公开(公告)号：US20120163708A1

公开(公告)日：2012-06-28

申请号：US13335077

申请日：2011-12-22

申请人： Wei FAN ,  Akihiro Minagawa ,  Jun Sun ,  Yoshinobu Hotta ,  Satoshi Naoi

发明人： Wei FAN ,  Akihiro Minagawa ,  Jun Sun ,  Yoshinobu Hotta ,  Satoshi Naoi

IPC分类号： G06K9/62

CPC分类号： G06K9/56 ,  G06K9/4642 ,  G06K9/6231 ,  G06K2009/4666

摘要： There provides an apparatus for and a method of generating a classifier for detecting a specific object in an image. The apparatus for generating a classifier for detecting a specific object in an image includes: a region dividing section for dividing, from a sample image, at least one square region having a side length equal to or shorter than the length of shorter side of the sample image; a feature extracting section for extracting an image feature from at least a part of the square regions divided by the region dividing section; and a training section for performing training based on the extracted image feature to generate a classifier. By using the apparatus for and method of generating the classifier, it becomes possible to make full use of recognizable regions of objects to be recognized with variable aspect ratios and improve speed and accuracy for recognizing in complex backgrounds.

摘要翻译： 提供了一种用于生成用于检测图像中的特定对象的分类器的装置和方法。 用于产生用于检测图像中的特定物体的分类器的装置包括：区域分割部分，用于从样本图像中分离至少一个方边区域，其具有等于或短于样本的短边长度的边长 图片; 特征提取部分，用于从由所述区域划分部分划分的所述正方形区域的至少一部分中提取图像特征; 以及训练部，用于基于所提取的图像特征进行训练以生成分类器。 通过使用生成分类器的装置和方法，可以充分利用被识别的可识别区域的可变长宽比，并提高用于在复杂背景中识别的速度和精度。

公开(公告)号：US20120045129A1

公开(公告)日：2012-02-23

申请号：US13067247

申请日：2011-05-18

申请人： Jun Sun ,  Yoshinobu Hotta ,  Hao Yu ,  Satoshi Naoi

发明人： Jun Sun ,  Yoshinobu Hotta ,  Hao Yu ,  Satoshi Naoi

IPC分类号： G06K9/34

CPC分类号： G06K9/3283 ,  G06K9/00456 ,  G06K9/00463

摘要： A method for processing a document image includes: performing horizontal and vertical text line extraction on the document image; providing an overlapping matrix, a value of an element of the overlapping matrix indicating an overlapping relation between horizontal and vertical text lines; merging the overlapping matrix in the vertical and horizontal direction; determining one or more text overlapping regions in the document image, based on the values of the elements of the merged overlapping matrix; counting the total number of strokes or pixel points in the horizontal and vertical text lines, respectively, within one of the one or more text overlapping regions; and determining an orientation of the text overlapping region is horizontal if the total number of strokes or pixel points in the horizontal text lines is larger than that in the vertical text lines, otherwise, determining the orientation is vertical.

摘要翻译： 一种用于处理文档图像的方法包括：在文档图像上执行水平和垂直文本行提取; 提供重叠矩阵，所述重叠矩阵的元素的值指示水平和垂直文本行之间的重叠关系; 在垂直和水平方向上合并重叠矩阵; 基于所述合并的重叠矩阵的元素的值来确定所述文档图像中的一个或多个文本重叠区域; 在一个或多个文本重叠区域之一内分别计算水平和垂直文本行中的笔画或像素点的总数; 并且如果水平文本行中的笔画或像素点的总数大于垂直文本行中的大小，则确定文本重叠区域的取向是水平的，否则确定方向是垂直的。

公开(公告)号：US08345977B2

公开(公告)日：2013-01-01

申请号：US13067247

申请日：2011-05-18

申请人： Jun Sun ,  Yoshinobu Hotta ,  Hao Yu ,  Satoshi Naoi

发明人： Jun Sun ,  Yoshinobu Hotta ,  Hao Yu ,  Satoshi Naoi

IPC分类号： G06K9/34 ,  G06K9/00

CPC分类号： G06K9/3283 ,  G06K9/00456 ,  G06K9/00463

摘要： A method for processing a document image includes: performing horizontal and vertical text line extraction on the document image; providing an overlapping matrix, a value of an element of the overlapping matrix indicating an overlapping relation between horizontal and vertical text lines; merging the overlapping matrix in the vertical and horizontal direction; determining one or more text overlapping regions in the document image, based on the values of the elements of the merged overlapping matrix; counting the total number of strokes or pixel points in the horizontal and vertical text lines, respectively, within one of the one or more text overlapping regions; and determining an orientation of the text overlapping region is horizontal if the total number of strokes or pixel points in the horizontal text lines is larger than that in the vertical text lines, otherwise, determining the orientation is vertical.

摘要翻译： 一种用于处理文档图像的方法包括：在文档图像上执行水平和垂直文本行提取; 提供重叠矩阵，所述重叠矩阵的元素的值指示水平和垂直文本行之间的重叠关系; 在垂直和水平方向上合并重叠矩阵; 基于所述合并的重叠矩阵的元素的值来确定所述文档图像中的一个或多个文本重叠区域; 在一个或多个文本重叠区域之一内分别计算水平和垂直文本行中的笔画或像素点的总数; 并且如果水平文本行中的笔画或像素点的总数大于垂直文本行中的大小，则确定文本重叠区域的取向是水平的，否则确定方向是垂直的。

公开(公告)号：US20120045131A1

公开(公告)日：2012-02-23

申请号：US13067389

申请日：2011-05-27

申请人： Yuan He ,  Jun Sun ,  Hao Yu ,  Satoshi Naoi ,  Yoshinobu Hotta

发明人： Yuan He ,  Jun Sun ,  Hao Yu ,  Satoshi Naoi ,  Yoshinobu Hotta

IPC分类号： G06K9/46

CPC分类号： G06K9/00449

摘要： Ruled lines are extracted and fitted into a real 2-D space. Correspondence between fitted cells and template cells of a ruled line template is determined. For each pair of cells corresponding to each other, the position of each pixel in the template cell is mapped into a real position in the real 2-D space based on an affine transformation between the cells. A pixel value based on pixel values of a plurality of pixels in the image with positions adjacent to the real position is generated as a pixel value of the pixel in the template cell corresponding to the real position. A synthesized image corresponding to the image is generated by merging the ruled lines of the ruled line template with the pixels in the template cells having the pixel values as generated. A form template is obtained based on the synthesized images corresponding to the plurality of images.

摘要翻译： 规则线被提取并拟合到真实的2-D空间中。 确定拟合细胞和格线模板的模板细胞之间的对应关系。 对于彼此对应的每对单元，基于单元之间的仿射变换，将模板单元中的每个像素的位置映射到实际2-D空间中的真实位置。 基于与实际位置相邻的位置的图像中的多个像素的像素值的像素值被生成为与实际位置对应的模板单元中的像素的像素值。 通过将格线模板的划线与具有生成的像素值的模板单元中的像素合并来生成与图像对应的合成图像。 基于与多个图像对应的合成图像获得表单模板。

公开(公告)号：US08805080B2

公开(公告)日：2014-08-12

申请号：US13472790

申请日：2012-05-16

申请人： Jun Sun ,  Satoshi Naoi

发明人： Jun Sun ,  Satoshi Naoi

IPC分类号： G06K9/62

CPC分类号： G06K9/3208 ,  G06K9/342 ,  G06K2209/01

摘要： The present embodiments disclose a method of and device for identifying the direction of characters in an image block. The method includes: performing optical character recognition processing on the image block by assuming various directions as assumed character directions to obtain sub image blocks, recognized characters corresponding to the sub image blocks and correctness measures thereof in each assumed character directions; in sub image blocks in the assumed character directions with 180° mutual relation, searching for a minimum matching pair of the sub image blocks; adjusting the sub image blocks in the searched minimum matching pair to eliminate the effect, on an identification result, of different numbers of sub image blocks in various assumed character directions; calculating an accumulative correctness measure in each assumed character directions based on the adjusted sub image blocks; and identifying the direction of characters in the image block according to the accumulative correctness measures.

摘要翻译： 本实施例公开了用于识别图像块中的字符的方向的方法和装置。 该方法包括：通过假定各个方向作为假定的字符方向，对每个假定的字符方向进行图像块的识别处理，对应于子图像块的识别字符及其正确性度量，对图像块执行光学字符识别处理; 在具有180°相互关系的假定字符方向的子图像块中，搜索子图像块的最小匹配对; 调整搜索到的最小匹配对中的子图像块以消除识别结果对各种假定字符方向的不同数量的子图像块的影响; 基于经调整的子图像块计算每个假设字符方向上的累积正确性度量; 并根据累积的正确性度量来识别图像块中的字符的方向。

公开(公告)号：US08717632B2

公开(公告)日：2014-05-06

申请号：US13471977

申请日：2012-05-15

申请人： Danian Zheng ,  Jun Sun ,  Satoshi Naoi

发明人： Danian Zheng ,  Jun Sun ,  Satoshi Naoi

IPC分类号： G06T5/00 ,  H04N1/407 ,  H04N1/409

CPC分类号： H04N1/387 ,  G06K9/3275 ,  G06K2009/363 ,  G06T3/0031

摘要： An image processing method generally includes: obtaining a vanishing point on a curved surface in a two-dimension image; extracting all the straight line segments between a top contour line and a bottom contour line of the curved surface by the vanishing point; removing a perspective distortion to get parallel straight line segments; obtaining the lengths of the straight line segments, obtaining the true width of each of the straight line segments in a three-dimension space and the depth increment of the straight line segments according to the lengths; obtaining the expanded width of each straight line segment according to the true width and the depth increment; obtaining the total expanded width of the curved surface to transform it into a flat surface; transforming image contents on the curved surface onto the flat surface.

摘要翻译： 图像处理方法通常包括：在二维图像中的曲面上获得消失点; 通过消失点提取弯曲表面的顶部轮廓线和底部轮廓线之间的所有直线段; 去除透视失真以获得平行的直线段; 获得直线段的长度，根据长度获得三维空间中每个直线段的真实宽度和直线段的深度增量; 根据真实宽度和深度增量获得每条直线段的扩展宽度; 获得弯曲表面的总扩展宽度以将其变形成平坦表面; 将曲面上的图像内容转换成平面。

公开(公告)号：US08478045B2

公开(公告)日：2013-07-02

申请号：US13156688

申请日：2011-06-09

申请人： Jun Sun ,  Hao Yu ,  Satoshi Naoi

发明人： Jun Sun ,  Hao Yu ,  Satoshi Naoi

IPC分类号： G06K9/00

CPC分类号： G06K9/6814 ,  G06K9/6224

摘要： Method and apparatus for processing an image including a character are disclosed. The method may include: searching in a set of characters one or more characters having highest similarities of shape to a character in the set of characters, hereinafter the character being referred to as a first character, the one or more searched characters forming a similar character list of the first character; searching in the set of characters one or more characters having highest similarities of shape to each character in the similar character list of the first character, to form a similar character list of each character in the similar character list of the first character; and selecting in the similar character lists one or more characters having a high mutual similarity between each other, as a character cluster.

摘要翻译： 公开了用于处理包括字符的图像的方法和装置。 该方法可以包括：在一组字符中搜索具有与该组文字中的字符具有最高相似度的一个或多个字符，此后字符被称为第一个字符，该一个或多个搜索到的字符形成相似的字符 第一个字符的列表; 在所述一组字符中搜索与所述第一字符的相似字符列表中的每个字符具有最高相似度形状的一个或多个字符，以形成所述第一字符的相似字符列表中每个字符的相似字符列表; 并且在相似的字符中选择一个或多个彼此之间具有高相互相似性的字符作为字符簇。

公开(公告)号：US20120288203A1

公开(公告)日：2012-11-15

申请号：US13466538

申请日：2012-05-08

申请人： Yifeng PAN ,  Jun Sun ,  Yuanping Zhu ,  Pan Pan ,  Yuan He ,  Satoshi Naoi

发明人： Yifeng PAN ,  Jun Sun ,  Yuanping Zhu ,  Pan Pan ,  Yuan He ,  Satoshi Naoi

IPC分类号： G06K9/46

CPC分类号： G06K9/2054 ,  G06F16/5846 ,  G06K2209/01

摘要： Locating text areas in an image and recognizing text contents in the text areas through optical character recognition, OCR; selecting a first class of pending keywords from the recognized text contents to search for webpages; extracting a second class of pending keywords from the retrieved webpages; and determining one or more keywords corresponding to the image from at least the second class of pending keywords. With the embodiment, both OCR and webpage searching can be combined so that the webpages can be retrieved based upon the first class of pending keywords recognized and selected through OCR to ensure convergence of the keywords and then the second class of pending keywords can be selected from the retrieved webpages to ensure correctness of the keywords.

摘要翻译： 在图像中定位文本区域，并通过光学字符识别OCR识别文本区域中的文本内容; 从识别的文本内容中选择一类待处理的关键字来搜索网页; 从检索的网页中提取第二类待处理的关键字; 以及从至少所述第二类未决关键字确定与所述图像相对应的一个或多个关键字。 利用该实施例，可以组合OCR和网页搜索，使得可以基于通过OCR识别和选择的第一类未决关键字来检索网页，以确保关键字的收敛，然后可以从第 检索的网页以确保关键字的正确性。

公开(公告)号：US07929765B2

公开(公告)日：2011-04-19

申请号：US12778336

申请日：2010-05-12

申请人： Jun Sun ,  Yutaka Katsuyama ,  Satoshi Naoi

发明人： Jun Sun ,  Yutaka Katsuyama ,  Satoshi Naoi

IPC分类号： G06K9/18

CPC分类号： G06K9/3266 ,  G06K2209/01

摘要： Video frames that contain text areas are selected from given video frames by removing redundant frames and non-text frames, the text areas in the selected frames are located by removing false strokes, and text lines in the text areas are extracted and binarized.

摘要翻译： 通过删除冗余帧和非文本帧，从给定的视频帧中选择包含文本区域的视频帧，通过删除假笔划来选择所选帧中的文本区域，文本区域中的文本行被提取和二值化。

公开(公告)号：US20100220930A1

公开(公告)日：2010-09-02

申请号：US12778336

申请日：2010-05-12

申请人： Jun Sun ,  Yutaka Katsuyama ,  Satoshi Naoi

发明人： Jun Sun ,  Yutaka Katsuyama ,  Satoshi Naoi

IPC分类号： G06K9/46

CPC分类号： G06K9/3266 ,  G06K2209/01

摘要： Video frames that contain text areas are selected from given video frames by removing redundant frames and non-text frames, the text areas in the selected frames are located by removing false strokes, and text lines in the text areas are extracted and binarized.

摘要翻译： 通过删除冗余帧和非文本帧，从给定的视频帧中选择包含文本区域的视频帧，通过删除假笔划来选择所选帧中的文本区域，文本区域中的文本行被提取和二值化。