使得其北极 - 南极轴线将大致平行于文件（D）的平面延伸以被读取。

10. 发明申请

WO1991011780A1 CORRELATED MASKING PROCESS FOR DESKEWING, FILTERING AND RECOGNITION OF VERTICALLY SEGMENTED CHARACTERS 审中-公开
标题翻译：用于垂直分割字符的去除，过滤和识别的相关掩蔽过程
公开(公告)号：WO1991011780A1
公开(公告)日：1991-08-08
申请号：PCT/US1991000652
申请日：1991-01-31
申请人： EASTMAN KODAK COMPANY
发明人： EASTMAN KODAK COMPANY , GABORSKI, Roger, Stephen
IPC分类号： G06K09/18
CPC分类号： G06K9/183
摘要： Vertically segmented characters are pre-processed for optical character recognition so as to remove skew and noise by convolution with a mask consisting of parallel vertical lines with a one-pixel spatial frequency. The horizontal alignment of the mask with respect to each segmented character is adjusted until a maximum cross-correlation is found. Then, the product of the mask and the character image is formed. A histogram of the number of pixels in the product image along each vertical line of the mask is compared with a library of such histograms of known symbols and the best match identifies the symbol represented by the character image. Alternatively, the histogram is converted to a binary codework which is compared with a library of such codewords corresponding to a set of known symbols.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式