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. 发明公开

EP3258422A1 CHARACTER SEGMENTATION AND RECOGNITION METHOD 审中-公开
标题翻译：特征分割和识别方法
公开(公告)号：EP3258422A1
公开(公告)日：2017-12-20
申请号：EP15881706.4
申请日：2015-06-26
申请人： GRG Banking Equipment Co., Ltd.
发明人： WANG, Weifeng , QIU, Xinhua , WANG, Rongqiu , WANG, Kun
IPC分类号： G06K9/20
CPC分类号： G06K9/344 , G06K9/346 , G06K9/348 , G06K2209/01
摘要： A character segmentation and recognition method, which is used for resolving the problems that the prior-art segmentation method is low in capacity for recognizing characters under a complex background and poor in contamination interference resistance. The method comprises: collecting image data to obtain an image to be recognized; locating a character-row candidate area on the image to be recognized; acquiring preset character-row prior information, wherein the character-row prior information comprises the number of characters, a character pitch and a character size; acquiring a corresponding segmentation point template according to the character-row prior information; acquiring reliability of the segmentation point template on different positions when the character-row candidate area is traversed; determining a position with the highest reliability as an optimum segmentation position; segmenting the character-row candidate area according to the segmentation point template and the optimum segmentation position to obtain a plurality of single character areas; and performing character recognition on the single character area to obtain a corresponding recognition result.
摘要翻译：一种字符分割识别方法，用于解决现有分割方法在复杂背景下字符识别能力较低，抗污染干扰能力较差的问题。该方法包括：收集图像数据以获得待识别的图像; 在要识别的图像上定位字符行候选区域; 获取预设的字符行先验信息，所述字符行先验信息包括字符数，字符间距和字符大小; 根据所述字符行先验信息获取对应的分割点模板; 当遍历所述字符行候选区域时，获取所述分割点模板在不同位置的可靠性; 确定具有最高可靠性的位置作为最佳分割位置; 根据分割点模板和最佳分割位置对字符行候选区域进行分割，得到多个单个字符区域; 并对单个字符区域进行字符识别以获得相应的识别结果。

2. 发明公开

EP2214124A2 Character recognition device 审中-公开
标题翻译： Zeichenerkennungsvorrichtung
公开(公告)号：EP2214124A2
公开(公告)日：2010-08-04
申请号：EP10160939.4
申请日：2009-03-30
申请人： Fujitsu Frontech Limited
发明人： EGUCHI, Shinichi , KAWASHIMA, Hajime , KANAMOTO, Kouichi , HASEGAWA, Shohei , KOBARA, Katsutoshi , YABUKI, Maki
IPC分类号： G06K9/34 , G06K9/20
CPC分类号： G06K9/346 , G06K9/00449 , G06K2209/01 , G06K2209/015
摘要： A character frame specification device for specifying a character frame in a form comprises: a line segment extracting unit for extracting a line segment corresponding to a character frame from image data in the form; a space table generation unit for calculating a space between two arbitrary vertical lines in each of all combinations in the case where the two arbitrary vertical lines are selected from respective vertical lines of a line segment extracted by the line segment extraction unit and generating a space table indicating the calculated spaces; a vote table generation unit for counting the number of spaces which are the same from among the spaces indicated in the space table, and generating a vote table indicating the total count value for each of the spaces; a space assumption unit for assuming a space whose total number indicated in the vote table is largest as the space of the character frame; a space modification unit for modifying the or each space in a shape pattern of a character frame registered in advance to the space assumed by the space assumption unit; and a specification unit for matching the shape pattern of the character frame whose space is modified by the space modification unit with a pattern of a line segment extracted by the line segment extraction unit and specifying a character frame on the basis of a result of the pattern matching.
摘要翻译：用于指定形式的字符帧的字符帧指定装置包括：线段提取单元，用于从形式的图像数据中提取与字符帧相对应的线段; 空间表生成单元，用于在从线段提取单元提取的线段的各垂直线中选择两条任意的垂直线的情况下，计算所有组合中的两个任意垂直线之间的间隔，并生成空格表指示计算的空间; 投票表生成单元，用于从空格表中指示的空格中对相同的空格数进行计数，并生成表示每个空格的总计数值的投票表; 用于假设投票表中指示的总数最大的空格的空间假设单元作为字符框的空格; 空间修改单元，用于将预先登记的字符帧的形状图案中的每个空间修改为由空间假设单元假设的空间; 以及规格单元，用于将由空间修改单元修改的空格的字符框的形状图案与由线段提取单元提取的线段的图案进行匹配，并基于图案的结果指定字符帧匹配。

3. 发明公开

EP1842154A2 METHOD AND SYSTEM FOR IDENTIFYING ILLUMINATION FLUX IN AN IMAGE 有权转让
标题翻译： VERFAHREN UND SYSTEM ZUR IDENTIFIZIERUNG DES BELEUCHTUNGSFLUSSES IN EINEM BILD
公开(公告)号：EP1842154A2
公开(公告)日：2007-10-10
申请号：EP06733976.2
申请日：2006-01-27
申请人： Tandent Vision Science, Inc.
发明人： FRIEDHOFF, Richard, Mark , MAXWELL, Bruce, Allen , SMITH, Casey, Arthur
IPC分类号： G06K9/48
CPC分类号： G06K9/4661 , G06K9/2018 , G06K9/346 , G06K9/4638 , G06K9/4652 , G06T7/12
摘要： In a first exemplary embodiment of the present invention, an automated, computerized method is provided for determining an illumination boundary in an image. The method comprises the steps of performing dynamic sampling in preselected local areas (fig. 4A element 202) of the image to determine spectral ratio information for the image at each of the preselected local areas (fig. 4A element 206), and utilizing the spectral ratio information to identify an illumination boundary (fig. 4A element 208).
摘要翻译：在本发明的第一示例性实施例中，提供了一种用于确定图像中的照明边界的自动计算机化方法。该方法包括以下步骤：在图像的预选局部区域中执行动态采样，以确定每个预选局部区域处的图像的频谱比信息，并利用频谱比信息来识别照明边界。

4. 发明公开

EP1668886A1 STATISTICAL SELF-CALIBRATING DETECTION AND REMOVAL OF BLEMISHES IN DIGITAL IMAGES 有权
标题翻译：统计自校准检测和数字图像点位的拆除
公开(公告)号：EP1668886A1
公开(公告)日：2006-06-14
申请号：EP04765121.1
申请日：2004-09-13
申请人： Fotonation Vision Limited
发明人： STEINBERG, Eran , PRILUTSKY, Yury , CORCORAN, Peter , BIGIOI, Petronel , ZAMFIR, Adrian , BUZULOIU, Vasile , URSU, Danutz , ZAMFIR, Marta
IPC分类号： H04N1/409 , H04N5/217
CPC分类号： H04N1/4097 , G06K9/346 , G06K2209/01 , H04N5/3675
摘要： A method automatically corrects dust artifact within images acquired by a system including a digital acquisition device including a lens assembly. Multiple original digital images are acquired with the digital acquisition device. Probabilities that certain pixels correspond to dust artifact regions within the images are determined based at least in part on a comparison of suspected dust artifact regions within two or more of the images. Probable dust artifact regions are associated with extracted parameter values relating to the lens assembly when the images were acquired. A statistical dust map is formed including mapped dust regions based on the determining and associating. Pixels corresponding to correlated dust artifact regions are corrected within further digitally-acquired images based on the associated statistical dust map.

5. 发明公开

EP1237116A3 Method and apparatus for detecting character frames using voting process 有权
标题翻译：弗拉基米尔和德黑兰
公开(公告)号：EP1237116A3
公开(公告)日：2004-12-15
申请号：EP02004286.7
申请日：2002-02-27
申请人： NEC CORPORATION
发明人： Masato, Hayashi , Tatsuo, Akiyama , Takeshi, Bashomatsu
IPC分类号： G06K9/32
CPC分类号： G06K9/00449 , G06K9/346 , G06K2209/01
摘要： In a character frame detecting method, edge pixels at edges of lines of character frames are extracted from gray pixels of a gray pixel image. Then, gray gradients (θ(i, j)) are calculated at the edge pixels. Then, voting regions (1201, 1202, 1203, 1204) with respect to the edge pixels are set in accordance with the gray gradients thereof. Then, voting processes are carried out by incrementing voting counts at coordinates within the voting regions. Finally, peak coordinates (PP(i, j)) having peak voting counts are detected from the coordinates, so that the peak coordinates are determined to be centers of the character frames (Fig. 8).
摘要翻译：在字符帧检测方法中，从灰度像素图像的灰度像素中提取字符帧的边缘边缘像素。然后，在边缘像素处计算灰度梯度（＆Thetas;（i，j））。然后，相对于边缘像素的投票区域（1201,1202,1203,1204）根据灰度梯度被设定。然后，通过在表决区域内的坐标上增加投票数来进行表决程序。最后，从坐标中检测具有峰值投票计数的峰值坐标（PP（i，j）），从而将峰值坐标确定为字符帧的中心（图8）。

6. 发明公开

EP1310904A1 METHOD FOR PROCESSING DOCUMENT, RECORDED MEDIUM ON WHICH DOCUMENT PROCESSING PROGRAM IS RECORDED AND DOCUMENT PROCESSOR 审中-公开
标题翻译：澳大利亚科学技术大学，澳大利亚新奥尔良，澳大利亚德黑兰共和国科学技术大学，澳大利亚科技大学
公开(公告)号：EP1310904A1
公开(公告)日：2003-05-14
申请号：EP01936835.6
申请日：2001-06-01
申请人： Japan Science and Technology Corporation
发明人： NAKAGAWA, Masaki
IPC分类号： G06K9/20 , G06K9/62
CPC分类号： G09G3/3413 , G06K9/346 , G06K2209/01 , G09G2310/0235 , G09G2320/0247
摘要： The color display according to the invention relates to a color display device and a color display method for color display by monochromatic display of data of respective colors in synchronism with sequential ON/OFF switching of three light sources for red, green and blue, respectively. In a preferred embodiment, the red, green and blue light sources are controlled to turn on and off in the order of red, green, blue and green in each unit period and to repeat the unit period.
摘要翻译：根据本发明的彩色显示器涉及彩色显示装置和彩色显示方法，用于通过分别与用于红色，绿色和蓝色的三个光源的顺序ON / OFF切换同步地通过单色显示各种颜色的数据进行彩色显示。在优选实施例中，红色，绿色和蓝色光源被控制为在每个单位周期内以红色，绿色，蓝色和绿色的顺序打开和关闭，并重复单位周期。

7. 发明授权

EP0567344B1 Method and apparatus for character recognition 失效
标题翻译：用于字符识别的方法和装置
公开(公告)号：EP0567344B1
公开(公告)日：2002-11-06
申请号：EP93303194.0
申请日：1993-04-23
申请人： CANON KABUSHIKI KAISHA
发明人： Wang, Shin-Ywan, c/o Canon Information Systems , Vaezi, Mehrzad R., c/o Canon Information Systems , Sherrick, Christopher A., c/o Canon Info. Systems
IPC分类号： G06K9/20 , G06K9/34
CPC分类号： G06K9/00463 , G06K9/00456 , G06K9/342 , G06K9/346 , G06K2209/01

8. 发明授权

EP0717365B1 Linear line detection apparatus using projection image of character string including linear line 失效
标题翻译：用于从线的含有投影图像中检测直线串设备
公开(公告)号：EP0717365B1
公开(公告)日：2002-02-27
申请号：EP95119237.6
申请日：1995-12-06
申请人： KABUSHIKI KAISHA TOSHIBA
发明人： Sano, Tikara, c/o K.K. Toshiba , Nakamura, Yoshikatu, c/o K.K. Toshiba
IPC分类号： G06K9/34
CPC分类号： G06K9/346 , G06K2209/01

9. 发明公开

EP0854434A1 Ruled line extracting apparatus for extracting ruled line from normal document image and method thereof 失效
标题翻译：装置和方法用于正常文档图像内的表线的提取
公开(公告)号：EP0854434A1
公开(公告)日：1998-07-22
申请号：EP97306162.5
申请日：1997-08-13
申请人： FUJITSU LIMITED
发明人： Katsuyama, Yutaka
IPC分类号： G06K9/20 , G06K9/34
CPC分类号： G06K9/346 , G06K2209/01
摘要： A ruled line extracting apparatus obtains circumscribed rectangles of pixel concatenation regions included in an input pattern, and calculates the most frequent value of their heights. Additionally, the apparatus integrates segments by ignoring a wild card segment, and calculates the most frequent value of height/width of extracted straight lines and segments structuring the straight line. Next, it performs a process for integrating/deleting straight lines using each threshold value based on the highest frequency value. Then, it checks/deletes a straight line according to a distribution of black pixels around the straight line, and recognizes the remaining straight lines as ruled line candidates.
摘要翻译：的格线提取装置在输入模式获得包括像素级联区域的外接矩形的，并且计算它们的高度的最频繁的值。另外，该装置通过忽略通配符段集成段，并计算所提取的直线和段构造的直线高度/宽度的最频繁的值。接着，它执行用于INTEGRA婷过程/删除使用基于所述最高频率值中的每个阈值的直线。然后，它检查/删除的直线gemäß到黑色像素的周围的直线的分布，并认识到剩余的直线为格线的候选者。

10. 发明公开

EP0814422A3 System for extracting attached text from a table-cell frame 失效
标题翻译：用于从表格单元框架中提取附加文本的系统
公开(公告)号：EP0814422A3
公开(公告)日：1998-01-28
申请号：EP97304087.6
申请日：1997-06-11
申请人： CANON KABUSHIKI KAISHA
发明人： Shin-Ywan, Wang
IPC分类号： G06K9/20
CPC分类号： G06K9/00449 , G06K9/346 , G06K2209/01
摘要： A method for identifying and extracting text data from a table-cell frame. The method includes the steps of tracing connected components of a document image, tracing white contours within a connected component, defining a frame outline based on the white contours, identifying unattached character data inside the frame outline, and defining an initial rectangular area inside the frame outline. The method further includes detecting black pixels in a horizontal or vertical direction from the initial rectangular area in order to create an extended character area, locating boundary pixels lying inside the extended character area for each white contour, identifying black pixels positioned between boundary pixels lying inside the extended character area, combining black pixels positioned between boundary pixels lying inside the extended character area so as to form at least one connected component, recognizing the at least one connected component as a text component if it is not recognized as a vertical line, as a horizontal line, as part of a broken line, or as part of the frame, and defining a character node of a hierarchical tree structure corresponding to the extended character area and containing both the at least one connected component and any identified unattached connected components.
摘要翻译：一种从表格单元框架中识别和提取文本数据的方法。该方法包括以下步骤：跟踪文档图像的连通分量，跟踪连通组件内的白色轮廓，基于白色轮廓定义框架轮廓，识别框架轮廓内的未附加字符数据，以及在框架内定义初始矩形区域大纲。该方法还包括从初始矩形区域检测水平或垂直方向上的黑色像素，以便创建扩展字符区域，对于每个白色轮廓定位位于扩展字符区域内的边界像素，识别位于位于内部的边界像素之间的黑色像素所述扩展字符区域组合位于所述扩展字符区域内的边界像素之间的黑色像素以形成至少一个连接的组件，如果所述至少一个连接的组件未被识别为垂直线，则将所述至少一个连接的组件识别为文本组件，水平线，作为虚线的一部分或者作为框架的一部分，并且定义对应于扩展字符区域并包含至少一个连接组件和任何识别的未连接的连接组件的层级树结构的字符节点。

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