JPH06149881A - Dictionary consultant device, document processor, and digital copying machine - Google Patents

Abstract

PURPOSE:To acquire the comprehensive output of a retrieval result only for a character string(word) requiring retrieval with an easy operation. CONSTITUTION:A dictionary retrieval means 9 retrieves an electronic dictionary 8, and extracts the content of it. A display/output means 11 displays or outputs the content retrieved by the dictionary retrieval means 9. A document input means 2 inputs electronic document data. An enclosure retrieval mark detecting means 3 detects an enclosure retrieval mark decided in advance in the document data inputted by the document input means 2. A retrieval character string extraction means 6 extracts a character string enclosed with the enclosure retrieval mark detected by the enclosure retrieval mark detecting means 3. Retrieval is performed on the electronic dictionary 8 by the retrieval character string extraction means 6, and a pair of the character string and the content is displayed and outputted.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a dictionary lookup device, a document processing device, and a digital copying device, and more particularly to a dictionary lookup device, a document processing device, and a digital copying device in a natural language document processing technique. For example, it is applied to a dictionary lookup unit as a support function of a foreign language document subscription / creation device.

[0002]

2. Description of the Related Art A dictionary lookup function is incorporated in a single dictionary lookup device or a document creation device. In such a device, the character string to be looked up in the dictionary is directly specified. Further, in such a dictionary lookup device, it is possible to more efficiently search a normal book format dictionary and utilize the searched contents in an electronic state. However, the trouble of inputting the target word remains. On the other hand, in a device that performs dictionary lookup while subscribing or translating a document, there is a method of giving a target word by specifying a range or a pointing device on the display screen of the document to simplify the operation. In such a method, a simple operation cannot be used because the work of inputting the document is difficult if the document is not digitized.

However, if a so-called character reading device (or document reading device) is connected to the document input section, a non-electronic document can be easily digitized. Such a reading device is also incorporated in a machine translation device or the like that translates between natural languages, and is also used to improve efficiency in inputting a document in a source language. However, such interactive use is not suitable for the purpose of collectively searching a large number of words and creating a form such as a word list. However, digital copying equipped with a dictionary lookup function is not suitable. In combination with a character reading device, the machine extracts a word in the read document and makes that word (a part of it) the search target.
With this device, it is possible to eliminate the trouble of designating a word, but there is a problem for the user that even a search result of an unnecessary word is output. This is, for example,
Although the useless output can be reduced to some extent by suppressing the search for functional words, there are many outputs that are not really necessary for individual users.

9 to 11 are block diagrams of a conventional dictionary look-up device, in which 31 is an object character string input means, 32a and 3a.
2b is an electronic dictionary, 33 is a dictionary search unit, 34 is a target character string memory, 35 is a display / output unit, 36 is a search content memory, 37 is a document input unit, 38 is a target character string detection unit, and 39 is a document memory. , 40 is an original document (image), 41 is an image reading unit, 42 is a document image memory, 43 is a character pattern dictionary, 44a is a character recognition unit, and 44b is a document recognition unit.

FIG. 9 shows an example of a single dictionary lookup device.
The target character string input means 31 accepts the user's character string input, the dictionary search means 33 searches the digitized dictionary 32a for the character string, and the display / output means 35 displays or outputs the search result. It is a thing. For example, assume that there is an electronic dictionary partially having the contents shown in Table 1.

[0006]

[Table 1]

When the character string "prove" is entered as a search target in such a dictionary, the next search content is displayed as shown in Table 2.

[0008]

[Table 2]

FIG. 10 shows an example of a document creating apparatus or the like which can handle an electronic document, and is a target character string detecting means 38 equipped with a pointing device or the like for the document display screen.
, The target character string is detected from the document. The following is Fig. 9
It is similar to the case of. For example, it is assumed that the document shown in Table 3 is displayed on the screen.

[0010]

[Table 3]

When "proved" in the fourth line on the screen is designated as a search target with a pointing device or the like, for example, a sub-window screen appears on the screen and the search contents as shown in Table 4 are displayed.

[0012]

[Table 4]

At this time, in order to obtain the original form of a word that is utilized / changed, for example, if the object is English, the endings of "ed" or "d", "es" or "s" may be removed. It requires a simple morphological analysis means.

FIG. 11 shows an example in which a so-called OCR device is provided. Document data printed on paper is read by an OCR unit equipped with image reading means, character recognition means, document reading means, character pattern dictionary and the like. It can be computerized. The following is the same as the case of FIG. Some digital copying machines equipped with such a dictionary lookup device print out the search contents as shown in Table 5.

[0015]

[Table 5]

In the example of this English sentence, words called so-called functional words such as articles, auxiliary verbs and be verbs are excluded from the search target, but other words are excluded from the search target Japanese translation. One of them is shown in ruby format. Alternatively, it may be output in a word book format as shown in Table 6 according to the appearance order of the original text.

[0017]

[Table 6]

Of course, such a result is also possible with the arrangement of FIG. 10 without the OCR device. These dictionary lookup devices have the following problems. In the form of having only interactive input means for designating a search word (FIGS. 9 and 10), there is no method for collectively viewing a large number of search results. There is a function that can output many search results at once (Fig.
Even in (1)), the target is unnecessary for the user.

[0019]

[Object] The present invention has been made in view of the above situation, and in the dictionary lookup of words in an electronic document, a batch output of search results of only character strings (words) that need to be searched. To realize the acquisition by a simpler operation, and to simplify the specification of only adding a mark to the head of a dictionary lookup device that needs to add two marks surrounding a search target word. , Also facilitate the correspondence between the original document and the batch output of the obtained search results,
Another object is to provide a document processing device or a digital copying device.

[0020]

In order to achieve the above object, the present invention provides (1)
A dictionary lookup device including a computerized dictionary to be searched, dictionary retrieval means for retrieving the computerized dictionary and extracting contents, and display / output means for displaying or outputting the contents retrieved by the dictionary retrieval means In, document input means for inputting digitized document data, surrounding search mark detecting means for detecting a predetermined surrounding search mark for the document data input by the document input means, and the document input means A search character string extracting means for extracting a character string enclosed by the enclosed search mark detected by the enclosed search mark detecting means, from the document data input in Searching for the digitized dictionary by the dictionary searching means for the character string, and displaying or outputting a set of the character string and its contents by the display / output means. Further, (2) a leading search mark detecting means for detecting a leading leading search mark for the document data input by the document inputting means, and document data input by the document inputting means, A search character string extracting means for extracting a character string after the head search mark detected by the head search mark detecting means, wherein the dictionary searching means targets the character string extracted by the search character string extracting means. Performing a longest match search on the electronic dictionary and displaying or outputting a set of the search word and its contents by the display / output means, and (3) the word of the input document in (2) above. The leading search mark detection is performed based on the result of the word division means for the word division means for dividing and giving the original form of the inflection word and the document data input by the document input means. Search word extraction means for extracting a word starting from a character after the first search mark detected by the means, and targeting the character string extracted by the search word extraction means to the electronic dictionary by the dictionary search means. The display / output means displays or outputs a set of words and their contents, or (4) a document input means for inputting digitized document data and an electronic dictionary to be searched. A dictionary search means for searching the electronic dictionary to extract contents, a display / output means for displaying or outputting the contents searched by the dictionary search means, an image reading means, and an image reading means. A character recognition means for recognizing the obtained image data as a character by a character pattern dictionary, and a character data group obtained by the character recognition means as electronic document data. In a dictionary lookup device including document recognizing means for recognizing, a surrounding mark recognizing means for recognizing a position of a surrounding mark figure from the image data obtained by the image reading means, and a surrounding mark recognized by the surrounding mark recognizing means. A character string detecting means with a surrounding mark for detecting a marked character string in the character data group obtained by the character recognizing means from the position of the figure, and the character detected by the surrounding character string detecting means Searching the digitized dictionary by the dictionary searching means for a string, and displaying or outputting a set of a character string and its contents by the display / output means, further, (5) above (4) ),
A head mark recognition means for recognizing the position of the head mark figure from the image data obtained by the image reading means,
Character recognition means with head mark for detecting a marked character in the character data group obtained by the document recognition means from the position of the head mark figure recognized by the head mark recognition means, and the character recognition with head mark A character string after the character recognized by the means and a search character string extracting means for extracting the character string after the character recognized by the character recognition means with the leading mark, and the character string extracted by the search character string extracting means. Performing a longest match search on the digitized dictionary by the dictionary search means for the character string, and displaying or outputting a set of the search word and its contents by the display / output means; and (6) ) In the above (5),
Word division means for dividing the input document into words and giving the original forms of the utilized words, word mark recognition means for recognizing the position of the word mark graphic from the image data obtained by the image reading means, and the word mark recognition means And a word recognized by the marked word recognition means for detecting a marked word in the word string obtained by the word dividing means from the position of the word mark figure recognized by The electronic dictionary is searched by the dictionary searching means, and the set of the word and its contents is displayed or output by the display / output means. Further, (7) (1) to (1) to In any one of (6), the position information of the character string in the input document is added to the set of the target character string to be displayed or output and its contents and displayed or output. If, furthermore, (8) the (1) to
In any one of (6), displaying or outputting an input document in which position information of a target character string to be displayed or output and its contents is added to the target character string in the input document, or (9) above ( A document processing device including the dictionary lookup device according to any one of 1) to (8), or (10) a digital processing device including the dictionary lookup device according to any one of (4) to (8) above. It is characterized by being a copying machine. Hereinafter, description will be given based on examples of the present invention.

FIG. 1 is a block diagram for explaining an embodiment of a dictionary lookup apparatus according to the present invention. In the figure, 1 is an electronic document, 2 is a document input means, 3 is an enclosing search mark detecting means,
Reference numeral 4 is a document memory, 5 is an enclosing search mark position, 6 is a search character string extracting means, 7 is a target character string memory, 8 is an electronic dictionary, 9 is a dictionary searching means, 10 is a search content memory, and 11 is display / output. It is a means.

According to claims 1 to 3 described below, in order to output a large number of search results at a time, and in order to limit the users to only the necessary ones, the search target in the document is set in advance by the user. Is a method in which only the objects to which the mark is added are searched by adding a mark, and the search results are collectively output. Further, each claim is different in the way of adding a mark. In addition, the addition of the mark of each claim is performed by the document editing means of the device incorporating the dictionary lookup of this system,
Alternatively, the input is a document marked by a separate device.

[0023] claims 1 1, in the document, a character string range specified by a pair of enclosing marks predetermined searched. For example, suppose that the surrounding marks are “# <” and “>#” and the Japanese document is marked as shown in Table 7.

[0024]

[Table 7]

In response to this designation, for example, if the dictionary is a Japanese-English dictionary, a list in a word book format as shown in Table 8 is output as the search result.

[0026]

[Table 8]

As described above, the first aspect can be realized by sequentially extracting and retrieving the character string in the range designated by the mark.

Claim 2 In claim 1, the range of the target character string is specified, whereas in claim 2, only the beginning of the search target is specified.
That is, the longest match search is performed in the dictionary for the character string after the predetermined start mark. In the longest match search, the search target character string is simply extended one character at a time from one character, and the longest target for which the search is successful is set as the result. Actually, if the entries in the dictionary are sorted, and if the search with each character string that has been sequentially extended by one character fails, the entry that includes the character string at the beginning (prefix match) is No, or-The matching entry is behind the target string of the same length in the sort order. Then, the longest target up to that point may be the result. For example, assume that a Japanese document is marked as shown in Table 9 with the leading mark being "@>".

[0029]

[Table 9]

In this case, assume that the contents of the target Japanese-English dictionary are as shown in Table 10 in relation to the second head mark.

[0031]

[Table 10]

The search target character string after the head mark is
"Sentence", "Document", "In the document", "In the document", ... Are sequentially assigned, but the search is successful in "Sentence" and "Document", but fails in "In the document" and A search that includes a character string at the beginning also fails, so the "document" is the longest match. If there is no entry having “Claim” and “Search pair” in the front, the search result is similar to the example of Claim 1. That is, according to the second aspect, it is only necessary to specify the head, not the range. In addition, if the words are separated by white space like English, it is not necessary to use the longest match search, but if you include consecutive idioms of multiple words such as "at last", the longest match search is not necessary. A match search is required. In addition,
For example, if there is a "document search" in the sentence, but you want to search for "document", it is not possible to use the leading mark, and it is necessary to specify with the enclosing mark.

Claim 3 FIG. 2 is a diagram showing another embodiment (claim 3) of the dictionary lookup apparatus according to the present invention, in which 12 is a word dividing means, 13 is a head search mark detecting means, and 14 is Memory by word division, 1
Reference numeral 5 is the position of the head search mark, 16 is the target word memory, and other parts having the same operations as in FIG. In claim 3, the designation method is the same as in claim 2, but the word division result is used to specify the search target character string, and the marked word is set as the search target. For this reason, the word dividing means is required in claim 3. That is, the document is divided into words by the word dividing means in advance, and the original forms of the words that are utilized / changed are also obtained. This word division is a technique already established as morphological analysis. Also, for English, although it is divided into words by white space,
It is necessary to obtain word forms and plural forms. According to the word dividing means, the word dividing result as shown in Table 11 is obtained for, for example, an English sentence and a Japanese sentence.

[0034]

[Table 11]

For example, it is assumed that in the Japanese sentence, the head mark of claim 3 is set to "$>" and the specification shown in Table 12 is made.

[0036]

[Table 12]

In the second designation, since "document" is divided as a word, "document" is to be searched. In the third designation, "search" is the target, whereas in the fourth designation, "search" is the target. Therefore, in the above example, the search result as shown in Table 13 can be obtained.

[0038]

[Table 13]

As for the word to be used, such as "search", the original form cannot be specified in the sentence.
This is possible by referring to the word division result. Although three types of marks are shown above, the combination of the claims makes it possible to specify more flexibly by mixing these marks.

Claims 4 to 6, which will be described below, claim 1 when the OCR device as shown in FIG. 11 is provided.
In the same manner as ~ 3, a large number of search results can be output in a batch, and in order to limit the users to only the necessary ones, the user marks the search objects in the document in advance. Search only for objects with marks,
This is a method of outputting the search results all at once. The mark is placed on the original printed document that is input to the OCR device. In principle, fill in between lines so that it does not affect character recognition. In claims 4 to 6, a surrounding mark, a leading mark, and a word mark are given,
This is an instruction to search.

Claim 4 FIG. 3 shows a further embodiment (claim 4) of the dictionary lookup apparatus according to the present invention, in which 17 is an original document (image), 18 is an image reading means, and 19 is an image reading means. Is a document image memory, 20 is a character pattern dictionary, 21a is a character recognition means,
Reference numeral 21b is a document recognition means, and other parts having the same functions as those in FIG. 2 are denoted by the same reference numerals. In claim 4, an enclosing mark for designating the entire target character string is provided. Although various forms of the enclosing mark are possible, it is easy to draw a line or a wavy line at the bottom of the character string, as shown in FIG. 5, for example. Further, since the copying apparatus has a function of recognizing a line segment surrounded by a fluorescent marker for partial extraction, it is possible to specify that a character string is surrounded by a marker as shown in FIG. 6 (FIG. 6). Then, the dotted line encircling the fluorescent marker is shown). The character recognition result and the position of the recognition result of the enclosing mark are collated, and the character string given the enclosing mark is specified. In FIG. 5 and FIG. 6, 1 respectively
The line “request”, the second line “document” and “search” are specified. The following is the same as in claim 1.

Claim 5 In claim 5, a head mark for designating the head of the target character string is given. Actually, the head mark can have various formats. For example, as shown in FIG.
It is easy to put a check mark or a line at the bottom or top of the first letter. The position of the result of character recognition and the position of the result of recognition of the head mark are collated, and the character with the head mark is specified. In FIG. 7, “Contract” in the first line “Claim is ...” and “Sentence” and “Check” in the second line “Search target in document ...” are specified. The following is the same as in claim 2.

Claim 6 FIG. 4 is a diagram showing another embodiment (claim 6) of the dictionary lookup apparatus according to the present invention, in which 22 is a word mark recognition means, 23 is a word mark position, and 24 is a word mark position. In the attached word recognition means, other parts that have the same operation as in FIG. 3 are given the same reference numerals. In claim 6, a word mark for designating a target word is given. Further, in the sixth aspect, similarly to the third aspect, a means for dividing the document data subjected to the document recognition (character recognition) into words is provided. Although various forms of word marks can be used in practice, for example, as shown in FIG. 8, it is easy to write a line segment or a check mark on a part or the whole of a lower part or an upper part of a word character string. . The character recognition result and the position of the recognition result of the word mark are collated, the character or character string with the word mark is specified, and the word division result of the document data recognized by the document and the character or character specified by the mark are specified. From the column, the word that contains it can be identified. Document data of character recognition result. In FIG. 8, although depending on the result of word division, “request” on the first line, “document” and “search” on the second line, and “search” on the fourth line are specified. The following is the same as that of claim 3.
Although the example of the mark in the Japanese sentence is shown in the figure, in the English sentence, the word is divided by a space, so the claim 4
With the enclosing mark, if you specify the entire character string separated by white space, you do not have to specify the entire character string. Further, claim 5 can be substituted by the former method even if it is not particularly used, and there is not much inconvenience.

According to claims 7 and 8 described below, reference information to the other is added to one or both of the search result output and the input document to improve mutual reference efficiency. Claim 7 In claim 7, the position information of the search target in the input document is included in each search result output in the search result output. For example, Table 14 shows an example of correspondence when a page and a line in an input document are given as position information.

[0045]

[Table 14]

Claim 8 According to claim 8, the search target in the input document is also output including the position information of the search result in the output of the search result. For example, Table 15 shows an example of correspondence when a page and a number in a search result output are given as position information.

[0047]

[Table 15]

It should be noted that the eighth aspect includes the case where the retrieval result output in the seventh aspect is combined with the position information of the retrieval target in the input document.

Claims 9 and 10 described below are inventions of an application device incorporating the above dictionary lookup device. Claim 9 Claim 9 is a document processing apparatus incorporating the above dictionary lookup apparatus. Although there are already many document processing devices having a dictionary lookup function, the utility of the document lookup device of the present invention is further enhanced. For example, a foreign language sentence creation support system can be considered. For example, an English creation support system is equipped with the dictionary lookup device of the present invention having a Japanese-English dictionary as a dictionary. The document editing function of the system must be able to handle Japanese as well as English sentences. If you have the original Japanese sentence in advance, you can add various marks to it to output a list of parallel translations of the words you want to know in the wordbook format. Furthermore, Japanese OC
By including the R device, even if the original Japanese is printed on paper, a similar list can be obtained by writing a mark on the paper. Conversely, a foreign language text subscription support system is also conceivable. The English text subscription support system is equipped with an English-Japanese dictionary. When equipped with an OCR device, it is equipped with an English OCR device. This foreign language subscription support is
It is useful for subscribing a large amount of digitized foreign language texts in e-mail exchanges with foreign countries and electronic news systems. In addition, a word book creating system having an editing function or an OCR function is also possible.

Claim 10 Claim 9 is a digital copying apparatus incorporating the above dictionary lookup apparatus. Although a digital copying apparatus having a dictionary lookup function already exists, the dictionary lookup apparatus of the present invention makes the utility even more useful.

[0051]

As is apparent from the above description, the present invention has the following effects. (1) Effect corresponding to claim 1: In dictionary lookup of words in a digitized document, acquisition of batch output of search results of only words that need to be searched is realized by a simpler operation. (2) Effect corresponding to claim 2: Surrounding search target word 2
In addition to claim 1 in which it is necessary to add one mark, the specification is simply simplified by adding the mark to the head of the claim. (3) Effect corresponding to claim 3: The problem is further solved for claim 2 in which there is a possibility that there is a search failure in the conjugation word. (4) Effect corresponding to claim 4: In the dictionary lookup of words in a document that is not digitized, batch output of search results of only words that need to be searched can be obtained by a simpler operation. (5) Effect corresponding to claim 5: 2 surrounding a search target word
In addition to the claim 4 in which it is necessary to add one mark, the specification is simply simplified by adding the mark to the head part. (6) Effect corresponding to claim 6: The problem is further solved with respect to claim 4 in which there is a possibility that a search failure may occur in the conjugation word. (7) Effects corresponding to claims 7 and 8: It is realized that the original document and the batch output of the obtained search results are easily corresponded. (8) Effects corresponding to claims 9 and 10: To provide a document processing device or a digital copying device having the above-mentioned effects. That is, a document processing device or a digital copying device having a more useful dictionary lookup function is provided.

[Brief description of drawings]

FIG. 1 is a configuration diagram for explaining an embodiment (claim 1) of a dictionary lookup device according to the present invention.

FIG. 2 is a diagram showing another embodiment (claim 3) of the dictionary lookup apparatus according to the present invention.

FIG. 3 is a diagram showing still another embodiment (claim 4) of the dictionary lookup apparatus according to the present invention.

FIG. 4 is a diagram showing still another embodiment (claim 6) of the dictionary lookup apparatus according to the present invention.

FIG. 5 is a diagram showing an example of a mark (designation of a character string) in claim 4 of the present invention.

FIG. 6 is a diagram showing another example of a mark (designation of a character string) according to claim 4 of the present invention.

FIG. 7 is a diagram showing an example of a mark (designation of a leading character) in claim 4 of the present invention.

FIG. 8 is a diagram showing an example of a mark (designation of a word) according to claim 4 of the present invention.

FIG. 9 is a diagram showing a conventional dictionary lookup device.

FIG. 10 is a diagram showing another example of a conventional dictionary lookup device.

FIG. 11 is a diagram showing still another example of a conventional dictionary lookup device.

[Explanation of symbols]

1 ... Digitized document, 2 ... Document input means, 3 ... Enclosed search mark detection means, 4 ... Document memory, 5 ... Enclosed search mark position, 6 ... Search character string extraction means, 7 ... Target character string memory,
8 ... Electronic dictionary, 9 ... Dictionary search means, 10 ... Search content memory, 11 ... Display / output means.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical indication H04N 1/387 4226-5C

Claims (10)

[Claims] 1. An electronic dictionary to be searched, a dictionary search means for searching the electronic dictionary to extract contents, and a display for displaying or outputting the contents searched by the dictionary search means.
In a dictionary lookup device including output means, a document input means for inputting digitized document data, and a surrounding search for detecting a predetermined surrounding search mark for the document data input by the document input means A mark detecting means; and a search character string extracting means for extracting a character string enclosed by the enclosing search mark detected by the enclosing search mark detecting means from the document data input by the document inputting means, The dictionary search unit searches the electronic dictionary for the character string extracted by the search character string extraction unit, and the display / output unit displays or outputs a set of the character string and its contents. A dictionary lookup device characterized in that 2. A head search mark detecting unit for detecting a predetermined head search mark for the document data input by the document input unit, and the document data input by the document input unit for the document data input by the document input unit. A search character string extracting means for extracting a character string after the first search mark detected by the first search mark detecting means, wherein the dictionary searching means is provided for the character string extracted by the search character string extracting means. The longest match search is performed on the electronic dictionary and the above display /
2. The dictionary lookup apparatus according to claim 1, wherein the output means displays or outputs a set of search words and their contents. 3. The word division of the input document is performed, and the inflection word also gives the original form of the utilization word, and with respect to the document data input by the document input means, from the result of the word division means, the leading part is obtained. A search word extracting means for extracting a word starting from a character after the first search mark detected by the search mark detecting means, wherein the electronic dictionary is searched by the dictionary searching means for the character string extracted by the search word extracting means. 3. The dictionary lookup apparatus according to claim 2, wherein the dictionary is searched, and the display / output means displays or outputs a set of words and their contents. 4. A document input means for inputting electronic document data, an electronic dictionary to be searched, a dictionary search means for searching the electronic dictionary and extracting contents, and a search by the dictionary search means. Display / output means for displaying or outputting the displayed contents, image reading means, character recognition means for recognizing the image data obtained by the image reading means as a character by a character pattern dictionary, and the character recognition means. A dictionary recognizing device having a document recognizing means for recognizing the character data group as electronic document data, and a surrounding mark recognizing means for recognizing the position of the surrounding mark figure from the image data obtained by the image reading means. , A mark in the character data group obtained by the character recognizing means from the position of the surrounding mark figure recognized by the surrounding mark recognizing means A character string detecting means with an enclosing mark for detecting the attached character string, and searching the electronic dictionary by the dictionary searching means for the character string detected by the enclosing character string detecting means And displaying or outputting a set of a character string and its contents by the display / output means. 5. A head mark recognition means for recognizing the position of the head mark figure from the image data obtained by the image reading means, and a document recognition means for recognizing the position of the head mark figure recognized by the head mark recognition means. A character recognition means with a head mark for detecting a marked character in the obtained character data group, and a character recognition means with a head mark for a character after the character recognized by the character recognition means with a head mark A character string extracted by the search character string extraction means, and a character string extracted by the search character string extraction means. 5. The dictionary lookup apparatus according to claim 4, wherein a match search is performed, and the display / output means displays or outputs a set of search words and their contents. 6. A word dividing means for dividing a word of an input document and giving an original form of a utilized word, and a word mark recognizing means for recognizing a position of a word mark figure from image data obtained by said image reading means, A marked word recognition means for detecting a marked word in the word string obtained by the word dividing means from the position of the word mark figure recognized by the word mark recognition means, and the marked word recognition means 6. The word recognized by the above is searched for in the electronic dictionary by the dictionary searching means, and the set of the word and its contents is displayed or output by the display / output means. The dictionary lookup device described. 7. The character string position information in the input document is added to the set of the target character string to be displayed or output and its contents, and the character string position information is displayed or output.
6. The dictionary lookup device according to any one of 6 above. 8. The input document in which position information of a target character string in display or output and its content is added to the target character string in the input document is displayed or output. The dictionary lookup device according to item 1. 9. A document processing device comprising the dictionary lookup device according to claim 1. 10. A digital copying apparatus comprising the dictionary look-up device according to any one of claims 4 to 8.