CN1147807C - Automatic identifying method and system for name card - Google Patents

Automatic identifying method and system for name card

Info

Publication number
CN1147807C
CN1147807C CNB001196936A CN00119693A CN1147807C CN 1147807 C CN1147807 C CN 1147807C CN B001196936 A CNB001196936 A CN B001196936A CN 00119693 A CN00119693 A CN 00119693A CN 1147807 C CN1147807 C CN 1147807C
Authority
CN
China
Prior art keywords
block
image
layout
step
program
Prior art date
Application number
CNB001196936A
Other languages
Chinese (zh)
Other versions
CN1339775A (en
Inventor
潘卫军
何代水
蔡世光
Original Assignee
英业达集团(上海)电子技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 英业达集团(上海)电子技术有限公司 filed Critical 英业达集团(上海)电子技术有限公司
Priority to CNB001196936A priority Critical patent/CN1147807C/en
Publication of CN1339775A publication Critical patent/CN1339775A/en
Application granted granted Critical
Publication of CN1147807C publication Critical patent/CN1147807C/en

Links

Abstract

一种名片自动识别方法,包括影像输入程序、版面影像分割程序、字符识别程序以及资料分类程序。 One kind of automatic identification card comprising an image input program, program layout image segmentation, character recognition program, and a program classification information. 其中,影像输入程序取得名片的版面影像,版面影像分割程序将版面影像分割成多个区块影像,并判别版面影像所属的样板类型,字符识别程序将各区块影像识别为对应的文字资料,资料分类程序则分析文字资料以便分类储存。 Wherein the input program video image acquisition card layout, the layout image layout image dividing program into a plurality of image blocks, and determines the type of layout template image belongs, the character recognition programs to recognize the image corresponding to each block of text data, the data the classification program analyzes text data in order to classify storage. 本发明亦揭示了实现此方法的名片自动辨识系统。 The present invention also discloses an automatic identification system card implementation of this method.

Description

名片自动识别方法与系统 Automatic Identification card method and system

本发明涉及一种名片的自动识别方法与系统。 The present invention relates to an automatic identification card system and method.

随着经济交往的日益频繁,每个人所拥有的名片数量也大量增加,使得名片的保存、管理和查询等都相当困难。 With the increasingly frequent economic exchanges, the number of cards per person has also increased, making business cards preservation, management and query and so very difficult. 为了使用上的方便,人们往往把名片的资料记录在如个人电脑、手机或其它电子装置中,如此,可以简化管理或查询名片的程序。 For ease of use, people tend to put business cards record data in personal computers, mobile phones or other electronic device, so that simplifies management of a business card or query procedures. 若储存于体积较小的电子装置,如PDA(个人数字助理器)等,则更可以缩小储存大量名片资料所需的空间。 If stored in a smaller volume of the electronic device, such as a PDA (personal digital assistant) or the like, can be reduced even more space to store large amount of data required for the business card.

虽然目前市面上已有可将名片上的资料输入至某些特定电子装置中的装置,但是其大多仅单纯地以影像方式来记录名片上的资料。 Although the market has been on the business card data can be input to the device specific electronic devices, but it mostly just simply to record data in a way the image on the card. 这种记录方式并未能提供进一步的资料应用。 This recording mode and unable to provide further information on the application. 例如,对于移动电话而言,人们所需的资料大多只是姓名与电话号码。 For example, for a mobile phone, the information required for most people is just the name and phone number. 因此,对于手机使用者而言,最好可直接自一名片自动识别装置将姓名与电话号码记录于手机中,而不必以手动输入的方式,逐一地将名片上的姓名与电话号码资料输入。 Thus, for mobile phone users, the best directly from a means of automatic identification card name and telephone number recorded in the phone without having to manually input, one by one on the card with the name of the telephone number data input. 若以影像方式来记录名片上的资料,并无法达到上述的要求。 In terms of video mode to record data on the card, and can not achieve the above requirements. 又,仅以影像来记录名片的资料并无法达到对大量名片资料进行整理。 Also, only the image data recorded on the card and can not reach a large number of business cards to collate data. 例如,由于所输入的名片资料仅为影像资料,故无法直接输入至个人电脑的通讯录中,以进一步加以排序或群络化。 For example, because the business card image data input only, it can not be entered directly into the PC's address book to further sorting or group of the network. 由上述可知,已知以影像来记录名片上的资料的方式,对于使用者而言仍然相当不便。 From the above, in a known manner in order to record the image data on a business card, it is still quite inconvenient for the user.

中国发明专利申请第99113803.1号公开了“一种名片全自动识别录入与检索系统”,该申请虽然也揭示了名片的全自动识别录入方法,但是其无法对名片上的资料进行解析和分类。 Chinese invention patent application No. 99113803.1 discloses "an automatic identification card entry and retrieval system", although this application discloses a method for automatic identification entry card, but it can not be on the card data parsing and classification.

因此,如何将名片上的资料进行解析,并将使用所有资料进行记录,进而让使用者更为便利地将名片上的资料输入至各种电子装置,以克服名片资料输入上的不便已成为一亟待解决的重要课题。 Therefore, how to parse the data on the card, and will use all the information is recorded, thereby allowing the user to more conveniently input the information on the card to a variety of electronic devices, to overcome the inconvenience has become a business card data entry urgently important to resolve the issue.

针对上述问题,本发明的目的为提供一种名片自动识别方法与系统,其可自动识别名片上所记载的资料,并可将其分类储存,以克服名片资料输入上的不便。 For the above problems, an object of the present invention is to provide an automatic identification card system and method, which can automatically identify the data card described in, and may be stored in categories, in order to overcome the inconvenience of the input data on the card.

为达上述目的,本发明提供的名片自动识别方法,包含:影像输入程序,取得名片的版面影像;版面影像分割程序,将所述版面影像分割成多个区块影像,并判别所述版面影像所属的样板类型;以及字符识别程序,将各所述区块影像识别为对应的文字资料;所述版面影像分割程序包含:第一步骤,对所述版面影像进行横向投影以判断所述版面影像是否为可分割,并当所述版面影像为可分割时,找出所述版面影像的第一区块;第二步骤,对所述第一区块以外的区域进行横向投影以分割所述第一区块以外的区域,并在所述第一区块以外的区域为可分割时,找出所述版面影像的第二区块与第三区块;第三步骤,对所述第三区块进行纵向投影,以判断所述第三区块是否为可分割;第四步骤,在所述第三区块为可分割时,对所述第一区块进行纵向投影,以判 To achieve the above object, an automatic identification card of the present invention provides a method, comprising: an image input program, an image acquisition card of the layout; layout image segmentation program, the layout dividing the image into a plurality of image blocks, and determines the layout image template type belongs; and a character recognition program, each of the image blocks identified as corresponding text data; a layout image segmentation program comprising: a first step of said image layout lateral projection to determine whether the layout image whether to be divided, and when the image layout to be split, to identify a first block of the layout image; a second step of block regions other than the first lateral projection to divide the first a region other than a block, and a region other than the first block is divisible, the layout of the image to identify a second block and the third block; a third step of the third region block longitudinal projection, the third block to determine whether to be divided; a fourth step of, when said third block to be divided, the first longitudinal projection block to judge 所述第一区块是否为可分割;第五步骤,在所述第三步骤后,当所述第三区块为可分割时,对所述第一区块进行纵向投影,以判断所述第一块是否为可分割;在所述第二步骤中,当所述第一区块以外的区域为不可割时,将所述第一区块以外的区域视为第四区块;第六步骤,对所述第四区块进行纵向投影,以判别所述第四区块是否可进一步分割;以及第七步骤,当所述第四区块为可分割时,对所述第一区块进行纵向投影,以判别所述第一区块是否为可分割。 The first block is an available segmentation; a fifth step, after the third step, when the third block is divisible, the first longitudinal projection block to determine whether the dividing the first block is an available; in the second step, when a region other than the first block is not cut, a region other than the first block of the fourth block considered; sixth step, the fourth longitudinal projection block, to discriminate whether the fourth block may be further divided; and a seventh step of, when the fourth block is divisible, the first block longitudinal projection, to discriminate whether the first block is divisible.

本发明亦提供一种名片自动识别系统,包含:影像输入装置,读取名片的版面影像:以及处理单元,执行将该版面影像分割成多个区块影像的版面影像分割程序,以及将各所述区块影像识别为对应的文字资料的字符识别程序,所述版面影像分割程序包含:第一步骤,对所述版面影像进行横向投影以判断所述版面影像是否为可分割,并当所述版面影像为可分割时找出所述版面影像的第一区块;第二步骤,对所述第一区块以外的区域进行横向投影以分割所述第一区块以外的区域,并于所述第一区块以外的区域为可分割时,找出所述版面影像的第二区块与第三区块:第三步骤,对所述第三区块进行纵向投影,以判断所述第三区块是否为可分割;第四步骤,在第三区块为可分割时,对所述第一区块进行纵向投影,以判断所述第一区块是否为可分割;第五步 The present invention also provides an automatic card identification system, comprising: image input means for reading card layout image: and a processing unit that performs the division layout image into a plurality of blocks of video image segmentation program sections, and each of the said image block identified as corresponding text data of character recognition programs, the layout image segmentation program comprising: a first step of said image layout lateral projection image to determine whether the layout can be divided, and when the layout image is dividable identify a first block of the layout image; a second step, a region other than the region of the first block for the lateral projection to divide the block other than the first, and to the region other than said first block is divisible, the layout of the image to identify a second block and the third block: a third step, the third longitudinal projection block to determine whether the first three split block is an available; and a fourth step of, when the third block to be divided, the first longitudinal projection block to determine whether the first block is divisible; a fifth step ,在第三步骤后,当所述第三区块为不可分割时,对所述第一区块进行纵向投影,以判断所述第一区块是否为可分割;在所述第二步骤中,当所述第一区块以外的区域为不可分割时,将所述第一区块以外的区域视为第四区块;第六步骤,对所述第四区块进行纵向投影,以判别所述第四区块是否可进一步分割;以及第七步骤,当所述第四区块为可分割时,对所述第一区块进行纵向投影,以判别所述第一区块是否为可分割。 , After the third step, when the third block is an inseparable, the first longitudinal projection block to determine whether the first block is divisible; in the second step , when the region other than the first block is an inseparable, the region other than the first block of the fourth block considered; a sixth step, the fourth longitudinal projection block, to discriminate whether the fourth block is further divided; and a seventh step of, when the fourth block is divisible, the first longitudinal projection block, to discriminate whether to be the first block segmentation. 另一种依本发明的名片自动识别系统包括一影像输入装置、一版面影像分割装置、一字符识别装置以及一储存装置。 Another automatic card identification system under this invention includes an image input means, a layout image segmentation means, a character recognition device and a storage device. 影像输入装置读取一名片一版面影像,储存装置储存版面影像,版面影像分割装置将该版面影像分割成多个区块影像,并判别该版面影像所属的样板类型,字符识别装置则将各区块影像识别为对应的文字资料。 A video input means reads a card layout image, image storage means for storing the layout, the image layout apparatus into a plurality of blocks dividing the image layout image, and determines a type of the template layout image belongs, the character recognition apparatus will each block identified as corresponding image of text information.

依本发明的名片自动辨识方法与系统,使用者可更为便利地将名片上的资料输入至各种电子装置中,因此解决了传统名片资料输入不便问题。 Identification cards under this invention is an automatic method and system, the user can more conveniently input the data on the card to the various electronic devices, thereby solving the problem of inconvenient input traditional business card.

以下将参照相关附图,说明依本发明较佳实施例的名片自动辨识方法与系统。 Reference to related drawings will be described an automatic identification card system and method of the preferred embodiment under this invention. 其中相同的元件与步骤将以相同的参照符号表示。 Wherein the step of the same elements will be represented by the same reference numerals.

图1为本发明较佳实施例的名片自动识别方法的流程图;图2显示了储存名片原始影像资料的像素矩阵的示意图。 Automatic identification card flowchart preferred embodiment of the present invention. FIG. 1; FIG. 2 shows a schematic card storing original image data pixel matrix.

图3为显示本发明的发明人对名片版面的配置统计结果的示意图。 3 is a schematic configuration statistics card layout of the present invention invention.

图4为显示依本发明较佳实施例的名片自动识别方法中,版面影像分割程序的流程图。 FIG 4 is a method of automatic identification card embodiment, a flowchart of the layout image segmentation program under this preferred embodiment of the invention.

图5(A)为显示一对像素矩阵进行横向投影的例子的示意图。 Schematic diagram 5 (A) is a lateral projection display example of a pair of pixel matrix.

图5(B)为显示一对像素矩阵进行纵向投影的例子的示意图。 Schematic diagram 5 (B) is a longitudinal projection display example of a pair of pixel matrix.

图6为显示依本发明较佳实施例的名片自动识别方法中,对识别后的文字资料进行资料分类程序的结果的示意图。 FIG 6 is a preferred embodiment of the method under this invention is the automatic identification card embodiment, a schematic diagram of data classification result of program text information for the identification.

图7为显示依本发明较佳实施例的名片识别系统的架构的示意图。 7 is a schematic view of the architecture of the business card identification system according to the preferred embodiment of the display under this invention.

请参照图1,本发明较佳实施例的名片自动辨识方法1先对名片进行影像输入程序11,以取得名片的版面影像,再进行版面影像分割程序12,以将名片的版面影像分割成几个区块影像。 Referring to FIG 1, an example of an automatic card identification method of the present invention, a first preferred embodiment of the card 11 for image input program to obtain the layout of the business card image, then the image layout segmentation program 12, to divide the layout of the business card image into several blocks image. 接着,进行字符识别程序13,将各个区块影像识别为文字资料。 Next, the character recognition program 13, the image of each block is identified as text information. 然后,进行资料分类程序14,对字符辨识程序I3所得的文字资料加以分析,以便分类储存。 Then, data classification program 14, to analyze the obtained character recognition program I3 text data in order to classify storage. 以下将对名片自动识别方法1中的各程序进行详细说明。 Automatic identification of the card 1 will in the respective procedures described in detail.

在影像输入程序11中,先取得一名片的原始影像资料,并将原始影像资料如图2所示,以二级灰阶格式的像素矩阵的形式储存。 In the image input program 11, to obtain original image data of a business card, and the original image data shown in FIG. 2, stored in the form of two grayscale pixel matrix format. 在图2中,(xm,yn)表示影像中各像素的座标位置,Pmn(xm,yn)则表示该像素的有或无。 In FIG. 2, (xm, yn) represents the coordinates of each pixel in the image position, Pmn (xm, yn) of the pixel is expressed or absent.

接着进行版面影像分割程序12,将名片的版面影像先分割为多个区块影像,以便进行后续的字符识别程序13与资料分类程序14。 Followed by layout image segmentation program 12, the card layout image into a plurality of blocks of the first image for subsequent character recognition program 13 and program classification information 14.

关于名片的版面配置,本发明的发明人在分析过500张不同的名片后发现,相对于普通文章的版面,名片的版面配置有其与众不同的特征。 About business card layouts, the present inventors analyzed after 500 different card found general layout of the article, relative to the business card layouts have their distinctive features. 首先,名片上面的各种资料多半会互相以较多的空白来分隔,因此,可以将名片的版面配置分成不同的区块,如单位名称区块、姓名区块、职称区块或地址区块等。 First, the card will likely be above all kinds of information with each other more blanks to separate, therefore, can be a business card layout is divided into different blocks, such as the name of the unit block, block name, title block or address block Wait. 再者,各区块的配置具有规律性。 Further, each block having a regular configuration. 例如,姓名区块常与职称区块放在一起,地址区块多半位于名片的下半部,单位区块则多半位于名片的上半部。 For example, the names and titles of the block is usually placed together with the block, the address block is located mostly in the lower half of the card, the unit block is mostly located in the upper half of the card. 此外,名片为了美观,还可能印有其它装饰性的要素,例如单位的商标或分隔用的水平线等。 Further, business cards for beauty, other elements may also be printed with decorative, for example, units separated by a horizontal line mark or the like.

基于上述名片的版面配置特征,本发明的发明人运用统计学原理,对一般的名片进行分析之后,将名片上的版面配置分为如图3所示的七种不同的样板。 Based on the business card layout features described above, the present invention is the use of statistical theory, general business card after analysis, the layout on the card configuration is divided into seven different model shown in Fig.

欲判别名片的版面配置是属于此七种样板中的哪一种,可由下面述的三个条件来判断:第一为判断名片的版面影像是否为可横向分割,并找出横向分割之后,版面影像所分割成的横向的区块数目,如样板TI、T2、T3与T4为可横向分割为三个区块,T5与T6则可横向分割为两个区块;第二为判断第一区块,亦即在图3所示的各样板中最上方的区块,是否为可纵向分割,如样板T1的第一区块为不可纵向分割,样板T3的第一区块则为可纵向分割;第三则为判断距离第一区块最远的区块,亦即在图3所示的各样板中最下方的区块,是否为可纵向分割,如样板T1中距离第一区块最远的区块为可纵向分割,样板T3中距离第一区块最远的区块则为不可纵向分割。 Is determined to be a business card layout which belongs to this model of the seven kinds, the following three conditions can be determined later: After the first image of the layout is determined whether the card is laterally divided, and divided transversely to identify, forum the number of image blocks divided laterally into such model TI, T2, T3 and T4 are laterally divided into three blocks, T5 and T6 may be divided into two lateral blocks; determining a first region of a second block, i.e., the uppermost in the template shown in FIG. 3 blocks, whether or longitudinally split, as a first template T1 is not longitudinally divided block, the first block of the template T3 was longitudinally split ; third, compared with the first block is determined from the farthest block, i.e., the lowermost block in the template shown in FIG. 3, whether or longitudinally split, as the template T1 in the first block from the most far block is longitudinally split, was not divided longitudinally furthest from the first block in the block template T3.

请参照图4,图4为依上述判断条件对像素矩阵进行影像分割的流程图。 Referring to FIG. 4, FIG. 4 according to the above determination condition is a pixel matrix is ​​a flowchart of the image segmentation. 首先进行第一步骤121,其是对像素矩阵中的资料进行横向投影。 First, a first step 121, which corresponds to the pixel data matrix lateral projection. 有关横向投影的说明请参照图5(A)。 For a description of lateral projection Referring to FIG. 5 (A). 在图5(A)中。 In FIG. 5 (A) in the. 若像素矩阵第n列中存在任何像素,则该列的投影结果即为非望白,反之若像素矩阵第n列中未存在有任何像素,则该列的投影结果即为空白。 If there is any pixel in the n-th column of the pixel matrix, the projection column is the result of the non-looking white, whereas if the n-th column in the pixel matrix there is not any pixel, the projector is the result of the blank column.

在第一步骤121结束后,若在横向投影的结果中发现一空白区域,亦即,在投影结果中发现一大于某一预定值的相邻列均无像素存在(例如图5(A)中,连续五列均无像素存在),则视为找到第一区块,否则将名片版面配置的样板视为T7,并离开版面影像分割程序12。 After the end of the first step 121, if the results found in a blank area in the transverse projection, i.e., the projection results found in a row is greater than a predetermined value of adjacent pixels were not present (e.g., FIG. 5 (A) in , five consecutive no pixels exist), it is considered to find the first block, otherwise it will be considered a model business card layout T7, and leave the layout image segmentation program 12. 换言之,若在横向投影的结果中发现一空白区域,则像素矩阵至少可以横向分割为两个区域,此时,便将可分离出来的第一个区块,视为该第一区块。 In other words, if the results found in a blank area of ​​the lateral projection, the lateral pixel matrix may be divided into at least two areas, then, put out of separable first block, the first block considered.

接着进行第二步骤122,对第一区块以外的区域进行横向投影。 Followed by a second step 122, a region other than the first block lateral projection. 投影结果若发现另一空白区域,则视为找到第二区块与第三区块,且版面配置的样板可能为T1、T2、T3或T4。 If the result of the projection further found an empty area, it is considered to find the second block and the third block, and the template layout possibilities for T1, T2, T3 or T4. 若无法找到另一空白区域,则视为仅找到第四区块,且版面配置的样板可能为T5或T6。 If you can not find another empty area, it is considered to find only the fourth block, and the template layout possibilities for T5 or T6.

若版面配置的样板可能为T1、T2、T3或T4,则进行第三步骤123,对第三区块进行纵向投影。 If the layout of the template may be T1, T2, T3 or T4, the third step 123, a third longitudinal projection block. 有关纵向投影的说明请参照图5(B)。 For a description of the longitudinal projection Referring to FIG. 5 (B). 在图5(B)中,与前述横向投影相似地,若第三区块所对应的像素矩阵的第m行中存在有任何像素,则该行的投影结果即为非空白,反之若像素矩阵第m行中未存在有任何像素,则该行的投影结果即为空白。 (B), with the lateral projection Similarly, if the m-th row of the pixel matrix corresponding to the third block is present in any pixel in FIG. 5, the projector is the result of the non-blank lines, and vice versa if the pixel matrix m-th row is not present in any pixel, the projection line is the result of the blank.

在第三步骤123结束后,若在纵向投影的结果中发现一空白区域,亦即,在投影结果中发现一大于某一预定值的相邻列均无像素存在,则第三区块可纵向分割,且版面配置的样板可能为T1或T4。 After the end of the third step 123, if found in a blank area in the longitudinal projection of the results, i.e., found in the projection result is greater than a predetermined value of a column adjacent pixels were not present, then the third block is longitudinally segmentation and layout template may be T1 or T4. 若无法找到一空白区域,则第三区块无法分割,且版面配置的样板可能为T2或T3。 If a blank area can not be found, then the third block can not be divided, and the layout template may be T2 or T3.

若第三区块为可纵向分割,则进行第四步骤124;若第三区块为不可纵向分割,则进行第五步骤125。 If the third block is divided longitudinally, the fourth step 124 is performed; if the third block is not divided longitudinally, a fifth step 125 is performed. 第四步骤124与第五步骤125均为对第一区块进行纵向投影,以判别第一区块是否可进一步纵向分割。 A fourth step 124 and fifth step 125 are the longitudinal projection of the first block, to discriminate whether the first block is further divided longitudinally. 在第四步骤124中若第一区块为可纵向分割,则版面配置的样板为T4,若第一区块为不可纵向分割,则版面配置的样板为T1。 In a fourth step, when the first block 124 to be longitudinally divided, the layout of the template to T4, if the first block is not longitudinally divided, the layout of the template T1. 而在第五步骤125中,若第一区块为可纵向分割,则版面配置的样板为T3,若第一区块为不可纵向分割,则版面配置的样板为T2。 In a fifth step 125, if the first block is divided longitudinally, the layout of the template T3, when the first block is not longitudinally divided, the layout of the template T2.

若版面配置的样板可能为5S或T6,则进行第六步骤126,对第四区块进行纵向投影,以判别第四区块是否可进一步纵向分割。 If the layout template may be 5S or T6, then a sixth step 126, the fourth block of the longitudinal projection, to discriminate whether or not the fourth block may be further divided longitudinally. 若纵向投影的结果显示第二区块为不可纵向分割,则将名片版面配置的样板视为T7,并离开版面影像分割程序12。 If the result of the longitudinal projection of the second block is displayed is not longitudinally divided, then the card layout model considered T7, and exits the layout image segmentation program 12. 若为可纵向分割,则进行第七步骤127,对第一区块进行纵向投影以判别第一区块是否为可纵向分割。 If it is longitudinally split, then a seventh step 127, the first block of the first longitudinal projection to discriminate whether the block to be split lengthwise. 若第一区块为可纵向分割,则版面配置的样板为T6,若第一区块为不可纵向分割,则版面配置的样板为T5。 If the first block is divided longitudinally, the layout of the model is T6, the first block if the model is not longitudinally divided, the layout of T5.

上述版面影像分割程序12完成之后,即进行字符识别程序13。 After the above-described layout image segmentation program 12 is completed, i.e., character recognition program 13. 由于名片中可能会包含中文、英文、数字、标点甚至日文等多种文字及符号,所以,字符识别程序13可采用一种多语种混合识别程序。 Since the card may contain Chinese, English, numbers, punctuation characters and other Japanese and even symbols, the character recognition program 13 may employ a multi-lingual recognition program mixed. 例如,可采用几何特特(字符与笔画之间、各部分以及笔画与部分之间稳定的相对关系)和拓朴特征(笔画之间的特征点,如端点、折点、两笔画相接而成的歧点、以及两笔画相交而成的交点等)等来进行识别,这些特征在进行多语种混合办识时均具有稳定性与重要性。 For example, a very very special geometry (between characters and strokes, as well as portions of a stable relationship between the stroke and the relative portion) and topological feature (feature points between strokes, such as endpoints, vertices, and the two strokes in contact into a manifold point, and an intersection obtained by the intersection of two strokes, etc.) to be identified, have these characteristics and importance of stability during mixing do multilingual recognition.

资料分类程序14对识别后的文字资料进行分析,以便管理或查询名片的程序。 Data classification program text information 14 after the identification will be analyzed in order to manage or query business card program. 请参照图6,依照识别的结果,可依名片的版面所属的样板种类,得到将名片上的资料区分为个人资料、通讯资料与其它资料等,并将各种资料分类储存。 Referring to FIG 6, in accordance with results of the identification, to follow the type of card layout template belongs, obtain the data area on the card is divided into profile data communication with other information, etc., and various data stored in categories. 例如,将姓名、公司名称或职称等视为个人资料,将电话、地址、电子邮件或传真号码等视为通讯资料,公司的统一编号则可归类于其它资料中。 For example, such as name, company name or title regarded as personal data, telephone, address, e-mail or fax number and other information deemed Communications, the company's uniform number can be classified as other materials. 如此,使用者可更为便利地记录与整理名片上的资料。 Thus, the user can more conveniently organize recorded data on the card.

请参照图7,依本发明较佳实施例的名片自动识别系统2包括影像输入装置21、模拟/数字信号转换器22、数字信号处理器23、处理单元24以及储存装置25。 Referring to FIG 7, under this embodiment of the business card preferred embodiment of the invention the automatic identification system 2 includes an image input device 21, an analog / digital signal converter 22, a digital signal processor 23, the processing unit 24 and a storage device 25. 其中,影像输入装置21可采用一CCD或CMOS影像感测器以读取名片的影像并产生模拟影像信号。 Wherein the image input device 21 may employ a CCD or CMOS image sensor to read the image of the card and generates an analog video signal. 模拟/数字信号转换器22将影像输入装置21所获得的模拟影像信号转换为数字影像信号。 Analog video signal into an analog / digital signal converter 22 the video input device 21 is obtained a digital video signal. 数字信号处理器23则对数字影像信号进行滤波处理。 The digital signal processor 23 pairs of the digital video signal is filtered.

处理单元24可为CISC处理器、RISC处理器或任何可执行前述名片自动识别方法1中各程序的处理器,例如一般PC中的CPU。 The processing unit 24 may be a CISC processors, RISC processors or processor perform any of the method of automatic identification card 1 in each program, for example, in the general PC CPU. 储存装置25则储存前述名片自动识别方法1中各程序的对应程序码以及影像资料等,并可视需要使用如硬盘驱动器、RAM或ROM等常用的存储装置。 Storage device 25 and the image data corresponding to the program code of each program stored in the Automatic identification card 1 and the like, and optionally using conventional means such as a storage like a hard disk drive, RAM or ROM.

需注意的是,亦可将前述名片自动识别方法1中的各程序,直接内建于处理单元24中,即,处理单元24中的指令集可直接包合执行前述名片自动识别方法1中的各程序的指令集。 Note that, the foregoing procedure may also be a card in the automatic identification method, is built directly into the processing unit 24, i.e., the instruction set of the processing unit 24 may perform the direct inclusion automatic card recognition method 1 each program instruction set. 如此,储存装置25中即不需储存前述名片自动识别方法1中的各程序。 Thus, the storage device 25 is stored in the card i.e. without an automatic identification method in each program.

名片识别的后所得到的名片资料可直接储存于储存装置25中,亦可输出至其它电子装置或储存装置中。 The resulting identification card after the card data may be directly stored in the storage device 25 can also be output to other electronic devices or a storage device. 例如,可视实际需要,将名片识别之后所得到的名片资料传送至PDA或个人电脑中,并依一预定的格式分类储存,以利使用者管理。 For example, to actual needs of the business card identification obtained after the card data is sent to a PDA or a personal computer, and in accordance with a predetermined format stored in categories to facilitate user management. 又,若配合一般的手机使用时,可选择仅将姓名及电话号码储存于手机中,以简化使用者输入名片上资料的手续。 Also, if the phone with general use, you can select only the names and phone numbers stored in the phone in order to simplify procedures for user input on the business card data.

当然,依本发明的名片自动识别系统亦可以其它方法实施,而不脱离本发明的精神与范围。 Of course, under this system of automatic identification card to the invention also may be embodied in other ways without departing from the spirit and scope of the invention. 例如,可使用ASIC构成执行前述的版面影像分割程序与字符识别程序的特定硬件,亦即,针对前述的版面影像分割程序与字符识别程序,在名片自动识别系统中加入特定的版面影像分割装置与字符识别装置来执行。 For example, configuration may be performed using the ASIC specific hardware layout image segmentation and character recognition program of the program, i.e., the layout image for segmentation and character recognition program procedure, adding specific layout image segmentation device in card Recognition System performing character recognition apparatus. 如此,由于直接以硬件执行的速度,将比由处理单元来执行软件的速度快,所以加入特定硬件后,名片识别的效率将较高。 Thus, since the speed at a speed directly executed by hardware, software than performed by the processing unit, so that the addition of specific hardware, the identification card will be higher efficiency.

此外,本发明亦可配合电脑可读取的记录媒体实施。 Further, the present invention also with the computer-readable recording medium embodiment. 亦即,将前述的名片自动识别方法各个程序记录于电脑可读取的记录媒体上后,电脑将可藉由读取该记录媒体上的各个程序来进行前述的名片自动识别方法。 That is, the above-described method of automatic identification card after each program recorded on a recording medium in a computer-readable, computer will read each program by the recording medium on the card performs automatic identification method. 如此,使本发明的名片自动识别方法将具有更大的使用弹性以及产业上可利用性。 Thus, automatic identification card so that the method of the invention will have a greater flexibility and use INDUSTRIAL APPLICABILITY.

以上所述仅为本发明的较佳实施例,故其仅为举例性,而非用以限制本发明的专利保护范围。 The above are only preferred embodiments of the present invention, so that merely illustrative, and not to limit the patent scope of the present invention. 任何不脱离本发明的精神与范围,而对本发明所进行的等效修改或变更,均应包含于所附的权利要求书的范围内。 Any without departing from the spirit and scope of the present invention, but the present invention is carried out equivalent modifications or changes, the claims shall be included in the scope of the appended claims.

Claims (10)

1.一种名片自动识别方法,包含:影像输入程序,取得名片的版面影像;版面影像分割程序,将所述版面影像分割成多个区块影像,并判别所述版面影像所属的样板类型;以及字符识别程序,将各所述区块影像识别为对应的文字资料;所述版面影像分割程序包含:第一步骤,对所述版面影像进行横向投影以判断所述版面影像是否为可分割,并当所述版面影像为可分割时,找出所述版面影像的第一区块;第二步骤,对所述第一区块以外的区域进行横向投影以分割所述第一区块以外的区域,并在所述第一区块以外的区域为可分割时,找出所述版面影像的第二区块与第三区块;第三步骤,对所述第三区块进行纵向投影,以判断所述第三区块是否为可分割;第四步骤,在所述第三区块为可分割时,对所述第一区块进行纵向投影,以判断所述第一区块是否为 An automatic card recognition method, comprising: an image input program, an image acquisition card of the layout; layout image segmentation program, the layout dividing the image into a plurality of image blocks, and determines the type of the layout template image belongs; and a character recognition program, each of the image blocks identified as corresponding text data; a layout image segmentation program comprising: a first step of said image layout lateral projection to determine whether the layout image to be split, and when the image layout to be split, to identify a first block of the layout image; a second step of block regions other than the first lateral projection to divide the first block other than region, and a region other than the first block is divisible, the layout of the image to identify a second block and the third block; a third step, the third longitudinal projection block, the third block to determine whether to be divided; a fourth step of, when said third block to be divided, the first longitudinal projection block, the first block to determine whether 分割;第五步骤,在所述第三步骤后,当所述第三区块为可分割时,对所述第一区块进行纵向投影,以判断所述第一块是否为可分割;在所述第二步骤中,当所述第一区块以外的区域为不可割时,将所述第一区块以外的区域视为第四区块;第六步骤,对所述第四区块进行纵向投影,以判别所述第四区块是否可进一步分割;以及第七步骤,当所述第四区块为可分割时,对所述第一区块进行纵向投影,以判别所述第一区块是否为可分割。 Dividing; a fifth step, after the third step, when the third block is divisible, the first longitudinal projection block to determine whether the first block to be split; in in the second step, when a region other than the first block is not cut, a region other than the first block of the fourth block considered; a sixth step, the fourth block longitudinal projection, to discriminate whether the fourth block may be further divided; and a seventh step of, when the fourth block is divisible, the first longitudinal projection block, to discriminate the first whether a block to be split.
2.如权利要求1所述的名片自动识别方法,其特征在于,还包含:资料分类程序,将所述文字资料分类储存。 Automatic identification card 2. The method according to claim 1, characterized in that, further comprising: program classification information, the classification text data storage.
3.如权利要求1所述的名片自动识别方法,其特征在于,还包含:将所述文字资料传输至其它电子装置中。 Automatic identification card according to claim 1, characterized in that, further comprising: transmitting the text data to the other electronic devices.
4.一种名片自动识别系统,包含:影像输入装置,读取名片的版面影像:以及处理单元,执行将该版面影像分割成多个区块影像的版面影像分割程序,以及将各所述区块影像识别为对应的文字资料的字符识别程序,所述版面影像分割程序包含:第一步骤,对所述版面影像进行横向投影以判断所述版面影像是否为可分割,并当所述版面影像为可分割时找出所述版面影像的第一区块;第二步骤,对所述第一区块以外的区域进行横向投影以分割所述第一区块以外的区域,并于所述第一区块以外的区域为可分割时,找出所述版面影像的第二区块与第三区块:第三步骤,对所述第三区块进行纵向投影,以判断所述第三区块是否为可分割;第四步骤,在第三区块为可分割时,对所述第一区块进行纵向投影,以判断所述第一区块是否为可分割;第五步骤,在第三 An automatic card identification system, comprising: image input means for reading card layout image: and a processing unit that performs the division layout image into a plurality of blocks of video image segmentation program sections, each of said regions and image block identified as corresponding text data of character recognition programs, the layout image segmentation program comprising: a first step of said image layout lateral projection image to determine whether the layout can be divided, and when said image layout identify the first block of the layout image is dividable; a second step of block regions other than said first region for lateral projection to divide the block other than the first, and in the second a region other than the block is divisible, the layout of the image to identify a second block and the third block: a third step, the third longitudinal projection block to determine whether the third region whether a dividable block; and a fourth step of, when the third block to be divided, the first longitudinal projection block to determine whether the first block is divisible; a fifth step, the first three 骤后,当所述第三区块为不可分割时,对所述第一区块进行纵向投影,以判断所述第一区块是否为可分割;在所述第二步骤中,当所述第一区块以外的区域为不可分割时,将所述第一区块以外的区域视为第四区块;第六步骤,对所述第四区块进行纵向投影,以判别所述第四区块是否可进一步分割;以及第七步骤,当所述第四区块为可分割时,对所述第一区块进行纵向投影,以判别所述第一区块是否为可分割。 After quenching, when the third block is an inseparable, the first longitudinal projection block to determine whether the first block is divisible; in the second step, when the region other than the first block is an inseparable, the region other than the first block of the fourth block considered; a sixth step, the fourth longitudinal projection block, to discriminate the fourth whether the block can be further partitioned; and a seventh step of, when the fourth block is divisible, the first longitudinal projection block, to discriminate whether the first block is divisible.
5.如权利要求4所述的名片自动识别系统,其特征在于,还包含:模拟/数字信号转换器,将所述影像输入装置所获得的所述版面影像转换为数字影像信号。 5. The system of automatic identification card according to claim 4, characterized in that, further comprising: an analog / digital signal converter, the layout image of the obtained image input means into a digital video signal.
6.如权利要求4所述的名片自动识别系统,其特征在于,所述处理单元还执行:资料分类程序,将所述文字资料分类储存。 Automatic identification card system according to claim 6, wherein the processing unit further performs: program classification information, the classification text data storage.
7.如权利要求4所述的名片自动识别系统,其特征在于,所述文字资料系传输至其它电子装置中。 7. Automatic Identification card system according to claim 4, wherein said text-based data transmission to other electronic devices.
8.如权利要求4所述的名片自动识别系统,其特征在于,所述版面影像分割程序、所述字符识别程序与所述资料分类程序内建于所述处理单元中。 8. Automatic Identification card system according to claim 4, wherein said layout image segmentation procedure, the character recognition program and the program built in the classification information processing unit.
9.如权利要求4所述的名片自动识别系统,其特征在于,还包含:储存装置,储存所述版面影像。 Automatic identification card system according to claim 9, characterized in that, further comprising: storage means storing said image layout.
10.如权利要求9所述的名片自动识别系统,其特征在于,所述版面影像分割程序、所述字符识别程序与所述资料分类程序记录在所述储存装置中。 Automatic identification card 10. The system of claim 9, wherein said layout image segmentation procedure, the character recognition program from the data classification program recorded in the storage means.
CNB001196936A 2000-08-22 2000-08-22 Automatic identifying method and system for name card CN1147807C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB001196936A CN1147807C (en) 2000-08-22 2000-08-22 Automatic identifying method and system for name card

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB001196936A CN1147807C (en) 2000-08-22 2000-08-22 Automatic identifying method and system for name card

Publications (2)

Publication Number Publication Date
CN1339775A CN1339775A (en) 2002-03-13
CN1147807C true CN1147807C (en) 2004-04-28

Family

ID=4587930

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB001196936A CN1147807C (en) 2000-08-22 2000-08-22 Automatic identifying method and system for name card

Country Status (1)

Country Link
CN (1) CN1147807C (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7305129B2 (en) 2003-01-29 2007-12-04 Microsoft Corporation Methods and apparatus for populating electronic forms from scanned documents
CN100452814C (en) 2003-06-23 2009-01-14 英华达(上海)电子有限公司 Method of converting paper visiting card to electronic visiting card for communication device
CN1316418C (en) * 2004-04-16 2007-05-16 中国科学院自动化研究所 Automatic identifying system and method for house number
CN1328695C (en) * 2004-12-30 2007-07-25 北京中星微电子有限公司 Automatic searching and determining method for key words information in name card identification
CN1301490C (en) * 2004-12-30 2007-02-21 北京中星微电子有限公司 Method for deciding background color according to area in optical character recognition of mobile terminal
CN101739441B (en) 2009-12-01 2012-01-25 中国建设银行股份有限公司 Method of image information input and system thereof
CN103279743A (en) * 2013-05-28 2013-09-04 深圳市中兴移动通信有限公司 Business card recognition method and device

Also Published As

Publication number Publication date
CN1339775A (en) 2002-03-13

Similar Documents

Publication Publication Date Title
US7587412B2 (en) Mixed media reality brokerage network and methods of use
US5761344A (en) Image pre-processor for character recognition system
US8332401B2 (en) Method and system for position-based image matching in a mixed media environment
US7860312B2 (en) System and method for identifying and labeling fields of text associated with scanned business documents
US5809167A (en) Page segmentation and character recognition system
JP4926004B2 (en) Document processing apparatus, document processing method, and document processing program
US7769772B2 (en) Mixed media reality brokerage network with layout-independent recognition
JP4366108B2 (en) Document search apparatus, document search method, and computer program
CN101253514B (en) Grammatical parsing of document visual structures
CN1320485C (en) Image searching device and key word providing method therefor
US20070052997A1 (en) System and methods for portable device for mixed media system
CN102855906B (en) The image processing apparatus and an image processing method
US20110128288A1 (en) Region of Interest Selector for Visual Queries
JP4118349B2 (en) Document selection method and document server
US9514103B2 (en) Effective system and method for visual document comparison using localized two-dimensional visual fingerprints
EP0677812B1 (en) Document storage and retrieval system
US20070050411A1 (en) Database for mixed media document system
US8156427B2 (en) User interface for mixed media reality
US7672543B2 (en) Triggering applications based on a captured text in a mixed media environment
US7991778B2 (en) Triggering actions with captured input in a mixed media environment
CN1137430C (en) Handwritten data input deivce having coordinate detection image input tablet and method thereof
US20070047781A1 (en) Authoring Tools Using A Mixed Media Environment
JP4181892B2 (en) Image processing method
JP3425834B2 (en) Title extracting apparatus and method of the document image
CN102117269B (en) Apparatus and method for digitizing documents

Legal Events

Date Code Title Description
C10 Entry into substantive examination
C06 Publication
C14 Grant of patent or utility model
C56 Change in the name or address of the patentee

Owner name: INVENTEC APPLIANCES (SHANGHAI) ELECTRONICS CO., LT

Free format text: FORMER NAME OR ADDRESS: SHANGHAI ELECTRONIC TECHNOLOGY CO., LTD., YINGYEDA GROUP

CF01 Termination of patent right due to non-payment of annual fee