JP4991407B2 - Information processing apparatus, control program thereof, computer-readable recording medium storing the control program, and control method - Google Patents

Description

  The present invention identifies character strings representing names, company names, store names, etc. from among character strings described in a specific medium, and in particular, an e-mail address and a network described in the medium The present invention relates to an information processing apparatus for identifying a character string representing a name, a company name, a store name, and the like using address information, a control program for the information processing apparatus, a computer-readable recording medium on which the control program is recorded, and a control method.

  2. Description of the Related Art Conventionally, a technique is known in which the contents of a character string written on a specific medium such as a business card are read by a mobile phone with a camera function and recorded in a telephone directory. In addition, when recording the contents of a character string read with a mobile phone with a camera function in the phone book, there are items such as name and phone number in the phone book, so the contents of the character string written on the business card etc. are classified. It is necessary to record correctly in each item part.

  In recent years, a technique for performing the above classification automatically, not manually, is also known. For example, Patent Document 1 discloses a character recognition device that automatically classifies character strings using a character string that is a specific classification index such as “(stock)” or “stock” included in the company name. ing.

  In addition, by using the general feature that the character spacing on each name is relatively wide and the characters are large, the character string representing the name is identified and classified from the name card. There is also known a method for classifying items.

  Further, in Patent Document 2, a score is assigned to each character string obtained by character recognition, and a probable semantic attribute (name, address, telephone number, job title) of the character string based on the score. An information processing apparatus for determining and classifying information) is disclosed. In the information processing apparatus disclosed in Patent Document 2, based on a combination pattern of character strings of a character string group obtained by character recognition (that is, a character string group having a character string found on the same line as a pattern). Since character strings are classified into each semantic attribute, even if a character string that serves as a specific classification index is not included in the character string, it can be classified for each item with relatively high accuracy. It has become.

By the way, “business cards” taken up as a specific medium are often used in business. In recent years, there are many cases where business is done not only in Japan but also overseas, so “business cards” were written in various languages such as English, German, and French, as well as Japanese. Things exist.
JP-A-10-78997 (published March 24, 1998) JP 2006-185342 A (published July 13, 2006)

  However, the character recognition device disclosed in Patent Document 1 has a problem that it is difficult to automatically classify a character string obtained from a business card written in a language other than Japanese, such as an English business card. That is, a character string (such as “Corp.” or “Inc.” in an English business card) that is an index for a specific classification such as “(stock)” or “stock” in a business card written in Japanese (Japanese business card) ) Is often not listed on English business cards, and there is a problem that character strings cannot be automatically classified using a character string that is an index for a specific classification. .

  Further, the conventional item classification method has a problem that it is difficult to automatically classify a character string obtained from a business card written in a language other than Japanese such as an English business card. That is, although the name name in English business cards has a relatively wide interval between the last name and the first name, the character spacing of each character included in the last name or the first name is often the same as other items. There is a problem that it is difficult to identify and classify character strings representing names from English business cards using features.

  Further, the information processing apparatus disclosed in Patent Document 2 needs to have a recording medium such as a dictionary database (DB), a probability information DB, and a rule DB in advance when classifying names and the like from character strings obtained by character recognition. Therefore, a huge amount of recording medium is required. Further, since the score assignment using the probability model is performed, the processing becomes complicated. That is, the information processing apparatus disclosed in Patent Document 2 has a problem that it is not easy to classify names and the like from character strings obtained by character recognition. In addition, when the information processing apparatus disclosed in Patent Document 2 is applied to a character string obtained from an English business card, it is necessary to further include a database corresponding to English, so that the required capacity of the recording medium further increases. However, classification of names and the like from character strings obtained by character recognition cannot be realized more easily.

  The present invention has been made in view of the above-described conventional problems, and its purpose is to more easily at least a name, a company name, and a store name from a character string extracted from image information about a specific medium. It is an object of the present invention to provide an information processing apparatus capable of identifying a character string representing any one of the above, a control program thereof, a computer-readable recording medium recording the control program, and a control method.

  In order to solve the above problems, the information processing apparatus of the present invention performs character recognition using a character recognition dictionary based on image information of a medium on which a character string is described, and is described in the medium. An information processing apparatus that acquires a character string, and includes, from the character string obtained by the character recognition, at least mail address information that is a character string that represents an email address and network address information that is a character string that represents a network address. A plurality of character strings for comparison and collation that are character strings other than the mail address information and the network address information from the character information obtained by the character recognition and the address information obtaining means for obtaining address information that is any one of them Based on the address information acquired by the comparison verification character string acquisition means and the address information acquisition means. An identification character string that generates an identification character string that is one of a personal identification character string used to identify the name, a company identification character string used to identify the company name, and a store identification character string used to identify the store name. A character string comparison / collation unit that compares and collates the identification character string generated by the identification character string generation unit and each of the plurality of comparison / collation character strings acquired by the comparison / collation character string acquisition unit. The character string for comparison and collation determined to be similar to the personal identification character string based on the result of comparison and collation by the character string comparison and collation means is identified as a character string representing a name, and similar to the company identification character string. Then, the determined character string for comparison and collation is identified as a character string representing the company name, and the character string for comparison and collation determined to be similar to the store identification character string is used as a character string representing the store name. It is characterized in that it comprises a identification means constant to.

  In order to solve the above problems, the control method of the present invention performs character recognition using a character recognition dictionary based on image information of a medium on which a character string is described, and is described in the medium. A method of controlling an information processing apparatus for acquiring a character string, wherein the address information acquisition means is a mail address information that is a character string representing at least an e-mail address from the character string obtained by the character recognition, and a network address From the character string obtained by the character recognition by the address information obtaining step for obtaining address information which is any of the network address information which is a character string representing the character string and the comparison character string obtaining means, the mail address A string for comparison and collation that obtains multiple strings for comparison and collation that is a character string other than information and the above network address information And, based on the address information acquired by the address information acquisition means by the identification character string generation means, at least a personal identification character string used for name identification, a company identification character string used for company name identification, and a store name An identification character string generation step of generating an identification character string that is one of the store identification character strings used for identification of the store, and the identification character string generated by the identification character string generation means by a character string comparison / collation means; A character string comparison / collation step for comparing and collating each of the plurality of comparison / collation character strings acquired by the comparison / collation character string obtaining unit, and a comparison / collation result in the character string comparison / collation step by the identification unit. Based on the above, the comparison verification character string determined to be similar to the personal identification character string is identified as a character string representing a name, and the company identification character string An identification step of identifying a comparison matching character string determined to be similar as a character string representing a company name, and identifying a comparison matching character string determined to be similar to the store identification character string as a character string representing a store name; It is characterized by including.

  According to the above invention, at least an e-mail address is obtained from the character string obtained by performing character recognition using the character recognition dictionary based on the image information of the medium in which the character string is described, by the address information obtaining unit. Address information that is any one of mail address information that is a character string representing network address information and network address information that is a character string representing a network address, and at least based on the address information by the identification character string generation means An identification character string that is one of a personal identification character string used for identifying a name, a company identification character string used for identifying a company name, and a store identification character string used for identifying a store name is generated. . In addition, from the character string acquired by performing the character recognition, a plurality of comparison and verification character strings that are character strings other than the mail address information and the network address information are acquired by the comparison and verification character string acquisition means. Yes. Then, the character string comparison / collation means compares the identification character string with each of the plurality of comparison / collation character strings, and the identification means is similar to the personal identification character string based on the result of the comparison / collation. The determined comparison matching character string is identified as a character string representing the name, and the comparison matching character string determined to be similar to the company identification character string is identified as a character string representing the company name, and similar to the store identification character string. Then, since the determined comparison matching character string is identified as a character string representing a store name, it is necessary to provide a new dictionary and database in the information processing apparatus in order to identify a character string representing a name, a company name, and a store name. Absent.

  Also, an identification character string derived from at least one of mail address information and network address information obtained from character strings obtained through character recognition, and a character string other than mail address information and network address information By comparing and collating multiple character strings for comparison and extracting character strings for comparison and collation similar to identification character strings, name, company name and store name are identified, so complex dictionary data and complex processing This rule is not required for the information processing apparatus. Therefore, it is possible to identify a character string representing at least one of the name, the company name, and the store name from the image information about the medium in which the character string is described by a relatively simple process.

  Further, as described above, a personal identification character string used to identify a name, a company identification character string used to identify a company name, and a store name identification generated from at least one of email address information and network address information The comparison character string similar to the identification character string by comparing and collating the identification character string such as the store identification character string used for the item and a plurality of comparison character strings that are character strings other than the mail address information and the network address information The name, company name, and store name are identified by extracting, so characters that serve as indicators for a specific classification, such as “(Co),” “Co.,” “Corp.”, “Inc.” Even if the column is not included in the character string obtained by performing character recognition, it is possible to identify the character string representing the name, the company name, and the store name. Also, in particular, more easily from character strings extracted from image information about media such as business cards in English notation that the name, company name, store name rarely contain character strings that serve as specific classification indices, It becomes possible to identify a character string representing at least one of a name, a company name, and a store name.

  As a result, it is possible to more easily identify a character string representing at least one of a name, a company name, and a store name from a character string extracted from image information on a specific medium.

  In the information processing apparatus of the present invention, the address information acquisition unit uses at least one of at least a specific character and a specific character string included in the mail address information or the network address information. It is preferable to acquire either the mail address information or the network address information.

  As a result, by using at least one of the specific character and the specific character string included in the mail address information or the network address information, the address information acquisition unit can perform at least the mail address information and the network address information. It will be possible to get either. Therefore, if the character or character string characteristic of the mail address information or the network address information is the specific character or character string, the mail address information and the network address information can be easily determined and acquired by the address information acquisition means. It becomes possible.

  Moreover, in the information processing apparatus of the present invention, the address information acquisition unit is a character string representing a component other than the components of the email address and the network address in the acquired email address information or the network address information. It is preferable to delete the clear character string to obtain mail address information or network address information.

  As a result, the address information acquisition means deletes a character string that is clearly a character string representing elements other than the constituent elements of the mail address and the network address, so that the mail address information or the network address information is obtained. It is easy to generate an identification character string that does not include a character string that is clearly a character string that represents a component other than the above-described components by the identification character string generation unit. In addition, the fact that the identification character string used for the comparison and collation in the character string comparison and collation means does not include a character string that is clearly a character string representing elements other than the components of the mail address and the network address is As it increases accuracy and reduces unnecessary comparisons and collations, it is even easier to use at least one of the name, company name, and store name from the character string extracted from the image information about a specific medium. Can be identified, and the amount of processing in the information processing apparatus can be reduced.

  Further, in the information processing apparatus of the present invention, the comparison / matching character string obtaining unit delimits a character string other than the mail address information and the network address information among character strings obtained by the character recognition. It is preferable to obtain the plurality of character strings for comparison and collation by dividing the character before and after.

  As a result, character strings other than the mail address information and the network address information divided before and after the predetermined delimiter character are used as the comparison / collation character string acquired by the comparison / collation character string acquisition means. When a character string other than e-mail address information and network address information includes a character string that represents a name, company name, or store name, the character string that represents a name, company name, or store name is generally separated by a delimiter. Therefore, it is possible to obtain a comparison / matching character string including only a character string representing a name, a company name, and a store name. When a comparison / matching character string that includes only the character string representing the name, company name, and store name is acquired, the identification means represents the name, company name, or store name as a result of the comparison / matching by the string comparison / collation means. Since the comparison / matching character string to be determined as a character string is a comparison / matching character string that includes only the character string representing the name, company name, and store name, more accurately, at least of the name, company name, and store name It becomes possible to identify a character string representing either of the following.

  In the information processing apparatus according to the present invention, the comparison / matching character string acquisition unit may be a character string representing a name, a company name, and a store name other than the plurality of comparison / matching character strings. It is preferable to delete a character string that is clearly defined as a character string for comparison and collation.

  As a result, the comparison / matching character string acquisition unit deletes a character string that is clearly a character string representing a name other than the name, company name, and store name, and forms a comparison / matching character string. In addition, the comparison / matching character string used for the comparison / collation in the character string comparison / collation means does not include a character string that is clearly a character string other than the name, company name, and store name. As this will increase the accuracy of verification and reduce unnecessary comparison verification, it is even easier to use at least one of name, company name, and store name from a character string extracted from image information about a specific medium. It becomes possible to identify a character string representing one of them and reduce the amount of processing in the information processing apparatus.

  Further, in the information processing apparatus of the present invention, the identification character string generation unit uses the company identification character string included in the mail address information as a company when the network address information is not acquired by the address information acquisition unit. It is preferable to generate the identification character string.

  Thereby, even when the network address information is not acquired by the address information acquisition unit, the identification character string generation unit can generate the company identification character string from the mail address information.

  In the information processing apparatus according to the present invention, the identification character string generation unit includes a specific name that falls between the user name and the host name of the email address included in the email address information acquired by the address information acquisition unit. It is preferable to generate the personal identification character string and the company identification character string or the store identification character string by dividing the mail address information before and after the character.

  Thereby, the identification character string generating means divides the mail address information before and after a specific character that falls between the user name and host name of the electronic mail address, that is, “@”. E-mail addresses used in business often use the name before the “@”, and often use the company name and store name after the “@”. If it is the target, the character string representing the name is generated as a personal identification character string, the character string representing the company name is generated as the company identification character string, or the character string representing the store name is stored in the store. It can be generated as an identification character string. Therefore, it is possible to more accurately identify at least the name and the character string representing either the company name or the store name from the character string extracted from the image information on the specific medium.

  In the information processing apparatus of the present invention, the identification character string generation unit generates a character string ahead of the specific character in the mail address information as a personal identification character string or a store identification character string. Is preferred.

  As a result, the identification character string generation means generates the specific character, that is, the character string ahead of “@” as the personal identification character string or the store identification character string from the mail address information. E-mail addresses used for business name cards often use names in front of “@”, so when business name cards are targeted for character recognition, it is more accurate and a character string representing the name is personal. It can be generated as an identification character string. In addition, e-mail addresses that are published in magazines that contain store information often use store names before the “@”, so magazines such as those listed above are subject to character recognition. If it is, the character string representing the store name can be generated with higher accuracy as the store identification character string. Therefore, it is possible to more accurately identify the character string representing the name or the store name identification character string from the character string extracted from the image information about the specific medium.

  In the information processing apparatus according to the present invention, the identification character string generation unit generates a character string behind the specific character in the mail address information as a company identification character string or a store identification character string. Is preferred.

  As a result, the identification character string generation means generates the specific character, that is, the character string behind “@” as the company identification character string or the store identification character string from the mail address information. E-mail addresses used for business name cards often use the company name in the part after “@”, so when business name cards are targeted for character recognition, the character string that represents the company name is more accurate. Can be generated as a company identification character string. In addition, e-mail addresses published in magazines that contain store information often use store names in the part after “@”. If it is, the character string representing the store name can be generated with higher accuracy as the store identification character string. Accordingly, it is possible to more accurately identify a character string representing at least a company name or a store name identification character string from a character string extracted from image information on a specific medium.

  In the information processing apparatus of the present invention, the identification character string generation unit further includes the personal identification character string and the company identification character before and after a delimiter included in the mail address information acquired by the address information acquisition unit. It is preferable to divide the column or the store identification character string into a new personal identification character string, company identification character string, or store identification character string.

  As a result, a specific character that falls between the user name and host name of the email address, that is, a personal identification character string, company identification character string, or store identification generated by dividing the mail address information before and after “@” Personal identification character string generated by the identification character string generation means, company identification character string, company identification character string, or store identification character string further divided before and after the delimiter included in the email address information It can be a character string or a store identification character string.

  In the information processing apparatus of the present invention, the identification character string generation unit divides the network address information before and after a delimiter included in the network address information acquired by the address information acquisition unit, thereby It is preferable to generate an identification character string or the store identification character string.

  Thereby, the network address information divided before and after the delimiter included in the network address information is used as the company identification character string or the store identification character string generated by the identification character string generation means. If the network address information includes a character string representing a company name or store name, the character string representing the company name or store name is generally separated by a delimiter. It becomes possible to generate a company identification character string or a store identification character string including only a character string representing a name. When a company identification character string or store identification character string that includes only the character string representing the company name or store name is generated, only the character string representing the company name or store name is included in the comparison by the character string comparison and collation means. Since the company identification character string or the store identification character string is used for comparison and collation, the accuracy of comparison and collation can be further increased.

  In the information processing apparatus of the present invention, the character string comparison and collation unit includes the identification character string generated by the identification character string generation unit and the plurality of comparison and collation acquired by the comparison and collation character string acquisition unit. For each of the character strings, a comparison / collation from the characters in front of each other, a comparison / collation from the characters behind each other, and a comparison / collation by shifting the characters to be compared with each other one by one in order. It is preferable.

  This enables different types of comparison collation, such as comparison collation from the characters in front of each other, comparison collation from the characters behind each other, and comparison collation while sequentially shifting each other's comparison collation one character at a time. It is possible to make a determination by the identification means by combining the results. Therefore, it is possible to identify a character string representing at least one of a name, a company name, and a store name from the character string extracted from the image information about the specific medium with higher accuracy. .

  In the information processing apparatus according to the present invention, the character string comparison / collation unit may perform the comparison collation while sequentially shifting each other's counterparts for comparison / collation one character at a time. It is preferable to compare and collate the character string having the smaller number of characters with respect to the partner while sequentially shifting the characters one by one.

  As a result, when performing the comparison and collation while sequentially shifting the other party to be compared and collated one character at a time, the one having the smaller number of characters of the identification character string and the plurality of comparison and collation character strings is Comparison and collation can be performed while shifting character by character in order.

  In the information processing apparatus according to the present invention, the identification unit may determine the number of characters that match between the identification character string and each of the plurality of comparison / matching character strings as a result of the comparison / collation in the character string comparison / collation unit. It is preferable to determine whether or not the identification character string and each of the plurality of comparison matching character strings are similar to each other.

  As a result, the identification unit compares the identification character string with a plurality of comparisons based on the number of matching characters between the identification character string and each of the plurality of comparison matching character strings as a result of the comparison / collation in the character string comparison / collation unit. It is possible to determine whether or not each of the matching character strings is similar.

  In the information processing apparatus of the present invention, the identification unit determines that the comparison matching character string having the largest number of matching characters with respect to the personal identification character string is similar to the personal identification character string, and the matching The comparison / matching character string having the largest number of characters to be compared with the company identification character string is determined to be similar to the company identification character string, and the number of matching characters is the largest with respect to the store identification character string. It is preferable to determine that the column is similar to the store identification character string.

  Thereby, for each of the personal identification character string, the company identification character string, and the store identification character string, the comparison matching character string having the largest number of matching characters, the character string representing the name and the character string representing the company name, respectively. , And can be identified as a character string representing the store name.

  In the information processing apparatus of the present invention, when the identification unit determines that none of the plurality of comparison matching character strings is similar to the identification character string, as a result of the comparison matching in the character string comparison matching unit. It is preferable to identify a comparison / matching character string satisfying a predetermined condition among the plurality of comparison / matching character strings as a name and a company name or a store name.

  As a result of the comparison and collation by the character string comparison and collation means, even when it is determined that none of the plurality of comparison and collation character strings is similar to the personal identification character string, Thus, it is possible to identify a comparison / matching character string satisfying a predetermined condition as a name and a company name or a store name.

  The information processing apparatus may be realized by a computer. In this case, a control program that causes the information processing apparatus to be realized by the computer by causing the computer to operate as the respective means, and a computer that records the control program A readable recording medium falls within the scope of the present invention.

  According to the present invention, it is not necessary to provide a new dictionary and database in the information processing apparatus. Also, a personal identification character string used for identifying a name, which is generated from at least one of mail address information and network address information obtained from a character string obtained by performing character recognition, and used for identifying a company name The identification character string such as the company identification character string and the store identification character string used for identifying the store name is compared with a plurality of comparison and verification character strings that are other than the mail address information and the network address information. By extracting a character string for comparison and collation similar to the column, the name, company name, and store name are identified, so that complicated dictionary data and complicated processing rules are not required for the information processing apparatus, and “( Stock) ”,“ Inc. ”,“ Corp. ”,“ Inc. ”, etc., character strings that serve as indices for specific classifications are converted to character strings obtained through character recognition. Even in the case it was not rare, full name, company name, it is possible to identify a string that represents the store name.

  Therefore, it is possible to identify a character string representing at least one of the name, the company name, and the store name from the image information about the medium in which the character string is described by a relatively simple process. Also, in particular, more easily from character strings extracted from image information about media such as business cards in English notation that the name, company name, store name rarely contain character strings that serve as specific classification indices, It becomes possible to identify a character string representing at least one of a name, a company name, and a store name. Therefore, it is possible to more easily identify a character string representing at least one of a name, a company name, and a store name from a character string extracted from image information about a specific medium. Play.

[Embodiment 1]
An embodiment of the present invention will be described with reference to FIGS. 1 to 16 as follows. In the drawings used for the following description, the same members or the same functions are denoted by the same reference numerals. Therefore, detailed description thereof will not be repeated.

  First, the outline of the configuration of the information processing apparatus 1 will be described with reference to FIG. FIG. 1 is a functional block diagram showing a schematic configuration of an information processing apparatus 1 according to the present embodiment. As shown in FIG. 1, the information processing apparatus 1 includes an image data acquisition unit 11, a character recognition dictionary data storage unit 12, a character recognition unit 13, an address information acquisition unit (address information acquisition unit) 14, and a comparison verification character string acquisition. Part (comparison character string acquisition means) 15, identification character string generation part (identification character string generation means) 16, character string comparison collation part (character string comparison collation means) 17, and item classification determination part (identification means) 18 It has. The information processing apparatus 1 performs character recognition based on image information about a medium such as a business card, identifies a character string such as a name and a company name written on the medium, and items such as a name and a company name. Classification is performed.

  First, the image data acquisition unit 11 acquires image data (image information) of an image taken by a camera of a mobile phone with a camera function, image data (image information) of an image read by a scanner, and the like. The image data acquisition unit 11 sends the acquired image data to the character recognition unit 13. The image data acquired by the image data acquisition unit 11 is image data derived from an image obtained by reading a business card (medium on which a character string is written) with a scanner, or a business card is photographed with a camera of a mobile phone with a camera function. The following description will be made assuming that the image data is derived from the obtained image.

  The character recognition dictionary data storage unit 12 stores character string data (character recognition dictionary) used for the character recognition unit 13 to recognize a character string. For example, the character recognition dictionary data storage unit 12 may be configured to store character codes such as kanji, hiragana, katakana, numbers, alphabets, symbols, and standard feature vectors of characters corresponding to the character codes. In the following description, it is assumed that the character recognition dictionary data storage unit 12 stores a character code and a standard feature vector (character recognition dictionary data) of a character corresponding to the character code.

  The character recognition unit 13 refers to the character recognition dictionary data stored in the character recognition dictionary data storage unit 12 and performs character recognition on the image data sent from the image data acquisition unit 11. For example, the character recognition unit 13 extracts a feature vector from the image data sent from the image data acquisition unit 11 and compares it with the standard feature vectors of all character codes in the character recognition dictionary data in the character recognition dictionary data storage unit 12. Character recognition. The character recognition unit 13 may be configured to use a known OCR (Optical Character Recognition).

  The address information acquisition unit 14 extracts and acquires address information such as e-mail address information and URL (Uniform Resource Locator) information from a character string obtained by character recognition in the character recognition unit 13. For example, as a method of extracting e-mail address information, a method in which a character string is composed of single-byte characters and a character string (specific character string) such as “E-mail” or “mail” is included in the prefix is extracted. Since the character string indicating the e-mail address information should include “@”, a method of extracting a character string (specific character) including “@” may be used. In addition, as a method of extracting URL information (network address information), for example, a method of extracting a character string (a specific character string) starting from “http” or “www” where the character string is made up of single-byte characters. You may use, and the method of extracting the character string (specific character string) containing "http" or "www" may be used.

  The comparison collation character string acquisition unit 15 acquires a character string other than the character string corresponding to the address information from the character string obtained by the character recognition in the character recognition unit 13, and obtains “(space)”, “, The character string for comparison and collation used for comparison and collation in the character string comparison and collation unit 17 described later is obtained.

  The identification character string generation unit 16 generates a personal identification character string (identification character string) by dividing the character string indicating the e-mail address information acquired by the address information acquisition unit 14 with a specific delimiter. . The identification character string generation unit 16 generates a company identification character string (identification character string) by dividing the character string indicating the network address information acquired by the address information acquisition unit 14 with a specific delimiter. is there. Each specific delimiter here will be described in detail later.

  The character string comparison / collation unit 17 performs comparison / collation between the personal identification character string and company identification character string generated by the identification character string generation unit 16 and the comparison / collation character string obtained by the comparison / collation character string acquisition unit 15. Is what you do.

  Then, the item classification determination unit 18 classifies the items of the character string based on the result of the comparison and collation (comparison collation result) in the character string comparison and collation unit 17 (identifies the character string representing the name and company name). It is. The classification result (classification result) determined by the item classification determination unit 18 is sent to a display device provided outside the information processing apparatus 1, and the classification result is displayed on the display device or the like.

  Next, the operation flow in the information processing apparatus 1 will be described with reference to FIG. FIG. 2 is a flowchart illustrating an example of an operation flow in the information processing apparatus 1. Here, a case where processing is performed by the information processing apparatus 1 based on all image data of an English business card as shown in FIG. 3 will be described as an example.

  First, in step S1, the character recognition unit 13 performs character recognition on the image data acquired by the image data acquisition unit 11. Character recognition is performed in units of one line for vertical business cards and in units of one column for horizontal business cards. As a result of performing character recognition on the image data of an English business card as shown in FIG. 3, as shown in FIG. 4A, “Ray Smith”, “NML Market”, “XX Avenue, Δ △, [] [] ”,“ TEL (999) 888-5678 ”,“ FAX (999) 888-1234 ”,“ E-mail: rsmith@nmlMar.com ”,“ www.nml-Market.com ” Will be. As shown in FIG. 4A, a white square is originally applicable to the “[]” portion, and a white square is also applicable to the subsequent “[]” portion. And

  Subsequently, in step S2, the address information acquisition unit 14 obtains a character string indicating address information such as e-mail address information and URL (network address) information from a character string obtained as a result of character recognition by the character recognition unit 13. Extract and get. In this example, the address information acquisition unit 14 uses the e-mail address information “E-mail: rsmith@nmlMar.com” and the network address information “www.nml-Market. com ”. In addition, the character string “E-mail:” in the character string indicating the e-mail address information (a character string that is clearly a character string representing components other than the mail address and the network address component) is the e-mail address. Since the information is a common character string for each information and does not include a name, a company name, or the like and is a character string that represents a component other than the constituent elements of the mail address and the network address, the address information acquisition unit 14 “E-mail: The address information may be acquired by deleting the part “”. In this example, the character string “rsmith@nmlMar.com” as shown in the part surrounded by the broken line in FIG. It will be acquired as address information. Here, the configuration in which the “E-mail:” part is deleted is shown as an example, but the configuration is not necessarily limited thereto. For example, if it is clear that the character string represents a component other than the components of the email address and network address, such as a name and company name not included, the above deletion may be used. A configuration such as “” (a character string that is clearly a character string representing a component other than the constituent elements of the mail address and the network address) may be similarly deleted.

  Next, in step S3, the character string (comparison character string candidate) other than the character string corresponding to the address information obtained by the character recognizing unit 13 is further converted into a character string “(( (Space)) Divide before and after delimiters (predetermined delimiters) such as “,” (separate by delimiters). In this example, from the comparison / matching character string candidate (character string candidate 1 in FIG. 5 (a)) shown in FIG. 5 (a) to the comparison / matching character string (FIG. 5 (b)). The middle character string candidate 2) is obtained. Further, information on whether or not the divided character strings (comparison character strings) were originally written on the same line on the business card (specific medium) at the time of comparison and collation by the character string comparison and collation unit 17 is necessary. Therefore, when the character string (character string candidate for comparison and collation) is divided, the fact that it was originally written on the same line on the business card is held in some form. In this example, the comparison index character strings originally described in the same line are given the same index as shown in FIG. Note that the delimiter (predetermined delimiter) mentioned here includes "(", ")", "-", "_", etc. in addition to "(space)", ",", etc. However, any other character may be used as a delimiter. Moreover, the structure which uses only any of the above as a delimiter may be sufficient, and the structure which uses the above combination may be sufficient.

  Of the candidate strings for comparison and collation, a character string such as a number following the character string “TEL” or “FAX” (a character that clearly represents a character string other than the name, company name, and store name) Column) is often only a phone number or fax number, so the string “TEL” and “FAX” and the string following the string “TEL” or “FAX” It may be configured to be determined as a fax number and deleted from the comparison / matching character string, although it is followed by a specific character string of “TEL” or “FAX”, but the telephone number or fax number. A character string that does not apply as a feature may be configured as a comparison / matching character string.

  Subsequently, in step S4, the identification character string generation unit 16 acquires a personal identification character string and a company identification character string from the e-mail address information in the address information obtained by the address information acquisition unit 14, and step S5. Move on. In step S5, the identification character string generation unit 16 acquires a company identification character string from the network address information in the address information obtained by the address information acquisition unit 14, and proceeds to step S6. Detailed processing in step S4 will be described later.

  Here, with reference to FIG. 6, a detailed description will be given of the process of generating the personal identification character string and the company identification character string from the e-mail address information in step S4. FIG. 6 is a flowchart showing details of an example of processing for generating a personal identification character string and a company identification character string from the electronic mail address information.

  First, in step S41, the location of “@” (characters between the user name and host name of the email address) is detected from the email address information in the address information obtained by the address information acquisition unit 14. Then, the process proceeds to step S42. In step S42, a character string before “@” is acquired from the e-mail address information, and the process proceeds to step S43. Here, the character string before “@” is a candidate for a personal identification character string.

  In step S43, delimiters such as “_”, “−”, and “.” That may be included in the character string before “@” (delimiters included in the mail address information) are candidates for personal identification character strings. Determine if it exists in the string. If the delimiter is present (Yes in step S43), the process proceeds to step S44. If the delimiter does not exist (No in step S43), the process proceeds to step S45.

  In step S44, the character strings of the personal identification character string candidates are divided before and after the position where the delimiter character exists, and each character string obtained by the division is held as a new character string of the personal identification character string candidate. . In step S45, the personal identification character string candidate character string is held as it is as a personal identification character string.

  In this example, “rsmith@nmlMar.com” is obtained as the e-mail address information, so “@” is detected by the processing in step S41. Then, “rsmith” is acquired as a candidate for the personal identification character string by the process of step S42. Since “rsmith” does not include delimiters such as “_”, “−”, “.”, The process in step S44 is not performed, and the character string “rsmith” is obtained from the e-mail address information. This is a generated personal identification character string. If the e-mail address information is “r-smith@nmlMar.com”, there is a delimiter with “-” in “r-smith” that is a candidate for the personal identification character string. By the processing in S43, two personal identification character strings “r” and “smith” are generated.

  Subsequently, in the processing after step S46, the character string after “@” in the e-mail address information is acquired, and the process proceeds to step S47. A character string after “@” is a candidate for a company identification character string. In step S47, delimiters such as “_”, “−”, “.” That may be included in the character string after “@” (delimiters included in the mail address information) are candidates for the company identification character string. Determine if it exists in the string. If the delimiter is present (Yes in step S47), the process proceeds to step S48. If the delimiter does not exist (No in step S47), the process proceeds to step S49.

  In step S48, the company identification character string candidate character string is divided before and after the position where the delimiter character exists, and each character string obtained by the division is held as a new company identification character string candidate character string. . In step S49, the candidate character string of the company identification character string is held as it is as the company identification character string, and the flow ends.

  In this example, the company identification string is “nmlMar.com”, so “nmlMar.com” is divided into “nmlMar” and “com” before and after the delimiter “.”, And “nmlMar” and “com” These two company identification character strings are generated. However, since “com” is a character string commonly used for each e-mail address, the identification character string generation unit 16 may delete it. In addition, character strings commonly used for e-mail addresses such as “co”, “ne”, “go”, “jp”, and “ca” are optionally stored in a dictionary data storage unit (not shown). In addition, the identification character string generation unit 16 may delete the character string by collating the dictionary data in the dictionary data storage unit. In this example, “com” is deleted and “nmlMar” is generated as a company identification character string.

  Further, here, with reference to FIG. 7, a detailed description will be given of a process of generating a company identification character string from the network address information in step S5. FIG. 7 is a flowchart showing details of an example of processing for generating a company identification character string from network address information.

  First, in step S51, it is a process for determining whether or not a delimiter exists in the network address information of the address information obtained by the address information acquisition unit 14. Note that “−”, “.”, “_”, “/”, “:”, And the like are targeted as delimiters in the network address information (delimiters included in the network address information). If the delimiter is present (Yes in step S51), the process proceeds to step S52. If the delimiter character does not exist (No in step S51), the process proceeds to step S53.

  In step S52, the character string of the network address information is divided before and after the position where the delimiter character exists, and each character string obtained by the division is held as a new character string of the network address information. In step S53, the character string of the network address information is held as it is as the company identification character string, and the flow ends.

  In this example, since “www.nml-Market.com” is obtained as the network address information, four processes of “www”, “nml”, “Market”, and “com” are obtained by the processing in step S53. A company identification string will be generated. In addition, character strings such as “www” and “com” in the character string indicating the network address information (character strings that are clearly character strings other than names, company names, and store names) are network address information. The address information acquisition unit 14 deletes the parts such as “www” and “com” and acquires the company identification character string because it is clear that each character string is common and does not include the company name. It may be configured to. In this example, “www” and “com” are deleted, and “nml” and “Market” are generated as company identification character strings. Here, the configuration in which the portions “www” and “com” are deleted is shown as an example, but the present invention is not necessarily limited thereto. For example, if it is a character string that does not include a company name or the like (a character string that is clearly a character string that represents a name other than the name, company name, and store name), the deletion may be performed. Good. Further, as in the case of the e-mail address information, a character string commonly used for the network address is stored in a dictionary data storage unit (not shown) and the dictionary data in the dictionary data storage unit is collated, The identification character string generation unit 16 may delete the character string.

  That is, in this example, as a result of the process of step S4, as shown in FIG. 8A, “rsmith” as the personal identification character string and “nmlMar” as the company identification character string as shown in FIG. 8B. As a result of the processing in step S5, “nml” and “Market” are obtained as the company identification character string as shown in FIG. 8B.

  Subsequently, in step S6, the character string comparison / collation unit 17 compares the personal identification character string obtained by the identification character string generation unit 16 with a plurality of comparison / collation character strings obtained by the comparison / collation character string obtaining unit 15. Match. More specifically, the comparison matching character string and the personal identification character string are compared, and the number of characters that match is obtained. Note that the number of characters obtained here is an evaluation value obtained by the comparison and collation.

  Here, a specific example of the comparison and collation between the comparison / matching character string “Ray” and the personal identification character string “rsmith” will be described with reference to FIGS. 9A to 9C. Note that the comparison matching character string and the personal identification character string may be processed by converting the uppercase letter to lowercase by normalization, or may be processed without normalization. The case where uppercase letters are processed as lowercase letters by normalization will be described as an example. That is, “Ray” is processed as “ray”.

  First, as shown in FIG. 9A, “ray” and “rsmith” are compared and collated in order from the first character in the front (the first character in the front) (forward matching process). 1 is added as an evaluation value. In addition, when both front leading characters match, 1 is further added as an evaluation value. In the comparison / collation between “ray” and “rsmith”, since the first character in front matches, 2 is added as an evaluation value (a match between the first character in front and the character match). Subsequently, the subsequent characters are also compared and collated, but since there is no matching character, an evaluation value of +2 is finally obtained in the forward matching process of “ray” and “rsmith”. Note that when comparing and collating in order from the first character in front, “ith” of “rsmith” is not compared and collated due to the difference in the number of characters between “ray” and “rsmith”. In this example, the subtraction of the evaluation value by is not particularly performed. However, the configuration may be such that the evaluation value is subtracted depending on the number of characters.

  Subsequently, as shown in FIG. 9B, “ray” and “rsmith” are compared and collated in order from the rear first character (rear first character) (backward matching process), and each time the characters match. 1 is added as an evaluation value. In comparison comparison of “ray” and “rsmith”, “y” of “ray” is compared with “h” of “rsmith”, and if they match, 1 is added as an evaluation value, If they match, 1 is further added as the evaluation value. In the comparison collation between “ray” and “rsmith”, since the rear head characters do not match, 0 is added as the evaluation value (the evaluation value is not added). Subsequently, the subsequent characters are also compared and collated, but since there is no matching character, an evaluation value of 0 is finally obtained in the backward matching process of “ray” and “rsmith”.

  Finally, as shown in FIG. 9 (c), “ray” and “rsmith” are compared and collated while sequentially shifting the counterparts for comparison and matching one character at a time, and evaluated each time the characters match. 1 is added as a value. Here, of the comparison / matching character string and the personal identification character string, those having a small number of characters are compared and collated while being sequentially shifted one by one with respect to those having a large number of characters. That is, here, as shown in FIG. 9C, the comparison is made sequentially from the comparison collation between “r” of “ray” and “s” of “rsmith”. After the comparison and collation starting from “r” of “ray” and “s” of “rsmith” is completed, the comparison partner is shifted by one character, “r” of “ray” and “m” of “rsmith” The character string evaluation value of “ray” is compared with the type of comparison collation by repeating the comparison collation while sequentially shifting the comparison collation partner one character at a time, such as starting the comparison collation starting with “”. The maximum evaluation value is obtained as a final evaluation value when the comparison and collation is performed while sequentially shifting the counterparts to be compared one character at a time. However, when the evaluation value is 1 or less, the evaluation value is 0. The evaluation value 0 when the evaluation value is 1 or less is the same in the forward matching process and the backward matching process. The evaluation value finally obtained as a result of the comparison and collation between the character string for comparison and collation in the character string comparison and collation unit 17 is a forward match process, a backward match process, and a partner to be compared and collated. It is assumed that the maximum evaluation value is selected from the evaluation values obtained in each of the processes for performing comparison and collation while sequentially shifting the characters one by one. As described above, the process of step S6 is performed on all combinations of the comparison / matching character string and the personal identification character string, and the evaluation value of each comparison / matching character string is obtained.

  Here, FIG. 10 shows the evaluation values of the comparison matching strings “Ray”, “Smith”, “NML”, “Market”, “XX Avenue”, “△△”, and “[] []”. The result of calculating is shown. As shown in FIG. 10, an index and an evaluation value are associated with each comparison matching character string, the index of “Smith” is “1”, and the evaluation value is “+6” which is the maximum. I know that there is.

  Subsequently, in step S7, when the comparison / collation has been completed for all combinations of the comparison / collation character string and the personal identification character string (Yes in step S7), the process proceeds to step S8. If the comparison / collation is not completed for all combinations of the comparison / collation character string and the personal identification character string (No in step S7), the process returns to step S6 and the flow is repeated.

  In step S8, based on the comparison / collation result in the character string comparison / collation unit 17, the item classification determination unit 18 determines a character string classified into the name item (identifies a character string representing the name). Specifically, the character string having the maximum evaluation value obtained by the character string comparison / collation unit 17 is identified as a character string representing a name. In this example, as shown in FIG. 10, since the evaluation value of the comparison matching character string “Smith” is the maximum, the character string “Smith” is identified as a character string representing a name.

  Note that the full name is often written on the same line on the business card as the first name and last name, so the comparison / matching character string (the comparison / matching character with the same index assigned) Column) may be configured to be identified as a character string representing a name. That is, in this example, “Ray” to which the same index (1) as “Smith” is assigned is also identified as a character string representing the name. Note that the last name and first name can be determined based on the arrangement order of the character strings. In the case of English business cards, the character strings are generally arranged in the order of the first name followed by the last name. The first “Ray” may be used as a first name, and “Smith” after the arrangement order may be determined as a last name. This completes the process of identifying the character string representing the name (name item classification process).

  Next, in step S9, the character string comparison / collation unit 17 compares the company identification character string obtained by the identification character string generation unit 16 with a plurality of comparison / collation character strings obtained by the comparison / collation character string obtaining unit 15. Match. More specifically, the above-described comparison / matching character string and the company identification character string are compared to determine the number of characters that match. Note that the number of characters obtained here is an evaluation value obtained by the comparison and collation.

  In this example, there are multiple comparison matching strings and company identification strings, but here is a specific example of comparison matching between the comparison matching string “NML” and the company identification string “nmlMar” Indicates. In addition, for the comparison matching character string and the company identification character string, processing may be performed using uppercase letters as lowercase characters by performing normalization, or processing may be performed without performing normalization. The case where uppercase letters are processed as lowercase letters by normalization will be described as an example. That is, “NML” becomes “nml” and “nmlMar” becomes “nmlmar”.

  First, “nml” and “nmlmar” are compared and collated in order from the first character in the front (the first character in the front) (forward matching process). That is, the comparison collation is started by comparing “n” of “nml” with “n” of “nmlmar”. Note that when comparing and collating in order from the first character in front, “mar” in “nmlmar” is not compared and collated due to the difference in the number of characters between “nml” and “nmlmar”. In this example, the subtraction of the evaluation value by is not particularly performed. However, the configuration may be such that the evaluation value is subtracted depending on the number of characters. Further, the evaluation value is obtained in the same manner as in the case of the comparison and collation between the comparison collation character string and the personal identification character string. That is, as an evaluation value that is finally obtained as a result of the comparison and collation between the comparison / matching character string and the company identification character string in the character string comparison / collation unit 17, the forward matching process, the backward matching process, and the partner to be compared and collated It is assumed that the maximum evaluation value is selected from the evaluation values obtained in each of the processes for performing comparison and collation while sequentially shifting the characters one by one. Therefore, in this example, “+4” is the evaluation value of the comparison / matching character string “NML”.

  Subsequently, in step S10, if the comparison / collation has been completed for all combinations of the comparison / collation character string and the company identification character string (Yes in step S10), the process proceeds to step S11. If the comparison / matching has not been completed for all combinations of the comparison / matching character string and the company identification character string (No in step S10), the process returns to step S9 and the flow is repeated.

  In step S11, based on the comparison / collation result in the character string comparison / collation unit 17, the item classification determination unit 18 determines a character string classified into the company item (identifies a character string representing the company name). Specifically, the character string having the maximum evaluation value obtained by the character string comparison / collation unit 17 is identified as a character string representing the company name. In this example, the character string for comparison / collation shown in FIG. 11 (a) and the company identification character string shown in FIG. Such a comparison result (evaluation value) is obtained. As shown in FIG. 12, an index and an evaluation value are associated with each comparison / matching character string. As can be seen from FIG. 12, the comparison / matching character string having the largest evaluation value has an evaluation value of Since it is “Market” of “+7”, the character string “Market” is identified as a character string representing the company name.

  In many cases, the company name is written in the same line on the business card as the character string that forms the company name. The comparison and matching character string) may also be identified as a character string representing a company name. That is, in this example, “NML” assigned with the same index (2) as “Market” is also identified as a character string representing a company name, and “NML Market” represents a company name according to the arrangement order of the character strings. Identified as a string. This completes the process of identifying the character string representing the company name (company name item classification process), and the operation flow in the information processing apparatus 1 is also completed.

  In the above embodiment, the best one has been described, and the case where there is no error in the result of character recognition in the character recognition unit 13 has been described. Therefore, in the following, a case where the character recognition unit 13 misrecognizes e-mail address information as “l3mith@mnlMar.com” as shown in FIG. 13, for example, from the image data of the English business card shown in FIG. 3 will be described as an example. I do.

  When the process proceeds according to the flow shown in FIG. 2, as shown in FIG. 14 (a), “Ray”, “Smith”, “NML”, “Market”, “XX Avenue”, “△△”, and “[] []” Is obtained as a comparison / matching character string, and as shown in FIGS. 14B and 14C, the personal identification character string is “l3mith”, the company identification character string is “mnlMar”, It will be generated as “nml” and “Market”. Then, with respect to the generated personal identification character string and company identification character string, a forward matching process, a backward matching process, and a comparison / matching process are performed while sequentially shifting the counterparts to be compared and matched one character at a time.

  Here, FIG. 15 is used to show the result of comparing and collating the personal identification character string “l3mith” with the character string for comparison and collation. As shown in FIG. 15, since the evaluation value of the comparison matching character string “Smith” is “+5”, which is the maximum, the character string “Smith” is identified as a character string representing a name. Further, “Ray” to which the same index (1) as “Smith” is assigned is also identified as a character string representing the name.

  Next, with respect to all combinations of the comparison / matching character string shown in FIG. 14 (a) and the company identification character string shown in FIG. The comparison and collation process is performed while shifting the order one by one.

  Here, FIG. 16 is used to show the results of comparison and collation for all combinations of the comparison and collation character string shown in FIG. 14A and the company identification character string shown in FIG. As shown in FIG. 16, since the evaluation value of the comparison / matching character string “Market” is “+7”, which is the maximum, the character string “Market” is identified as a character string representing the company name. “NML” to which the same index (2) as “Market” is assigned is also identified as a character string representing the company name, and “NML Market” is identified as a character string representing the company name.

  As described above, according to the configuration of the present invention, even when there is an error in the result of character recognition by the character recognition unit 13, the character string representing the name and company name is more accurately identified (name and company name items). Can be classified).

[Embodiment 2]
The following will describe another embodiment of the present invention with reference to FIGS. Configurations other than those described in the present embodiment are the same as those in the first embodiment. For convenience of explanation, members having the same functions as those shown in the drawings of the first embodiment are given the same reference numerals, and explanation thereof is omitted.

  In the first embodiment, the image of the English business card shown in FIG. 3 is taken with a camera of a mobile phone or read with a scanner or the like, and all of the image data is acquired with the image data acquisition unit 11, followed by comparison and verification. A configuration is shown in which a plurality of character strings are obtained, and all of the obtained comparison and collation character strings are compared and collated with a personal identification character string and a company identification character string, and a character string representing a name and a company name is identified. On the other hand, in the present embodiment, the personal identification character string and the company identification character string are obtained for the comparison matching character string obtained from the acquired image data while the image data obtaining unit 11 obtains the image data. A configuration in which the character string representing the name and the company name is identified by comparison with the personal identification character string and the company identification character string will be described. The main configuration of the information processing apparatus that implements the present embodiment is the same as that shown in FIG.

  First, the flow of processing in the present embodiment will be described with reference to FIG. FIG. 17 is a flowchart illustrating an example of a processing flow in the present embodiment.

  First, in step S61, as shown in FIG. 18, scanning of an image is started from the lower side of the business card by an input device such as a scanner. Then, the image is read while scanning, and if a character string exists in the read image (Yes in step S61), the process proceeds to step S62. If no character string exists in the read image (Yes in step S61), the flow ends. In step S62, character recognition is started by the character recognition unit 13, and the process proceeds to step S63. In this case, it is assumed that the character recognition unit 13 performs character recognition in order from the lower side of the business card in units of one line described on the business card.

  In step S63, the character recognition unit 13 determines whether the character string obtained as a result of character recognition by the character recognition unit 13 is e-mail address information or network address information. If the character string obtained as a result of character recognition by the character recognition unit 13 is e-mail address information or network address information (Yes in step S63), the process proceeds to step S64. If the character string obtained as a result of character recognition by the character recognition unit 13 is not e-mail address information or network address information (No in step S63), the process proceeds to step S66.

  In step S64, the address information acquisition unit 14 acquires a character string indicating address information such as e-mail address information and network address information, and the identification character string generation unit 16 uses the address information acquired by the address information acquisition unit 14. A personal identification character string and / or a company identification character string is generated. Then, the process proceeds to step S65.

  In the business card image shown in FIG. 18, first, the character string on the lowermost side of the business card is recognized, so that “www.nml-Market.com” shown in FIG. become. Since the character string “www” is included, the character recognition unit 13 determines that the address is network address information, and the identification character string generation unit 16 performs “www” and “nml” shown in FIG. 19, and “-” between “nml” and “Market”, and “.” Between “Market” and “com”. “Nml” and “Market” are generated as company identification character strings as shown in FIG. “Com” is also obtained by the above division, but is deleted from the company identification character string as in the first embodiment. Further, since the process of step S64 when the character string is network address information is the same as the process described above with reference to FIG. 7, detailed description thereof will not be given here. In this example, a company identification character string is obtained in step S64, and the process proceeds to step S65.

  In step S65, the character recognition unit 13 determines whether there is a character string that can be recognized in the next line in the order from the lower side of the business card. If the character string exists (Yes in step S65), the process returns to step S62 and the flow is repeated. If there is no character string (No in step S65), the flow ends.

  In the business card image shown in FIG. 18, “E-mail: rsmith@nmlMar.com” shown in FIG. 20A in the next line of “www.nml-Market.com” in the order from the lower side of the business card. Is obtained by the character recognition unit 13. Then, because the prefix includes the character string “E-mail” and includes “@”, the character recognition unit 13 determines that the address information is e-mail address information by the process of step S63. Further, as a result of the processing in step S64, the identification character string generation unit 16 sets “rsmith” as the personal identification character string and “nmlMar” as the company identification character string as shown in FIGS. 20 (b) and 20 (c). Generated. “Com” is also acquired, but is deleted from the company identification character string as in the first embodiment. When the network address information mentioned above is obtained, “nml” and “Market” are already generated as company identification strings, so “nmlMar” must be added to the existing company identification string. become. Further, since the process of step S64 when the character string is electronic mail address information is the same as the process described above with reference to FIG. 6, detailed description thereof will not be given here.

  Further, in this example, the process proceeds to step S65, and the character recognition unit 13 determines whether or not there is a character string that can be recognized in the next line in the order from the lower side of the business card. In the business card image shown in FIG. 18, since there is a character string in the next line, the next line of “E-mail: rsmith@nmlMar.com” in the order from the lower side of the business card in step S62. , “FAX (999) 888-1234” is obtained by the character recognition unit 13. Since the character string “FAX” is included, the fax number is known, so that the character recognition unit 13 determines that it is not e-mail address information and network address information by the processing in step S63, and the process proceeds to step S66. .

  In step S66, since the personal identification character string or the company identification character string is necessary to perform the comparison / collation with the comparison / collation character string, is the personal identification character string or the company identification character string already obtained? The character string comparison / collation unit 17 determines whether or not. If the personal identification character string or the company identification character string has already been obtained (Yes in step S66), the process proceeds to step S67. If the personal identification character string or the company identification character string has not been obtained (No in step S66), the process returns to step S61 and the flow is repeated. In this example, since the personal identification character string and the company identification character string have already been obtained, the process proceeds to step S67.

  In step S67, a character string other than the character string corresponding to the address information (comparison character string candidate for comparison and verification) is further converted by the comparison and verification character string acquisition unit 15 into delimiters such as “(space)” and “,”. Divide before and after a predetermined delimiter. In this example, the character string is divided before and after “(”, “)”, “−” (predetermined delimiter) in “FAX (999) 888-1234” before and after the delimiter. However, since “FAX (999) 888-1234” is composed of a string of letters “FAX” and a series of numbers following the string of “FAX”, it is understood that it is not a name and company name. A configuration in which the comparison and collation processing after step S66 is omitted may be employed. In this example, the character string “TEL (999) 888-5678” obtained by the character recognition next to “FAX (999) 888-1234” is similarly called “TEL” and “TEL”. Since it consists of an enumeration of numbers following the character string, it can be understood that it is not a name or company name, and therefore, the comparison / collation processing after step S66 may be omitted.

  Subsequently, in the present example, the above-described processing is repeated, and in step S62, the character recognition unit 13 performs a character string (character for comparison and collation with “XX Avenue, ΔΔ, [] []” as shown in FIG. A case where a column candidate) is obtained will be described. In the process of step S63, since it is determined that “XX Avenue, ΔΔ, [] []” is not e-mail address information and network address information, the process proceeds to step S66. In the process of step S66, since the personal identification character string or the company identification character string has already been obtained, the process proceeds to step S67. In the process of step S67, the comparison / matching character string acquisition unit 15 performs “XX Avenue, Δ △, [] before and after the delimiter“, ”included in“ XX Avenue, Δ △, [] [] ”. The character string “[]” is divided, and the comparison character strings “XX Avenue”, “ΔΔ”, and “[] []” are obtained.

  In step S68, the character string comparison / collation unit 17 uses the personal identification character string and / or company identification character string obtained by the identification character string generation unit 16 and the comparison / collation character string obtained by the comparison / collation character string obtaining unit 15. Is compared. In addition, as described in the first embodiment, this process performs the forward matching process, the backward matching process, and the comparison matching process while sequentially shifting the counterpart to be compared and matched character by character. Is what you want.

  Subsequently, in step S69, the item classification determination unit 18 determines whether or not the comparison matching character string is similar to the personal identification character string or the company identification character string based on the result of the comparison matching. If the comparison / matching character string is similar to the personal identification character string or the company identification character string (Yes in step S69), the process proceeds to step S70. If the comparison / matching character string is not similar to the personal identification character string and the company identification character string (No in step S69), the process proceeds to step S65 to repeat the flow.

  In step S70, the item classification determination unit 18 indicates the name of the comparison / matching character string similar to the personal identification character string when the comparison / identification character string is similar to the personal identification character string. Identifies as a string. When the comparison / matching character string is similar to the company identification character string, the comparison / matching character string similar to the company identification character string is identified as a character string representing the company name. And it moves to step S65 and repeats a flow.

  In this example, from the comparison / collation results shown in FIGS. 22 (a) and 22 (b), the comparison / matching character strings “XX Avenue”, “ΔΔ”, and “[] []” all have evaluation values “ Since it is “0”, the item classification determination unit 18 determines that the comparison / matching character string is not similar to the personal identification character string and the company identification character string, and the process proceeds to step S65. In this example, since there is a character string on the line next to “XX Avenue, ΔΔ, [] []”, the process proceeds to step S62. Further, in the process of step S62, “NML Market” shown in FIG. 23 in the next line of “XX Avenue, ΔΔ, [] []” is generated by the character recognition unit 13 in the order from the bottom side of the business card. Is obtained and the process proceeds to step S63. In the process of step S63, since it is determined that “NML Market” is not e-mail address information and network address information, the process proceeds to step S66. In the process of step S66, since the personal identification character string or the company identification character string has already been obtained, the process proceeds to step S67. In the process of step S67, the character string for “NML Market” is divided before and after the delimiter character “(space)” included in “NML Market” by the comparison character string acquisition unit 15, and the character string for comparison and collation “ NML "and" Market "will be obtained.

  In this example, in the process of step S68, the comparison / matching character string “NML” and the personal identification character string “rsmith” are first compared, and then the comparison / matching character string “Market” and the personal identification character string are identified. Compare with the string “rsmith”. As a result of the comparison and collation, the evaluation value “0” as shown in FIG. 24A is obtained for both. Therefore, the comparison and collation character strings “NML” and “Market” are similar to the personal identification character string. It is determined by the item classification determination unit 18 as not having been performed. Further, the comparison / matching character string “NML” is compared with the company identification character string “nml”. As a result of the comparison and collation, an evaluation value “+4” as shown in FIG. 24B is obtained. Since the comparison character string “NML” and the company identification character string “nml” have three characters, the comparison character string “NML” and the company identification character string “nml” are completely You can see that they match. Therefore, the item classification determination unit 18 determines that the comparison / matching character string “NML” and the company identification character string “nml” are similar. In the process of step S70, the comparison / matching character string “NML” is identified as a character string representing the company name.

  In this embodiment, since processing is performed in units of one line described on the business card, “Market” enumerated in the same line as “NML” is not performed without further comparison. Further, the added character string “NML Market” may be identified as a character string representing a company name. According to the above configuration, it is possible to obtain an effect that the processing amount in the information processing apparatus 1 is reduced and the processing time is shortened.

  The above-described series of processing is repeated until the uppermost character string on the business card is recognized. In this example, as a result of repeating the above-described series of processing until the uppermost character string of the business card is recognized, “NML Market” is identified as a character string representing the company name (as a company name item) "Ray Smith" is identified (classified as a name item) as a character string representing the name.

  In the configuration of the present embodiment, instead of identifying the comparison / matching character string that maximizes the evaluation value as a character string that represents the name or company name, the character string that evaluates to a predetermined value or more is selected. Therefore, it is necessary to identify the character string representing the name or company name. The predetermined value mentioned here can be arbitrarily set, for example, a value that is half or more of the number of characters of the name identification character string or the company name identification character string to be compared and collated. .

  Here, a case where the predetermined value is set to a value more than half the number of characters of the name identification character string or the company name identification character string to be compared and collated will be described using this example. In this example, the character string shown in FIG. 25 is obtained as the character string candidate for comparison and collation, and the character strings “Ray” and “Smith” are obtained as the character string for comparison and collation. Since the number of characters of the personal identification character string “rsmith” is six, the predetermined value is 3 (+3), which is half the number of characters of the personal identification character string “rsmith”. Since the evaluation values obtained as a result of the comparison and collation between the comparison and verification character strings “Ray” and “Smith” and the personal identification character string “rsmith” are as shown in FIG. 26, the evaluation value is “+3” or more. “Smith”, which is “+6”, is selected as a character string representing the name, and “Ray Smith” including the comparative matching string “Ray” that was present on the same line as “Smith” "Is identified as a character string representing the name.

  Although the predetermined value is a value that can be arbitrarily set, it is desirable that the predetermined value is 2 (+2) or more in order to improve the accuracy of the process of identifying the character string representing the name or company name. This is because, when the first character of the compared and matched character strings is matched in the front matching process or the backward matching process, 2 is added to the evaluation value due to the matching of the leading character and the matching of the character. This is because there is a possibility that even if the character strings other than are not matched at all, they are selected as a character string representing the name or company name.

  In business name cards, e-mail address information and network address information are rarely described above the name and company name. Specifically, as shown in FIG. 3, even in the case of an English business card, e-mail address information and network address information are often described below the name and company name. Therefore, according to the above configuration, the character recognition starts in order from the lower character string to the upper character string of the business card image, and the identification character string is generated before the name and company name items. It is possible to reduce the processing amount and increase the processing speed in the processing apparatus 1. That is, in the information processing apparatus 1 according to the first embodiment, all of the business card from the upper side to the lower side is scanned, character recognition is performed on all the scanned cards, and after the character recognition results are obtained, the personal identification character string and company identification are performed. In contrast to generating a character string, the information processing apparatus 1 according to the second embodiment performs character recognition line by line from the lower side of the business card, and when the e-mail address information or the network address information is found, the personal identification character Since the information processing apparatus 1 according to the second embodiment scans everything from the upper side to the lower side of the business card and performs character recognition for all the scanned items, since the information processing device 1 according to the second embodiment is configured to generate the column and the company identification character string. Personal identification character strings and company identification character strings can be generated, and the amount of processing in the information processing apparatus 1 can be reduced and the processing speed can be increased.

  In both embodiments, the identification (item classification) of the character string other than the name and company name is not specified, but as a result of comparison with the personal identification character string and the company identification character string, either the name or the company name The comparison / matching character string that has not been determined may be stored in another storage area, and finally the item classification may be performed on items other than the name and company name. In addition, character strings such as telephone numbers and fax numbers can be easily determined as character numbers such as “TEL” and “FAX”, such as “TEL” and “FAX”. The information processing apparatus 1 may also be configured to identify a character string representing a telephone number, a fax number, or the like.

  Further, in both embodiments, as a result of comparison and matching between the name identification character string and the company identification character string and the comparison matching character string, the comparison matching character string does not include a character string representing the name and the company name. If it is determined, the character string representing the name other than the name and company name is identified first (items other than the name and company name are classified first), and remains in the classification of items other than the name and company name. A configuration may be adopted in which candidates that are frequently used for names (comparison character strings for comparison that satisfy a predetermined condition) are classified into items of names from the comparison matching character strings. On the other hand, with respect to the company name, after the name item is classified, the remaining comparison matching character strings (comparison character strings satisfying predetermined conditions) may be classified into the company name items. . Note that items other than the name and company name mentioned here include, for example, the above-described telephone number, fax number, and address. Further, candidates that are frequently used for names may be determined by holding the first and last name dictionary data in the information processing apparatus 1 and comparing and comparing with the first and last name dictionary data. Even if it does not hold first-name dictionary data, it keeps the condition such as “a relatively large written character string as a name”, which is a well-known technique, and a name that matches the condition is a name. Good.

  In both embodiments, the configuration in which the information processing apparatus 1 includes the character recognition dictionary data storage unit 12 is shown, but the present invention is not limited to this. For example, the character recognition dictionary data storage unit 12 may be stored in a storage device, a server device, or the like outside the information processing apparatus 1. In this case, the character recognition unit 13 may refer to the character recognition dictionary data stored in the character recognition dictionary data storage unit 12 by accessing the external storage device, the server device, or the like.

  In both embodiments, the case where the character recognition is performed on the image data of the English business card is described. However, the present invention is not limited to this, and the case where the character recognition is performed on the image data of the European business card is not limited to this. The invention is applicable. Because e-mail addresses and network addresses are always written in alphanumeric characters, even business cards in Western languages have ligatures or German umlauts, etc. Such special characters are not included. Accordingly, it is possible to proceed with processing in the same manner as when character recognition is performed on image data of an English business card.

  However, since the name and the like in the European business card may include special characters as described above, if the special character is included in the European business card, the conversion is performed by replacing the special character with the corresponding alphabetic character. The information processing apparatus 1 may be further provided with a member that performs the above, so that the present invention can be applied to image data of European business cards. Moreover, the structure which applies this invention with respect to the data after performing the conversion which replaces a special character with a corresponding alphabetic character may be sufficient.

  The present invention can also be applied to business cards written in Roman letters by performing the same processing as described above. In addition, even in Japanese business cards, the information processing apparatus 1 is further provided with a member that converts Japanese into Roman characters, so that the present invention can be applied to image data of Japanese business cards. The present invention can also be applied to data after conversion to be replaced with notation.

  In both the embodiments, a business card is used as a specific medium, and the information processing apparatus 1 identifies the character string representing the name and company name written on the business card. However, the present invention is not limited to this. . For example, the specific medium may be a magazine as well as a business card. In this case, the information processing apparatus 1 can identify a character string representing a store name published in a magazine or the like in the same manner as the process of identifying a character string representing a company name. In other words, if the store name is included in the network address and the e-mail address representing the store's homepage described in a magazine, etc., by using the network address and / or e-mail address, By generating a store identification character string (identification character string) similar to the identification character string and performing comparison and comparison with the comparison and verification character string, it is possible to identify the character string representing the store name.

  In both embodiments, when identifying a character string representing a name or company name, character strings recognized on the same line are identified as one character string (classified as the same item). However, it is good also as a structure which selects the character string whose score is more than a certain fixed value. It is not necessary to use what is specifically defined as the way of assigning the score here. In addition, when selecting a character string whose score is equal to or greater than a certain value, the data of the surname dictionary is separately stored, and a high score is assigned to a frequent name so that the character string corresponding to the frequent name is Processing such as sorting may be performed.

  In the present invention, a character string before “@” of the e-mail address is used for the item classification of the name, and a character string of a specific part after “@” is used for the item classification of the company name. In addition, a character string of a specific part delimited by delimiters (delimiters included in network address information) such as “.”, “_”, “-”, “/”, And “:” included in network address information Is used to classify company name items.

However, the personal identification string used in the email address is unlikely to be the first and last name itself,
A) Last name + Separator + First name + (Others)
B) First name + Separator + Last name + (Other)
C) Part of surname + (separator) + first name + (other)
D) Part of first name + (separator) + surname + (other)
E) Surname (part) + (Other)
F) Name (part) + (Other)
G) (others) + surname (part of)
H) (others) + name (part of)
B) Since there is a pattern such as irrelevant first name and last name, it is difficult to match with the comparison matching character string even if the comparison matching character string is simply compared.

  Therefore, in the present invention, the personal identification character string before “@” of the obtained e-mail address is converted into delimiters (delimiters included in the e-mail address information) such as “.”, “_”, And “-”. By dividing with, a plurality of or one personal identification character string is obtained, compared with the already obtained comparison collation character string, and the most similar comparison collation character string is identified as a name. . In addition, the character string of the specific part after the “@” in the email address is similarly cut out by using a delimiter to make a company identification character string, which is compared and collated with the comparison character string already obtained. The comparison matching character string is identified as the company name. Furthermore, the network address is also extracted as a company identification character string by cutting out a character string of a specific part, compared with a comparison character string already obtained, and the most similar comparison matching character string is identified as a company name. It is configured.

  In the case of the item classification of the store name, a character string before “@” of the e-mail address may be used, or a character string of a specific part after “@” may be used. This is because, in the case of an e-mail address published in a magazine or business name card that contains store information, the store sets its own e-mail address. And the case where the store name is used in the part after “@”.

  In the present invention, the e-mail address and / or the network address used for the classification of the name and / or company name items are obtained as a result of character recognition such as a business card. Therefore, it is necessary to newly provide a dictionary or database. Absent. In addition, as described above, in the present invention, since the configuration does not hold a complicated dictionary and rules, the name and company name written on a specific medium such as a business card can be identified by a relatively simple process. Can do. Furthermore, in the present invention, even if there is no company identification character string for specifying a company name such as “(stock)”, “corporation”, “Corp.”, “Inc.”, the name and company name items are changed. In particular, the name, company name, and store name must be accurately identified in English, where the name, company name, and store name contain few character strings that can be used as indicators for specific classification. Can do.

  In the present invention, in order to make it possible to identify a character string representing a name with higher accuracy, even when the above-described first-name dictionary data is further stored in the information processing apparatus 1, the stored first-name dictionary is stored. The rules do not have to be difficult, you only need to keep a number of first and last names. Therefore, even if the first and last name dictionary data is configured to be further stored in the information processing apparatus 1, it is not necessary to provide a new complicated dictionary or database, and characters extracted from image information about a specific medium. From the column, it is possible to more easily identify a character string representing at least one of a name, a company name, and a store name with high accuracy.

  Finally, each block of the information processing apparatus 1 may be configured by hardware logic, or may be realized by software using a CPU as follows.

  That is, the information processing apparatus 1 includes a CPU (central processing unit) that executes instructions of a control program that realizes each function, a ROM (read only memory) that stores the program, and a RAM (random access memory) that expands the program. And a storage device (recording medium) such as a memory for storing the program and various data. An object of the present invention is to provide a recording medium in which a program code (execution format program, intermediate code program, source program) of a control program of the information processing apparatus 1 which is software that realizes the above-described functions is recorded so as to be readable by a computer. This can also be achieved by supplying the information processing apparatus 1 and reading and executing the program code recorded on the recording medium by the computer (or CPU or MPU).

  Examples of the recording medium include a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a floppy (registered trademark) disk / hard disk, and an optical disk such as a CD-ROM / MO / MD / DVD / CD-R. Card system such as IC card, IC card (including memory card) / optical card, or semiconductor memory system such as mask ROM / EPROM / EEPROM / flash ROM.

  Further, the information processing apparatus 1 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network. The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available. Also, the transmission medium constituting the communication network is not particularly limited. For example, even in the case of wired such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL line, etc., infrared rays such as IrDA and remote control, Bluetooth ( (Registered trademark), 802.11 wireless, HDR, mobile phone network, satellite line, terrestrial digital network, and the like can also be used. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the claims, and the technical means disclosed in different embodiments can be appropriately combined. Such embodiments are also included in the technical scope of the present invention.

  As described above, the information processing apparatus of the present invention, the control program thereof, the computer-readable recording medium recording the control program, and the control method are obtained from a character string extracted from image information about a specific medium. It is possible to easily identify a character string representing at least one of a name, a company name, and a store name. Therefore, the present invention is suitable for an industrial field related to an apparatus that identifies a name from a character string extracted from image information obtained by photographing a medium such as a business card or a magazine, and classifies the name by item such as a name. It can also be used suitably for a mobile phone with a camera function that reads the contents of a character string included in a photographed image and records it in a telephone directory.

It is a functional block diagram which shows the schematic structure of the information processing apparatus in this invention. It is a flowchart which shows an example of the operation | movement flow in the said information processing apparatus. It is the figure which showed an example of an English business card. (A) is a figure which shows an example of the character string obtained by the character recognition part of the said information processing apparatus, (b) is an example of the address information acquired by the address information acquisition part of the said information processing apparatus. FIG. (A) is a figure which shows an example of character strings other than the character string applicable to the address information in this invention, Comprising: (b) is the comparison collation obtained in the character string acquisition part for comparison collation of the said information processing apparatus It is a figure which shows an example of the character string for use. It is a flowchart which shows the detail of an example of the process which produces | generates a personal identification character string and a company identification character string from the email address information in this invention. It is a flowchart which shows the detail of an example of the process which produces | generates a company identification character string from the network address information in this invention. (A) is a figure which shows an example of the personal identification character string in this invention, (b) is a figure which shows an example of the company identification character string in this invention. (A) thru | or (c) is a schematic diagram which shows the specific example of the comparison collation with the character string for comparison and collation in this invention, and a personal identification character string. It is a figure which shows an example of the comparison collation result of the character string for comparison and collation in this invention, and a personal identification character string. (A) is a figure which shows an example of the character string for a comparison collation in this invention, (b) is a figure which shows an example of the company identification character string in this invention. It is a figure which shows an example of the comparison collation result with the character string for comparison and collation in this invention, and a company identification character string. It is a figure which shows an example of the character string obtained in the character recognition part of the said information processing apparatus. (A) is a figure which shows an example of the character string for comparison and collation in this invention, (b) is a figure which shows an example of the personal identification character string in this invention, (c) is this invention It is a figure which shows an example of the company identification character string in. It is a figure which shows an example of the result of having compared and collated the character string for comparison and collation with the personal identification character string in this invention. It is a figure which shows an example of the result of having compared and collated the character string for comparison and the company identification character string in this invention. It is a flowchart which shows an example of the flow of the process in this invention. It is the figure which showed an example of the order in the case of performing the character recognition in this invention. (A) is a figure which shows an example of the network address information in this invention, (b) is a figure which shows an example of the company identification character string in this invention. (A) is a figure which shows an example of the email address information in this invention, (b) is a figure which shows an example of the company identification character string in this invention, (c) is a figure in this invention It is a figure which shows an example of a personal identification character string. It is a figure which shows an example of the character string candidate for a comparison collation in this invention. (A) is a figure which shows the result of the comparison collation with the character string for comparison collation and personal identification character string in this invention, Comprising: (b) is the character string for comparison collation in this invention, a company identification character string, It is a figure which shows the result of these comparison collation. It is a figure which shows an example of the character string candidate for a comparison collation in this invention. (A) is a figure which shows the result of the comparison collation with the character string for comparison collation and personal identification character string in this invention, Comprising: (b) is the character string for comparison collation in this invention, a company identification character string, It is a figure which shows the result of these comparison collation. It is a figure which shows an example of the character string candidate for a comparison collation in this invention. The figure which shows the result of the comparison collation with the character string for comparison collation and personal identification character string in this invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 11 Image data acquisition part 12 Character recognition dictionary data storage part 13 Character recognition part 14 Address information acquisition part (address information acquisition means)
15 Character string acquisition unit for comparison and collation (character string means for comparison and collation)
16 Identification character string generation unit (identification character string generation means)
17 Character string comparison / collation section (character string comparison / collation means)
18 Item classification determination unit (identification means)

Claims (19)

An information processing apparatus that performs character recognition using a character recognition dictionary based on image information of a medium in which a character string is described, and acquires a character string described in the medium,
An address for acquiring address information that is at least one of mail address information that is a character string that represents an e-mail address and network address information that is a character string that represents a network address, from the character string obtained by the character recognition. Information acquisition means;
A comparison verification character string acquisition means for acquiring a plurality of comparison verification character strings that are character strings other than the mail address information and the network address information from the character string obtained by the character recognition;
Based on the address information acquired by the address information acquisition means, for generating an identification string is one of a company identification string for use in at least name personal identification string used to identify an, and identification of the company name An identification character string generating means;
A character string comparison and collation means for comparing and collating the identification character string generated by the identification character string generation means and each of the plurality of comparison and collation character strings acquired by the comparison and collation character string acquisition means;
Based on the result of comparison and collation by the character string comparison and collation means, the comparison and collation character string determined to be similar to the personal identification character string is identified as a character string representing a name, and similar to the company identification character string. the comparison collation string is determined, the information processing apparatus characterized by comprising a identification means you identified as a character string representing the company name.   The address information acquisition unit uses at least one of the specific character and the specific character string included in the mail address information or the network address information, so that at least the mail address information and the network address information The information processing apparatus according to claim 1, wherein any one of them is acquired.   The address information acquisition unit further deletes a character string that is clearly a character string representing a component other than a component of the mail address and the network address from the acquired mail address information or the network address information. The information processing apparatus according to claim 2, wherein address information or network address information is used.   The comparison verification character string acquisition means divides a character string other than the mail address information and the network address information, before and after a predetermined delimiter, among the character strings obtained by the character recognition. The information processing apparatus according to any one of claims 1 to 3, wherein a plurality of character strings for comparison and collation are acquired. The comparison collation string obtaining means further from the plurality of comparative collation string, name and company name except for comparison and collation by deleting a clear string that is a character string representing the The information processing apparatus according to claim 4, wherein the information processing apparatus is a character string.   The identification character string generation unit generates the company identification character string included in the mail address information as a company identification character string when network address information is not acquired by the address information acquisition unit. The information processing apparatus according to any one of claims 1 to 5. The identification character string generation means includes the mail address information before and after specific characters included between the user name and host name of the email address included in the mail address information acquired by the address information acquisition means. by splitting apparatus according to any one of claims 1 to 6, characterized in that for generating said company identification string and the personal identification string. The identification string generating means, the information processing apparatus according to claim 7, wherein said one of the mail address information, be generated by the personal identification string in front of the string than the specific character . The identification string generating means, information according to claim 7 or 8, characterized in that said one of the mail address information is generated as a company identification string behind the string than the specific character Processing equipment. The identification string generation means, further the personal identification character string before and after the delimiter included in the e-mail address information acquired by the address information acquisition means or dividing the company identification string, a new personal identification The information processing apparatus according to claim 8, wherein the information processing apparatus is a character string or a company identification character string. The identification string generating means, by dividing the network address information before and after the delimiter included in the network address information acquired by the address information acquisition means, and generates said company identification string The information processing apparatus according to any one of claims 1 to 10.   The character string comparison and collation unit is configured to forward each of the identification character string generated by the identification character string generation unit and each of the plurality of comparison and collation character strings acquired by the comparison and collation character string acquisition unit. 12. The comparison and collation from each other character, the comparison and collation from characters behind each other, and the comparison and collation while sequentially shifting each other's counterparts for comparison and collation one character at a time. The information processing apparatus according to any one of the above.   The character string comparison / collation means, when performing comparison / collation while sequentially shifting each other's comparison / collation partners one character at a time, one of the identification character string and the plurality of comparison / collation character strings having the smaller number of characters The information processing apparatus according to claim 12, wherein the information is compared and collated while sequentially shifting the characters one by one with respect to the other party.   The identification means, as a result of comparison and collation by the character string comparison and collation means, based on the number of characters that match between the identification character string and each of the plurality of comparison and collation character strings, The information processing apparatus according to any one of claims 1 to 13, wherein it is determined whether or not each of the plurality of character strings for comparison and collation is similar. The identification means determines that the matching character string having the largest number of matching characters with respect to the personal identification character string is similar to the personal identification character string, and the number of matching characters corresponds to the company identification character string. the most common comparison collation string Te an information processing apparatus according to claim 14, characterized that you determined to be similar to the company identification string. When the identification means determines that none of the plurality of comparison matching character strings is similar to the identification character string as a result of the comparison matching in the character string comparison matching means, the plurality of comparison matching character strings out of the satisfying comparison collation string respective predetermined information processing apparatus according to any one of claims 1, characterized in that the identified with name and company name to 15. A method for controlling an information processing apparatus that performs character recognition using a character recognition dictionary based on image information of a medium in which a character string is described, and acquires the character string described in the medium,
From the character string obtained by the character recognition by the address information acquisition means, at least one of mail address information that is a character string representing an e-mail address and network address information that is a character string representing a network address An address information acquisition step for acquiring address information;
A comparison / matching character string for obtaining a plurality of comparison / matching character strings that are strings other than the mail address information and the network address information from the character string obtained by the character recognition by the comparison / matching character string acquisition means. Acquisition process;
The identification string generating means, based on the address information acquired by the address information acquisition means, any of the company identification string used to identify at least the name personal identification string used to identify an, and company name An identification character string generation step for generating an identification character string,
A character string for comparing and collating the identification character string generated by the identification character string generating unit and each of the plurality of comparison and collating character strings acquired by the comparison and matching character string acquiring unit by a character string comparison and collating unit A comparison and verification process;
Based on the result of comparison and collation in the character string comparison and collation step, the identification means identifies the comparison and collation character string determined to be similar to the personal identification character string as a character string representing a name, and the company identification character control method for a comparative collation string is determined to be similar to the column, characterized in that it comprises a identification step you identified as a character string representing the company name.   The control program for operating a computer as said each means with which the information processing apparatus of any one of Claim 1-16 is provided.   The computer-readable recording medium which recorded the control program of Claim 18.

Images