JP2003018393A - Image processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing system that surely acquires information added to a print image. SOLUTION: The image processing system is configured such that the system divides a print original image into areas in advance, stores divided image information, adds and prints out division image information to a print document as a bar code at the same time, reads and interprets the bar code from a document added to the print document to acquire divided image information, divides the added document image by the divided image information, positions the added image divided and the original document image divided at a bar code read position so as to process the image thereby acquiring the added information.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing system, and more particularly, to a print image in a document information service to a customer, a document information retrieval system, a handwritten information separation system, a document management system, a proofreading system, and the like. The present invention relates to an image processing system that correctly and efficiently obtains information superimposed on a print image by adding or the like.

[0002]

2. Description of the Related Art At present, so-called digitization and digitization of information is widely performed, and the editing work is often performed on a computer. However, on the other hand, editing work on a computer can perform advanced work as such, but it takes time to learn the handling method of the user interface / man-machine interface peculiar to the software to be used, and character input etc. However, there is a problem in that character conversion work is required and a lot of trouble is required to correct an input error. As a result, it is often difficult to immediately reflect the ideas developed in the user's mind in the electronic document. On the other hand, the work of "writing" (handwriting), which is a basic motion of human beings, can only be simple, but has the characteristic that it can sufficiently cope with the speed of human thinking.

In recent years, a large number of configurations have been devised to realize a system having an excellent user interface by causing a computer to execute a complicated process and adopting a handwriting action as an input for the process to follow the speed of human thinking. .

For example, Japanese Unexamined Patent Application Publication No.
In the 2000-57260 publication, "extraction of information to be entered" is disclosed, and there is shown a method for accurately extracting the input information from a completed form containing a color image.

Further, Japanese Laid-Open Patent Publication No. 11-328309 discloses "Optical character reading method and its device", in which character ruled lines and character patterns are separated from an electronic image of a completed form and then the character patterns are extracted. A method of statistically extracting the feature amount and determining the category to which the character pattern belongs is shown.

Further, Japanese Patent Publication No. 61-005195 discloses a "character separation device" and Japanese Patent Publication No. 61-041029 discloses an "image input device", which recognizes handwritten characters written on a form. A method of using a field mark for this is shown.

Further, Japanese Patent Laid-Open No. 9-2 by the same applicant as the present application
The 45173 publication discloses an "image processing method",
There, a configuration is shown in which an element used for alignment is automatically selected from a filled-in image, and the alignment is accurately performed between the blank image and the filled-in image.

Further, Japanese Patent Laid-Open No. 10-09 filed by the same applicant as the present application
In Japanese Patent No. 1783, an "image registration method" is disclosed, in which line segments are crossed from both a reference image before additional writing and an image after additional writing as a preprocessing for recognizing an additional character from a form. There is disclosed a configuration in which a cross point is detected, and alignment / rotation correction is performed by the cross points corresponding to both.

Furthermore, the information of the document information format and the check box checked by the user by adding is embedded in the document to be printed as a barcode, and the document checked by the user is read by the scanner and embedded in the barcode. A system has been proposed in which the desired service instructed by the user is performed by decoding the existing information and detecting whether or not the check box is added.

As described above, in order to effectively process the handwritten information, in addition to the layout information of the printed matter and the function of recording and calling the handwriting action information in the system, how it is superimposed (added) on the manuscript. The issue is how to obtain the editing result (image).

Regarding the acquisition of the retouched image, the image is optically read and character recognition is performed from the read image.
CR technology is common.

[0012]

According to these methods, an objectized area such as a frame, a table, a figure, and a mark is first recognized from a read image, and this area is set as an extraction range, and a document existing in the extraction range is recognized. An added image is obtained by performing image processing on the image and the added superimposed image, and the added content is recognized by OCR from the added image. Therefore, in the case of this method, it is difficult to separate the added content overlaid on the sentence in which a general object does not exist because the extraction area cannot be specified.

The present invention has been made in view of the above problems, and an object of the present invention is to appropriately form a background image (original document image) even when an original is composed of only text. It is possible to manage and correctly extract the editing information even in the situation where there are many background images such as proofreading.
It is to provide a system capable of proofreading a sentence or the like on a paper as electronic information.

[0014]

According to a first aspect of the present invention, means for dividing a print image into a plurality of areas at the time of printing a document and the resulting document image division information are associated with the divided document images. Means for storing.

According to the second aspect of the invention, the document image division information is composed of the size of the division area and the coordinate position of the division area from a specific reference position.

According to the third aspect of the invention, the document image division information further comprises divided document images.

The invention according to claim 4 further comprises means for additionally printing a predetermined code storing the document image division information and the link information to the document information on the print document.

In a fifth aspect of the invention, the predetermined code further stores the document image division information itself.

According to the sixth aspect of the invention, the coordinate information contained in the document image division information is based on the coordinate system with at least one position of the predetermined code to be printed as a reference.

According to a seventh aspect of the present invention, a step of selecting a color that is not used in the document at the time of printing the document and a step of additionally printing a mark of the selected color on the print document in a mode related to the document image division. And consists of.

The invention according to claim 8 further comprises means for storing color information adopted for a mark additionally printed in a mode related to document image division at the time of document printing together with corresponding document image division information and document information. Become.

According to a ninth aspect of the present invention, the mark has a size that is three times or more the size of the smallest dot that can be printed by a predetermined printing device.

According to a tenth aspect of the invention, means for inputting a document image, means for reading a predetermined code from the input document image, means for decoding the read code, and a document of the input document image based on the decoding result A unit for reading the image division information, a unit for dividing the input document image based on the read document image division information, a divided document image obtained by dividing the input document image, and an original document image corresponding to the input document image. And a means for obtaining an image superimposed on the input document image from the divided document image.

According to the eleventh aspect of the present invention, the means for acquiring the image superimposed on the input document image is based on the process of obtaining the image difference between the divided document image of the input document image and the corresponding divided document image of the original document image. Become.

According to the twelfth aspect of the present invention, a mark additionally printed on the print document is used for alignment between the input document image and the original document image in a manner related to the document image division of the divided document image information. This is the configuration.

According to the thirteenth aspect of the invention, an arbitrary image included in the print document image is used for the alignment between the input document image and the original document image.

The invention described in claim 14 is the image superimposed by using a predetermined threshold from the image superimposed on the input document image acquired by the means for acquiring the image superimposed on the input document image. And means for determining the line width of.

According to a fifteenth aspect of the present invention, there is provided means for inputting a document image, means for printing and outputting the input document image, means for communicating with the outside, and means for controlling the entire apparatus. Performs one of the functions.

[0029]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

First, according to the present invention, when an arbitrary manuscript, particularly a manuscript in which sentences are described, is rewritten by a writing instrument such as a pen or a pencil, a digital image of both (the original document and the rewritten document) From the predetermined image processing,
It provides a system that accurately extracts the image of the part that has been retouched (the retouched image), performs optical character recognition processing (OCR) on the retouched image, and acquires the retouched content (the retouched information). The present invention is area management for cutting out a corresponding area between an image of a document before being edited (a document / original image) and a document image after being edited (a superimposed image for editing) that is performed in a pre-process when extracting a modified image. It is characterized by

First of all, a method of acquiring the correction information by OCR which is now generalized will be described with reference to FIGS.

On the left side of FIG. 1, there is shown an order table 2 in a standard format defined by a correction information acquisition system (not shown), and a handwritten input field 2a for entering orderer information is provided below the order table 2. There is. The orderer enters predetermined items in the entry field 2a.

In this correction information acquisition system, in order to acquire the contents (correction information) corrected by the user on the manuscript, first, the image of the order table 2 that has been corrected is converted into image data by an input device such as a scanner. Then, a touch-up superimposed image is obtained. Next, the system recognizes the object area (extraction area 100) in which the editing area exists (area recognition).

Here, in specifying the extraction area 100,
As in the method described in Japanese Patent Laid-Open No. 2000-57260 described above, a configuration is used in which the system extracts in advance by a base color that is added in advance when creating a standard format, and a standard method such as the method disclosed in Japanese Patent Laid-Open No. 11-328309. The system extracts the area by recognizing the ruled line which is a character frame based on the position of the ruled line set by the system at the time of creating the format, or the system at the time of creating the standard format like the method shown in Japanese Patent Publication No. 61-005195. A configuration or the like for recognizing the field mark set by and extracting the area is applied. In any configuration, the system uses the format information as defined in the order table 2 to determine the feature quantity of the area (in the case of this example, as shown in FIG.
It is determined by image processing using the thick black frame 2a in the center.

Next, the system uses the extraction area 100 to take out an image of an area in which character recognition is performed from the entire touched-up superimposed image in this area (touched-up image area cutting). In addition, in the case of addition to the format determined in this way, the target area on the original is often blank (that is, blank) originally, and thus the original superimposed image may not exist in the added-superimposed image of the extraction area. Many. Therefore, in general, this system regards the image extracted here as it is as a retouched image, and performs OCR on the entire image acquired here to obtain retouched information.

As a special case, there may be a document image that is not originally blank in the target area on the document, such as "address" and "name" in the upper right of FIG. In such a case, the system prepares a document image 101 corresponding to this area in advance, and aligns these two images including the image in the state after the correction with a predetermined algorithm. As the image alignment technique, for example, the methods disclosed in JP-A-9-245173 and JP-A-10-091783 are applied. This image alignment is performed by detecting a predetermined mark or by detecting and utilizing an appropriate alignment element from a character image in the area.

Next, the system obtains the modified image 102 shown in the lower right part of FIG. 1 (approved image acquisition) by appropriate image processing, for example, image difference calculation between the two.

Finally, the system performs character recognition on the character area of the thus-corrected image to be added, and acquires the added content (addition information) 103 (addition information acquisition).

In the above-mentioned method, the characteristic amount such as the table, figure, mark, color, and density determined by the system is referred to before the acquisition of the retouching information, the object corresponding to this is recognized, and the retouching is performed as a result. The region where the superimposed image is added is cut out. Such a method of acquiring the retouching information is effective mainly for a manuscript having a standard format.

On the other hand, in the above-mentioned method, when the manuscript is a general sentence as shown in FIG. 2, unless the feature amount is forcibly added, the retouched contents superimposed on it are acquired. It is difficult.

FIG. 2 shows an example in which the target is a sentence, and as the contents of addition to such a sentence, sentence proofreading and memo writing are often performed. Here, "grammar correction" is shown as an example. 9A shows a printed document before proofreading, and FIG. 9B shows a document (handwritten overwritten) over the document of FIG.

As described above, in order to perform character recognition, it is necessary to add a mark having some characteristic to the original document in advance in order to obtain the extraction area. Here, the mark for that purpose is called a position mark 104. The illustrated example is an example in which a cross (so-called “dragonfly”) is written as the position mark. If there is a position mark like this,
First, the system converts a touch-up overlay original into a touch-up overlay image by a reading device such as a scanner, and further obtains the position of the position mark by a predetermined algorithm, and divides it as shown in FIG. The extracted area 123 is obtained (area recognition unit 12). After that, the content of the editing is extracted by various techniques as described above (cutting of the editing area, acquisition of the additional image, acquisition of additional information).

However, when the position marks are arranged in the irregular format as shown in the example of FIG. 2, the following problems occur. (1) The position mark 104 needs to have a distinctive image feature with respect to characters and symbols used in a sentence. That is, it is necessary to have a conspicuous shape. (2) When a position mark is placed in the text area 122,
This makes it difficult to read the characters and, as a result, makes the text difficult to read. (3) On the other hand, when the position mark is arranged outside the text area 122 (margin), the position mark may be overwritten by the retouching for proofreading, and as a result, the position mark itself may not be imagewise detected.

Further, in a general document management system, maintenance / update management processing is performed at the timing of document storage or when a user registers a document in the document management system. That is, since the document printing is performed in the middle of the document management timing, the print image generated from the digital document finally registered in the document management system may be different from the actually printed image. In other words, this is the case when some kind of edit is made and then saved after printing. As an example of this, it is conceivable to slightly change the font size or height of a character after printing. Therefore, when performing some kind of editing operation on the information printed on paper, inconsistency with the electronic data may occur in the management form performed by the existing document management system. In such a case, inconsistency will occur between the retouched document image added to the printed document and the original image saved in the system, and as a result, proper matching between the two will not be possible, and thus the retouched information will be acquired correctly. You may not be able to do it.

Therefore, in the embodiment of the present invention described below, in the situation such as "text proofreading", there are problems in arranging the position mark, and the printed document which is the object of the text proofreading. Implemented a configuration and functions to solve problems such as inconsistency with electronically stored documents and to always perform stable processing on irregularly printed documents such as proofreading for printed documents did.

More specifically, the timing for managing information is a print command, and the means for cutting out the editing area does not use position marks or objects. These are document management independent of applications such as grammar correction. By configuring the system and the correction information acquisition system so that they can be performed, it is possible to provide a system that can acquire correction information by a method common to various applications.

First, the schematic configuration of the system according to the embodiment of the present invention will be described with reference to FIG.

The embodiment of the present invention roughly comprises a document management system 1 and a correction information acquisition system 2, and the document management system 1 stores a print information of a print document original and a storage unit 11 for storing the print information. A printing unit 12 that prints a print document original on the basis of a print document, a dividing unit 13 that divides print information (print image) of the print document original into a plurality of regions, and attributes of each region thus divided (print image division Information) and the divided print images thereof are associated with each other, and further associated with the corresponding document information and stored.

In addition, the correction information acquisition system 2 uses the input unit 21 for scanning and inputting the correction superimposed document in which the correction information is superimposed on the original print document, and the input image data (scan data) to generate the correction superimposed document. A bar code specifying section 22 for specifying the existing position of the included bar code and its area, a bar code decoding section 23 for decoding the bar code, and the printing of the original document to be printed based on the information obtained by the bar code decoding. B. A division information reading unit 24 for reading the image division information (from the division information storage unit 14), a division unit 25 for dividing the input image data based on the read print image division information, and an input image divided. The divided input image data and the divided original print image data stored in the divided information storage unit 14 are subjected to image processing such as comparison and collation calculation. The more Te for example, comparison and collation processing unit 26 to obtain the revision information as the difference information.

Next, the embodiment of the present invention will be described in more detail with reference to FIG.

FIG. 9C is a hierarchical structure of an application such as "sentence proofreading" that the user directly operates through the program, and a configuration in which the application operates in cooperation with the document management system and the editing information acquisition system within the program. Is shown in.

In the figure, the top is the user interface, the second is the logic layer, the third is the functional layer, and the bottom is the database.

The logic layer realizes a desired function by putting various functions starting from the printing function existing in the function layer on the flow. Since the present invention relates to the document management system and the additional information system shown in FIG. 1 and the printing method for printing them, the application actually used therein is not specified.

FIG. 4A shows an example of a management table for explaining one function of the document management system according to the embodiment of the present invention shown in FIG. 4C.

In this document management system, similarly to a general document management system, a command existing from an application, such as a document storage / drawing / updating command, is received, and the management existing in the database layer while using the function of the function layer. Maintain / update table. FIG. 4A shows a typical example of the management table. Such a management table exists in the database, and a unique ID is automatically given to the input / output document, and its attribute (creation date / version / document type / document name / ... ) Manage. In such a document management system, as described above,
The operation is triggered by the document storage command from the application. Therefore, there is a time lag in the generation of the information printed by the application and the digital document information held by the document management system.

In the modification information acquisition system according to the embodiment of the present invention, as shown in the operation explanatory view of FIG.
Image processing such as position adjustment, contour enhancement, and contrast adjustment of the two is performed using two images (the same figure (A)) and the added image 301 (the same figure (B)). (D)
It has a function to obtain a retouched image as shown in.

It should be noted that this system is configured to control various functions existing in the functional layer from the logic layer as shown in FIG. 4 (C).

Further, the purpose of this system is to extract the retouched portion as a retouched image, perform character recognition on the retouched image as necessary, and obtain digital information such as a character code. Therefore, as shown in FIG. 5D, when the alignment between the two base images 300 and 301 is poor, the afterimage of the original 300 remains in the result after image processing, and character recognition cannot be correctly applied. Become. Therefore, what is required in the present system is to reliably acquire the original image 300 and the touch-up superimposed image 301 and to perform the alignment between them.

The operation / principle of the embodiment of the present invention for achieving such an object will be individually described below.

First, it is similar to the means for solving the problem caused by the time difference between the print timing and the post-update storage timing, and the means for managing the information for surely obtaining the correction information by the document management system. explain about.

The first feature is to operate the document management logic with a print command of an application as a trigger, and the second feature is to take out a document instructed to print as a digital image based on this trigger,
The digital image of the document thus taken out is divided by a predetermined method (dividing unit 13 in FIG. 3), and this division information is maintained by the document management system (storage unit 14).

FIG. 4 shows an example of a management table of the above division information.
It is shown in (B). In this example, a divided document ID is assigned at each print instruction timing, and the print target document ID, print time (creation date and time), and actual print image division information are stored in the record thus obtained. The attributes such as file path to be managed are managed in a table.

Next, the print image division information will be described with reference to FIGS. 6, 7, 8, 9, and 10. 6 to 10 show an example of the case where "sentence proofing" is used as the application.

FIG. 6 is a manuscript in which proofreading correction information is added, and FIG. 7 is a proofreading process after electronically extracting the proofreading correction information and issuing a change command to the system from this information. FIG. 8 shows the divided information managed by the system in a visualized manner (divided area: indicated by a broken line), and FIG. 9 shows the original image and the overwriting overlay on a part (corner) of the divided area. An example in which a mark for performing alignment with an image is additionally printed is shown, and FIG. 10 shows a description example (electronic data) of the print image division information.

The operation of transitioning from the state of FIG. 6 to the state of FIG. 7 can be realized by implementing a function of treating a system-defined additional character as a command,
Since such processing can be performed by a known method, its description is omitted here.

In the embodiment of the present invention, the image division information as shown in FIG. 8 is realized by the print image division function 1207 of FIG. 4C, which is referred to as “reference file path” of FIG. 4B. As the contents of the file specified as
For example, it has a function of holding information as shown in FIG. 10 as data.

Many other information holding methods and print image dividing means can be considered in this document management system, but the form thereof is not particularly limited. The print image can be divided into any size and shape. For example, the divided areas may be overlapped as necessary. In any case, the position / shape information / color information for each divided area unit, for example, if the area is a rectangle, the coordinates, width, height, etc. of each vertex of the rectangle should be registered as print image division information. You can

By including a new function in the document management system as described above, the digital data management that is always synchronized with the printed information and the content position management on the paper that can be used for acquiring the correction information are It can be managed independently of applications that handle non-standard documents such as proofreading and individual applications that use standard documents such as questionnaires.

Further, the features of the embodiment of the present invention will be described below. This feature relates to the characteristics of the print image division information (document image division information) (particularly the management origin). In reality, the size, shape, color, relative position, etc. of the area should be managed as the print image division information. In the case of "sentence proofing" as an example, image information obtained from the print document. By accurately recognizing the positional relationship with the corresponding area of the digital document as the original document, which is stored in advance in the system, it is possible to obtain desired retouching information.

For example, by recognizing the first character of the printed document, searching for the character corresponding to the division information of the digital document corresponding to this position, and associating it with the position of the division information from the edge of this character. The positional relationship between the entire page of the printed document and the division information of the digital document can be uniquely determined. However, in the case of such a method, a separate calculation is required to obtain the positional relationship between the entire page and the division information of the digital document.

Therefore, in the above-described embodiment of the present invention, since such a calculation process is unnecessary, it is possible to configure the system in advance to specify the management origin of the print image division information managed by the system. Is. If the management origin is determined in this way, the position on the entire surface of the printed document and the division position of the digital document are directly linked,
It is possible to speed up a series of processes. As the management origin of the present invention, a part that can be managed as a print document can be designated.
For example, the edge or corner of the printing paper, or any character in the printed text can be used as a reference.

As described above, in the embodiment of the present invention, the print image division information is for maintaining / holding the image of the actually divided document in the document management system.
Therefore, by retaining the divided image in advance, it is not necessary to dynamically create the image image used for acquiring the additional information, and the divided image image prepared in advance can be provided immediately.

Since the document management system according to the embodiment of the present invention manages various information of printed documents,
From the image of the printed document, it has a function of determining which record in the management table the image corresponds to, and appropriately extracting desired related information regarding the record as digital information. This function can be generally realized by the cooperation information.

Here, the cooperation information means data that maintains at least unique information that can identify a document in the table managed by the document management system, and the cooperation information is printed on a sheet of paper in some form. The above function can be realized by adding In the case of the configuration of the above-described embodiment of the present invention, this linkage information is in some form, for example, a method of describing an ID such as a document management ID or a divided document at the position of the header or footer of the document to be printed. Can be realized with. As an ID recognition method in this case,
An automatic identification method such as character recognition can be applied. This method may be one that is commonly used.

Further, a bar code may be used to acquire this cooperation information. The barcode used here may be either one-dimensional or two-dimensional. The barcode is a technology for encoding (encoding) information into an image, and the information stored in the barcode can be kept unknown only by a third party viewing it. For this reason, it is an effective means in terms of improving security, and is a technique generally used for linking physical paper and digital information.

The characteristic feature here is that, in the case of a normal document management system, the related information for the entity of the digital document being managed is stored as a bar code, whereas in the embodiment of the present invention it is generated at the time of printing. Store related information to print image division information as a barcode (print the barcode additionally to the print document)
It is possible. An example of this configuration is Document I
The method of storing D and the version number in the barcode and storing the information that can refer to the print image division information in the record of the document ID, or dividing directly using the table as shown in FIG. It is assumed that the document ID is stored in the barcode. The restoration of the cooperation information from the barcode can be realized by incorporating a decoding routine corresponding to the type of barcode in the document management system.

In the case of the embodiment of the present invention, new management information is generated each time one digital document is printed. Therefore, the database of the document management system that maintains these management information is loaded with an increase in the amount of data. In addition, as a problem of possessing management information as electronic data for each sheet of paper, when the paper is discarded, the management information corresponding to the database is "garbage" (it is virtually unnecessary but should be deleted from the database. However, there is a problem that the data becomes unreadable.

Therefore, the embodiment of the present invention has the following configuration in order to solve these problems.

That is, it is assumed that the print image division information itself of the printed document is stored in the bar code and additionally printed on the print document. In this case, a two-dimensional barcode having a large storage capacity is used as the barcode. As a result, the encrypted management information (that is, the bar code) is inevitably discarded along with the destruction of the paper surface, and the above problems can be solved. The type of the two-dimensional barcode used to realize this function is not specified.

Further, in the embodiment of the present invention, the management origin of the print image division information may be detected based on the bar code printed on the paper surface. In general, barcodes have a special shape and can be recognized as objects by the processing program.

FIG. 11 shows a state in which the printing area is misaligned with respect to the paper edge surface that may occur during printing. Generally, in a printer that controls printing, printing is performed while aligning the printing paper with the document area, but the document area is tilted with respect to the end surface of the paper due to mechanical misalignment or skew of the printing paper. There are cases (B) and (C) where an offset occurs. In these situations, when the division information at the time of the print image division is based on the end face of the paper, the information in the division area is also affected by the deviation of the reference position, which causes a problem.

In order to solve this problem, by managing the coordinate system with an arbitrary portion of the bar code as a reference,
It is possible to perform position management that does not depend on the positional relationship between the printed text and the paper. According to this method, each area of the print image division information can be managed by relative coordinates using only the print information. Further, when the barcode is read, the position of the barcode with respect to the entire touch-up superimposed image passed to the document management system by barcode recognition can be easily calculated, so that the required processing time can be shortened.

Next, a modification information acquisition system as another embodiment of the present invention will be described.

Like the above-described document management system, this modification information acquisition system can also execute modification information acquisition processing independently of individual applications.

In this system, as described above, the original (original) image and the overlay image for additional editing are accurately cut out and aligned, the additional image is extracted by the processing of these two images, and the character recognition processing is performed on it. Give and get the correction information. The first step in this series of processes is the clipping of the touch-up superimposed image. In this system,
The various functions described above can be used for obtaining the region in the cutout.

That is, the acquisition of the divided area of the original image is cut out from the original image based on the print image division information, or the print image division information itself has an image of the divided area of the original image as described above. In some cases it can be called directly from the system. That is, unlike the prior art, the area information can be directly extracted from the print image division information without depending on the objects such as the position mark and the entry field.

In the case of the conventional technique, the region information cannot be acquired due to the loss of additional recording of the position mark, and the region information is inaccurate due to the recognition error of the position mark. However, in the case of the present invention, these problems occur. Does not occur.

In the second step, the above two images are aligned. The alignment can be performed based on a predetermined mark added by the document management system at the time of printing, a configuration in which any characteristic image element existing in the image is detected and used, and the like. As the characteristic image element used here, an image itself of a character included in a print image such as a punctuation mark can be used. Known techniques can be used for the alignment algorithm using these,
It will not be detailed here.

Here, the predetermined mark to be added will be described. As described above, it is necessary to detect the position of the position mark used as the cut-out mark of the area by image processing, and therefore it is necessary to stand out imagewise. However, if the position mark is not used for cutting out the image area, the position mark does not need to be so conspicuous, and the location can be detected only from the feature amount. Therefore, according to the present invention, it is possible to easily perform the above-mentioned alignment by adding an arbitrary mark to the intersection of the regions generated in the document image division and printing (see FIG. 9).

Any mark can be designated as the mark used here. Furthermore, if the mark is different in color from the text image, the position mark can be detected more easily. As a specific example,
At the time of printing the document, the color information used in the document managed by the processing program is detected, and the color information not used here is used as the mark color. The attributes such as the mark shape and color may be fixed in the document management system, but the mark attribute information is dynamically generated at the time of a print command, and this information is associated with the document image division information and the document information to document. By managing with a management system, a highly flexible system can be constructed.

The example of FIG. 9 shows an example in which, for example, light blue rectangles are arranged as alternate marks at the intersections of the divided areas.
Marking at intersections as in this example does not necessarily have to be applied to all divided areas, and the marks can be changed according to the characteristics of the divided areas. As described above, since the mark added for alignment need not be particularly noticeable, a small mark can be used. It is preferable that the size is at least three times as large as the minimum printable dot so that it can be distinguished from toner or ink dust that causes background stains during printing.

Next, the third step is AND, OR, XOR, N which is generally used as an image operation for obtaining a retouched image.
Operations such as AND, addition, multiplication, division, and difference can be used. When the difference is used as the final calculation after the alignment, the width of the image obtained by the difference can be used to detect the thickness of the added line. In this case, by setting a certain threshold value in the system, it becomes possible to identify the thickness of the added line as stepwise information.

As described above, according to the present invention,
While the document management system can manage the correlation between the areas corresponding to the original image and the added and superimposed original image, the original image is printed without the large and conspicuous position marks for clipping, so when proofreading It is possible to improve the appearance of the manuscript such as the above and to allow the user to use the blank area used for proofreading the document without limitation. Further, since the document management system and the editing information processing system exist independently of other individual applications such as "document proofreading", it is possible to provide a management form that does not depend on the application.

Further, it is possible to construct a hardware system having both the document management system and the functions of a general processing program. Here, the controller having a CPU and a memory function accesses the document management system / correction information acquisition system by a communication means such as a wired / wireless network or BlueTooth to perform desired processing. Desirably, a display device is provided for displaying to the user the display contents generated in a series of processes during printing. In addition, a scanner is provided for digitizing the overwritten original that is required for a series of operations. As an apparatus that enables a series of operations of various devices including these, a computer system including a copier / FAX, a printer, and a scanner can be applied.

[0095]

As described above, according to the present invention, a large and conspicuous position mark for image cutout can be managed in the document management system while the correlation between the corresponding regions of the original image and the print-added superimposed image can be managed. Since the original image can be printed without marking, the appearance of the manuscript can be improved even at the time of proofreading, and the blank area that can be used at the time of proofreading can be used without restricting the user. . Further, since the timing of printing is the timing of document management, it is possible to provide a highly reliable system in which matching and matching can be achieved between digital information and information on paper. Furthermore, since such a document management system and an editing information processing system can exist and function independently of other individual applications, it is possible to provide an application-independent management form (claim 1).

Further, since the document management system manages the coordinate information for dividing the image at the time of printing, for example, when the present invention is applied to the client-server system in which the processing execution part and the service providing part are separated. ,
It is possible to reduce the burden of mutual communication (claim 2).

Further, by managing the divided image of the image image at the time of printing by the document management system, it is not necessary to dynamically generate the divided image of the original (original) image each time in the process of extracting the additional information. The processing can be performed at high speed (claim 3).

Furthermore, since the content of the document image division information converted into a predetermined code such as a bar code is unknown to a third party, it is possible to provide a highly secure system (claim 4). ).

Further, by storing the document image division information itself as a predetermined code such as a bar code in the print document, the operation of the present invention can be locally processed in the client-server system. In other words, since the processing can be performed regardless of the communication speed or the congestion status of the server, it is possible to perform the high-performance addition information acquisition processing. Further, since the management information generated for each printing is stored on the paper surface, the data management load on the database is small, and meaningless data does not remain on the database even if the paper surface is discarded (claim 5). .

Further, by printing a barcode in which data for cooperation with the document management system is registered on a printing paper and using coordinates from an arbitrary portion of the printed barcode as division information. Also, even when the relative position shift between the divided document area and the edge of the sheet occurs due to the paper conveyance or the like at the time of printing, the reliable image cutout information can be obtained. Further, since the barcode can be easily recognized as an object graphically, it is more reliable than decoding the ID managed by a document management system or the like using OCR (claim 6).

Furthermore, by marking the intersections of the divided areas and printing, it becomes easy to align the image when extracting the additional information. Since this mark is not used for cutting out the image area, it can be a small mark and can be marked without impairing the appearance of the printed document (claim 7).

Furthermore, by printing by attaching a mark of a color not used in the document to the intersection of the divided areas, it is possible to more easily align the image with the mark.

Further, by causing the document management system to manage the mark attribute together with the document image division information and the document information, the mark attribute can be dynamically changed appropriately. As a result, even if an attribute in the document, for example, the color of the font is changed, the color of the mark can be appropriately changed according to the change, and stable image alignment can be stably performed with the mark. 8).

Further, by appropriately setting the minimum value of the mark size, the mark can be surely detected without being affected by the background stain that may occur during printing, and the image can be surely aligned (claim). Item 9).

Furthermore, by using the difference processing at the final stage of the image processing, the thickness of the line added can be recognized. Thereby, the thickness of the line can be used as a special image handled by the processing program, and various commands added by the processing can be processed by the processing program (claim 11).

Further, by performing the image registration by the mark, it is not necessary to search the input image for the element used for the image matching, so that the processing speed is increased (claim 12).

Furthermore, in a system that employs image registration using marks in this way, even if the marks cannot be recognized due to the superimposed retouched images, the images will not be recognized from the predetermined or arbitrary images included in the printed document. Since appropriate matching elements can be extracted and used, it is possible to stably extract the additional information (claim 13).

Further, the line width can be extracted by the stepwise method by determining the line width from the superimposed image obtained by the image difference with a predetermined threshold value.
4).

[Brief description of drawings]

FIG. 1 is a diagram for explaining a method (part 1) of extracting a correction image.

FIG. 2 is a diagram for explaining a method (No. 2) of extracting a correction image.

FIG. 3 is a document management system according to the embodiment of this invention (a).
FIG. 3 is a block diagram showing a schematic configuration of a correction information acquisition system (b).

FIG. 4 is a diagram for explaining the configuration and operation of the document management system and the correction information acquisition system according to the embodiment of the present invention in more detail.

FIG. 5 is a diagram for explaining the operation (particularly, the importance of alignment) of the correction information acquisition system according to the embodiment of the present invention.

FIG. 6 is a diagram (No. 1) for explaining the operation in the document management system and the correction information acquisition system according to the embodiment of the present invention in more detail.

FIG. 7 is a diagram (No. 2) for explaining the operation in the document management system and the correction information acquisition system according to the embodiment of the present invention in more detail.

FIG. 8 is a diagram (No. 3) for explaining the operation in the document management system and the correction information acquisition system according to the embodiment of the present invention in more detail.

FIG. 9 is a diagram (No. 4) for explaining the operation in the document management system and the correction information acquisition system according to the embodiment of the present invention in more detail.

FIG. 10 is a diagram (No. 5) for explaining the operation in the document management system and the additional information acquisition system in the embodiment of the present invention in more detail.

FIG. 11 is a diagram for illustrating and explaining a print state of a print document.

[Explanation of symbols]

1 Document management system 2 Editing information acquisition system 13 Original image area division unit 14 Division information storage 22 Bar code identification section 23 Bar Code Decoding Unit 24 division information reading unit 25 Input image division unit 26 Image comparison / collation unit

Continued front page    F term (reference) 2C087 AA03 AA09 AA13 AA14 BA03                       BB10 BD07 CA12 CB07                 5B029 AA01 BB03 EE08                 5C076 AA02 AA14 AA36 BA03 BA06                       CA07 CA10                 5L096 BA17 DA01 EA37 GA08 HA00

Claims (15)

[Claims] 1. An image processing system comprising: means for dividing a print image into a plurality of areas when a document is printed; and means for storing the document image division information at that time in association with the divided document images. 2. The image processing system according to claim 1, wherein the document image division information includes a size of the divided area and a coordinate position of the divided area from a specific reference position. 3. The image processing system according to claim 2, wherein the document image division information further comprises divided document images themselves. 4. The image processing system according to claim 1, further comprising means for additionally printing a predetermined code storing the document image division information and the link information to the document information on the print document. 5. The image processing system according to claim 4, wherein the predetermined code further stores document image division information itself. 6. The coordinate information included in the document image division information is based on a coordinate system based on at least one location of a predetermined code to be printed.
The image processing system according to any one of 1. 7. An image processing method comprising: a step of selecting a color that is not used during document printing; and a step of additionally printing a mark using the selected color on a print document in association with document image division. 8. The method according to claim 1, further comprising means for managing color information used for a mark additionally printed in association with document image division during document printing, together with corresponding document image division information and document information. The image processing system according to any one of claims. 9. The image processing method according to claim 7, wherein the mark has a size that is three times or more as large as a minimum dot that can be printed by a predetermined printing device. 10. A unit for inputting a document image, a unit for reading a predetermined code from the input document image, a unit for decoding the read code, and a unit for acquiring document image division information of the input document image from the decoding result. And a means for dividing the input document image based on the acquired document image division information, and input from the divided document image obtained by dividing the input document image and the divided document image of the original document image corresponding to the input document image. An image processing system comprising means for acquiring an image superimposed on a document image. 11. The method according to claim 10, wherein the means for acquiring the image superimposed on the input document image executes a process of obtaining an image difference between the divided document image of the input document image and the corresponding divided document image of the original document image. Image processing system. 12. The mark added to the print document in association with the document image division of the divided document image information is used for alignment between the input document image and the original document image. Image processing system. 13. The image processing system according to claim 10, wherein an arbitrary image included in the print document image is used for alignment between the input document image and the original document image. 14. The line width of the superimposed image is determined using a predetermined threshold from the image superimposed on the input document image, which is acquired by the means for acquiring the image superimposed on the input document image. Claim 10 consisting of means
The image processing system described in 1. 15. A device for inputting a document image, a device for printing out an input document image, a device for communicating with the outside, and a device for controlling the entire apparatus, and at least one of claims 1 to 14. An image forming apparatus capable of executing the function described in (1).