JP2010102734A - Image processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system allowing prompt finding of original image data corresponding to the original of image data added with a postscript. <P>SOLUTION: A control part 11 analyzes contents of an image represented by the input image data, and extracts a characteristic of the image. Next, the control part 11 searches for the original image data having a characteristic nearly according with the extracted characteristic from the original image data stored in a nonvolatile storage part 12. Next, the control part 11 compares the original image data obtained by the search and the input image data, and extracts a difference between both sides as the postscript image data. The control part 11 makes the extracted postscript image data be stored in the nonvolatile storage part 12 associatively to the original image data obtained by the search. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for searching for an original document to which additional writing has been made.

  Various kinds of comments, proofreading or correction instructions, etc. may be written on the document, or an approver's signature or seal may be added to indicate that the content of the document has been approved. A document in which such various types of additional writing (hereinafter referred to as an additional writing document) are read by a scanner, and the added content (hereinafter referred to as an additional writing portion) and the content of the original document before the additional writing are stored in association with each other. If this is done, it can be used in various usage scenes. For example, it is possible to perform a search operation for only the added part, such as searching for comments of a certain approver all at once. Further, it is also possible to use such as switching between a screen displaying only the original document and a screen displaying the added portion superimposed on the original document according to a user operation.

  Therefore, conventionally, a technique for extracting only such an additional portion has been proposed. For example, Patent Document 1 describes a technique in which only an added part is extracted by comparing and collating a document original and an additionally-written document added to the original. In Patent Document 2, if a certain document is a recordable document B of the document A that has already been registered in the database, the recordable document B and the document A are compared to extract image data of a recordable part and extract it. A mechanism for associating the processed image data with the document A has been proposed. In Patent Document 3, as described in Patent Document 2, after comparing the postscript document B and the document A and extracting the image data of the postscript portion, information on the image data of the postscript portion (for example, It is described that the user inputs a keyword and update date and that the input information is stored together with the image data of the additional portion. Thereby, in the search operation after saving, for example, a desired document can be extracted by performing a keyword search.

  As described in Patent Documents 1 to 3, when the method of comparing the document before appending and the document after appending and specifying the difference as the appending part is used, only the appending part is reliably extracted. be able to. However, assuming an actual usage scene, it is not so easy to find out how to find the document before the additional writing for the document after the additional writing. For example, in Patent Documents 1-3, the user designates a document before appending by a file name. However, in this method, the file name must be remembered for a huge number of all documents. It is very unusable.

JP-A-61-67177 JP-A-8-317155 JP 2000-148790 A

  The present invention has been made in view of such a background, and an object of the present invention is to quickly find original image data corresponding to the original of a specific postscript document from a plurality of original image data. To provide a mechanism.

  The invention according to claim 1 is a storage means for storing a plurality of original image data, an input means for inputting image data, a plurality of original image data stored in the storage means, and an input to the input means. Extraction means for extracting difference images from the input image data obtained, and pixel distribution of each of the difference images based on each of the difference images extracted by the extraction means, and stored by the storage means From among a plurality of original image data, original image data in which the obtained pixel distribution is biased to a specific area with respect to other original image data is specified as original image data corresponding to the original of the input image data A difference means between the original image data specified by the specifying means and the specifying means and the input image data is added to the additional image data for the original image data. An image processing apparatus and a storage control means for storing in said memory means as a data.

  The invention according to claim 2 is a storage means for storing a plurality of original image data, an input means for inputting image data, a plurality of original image data stored in the storage means, and an input to the input means. Extracting means for extracting difference images from the input image data, and performing layout analysis on each of the difference images extracted by the extracting means, and obtaining the size or position of the layout unit region obtained thereby Based on the plurality of original image data stored by the storage means, specifying means for specifying original image data corresponding to the original of the input image data, original image data specified by the specifying means, The difference image from the input image data is stored in the storage means as additional image data for the original image data. An image processing apparatus and a 憶制 control means.

  According to a third aspect of the present invention, in the configuration according to the first or second aspect, the storage control means, when there are a plurality of the additional image data, the additional images that are close to each other in the image represented by the input image data The image processing apparatus is characterized in that additional image data representing the above is grouped and the additional image data is stored in the storage means in units of groups.

  According to a fourth aspect of the present invention, in the configuration according to the third aspect, the storage control means stores the additional image data as electronic sticky note data including the position of the additional image in the image represented by the input image data. The image processing apparatus is characterized in that it is stored in the image processing apparatus.

  According to a fifth aspect of the present invention, there is provided an extraction function for comparing a plurality of original image data stored in a storage means and input input image data in a computer and extracting a difference image between the two, and the extraction A distribution of pixels of each of the difference images is obtained based on each of the difference images extracted by the function, and the obtained pixel distribution is selected from the plurality of original image data stored in the storage unit. A specifying function that specifies original image data that is biased to a specific area from the image data as original image data corresponding to the original of the input image data; the original image data specified by the specifying function; and the input image data And a storage control function for storing the difference image in the storage unit as additional image data with respect to the original image data. Is a program.

According to a sixth aspect of the present invention, the computer compares each original image data stored in the storage means with the input image data inputted, and extracts a difference image between the two, and the extraction A layout analysis is performed on each of the difference images extracted by the function, and based on the size or position of the layout unit area obtained thereby, the plurality of original image data stored in the storage unit is selected. A specifying function for specifying original image data corresponding to the original of the input image data, and a difference image between the original image data specified by the specifying function and the input image data as additional image data for the original image data A program for realizing a storage control function stored in the storage unit.

1 is a block diagram illustrating a configuration of an image processing apparatus according to a first embodiment of the present invention. 4 is a flowchart illustrating a procedure executed by the image processing apparatus according to the embodiment. It is the figure which illustrated the content of the postscript document in the embodiment. It is the figure which illustrated the contents of the original document in the same embodiment. It is a figure explaining the principle which the image processing apparatus which concerns on the embodiment specifies original image data. It is a figure explaining the principle which the image processing apparatus which concerns on the embodiment specifies original image data. It is a figure explaining the principle which the image processing apparatus which concerns on the embodiment specifies original image data. It is a figure explaining the principle which the image processing apparatus which concerns on the embodiment specifies original image data. It is a flowchart which shows the procedure which the image processing apparatus which concerns on 2nd Embodiment of this invention performs. It is a figure explaining the process of the image processing which the image processing apparatus which concerns on the same embodiment performs. It is a figure explaining the principle which the image processing apparatus which concerns on the embodiment specifies original image data. It is the figure which illustrated a mode that the image processing apparatus which concerns on the embodiment extracts the difference of input image data and original image data.

First, main terms used in the embodiment of the present invention are defined.
The “original document” means a document that has not been additionally written by an adder after an image is formed by an image forming apparatus such as a printer. “Original image data” is image data representing the contents of the original document. In addition, the term “original image data” here means that, in addition to image data in which the content of a document is expressed as an image, for example, it is expressed using unique code information by a document creation application program. Document data is also included. This is because it is easy to appropriately convert such document data into image data, and particularly in the present invention, whether it is image data or document data, the contents of the original document are analyzed and its characteristics are analyzed. This is because these data are common in that they can be used to extract.
On the other hand, the “additional document” means a document in a state where the original document is additionally written. Further, the “additional image” means an image of the additional portion. For example, when a certain character string is underlined with a pen, the underline itself is a postscript image. “Additional image data” is image data representing the contents of the additional image.

(1) First Embodiment FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus 1 according to a first embodiment. The image processing apparatus 1 is a personal computer, for example, and includes a control unit 11, a nonvolatile storage unit 12, an input unit 13, a display unit 14, and an operation unit 15. An image reading device 2 such as a scanner is connected to the input unit 13 via a communication cable.

The control unit 11 includes a CPU (Central Processing Unit) 111, a ROM (Read Only Memory) 112, and a RAM (Random Access Memory) 113. The control unit 11 converts image data input from the image reading device 2 through the input unit 13. On the other hand, various image processing is executed. The non-volatile storage unit 12 is, for example, a hard disk, and stores an image processing program PRG describing a procedure of image processing executed by the control unit 11. In addition, the nonvolatile storage unit 12 stores a plurality of original image data of original documents before additional writing is performed. The operation unit 15 includes, for example, a keyboard and a mouse, receives an input operation by a user, and supplies a signal corresponding to the operation to the control unit 11. The display unit 14 is a liquid crystal display device, for example, and displays various images under the control of the control unit 11.

FIG. 2 is a flowchart showing a procedure described in the image processing program PRG.
The appender adds to the original document. The additionally written document that has been additionally written is read by the image reading device 2. The image reading device 2 inputs image data representing the read image to the input unit 13 of the image processing device 1.

  When image data is input to the input unit 13 (step S1), the control unit 11 analyzes the content of the image represented by the input image data (hereinafter referred to as input image data) according to a predetermined algorithm, The feature of the image is extracted (step S2). A specific algorithm for extracting image features will be described in detail later. Next, the control unit 11 searches the original image data stored in the non-volatile storage unit 12 for original image data having features that substantially match the features extracted in step S2 (step S3). Here, it is not necessary for the features of the two to completely coincide with each other, and any original image data having features closest to the features extracted in step S2 among the original image data stored in the nonvolatile storage unit 12 may be used. . Here, when the original image data stored in the nonvolatile storage unit 12 is not so-called “image data” but document data as described at the beginning, the document data is stored as necessary. What is necessary is just to rasterize and develop into an image. Of course, this may be converted when necessary, and when a character string is used as a feature as will be described later, it is not necessary to develop document data into an image.

  Next, the control unit 11 compares the original image data obtained by the search in step S3 with the input image data obtained in step S1 (step S4), and extracts the difference between the two as additional image data (step S4). Step S5). The control unit 11 stores the extracted additional image data in the nonvolatile storage unit 12 in association with the original image data obtained by the search in step S3 (step S6).

In step S6, when there are a plurality of additional image data, the control unit 11 groups the additional image data representing additional images adjacent to each other in the image represented by the input image data, and the additional image data is grouped. You may make it memorize | store in the non-volatile memory | storage part 12. FIG. Desirably, the control part 11 memorize | stores this additional image data in the non-volatile memory | storage part 12 as electronic tag data. The electronic sticky note data here is data including position information indicating where the additional image is arranged in the image represented by the input image data.
The above is the basic operation of the control unit 11.

Next, a method for extracting image features in step S2 described above will be described. The specific methods are enumerated as follows.
(1-1) Method for Extracting Visible or Invisible Code Image as Feature (1-2) Method for Extracting Feature, Position, Shape, and Size of Each Layout Unit Area Obtained by Layout Analysis Processing (1-3 ) Method for Extracting Character String Obtained by Character Recognition Processing as a Feature (1-4) Obtained by performing character recognition processing on a specific layout unit region among layout unit regions obtained by layout analysis processing (1-5) Method for Extracting Character Strings as Features (1-5) Method for Dividing Image into Multiple Regions and Extracting Frequency Distribution Obtained by Frequency Analysis for Each Region as Features (1-6) Method for extracting projection distribution of image as feature

Hereinafter, these methods will be described in order.
(1-1) Method for Extracting Visible or Invisible Code Image as Feature FIG. 3 is a diagram showing an example of an additional-write document. In the postscript document D, an instruction is added by the postscriptor to the original document Do as shown in FIG. 4 to correct the part of “Tanaka” to “Nakamura” in the character string “General Affairs Department Tanaka”. ing. These original document Do and additional document D
A barcode C (code image), which is identification information uniquely assigned to each original document, is formed at a predetermined location (upper right corner in FIGS. 3 and 4). The barcode C may be printed in advance on a recording paper on which an original document image is to be formed, or at the same time when an image forming apparatus such as a printer forms an original document image on the recording paper. It may be formed.

  Accordingly, in FIG. 2, the control unit 11 extracts the feature of the additional writing document by analyzing the barcode formed at a predetermined location of the image data of the additional writing document and extracting the identification information (step S2). Next, the control unit 11 may search the non-volatile storage unit 12 for original image data on which a barcode representing identification information that matches the extracted identification information is formed (step S3).

  The code image is not limited to a bar code, but may be any code image that is represented by a combination of figures or characters. For example, instead of a bar code, a code image such as a two-dimensional code such as a QR code may be used, or a character code may be used. . Further, the code image does not need to be a visible image, and the code image may be invisible when the image reading apparatus 2 includes a sensor that reads an invisible image formed with invisible toner.

(1-2) Method for Extracting Features of Position, Shape, and Size of Each Layout Unit Area Obtained by Layout Analysis Processing FIG. 5 is a diagram for explaining a mechanism for searching original image data by layout analysis. The layout structure R1 shown in FIG. 5 shows an example in which layout analysis is performed on an additional writing document (input image data) and divided into layout unit areas a1 to a7 such as a character area or a photo area. In step S2 of FIG. 2, the control unit 11 of the image processing apparatus extracts the positions, shapes, and sizes of the respective layout unit areas a1 to a7 obtained by such layout analysis as features. On the other hand, the control unit 11 performs the layout analysis as described above on each of the original image data before starting the processing, and stores the position, shape, and size of each layout unit area a1 to a7 in the nonvolatile storage unit 12. is doing.

  In step S3, the control unit 11 searches for original image data having a layout unit area that matches the position, shape, and size of the extracted layout unit areas a1 to a7. For example, the layout structure R1 of FIG. 3 does not match the layout structure R2 or the layout structure R3, but the layout structure R1 and the layout structure R1 'have substantially the same structure. Here, the reason why the layout structure R1 and the layout structure R1 'do not completely match is that the postscript image itself is handled as one layout unit area in the layout structure R1 of the postscript document. In the example shown in FIG. 5, the layout unit area a7 of the layout structure R1 is a layout unit area corresponding to the additional image, but naturally the layout unit area a7 exists in the layout structure R1 ′ of the original document before the additional recording. Not done. Therefore, the control unit 11 may search for original image data having a feature closest to the feature of the layout structure extracted in step S2 from the original image data stored in the nonvolatile storage unit 12.

(1-3) Method for Extracting Character String Obtained by Character Recognition Processing as a Feature FIG. 6 is a diagram for explaining a mechanism for searching original image data based on the result obtained by the character recognition processing. In step S2 of FIG. 2, the control unit 11 of the image processing apparatus performs character recognition of the postscript document D1 (input image data) having the contents as shown in FIG. 6, and extracts the result as a feature of the input image data. On the other hand, it is assumed that the control unit 11 performs character recognition in advance on each of the original image data and stores the result in the nonvolatile storage unit 12 before the start of the processing in FIG.

  In step S3, the control unit 11 searches for original image data of a character recognition result that matches the character recognition result obtained in step S2. For example, in the character recognition result of the postscript document D1 and the character recognition result of the original document D1 ′ shown in FIG. 6, the character string “corrected to Nakamura” of the additional document D1 is included in the original document D1 ′ (area T). Except that they are not, the two are almost identical. That is, the only difference between the two is that the character recognition result of the postscript document D1 includes the character string of the postscript portion. On the other hand, in the character recognition result of the postscript document D1 and the character recognition result of the document original D2, the character string “corrected to Nakamura” of the postscript document D1 is included in the character recognition result (region T) of the document original D2. In addition to the fact that there is no character string, the contents of the character strings t1 to t5 are different. As described above, the control unit 11 compares the character recognition result of the postscript document (input image data) with the character recognition result of the original image data, and among the original image data stored in the nonvolatile storage unit 12, The original image data closest to the character recognition result extracted in step S2 may be searched.

(1-4) A method of extracting a character string obtained by performing character recognition on a specific layout unit area from among layout unit areas obtained by layout analysis processing as a feature of (1-3) described above When character recognition processing is performed as in the method, there is no problem if all of the processing objects are character areas. However, for example, when a photo area or a graphic area is included, characters are not included in those areas. Since it cannot be recognized, even if an attempt is made to search for original image data with a matching character recognition result, a more accurate search may be difficult. In particular, the postscript portion is often handwritten, but the character recognition accuracy of handwritten characters is significantly lower than that of printed characters.

  Therefore, in step S2 in FIG. 2, the control unit 11 performs a layout analysis on the postscript document as shown in FIG. 7, and classifies each layout unit area into a character area and other areas. Specifically, the control unit 11 may regard a region where the character recognition process is attempted and the process can be accurately executed as a character area. And the control part 11 extracts the character recognition result with respect to the area | region considered to be a character area as a characteristic of input image data. In step S3, the control unit 11 may search for original image data of a character recognition result that matches the result of character recognition obtained in step S2.

(1-5) Method of dividing an image into a plurality of regions and extracting the frequency distribution obtained by performing frequency analysis on each region as a feature. The control unit 11 of the image processing apparatus 1 performs steps shown in FIG. In S2, the image represented by the input image data is divided into a plurality of regions, and frequency analysis is performed on each region. And the control part 11 extracts the frequency distribution of the whole image as a characteristic using the frequency acquired from each area | region. Further, the control unit 11 calculates the frequency distribution as described above for each of the original image data, and stores this in the nonvolatile storage unit 12. In step S3, the control unit 11 searches for original image data that substantially matches the frequency distribution obtained in step S2.

(1-6) Method for Extracting Image Projection Distribution as Feature The control unit 11 of the image processing apparatus 1 projects an image of an additional record document (input image data) in a predetermined direction in step S2 of FIG. The projection distribution is extracted as a feature of the postscript document. For example, when the postscript document D3 shown in FIG. 8 is projected in the x direction (right side of the drawing) and the number of pixels is counted in that direction, the projection distribution is as shown in the figure. On the other hand, the control unit 11 calculates a projection distribution in advance for each of the original image data, and stores the result in the nonvolatile storage unit 12. Then, in step S3 of FIG. 2, the control unit 11 searches for the original image data of the projection distribution that substantially matches the result of the projection distribution obtained in step S2. However, since the projection distribution differs for the portion of the postscript image, the control unit 11 is the projection distribution closest to the projection distribution extracted in step S2 among the original image data stored in the nonvolatile storage unit 12. What is necessary is just to search for original image data having.

  According to the first embodiment, the control unit 11 analyzes the content of the image represented by the input image data and extracts the feature of the image. Next, the control unit 11 searches the original image data stored in the non-volatile storage unit 12 for original image data having a feature that substantially matches the extracted feature. Next, the control unit 11 compares the original image data obtained by the search with the input image data, and extracts the difference between the two as additional image data. Then, the control unit 11 stores the extracted additional image data in the nonvolatile storage unit 12 in association with the original image data obtained by the search. In this way, it is not necessary for the operator of the image processing apparatus 1 to bother to specify the original image, and it is easy to use.

  In addition, the methods (1-1) to (1-6) described above may be combined with several methods without being used individually. In this way, the original image data can be found more accurately. In addition, the image processing apparatus 1 calculates the characteristics of the original image data in advance and stores them in the nonvolatile storage unit. However, the present invention is not limited to this, and every time an attempt is made to search for original image data, The characteristics of the input image data and the original image data may be calculated.

  Further, the image processing apparatus 1 may be realized by a personal computer, or may be realized by a multifunction machine having the function of the image processing apparatus 1. Further, part of the functions of the image processing apparatus 1 may be realized by the image reading apparatus 2. In addition, the control unit 11 of the image processing apparatus 1 does not store the extracted additional image data to be stored in the nonvolatile storage unit 12 built in the image processing apparatus 1, but a floppy (registered trademark) disk or other recording medium. You may make it memorize. The image processing program PRG can be provided in a state of being recorded on various recording media such as a magnetic disk, a floppy (registered trademark) disk, a CD (Compact Disk), a DVD (Digital Versatile Disk), and a RAM. .

(2) Second Embodiment In the first embodiment described above, the process is executed in such a manner that the original image data (additional document) is searched by comparing the features of the original image data and the input image data. It was. In the second embodiment described below, rather than calculating the characteristics of the original image data and the input image data, first, the original image data and the input image data are compared at the pixel level, and a difference between them is extracted. Based on the difference, the process is performed by searching for the original input image data (additional document). Since the configuration of the image processing apparatus 1 according to the second embodiment is the same except for a part of the procedure described in the image processing program PRG, detailed description thereof is omitted.

FIG. 9 is a flowchart showing a procedure described in the image processing program PRG according to the second embodiment.
In FIG. 9, when the postscripted document is read by the image reading device 2, the image reading device 2 inputs image data representing the read image to the input unit 13 of the image processing device 1 (step S <b> 11). The control unit 11 compares the input image data with each original image data stored in the non-volatile storage unit 12, and extracts a difference image that is the difference (step S12). Then, the control unit 11 specifies original image data corresponding to the original of the input image data based on the extracted difference image (step S13). The control unit 11 stores the difference image extracted in step S12 as the additional image of the original image data specified in step S13 in association with each other in the nonvolatile storage unit 12 (step S14).

Here, a specific algorithm for extracting the difference and specifying the original based on the difference will be described.
(2-1) Method of specifying original image data based on ratio of number of pixels of difference image to number of pixels of foreground image represented by input image data Foreground image represented by additional image data (image formed on background image) ) Is, for example, an image as shown in the postscript document D1 in FIG. In this case, the difference between the postscript document D1 and the original document D1 ′ shown in FIG. 6 is the difference image d1 ′, and the difference between the postscript document D1 and the original document D2 shown in FIG. 6 is the difference image d2. . Although simply illustrated in FIG. 10, in more detail, the difference image is an absolute value of a difference between images to be compared. Therefore, for example, when the difference between the image “May 30” of the original document D2 and the image “March 29” of the postscript document D1 is extracted, as shown in FIG. The absolute value of the difference between “3” and “3” has a shape like the shaded portion m. Similarly, the absolute value of the difference between “3” and “2” and the difference between “0” and “9” also has a shape like the shaded portion m. It should be noted that the character images “month” and “day” are not extracted as difference images because the original document D2 and the postscript document D1 completely match the shape and position of the image. As shown in FIG. 11, the difference images d2-1 to d2-4 shown in FIG. 10 are not actually images that can be identified as characters, but are not readable as shown in FIG. Will be expressed as

  The control unit 11 extracts such difference images for all original image data stored in the nonvolatile storage unit 12. Then, the control unit 11 calculates the ratio of the number of pixels of each difference image to the number of pixels of the foreground image represented by the input image data, and specifies the original image data corresponding to the difference image having the smallest calculated ratio. In the example of FIG. 10, the difference image between the postscript document D1 and the original document D1 ′ is only an image “changed to Nakamura”, but the differential image between the postscript document D1 and the original document D2 is the difference image d2-1˜ It becomes like d2-4. Therefore, the original image data having a smaller ratio of the number of pixels of the difference image to the number of pixels of the foreground image represented by the input image data, that is, the document original D1 'in FIG. 6 is specified.

(2-2) Method of Specifying Original Image Data Based on Distribution of Pixels of Difference Image with respect to Number of Pixels of Foreground Image Represented by Input Image Data The method (2-1) described above determines the ratio of the number of pixels of the difference image. Although used, the following method uses the pixel distribution of the difference image.
As shown in FIG. 10, the difference image d1 ′ between the postscript document D1 and the original document D1 ′ and the differential image d2 between the postscript document D1 and the original document D2 have different pixel distributions. That is, since the difference image d1 ′ is equal to the postscript image, the pixel tends to be biased toward a specific area. On the other hand, the difference image d2 does not have its pixels biased to a specific area, and tends to be evenly distributed over the entire image. In particular, since the difference image d2 is a difference between the original document very similar to the postscript document D1 shown in FIG. 6, the pixel distribution of the difference image is evenly distributed over the entire image. However, if the postscript document D1 and the original document to be compared are completely different documents, the pixel distribution of the difference image should be evenly distributed over the entire image. The control unit 11 calculates the pixel distribution of such a difference image for all the original image data stored in the non-volatile storage unit 12, and the difference in which the calculated pixel distribution is most likely to be biased toward a specific region. The original image data corresponding to the image is specified.

(2-3) Method of Specifying Original Image Data Based on Layout Structure of Difference Image between Input Image Data and Original Image Data The control unit 11 performs a layout analysis on a difference image between additional image data and original image data, When classified into layout unit areas, for example, a layout unit area where the difference image d1 ′ exists and a layout unit area where the difference images d2-1 to d2-4 exist as shown in FIG. The control unit 11 specifies original image data corresponding to the original of the input image data based on the size and position of the layout unit area obtained by performing the layout analysis. Specifically, the added portion is relatively small or tends to be biased toward a specific area such as a page edge of the document. Therefore, the control unit 11 is biased toward a specific area such as a small size of each layout area of the difference image. What is necessary is just to specify the original image data corresponding to the difference image, such as a strong tendency.
Note that the second embodiment described above can be modified in the same manner as the first embodiment described above.

DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 2 ... Image reading apparatus, 11 ... Control part, 111 ... CPU, 112 ... ROM, 113 ... RAM, 12 ... Nonvolatile memory | storage part, 13 ... Input part, 14 ... Display part, 15 ... Operation unit, PRG... Image processing program.

Claims (6)

Storage means for storing a plurality of original image data;
An input means for inputting image data;
Extraction means for extracting a difference image between a plurality of original image data stored in the storage means and input image data input to the input means;
A pixel distribution of each of the difference images is obtained based on each of the difference images extracted by the extraction unit, and the obtained pixel distribution is different from the plurality of original image data stored in the storage unit. Specifying means for specifying original image data that is biased to a specific area from the original image data as original image data corresponding to the original of the input image data;
An image processing apparatus comprising: a storage control unit that stores a difference image between the original image data specified by the specifying unit and the input image data in the storage unit as additional image data for the original image data. Storage means for storing a plurality of original image data;
An input means for inputting image data;
Extraction means for extracting a difference image between a plurality of original image data stored in the storage means and input image data input to the input means;
A layout analysis is performed on each of the difference images extracted by the extraction unit, and a plurality of original image data stored in the storage unit is stored on the basis of the size or position of the layout unit area obtained thereby. A specifying means for specifying original image data corresponding to the original of the input image data,
An image processing apparatus comprising: a storage control unit that stores a difference image between the original image data specified by the specifying unit and the input image data in the storage unit as additional image data for the original image data.   When there are a plurality of the additional image data, the storage control unit groups the additional image data representing additional images in the vicinity of each other in the image represented by the input image data, and sets the additional image data in groups. The image processing apparatus according to claim 1, wherein the image processing apparatus is stored in a storage unit.   The image processing apparatus according to claim 3, wherein the storage control unit stores the additional image data in the storage unit as electronic sticky note data including a position of the additional image in the image represented by the input image data. On the computer,
An extraction function for comparing a plurality of original image data stored by the storage means and the input image data that has been input, and extracting a difference image between the two,
A pixel distribution of each of the difference images is obtained based on each of the difference images extracted by the extraction function, and the obtained pixel distribution is different from the plurality of original image data stored in the storage unit. A specifying function for specifying original image data that is biased to a specific region from the original image data as original image data corresponding to the original of the input image data;
A storage control function for storing a difference image between original image data specified by the specifying function and the input image data in the storage unit as additional image data for the original image data. On the computer,
An extraction function for comparing each original image data stored in the storage means with the input image data that has been input, and extracting a difference image between the two,
A layout analysis is performed on each of the difference images extracted by the extraction function, and a plurality of original image data stored in the storage unit is stored on the basis of the size or position of the layout unit area obtained thereby. A specifying function for specifying as original image data corresponding to the original of the input image data,
A storage control function for storing a difference image between original image data specified by the specifying function and the input image data in the storage unit as additional image data for the original image data.

Images