WO2018229903A1 - Image processing device, image processing system, image processing method and image processing program - Google Patents

Abstract

Provided is an image processing device that allows the operator work efficiency to be improved. In the image processing device (10): a storage unit (101) stores a plurality of first images read by a first reading performed for an original document group consisting of a plurality of original documents and partially including an original document bundle that is a bundle of a plurality of original documents; an insertion position determination unit (102) determines an insertion position, with respect to the plurality of first images, of a plurality of second images that are the images of the original documents of the original document bundle and that are read by a second reading following the first reading; and an image insertion unit (103) inserts the plurality of second images among the plurality of first images according to the determined insertion position.

Description

Image processing apparatus, image processing system, image processing method, and image processing program

The present invention relates to an image processing apparatus, an image processing system, an image processing method, and an image processing program.

Documents that are bound and managed by a binder are widely converted to electronic data using a scanner. A document bound with a binder also includes a bundle of documents bound with staples. In view of this, an apparatus that automatically removes staples bound to a bundle of documents to be converted into electronic data (hereinafter may be referred to as a “staple removing apparatus”) has been studied. If the staples bound to the bundle of documents can be automatically removed, the burden on the operator is reduced when the document managed by binding with the binder is converted to electronic data.

JP 2004-280691 A JP 2014-199507 A

However, the staple removal may fail even when the staple removal device is used. For example, when the original bundle is double-bound with staples, it is difficult for the staple removing device to automatically remove all the staples from the original bundle. If automatic removal of the staples using the staple removing device fails, the operator manually removes the staples from the original bundle, and the original bundle after the staples are removed (that is, after being separated into individual originals). The document bundle is again read by the scanner. At this time, conventionally, since the operator manually specifies the insertion position of the original document additionally read by the scanner with respect to the original document, the operation efficiency of the operator is poor.

The disclosed technology has been made in view of the above, and aims to improve the work efficiency of the operator.

In the disclosed aspect, the image processing apparatus includes a storage unit, a determination unit, and an insertion unit. The storage unit is a group of documents formed by a plurality of documents, and includes a plurality of first documents read by a first reading on the document group including a document bundle that is a bundle of a plurality of documents. Store the image. The determination unit is a plurality of second images that are images of each document of the document bundle, and the plurality of second images read by a second reading subsequent to the first reading. The insertion position for the first image is determined. The insertion unit inserts the plurality of second images between the plurality of first images according to the determined insertion position.

According to the disclosed aspect, the operator's work efficiency can be improved.

FIG. 1 is a diagram illustrating a configuration example of an image processing system according to the first embodiment. FIG. 2 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 3 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 4 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 5 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 6 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 7 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 8 is a diagram for explaining an operation example of the image processing system according to the first embodiment. FIG. 9 is a diagram illustrating an example of a processing flow during first reading according to the first embodiment. FIG. 10 is a diagram illustrating an example of a document table at the time of first reading according to the first embodiment. FIG. 11 is a diagram illustrating an example of area division according to the first embodiment. FIG. 12 is a diagram illustrating an example of a pixel density table according to the first embodiment. FIG. 13 is a diagram illustrating an example of a processing flow during second reading according to the first embodiment. FIG. 14 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BC according to the first embodiment. FIG. 15 is a diagram illustrating an example of a distance value calculated at the time of the second reading for the document bundle BC according to the first embodiment. FIG. 16 is a diagram illustrating an example of a document table at the time of second reading of the document bundle BC according to the first embodiment. FIG. 17 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BE according to the first embodiment. FIG. 18 is a diagram illustrating an example of a distance value calculated at the time of the second reading for the document bundle BE according to the first embodiment. FIG. 19 is a diagram illustrating an example of a document table at the time of second reading of the document bundle BE according to the first embodiment. FIG. 20 is a diagram illustrating an example of a pixel density table according to the third embodiment. FIG. 21 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BC according to the third embodiment. FIG. 22 is a diagram illustrating an example of a distance value calculated at the time of second reading for the document bundle BC according to the third embodiment. FIG. 23 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BE of the third embodiment. FIG. 24 is a diagram illustrating an example of a distance value calculated at the time of the second reading for the document bundle BE according to the third embodiment.

Embodiments of an image processing apparatus, an image processing system, an image processing method, and an image processing program disclosed in the present application will be described below with reference to the drawings. Note that the image processing apparatus, the image processing system, the image processing method, and the image processing program disclosed in the present application are not limited by this embodiment. In addition, in the embodiment, the same reference numerals are given to configurations having the same functions and steps for performing the same processing.

[Example 1]
<Configuration of image processing system>
FIG. 1 is a diagram illustrating a configuration example of an image processing system according to the first embodiment. In FIG. 1, the image processing system 1 includes an image processing device 10, a control unit 11, a document reading unit 12, a document bundle detection unit 13, a staple detection unit 14, a staple removal unit 15, and an offset discharge unit 16. And have. The image processing apparatus 10, the document reading unit 12, the document bundle detection unit 13, the staple detection unit 14, the staple removal unit 15, and the offset discharge unit 16 operate under the control of the control unit 11. The image processing apparatus 10 includes a storage unit 101, an insertion position determination unit 102, an image insertion unit 103, and a file generation unit 104.

In the image processing apparatus 10, the storage unit 101 stores an image of a document read by the document reading unit 12 (hereinafter sometimes referred to as “document image”). The storage unit 101 stores a document image (hereinafter also referred to as “first document image”) read by the first reading of the document reading unit 12 with respect to the document group GR. For example, as shown in FIG. 2, the document group GR is formed of 14 documents A1, A2, B1, B2, B3, B4, C1, C2, C3, D1, D2, E1, E2, and F1. . However, the number of documents forming the document group GR to be processed by the image processing apparatus 10 is not limited to 14 sheets. FIG. 2 is a diagram for explaining an operation example of the image processing system according to the first embodiment. As shown in FIG. 2, the document group GR includes a document bundle BA, BB, BC, BD, BE, which is a bundle of a plurality of documents, as a part of the document group GR. Therefore, the number of originals forming the original bundles BA, BB, BC, BD, BE is smaller than the number of originals forming the original group GR. That is, the original bundle BA is a bundle of originals in which two originals A1 and A2 are bound together by a staple SP, and the original bundle BB is a four-part original B1, B2, B3, and B4. The original bundle BC is a bundle of originals, which are three originals C1, C2, and C3 bound together by staples SP, and the original bundle BD is the originals D1, D2. The two originals are a bundle of originals bound together by staples SP, and the original bundle BE is a bundle of originals obtained by binding two originals E1 and E2 together.

The insertion position determination unit 102 determines the insertion position of the document image (hereinafter, also referred to as “second document image”) read by the second reading of the document reading unit 12 with respect to the first document image. The second reading by the document reading unit 12 is a reading subsequent to the first reading by the document reading unit 12. For example, when the first reading is the first reading, the second reading is the second reading. Reading. However, other readings may exist between the first reading and the second reading. The first reading is a reading on the document group GR, and the second reading is a reading on any one or a plurality of document bundles BA, BB, BC, BD, BE included in the document group GR. . That is, the second document image is an image of each document included in one of the document bundles BA, BB, BC, BD, and BE. For example, when the second reading target is the document bundle BC, the second document image is an image of each of the documents C1, C2, and C3.

The image insertion unit 103 inserts the second document image into the first document image according to the insertion position determined by the insertion position determination unit 102.

The file generation unit 104 generates an image file including the second document image inserted into the first document image according to a predetermined file format. An example of the predetermined file format is PDF (Portable Document Format).

The original reading unit 12 reads the original group GR by the first reading, and any one or a plurality of original bundles BA, BB, BC, BD, BE included in the original group GR by the second reading. Is read.

The staple detection unit 14 detects the position of each staple SP binding the original bundles BA, BB, BC, BD, BE. Hereinafter, the position of the staple SP binding the bundle of documents may be referred to as “staple position”.

The staple removing unit 15 tries to remove the staple SP from each original bundle BA, BB, BC, BD, BE according to the staple position detected by the staple detecting unit 14. The removal of the staple SP by the staple removing unit 15 may be successful or unsuccessful. In the following, a document bundle in which the staple SP has been successfully removed by the staple removing unit 15 is referred to as a “removed successful document bundle”, and a document bundle in which the staple SP has failed to be removed by the staple removing unit 15 is referred to as a “removal failed document bundle”. Sometimes.

The document bundle detection unit 13 detects a removal failure document bundle among the document bundles BA, BB, BC, BD, BE in the document group GR. For example, the document bundle detection unit 13 detects a document bundle in which the staple SP still exists even after the staple removal unit 15 tries to remove the staple SP as a removal failure document bundle.

The offset discharge unit 16 offsets the removal failure original bundle in the original group GR and discharges it to a stacker (not shown). Thereby, in the document group GR which is discharged and stacked on the stacker after the first reading, the unsuccessful removal document bundle is offset and stacked with respect to the removal success document bundle.

<Operation of image processing system>
3 to 8 are diagrams for explaining an operation example of the image processing system according to the first embodiment.

At the time of the first reading by the document reading unit 12, as shown in FIG. 2, a document group GR formed from the document bundles BA, BB, BC, BD, BE and the document F1 is shot by a shooter (not shown). Placed in. The document group GR placed on the shooter is taken into the image processing system 1 by ADF (Auto Document Feeder). At this time, since the original bundles BA, BB, BC, BD, and BE are bound by the staple SP, the originals A1 and A2 bound together by the staple SP are conveyed to the staple detecting unit 14 as a unit. The The document is transported by a transport mechanism (not shown) included in the image processing system 1. Similarly, the originals B1, B2, B3, and B4 bound together by the staple SP are conveyed to the staple detection unit 14 as a unit, and the originals C1, C2, and C3 bound together by the staple SP are integrated as a staple detection unit. 14, the originals D1 and D2 bound together by the staple SP are transported together to the staple detection unit 14, and the originals E1 and E2 bound together by the staple SP are transported together to the staple detection unit 14. The Further, the document F1 not bound by the staple SP is conveyed to the staple detection unit 14 by one sheet. Therefore, the staple detection unit 14 detects the staple position for each of the document bundles BA, BB, BC, BD, BE, and the staple removal unit 15 detects the document bundles BA, BB, BC, BD, BE. An attempt is made to remove the staples SP for each original bundle.

Here, it is assumed that the staple SP has been successfully removed from all the original bundles BA, BB, BC, BD, and BE. Therefore, the originals A1, A2, B1, B2, B3, B4, C1, C2, C3, D1, D2, E1, E2, and F1 are fed one by one from the staple removing unit 15 via the original bundle detecting unit 13. It is conveyed to the document reading unit 12. At this time, since the staple SP has been successfully removed from all the document bundles BA, BB, BC, BD, and BE, the document bundle detection unit 13 does not detect the removal failure document bundle. Further, since the staple SP has been successfully removed from all the original bundles of the original bundles BA, BB, BC, BD, and BE, the original reading unit 12 has the originals A1, A2, B1, B2, B3, B4. Read each original of C1, C2, C3, D1, D2, E1, E2, and F1 one by one. The document reading unit 12 can read both sides of each document. Therefore, when the staple SP is successfully removed from all the original bundles of the original bundles BA, BB, BC, BD, and BE, the first original image read by the first reading and stored in the storage unit 101 is 3, the image A1f on the front surface of the document A1, the image A1b on the back surface of the document A1, the image A2f on the front surface of the document A2, the image A2b on the back surface of the document A2, the image B1f on the surface of the document B1, and the document B1. Image B1b on the back side of document B2, image B2f on the front side of document B2, image B2b on the back side of document B2, image B3f on the front side of document B3, image B3b on the back side of document B3, image B4f on the front side of document B4, and back side of document B4 Image B4b, image C1f on the front surface of document C1, image C1b on the back surface of document C1, image C2f on the surface of document C2, image C2b on the back surface of document C2, image C3f on the surface of document C3, original image C3 back image C3b, document D1 front image D1f, document D1 back image D1b, document D2 front image D2f, document D2 back image D2b, document E1 surface image E1f, document E1 The back image E1b, the front image E2f of the document E2, the back image E2b of the document E2, the front image F1f of the document F1, and the back image F1b of the document F1. That is, the images A1f, A1b, A2f, A2b, B1f, B1b, B2f, B2b, B3f, B3b, B4f, B4b, C1f, C1b, C2f, C2b, C3f, C3b, D1f, D1b, D2f, D2b, E1b, E1b , E2f, E2b, F1f, and F1b correspond to the first original image. The first document image is stored in the storage unit 101 in the order of images A1f to F1b.

Then, the file generation unit 104 outputs an image file corresponding to the document bundle BA (that is, an image file including all the images A1f, A1b, A2f, and A2b) and an image file corresponding to the document bundle BB (that is, the image B1f, B1b, B2f, B2b, B3f, B3b, B4f, B4b) and an image file corresponding to the original bundle BC (that is, an image including all of the images C1f, C1b, C2f, C2b, C3f, C3b). File), an image file corresponding to the document bundle BD (that is, an image file including all of the images D1f, D1b, D2f, and D2b), and an image file corresponding to the document bundle BE (that is, the images E1f, E1b, E2f, An image file including all of E2b) and an image file corresponding to the document F1 (that is, an image) 1f, an image file) and containing all the F1b, generates in accordance with a predetermined file format.

The above is an operation example when the staple SP is successfully removed from all the document bundles of the document bundles BA, BB, BC, BD, and BE.

Next, an example of operation when the removal of the staple SP from a part of the original bundles BA, BB, BC, BD, BE in the first reading is unsuccessful will be described. Here, for example, it is assumed that the removal of the staple SP from the original bundles BC and BE among the original bundles BA, BB, BC, BD, and BE has failed. Accordingly, the originals A1, A2, B1, B2, B3, B4, D1, D2, and F1 are conveyed one by one from the staple removing unit 15 to the original reading unit 12 via the original bundle detecting unit 13. The originals C1, C2, and C3 are transported together from the staple removing unit 15 to the original reading unit 12 via the original bundle detection unit 13 while being bound together by the staples SP (that is, as the original bundle BC), and the original E1. , E2 are conveyed together from the staple removal unit 15 to the document reading unit 12 via the document bundle detection unit 13 while being bound together by the staple SP (that is, as the document bundle BE). Therefore, the document bundle detection unit 13 detects the document bundles BC and BE as the removal failure document bundle.

Since the removal of the staple SP from each original bundle of the original bundles BA, BB, and BD has succeeded, the original reading unit 12 determines each of the originals A1, A2, B1, B2, B3, B4, D1, D2, and F1. Scan one original at a time. On the other hand, since the removal of the staple SP from the original bundles of the original bundles BC and BE has failed, the original reading unit 12 reads the original bundles BC and BE that are still bound by the staple SP. It is difficult to read the image C1b, the image C2f, the image C2b, and the image C3f in the document bundle BC that has been bound by the staple SP. Similarly, it is difficult to read the image E1b and the image E2f in the document bundle BE that has been bound by the staple SP. Therefore, when the document reading unit 12 reads the document bundle BC that has been bound by the staple SP, the image C1f that is the uppermost image of the document bundle BC and the image that is the lowermost image of the document bundle BC. C3b is read. Similarly, when the document reading unit 12 reads the document bundle BE that has been bound by the staple SP, the document reading unit 12 includes an image E1f that is an uppermost image of the document bundle BE and an image of the lowermost surface of the document bundle BE. The image E2b is read. Therefore, among the original bundles BA, BB, BC, BD, BE, when the removal of the staple SP from the original bundle BA, BB, BD is successful, but the removal of the staple SP from the original bundle BC, BE fails. As shown in FIG. 4, the first original image read by the first reading and stored in the storage unit 101 includes images A1f, A1b, A2f, A2b, images B1f, B1b, B2f, B2b, B3f. , B3b, B4f, B4b, images C1f, C3b, images D1f, D1b, D2f, D2b, images E1f, E2b, and images F1f, F1b. The images C1f and C3b and the images E1f and E2b include the image SPI of the staple SP. The first document image is stored in the storage unit 101 in the order of images A1f to F1b.

Then, the file generation unit 104 outputs an image file corresponding to the document bundle BA (that is, an image file including all the images A1f, A1b, A2f, and A2b) and an image file corresponding to the document bundle BB (that is, the image B1f, B1b, B2f, B2b, B3f, B3b, B4f, B4b), an image file corresponding to the original bundle BD (that is, an image file including all of the images D1f, D1b, D2f, D2b), An image file corresponding to the document F1 (that is, an image file including all of the images F1f and F1b) is generated according to a predetermined file format. Here, each of the document bundles BA, BB, and BD is a successfully removed document bundle.

Further, as shown in FIG. 5, the offset discharge unit 16 offsets the original bundles BC and BE, which are the removal failure original bundles, and discharges them to the stacker.

Then, when a removal failure original bundle occurs in the original group GR, the second reading is performed subsequent to the first reading. In the second reading subsequent to the first reading, the document bundles BC and BE from which the staple SP has been removed by the operator's manual work are sequentially placed on the shooter by the operator.

The document bundles BC and BE sequentially placed on the shooter by the operator are taken into the image processing system 1 by the ADF. At this time, since the staple SP has already been removed from the document bundle BC, the documents C1, C2, and C3 are one by one for the staple detection unit 14 and the staple removal unit 15 at the time of the second reading of the document bundle BC. Then, it is conveyed to the document reading unit 12 via the document bundle detection unit 13. Therefore, the document reading unit 12 reads each document of the documents C1, C2, and C3 one by one during the second reading of the document bundle BC. The document reading unit 12 can read both sides of each document. Therefore, as shown in FIG. 6, the second original image read by the second reading with respect to the original bundle BC includes an image C1f ′ on the front side of the original C1, an image C1b ′ on the back side of the original C1, and an image on the front side of the original C2. C2f ′, an image C2b ′ on the back side of the document C2, an image C3f ′ on the front surface of the document C3, and an image C3b ′ on the back surface of the document C3.

Similarly, since the staple SP has already been removed from the document bundle BE placed on the shooter by the operator, the documents E1 and E2 are scanned one by one when the document bundle BE is read. Then, it is conveyed to the document reading unit 12 via the staple removing unit 15 and the document bundle detecting unit 13. Therefore, the document reading unit 12 reads each document of the documents E1 and E2 one by one in the second reading of the document bundle BE. Therefore, as shown in FIG. 7, the second document image read by the second reading with respect to the document bundle BE is the image E1f ′ on the front surface of the document E1, the image E1b ′ on the back surface of the document E1, and the image on the surface of the document E2. E2f ′ and the image E2b ′ on the back side of the document E2.

That is, each of the images C1f ', C1b', C2f ', C2b', C3f ', C3b', E1f ', E1b', E2f ', E2b' corresponds to a second original image.

The insertion position determination unit 102 reads the image C1f read by the first reading (that is, the uppermost image of the document bundle BC (FIG. 4)) and the image E1f (that is, the uppermost image of the document bundle BE (FIG. 4). )) And each of the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′, E1f ′, E1b ′, E2f ′, and E2b ′ read by the second reading are compared. The degree of coincidence is calculated. Here, since both the image C1f and the image C1f ′ are images of the surface of the original C1, the image C1f and the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′, E1f ′, E1b ′, In the comparison with each of E2f ′ and E2b ′, the degree of coincidence between the image C1f and the image C1f ′ is maximized. Since both the image E1f and the image E1f ′ are images of the surface of the document E1, the image E1f and the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′, E1f ′, E1b ′, E2f In comparison with each of ', E2b', the degree of coincidence between the image E1f and the image E1f 'is maximized.

Therefore, the insertion position determination unit 102 determines the insertion positions of the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′ with respect to the first document image of the first document image stored in the storage unit 101. It is determined that this is the position of the image C1f (FIG. 4). Further, the insertion position determination unit 102 indicates the insertion position of the images E1f ′, E1b ′, E2f ′, E2b ′ with respect to the first document image, among the first document images stored in the storage unit 101, the image E1f (FIG. It is determined that the position is 4).

Therefore, the image insertion unit 103 converts the images C1f and C3b into images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′ in the first document image (FIG. 4) stored in the storage unit 101. By replacing, the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′ are inserted into the first document image. The image insertion unit 103 replaces the images E1f and E2b with the images E1f ′, E1b ′, E2f ′, and E2b ′ in the first document image (FIG. 4) stored in the storage unit 101, thereby Images E1f ′, E1b ′, E2f ′, E2b ′ are inserted into the original image. Therefore, the image group stored in the storage unit 101 after inserting the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, C3b ′ and the images E1f ′, E1b ′, E2f ′, E2b ′ is shown in FIG. As shown. In FIG. 8, all the images C1f ′, C1b ′, C2f ′, and B2b ′ of the document bundle BC are between the image B4b that is the image on the bottom surface of the document bundle BB and the image D1f that is the image on the top surface of the document bundle BD. C2b ′, C3f ′, and C3b ′ are inserted. Further, in FIG. 8, all the images E1f ′, E1b ′, E2f ′, E2f ′ of the document bundle BE are placed between the image D2b which is an image on the lowermost surface of the document bundle BD and the image F1f which is an image on the surface of the document F1. E2b ′ is inserted.

Further, the file generation unit 104 generates an image file corresponding to the document bundle BC (that is, an image file including all of the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′) and the document bundle BE. Corresponding image files (that is, image files including all of the images E1f ′, E1b ′, E2f ′, E2b ′) are generated according to a predetermined file format. Here, each of the document bundles BC and BE is a removal failure document bundle.

The above is an operation example when the removal of the staple SP from the original bundle BC, BE among the original bundles BA, BB, BC, BD, BE has failed.

<Processing of image processing system>
Hereinafter, the processing of the image processing system will be described separately for the first reading process and the second reading process.

<Processing at first reading>
FIG. 9 is a diagram illustrating an example of a processing flow during first reading according to the first embodiment.

First, in step ST01, the control unit 11 resets the document ID to “0”.

Next, in step ST03, a plurality of documents included in the document group GR placed on the shooter are sequentially conveyed to the staple detection unit 14 by the ADF.

Next, in step ST05, the staple detection unit 14 tries to detect the staple position on the document. If the staple position is detected on the document (step ST05: Yes), the process proceeds to step ST17. If the staple position is not detected on the document (step ST05: No), the process proceeds to step ST07. .

In step ST07, one original is conveyed to the original reading unit 12.

Next, in step ST09, the document reading unit 12 reads the document and stores the read image in the storage unit 101 as the first document image.

Next, in step ST11, the control unit 11 increments the document ID.

Next, in step ST13, the file generation unit 104 generates an image file including all the images read in step ST09, and stores the generated image file in the storage unit 101.

Next, in step ST15, the control unit 11 determines whether there is a document remaining on the shooter, that is, whether the document is not finished. If there is a document remaining on the shooter, that is, if the document is not finished (step ST15: No), the process returns to step ST03. On the other hand, when there is no document remaining on the shooter, that is, when the document is finished (step ST15: Yes), the control unit 11 finishes the first reading.

On the other hand, in step ST17, the staple removing unit 15 attempts to remove the staple SP from the document according to the staple position detected in step ST05.

Next, in step ST19, the document bundle detection unit 13 detects a removal failure document bundle by determining whether the staple removal unit 15 has successfully removed the staple SP. If the removal of the staple SP in step ST17 is successful (step ST19: Yes), the process proceeds to step ST21, and if the removal of the staple SP in step ST17 has failed (step ST19: No). The process proceeds to step ST31.

In step ST21, only one original is conveyed from the original bundle detection unit 13 to the original reading unit 12.

Next, in step ST23, the document reading unit 12 reads the document, and stores the read image in the storage unit 101 as a first document image.

Next, in step ST25, the control unit 11 determines whether or not conveyance of the document bundle from the document bundle detection unit 13 to the document reading unit 12 is completed. If no document remains in the document bundle detection unit 13, the control unit 11 determines that the conveyance of the document bundle from the document bundle detection unit 13 to the document reading unit 12 has been completed, and causes the document bundle detection unit 13 to send a document. Is left, it is determined that the conveyance of the document bundle from the document bundle detection unit 13 to the document reading unit 12 is not completed. If the conveyance of the document bundle from the document bundle detection unit 13 to the document reading unit 12 is not completed (step ST25: No), the process returns to step ST21, and the document from the document bundle detection unit 13 to the document reading unit 12 is returned. When the conveyance of the bundle is completed (step ST25: Yes), the process proceeds to step ST27.

In step ST27, the control unit 11 increments the document ID.

Next, in step ST29, the file generation unit 104 generates an image file including all the images read in step ST23, and stores the generated image file in the storage unit 101. After the process of step ST29, the process proceeds to step ST15.

On the other hand, in step ST31, the original is conveyed from the original bundle detection unit 13 to the original reading unit 12 with the original bundle bound with the staple SP.

Next, in step ST33, the document reading unit 12 reads the uppermost surface and the lowermost surface of the document bundle, reads the image of the uppermost surface (hereinafter sometimes referred to as “the uppermost image”), and the read lowermost surface. Are stored in the storage unit 101 as the first original image (hereinafter referred to as “lowermost image”).

Next, in step ST35, the control unit 11 increments the document ID.

Next, in step ST37, the insertion position determination unit 102 acquires the pixel density of the uppermost image. The pixel density is an example of an image feature amount.

Next, in step ST39, the offset discharge unit 16 offsets the document bundle bound with the staple SP and discharges it to the stacker. After the process of step ST39, the process proceeds to step ST15.

Here, in the same way as described above, in the document bundles BA, BB, BC, BD, BE and the document F1 forming the document group GR, the staple SP is successfully removed from the document bundles BA, BB, BD, but the document. Assume that the removal of the staple SP from the bundles BC and BE has failed.

Therefore, at the time of the first reading with respect to the document group GR, the processing of steps ST21 to ST29 is executed for the document bundle BA, BB, BD, and the steps ST31 to ST29 are performed for the document bundle BC, BE. The process of ST39 is executed, and the processes of steps ST07 to ST13 are executed for the document F1.

When the processing of steps ST21 to ST29 is executed for the document bundle BA, for example, as shown in FIG. 10, an image file “20171212_01_001.pdf” corresponding to “document ID = 1” is generated. FIG. 10 is a diagram illustrating an example of a document table at the time of first reading according to the first embodiment. This document table is stored in the storage unit 101. The image file “20171212_01_001.pdf” includes all images A1f, A1b, A2f, and A2b of the originals A1 and A2 that form the original bundle BA. The file generation unit 104 registers the image file “20171212_01_001.pdf” in the document table in association with “document ID = 1”. Further, since the document bundle BA is a successfully removed document bundle, the control unit 11 sets the status corresponding to “document ID = 1” to “OK”.

Further, when the processing of steps ST21 to ST29 is executed for the document bundle BB, for example, as shown in FIG. 10, an image file “20171212_01_002.pdf” corresponding to “document ID = 2” is generated. The image file “20171212_01_002.pdf” includes all the images B1f, B1b, B2f, B2b, B3f, B3b, B4f, and B4b of the originals B1, B2, B3, and B4 that form the original bundle BB. The file generation unit 104 registers the image file “20171212_01_002.pdf” in the document table in association with “document ID = 2”. Further, since the document bundle BB is a successfully removed document bundle, the control unit 11 sets the status corresponding to “document ID = 2” to “OK”.

Further, when the processing of steps ST31 to ST39 is performed on the document bundle BC, the document bundle BC is a removal failure document bundle. For example, as shown in FIG. The status corresponding to “” is set to “bundle”. Since the original bundle BC is a removal failure original bundle, an image file corresponding to “document ID = 3” is not generated at the time of the first reading.

Further, when the processing of steps ST21 to ST29 is performed on the document bundle BD, for example, as shown in FIG. 10, an image file “20171212_01_004.pdf” corresponding to “document ID = 4” is generated. The image file “20171212_01_004.pdf” includes all the images D1f, D1b, D2f, and D2b of the documents D1 and D2 that form the document bundle BD. The file generation unit 104 registers the image file “20171212_01_004.pdf” in the document table in association with “document ID = 4”. Since the original bundle BD is a successfully removed original bundle, the control unit 11 sets the status corresponding to “document ID = 4” to “OK”.

Further, when the processing of steps ST31 to ST39 is performed on the document bundle BE, the document bundle BE is a removal failure document bundle. For example, as shown in FIG. The status corresponding to “” is set to “bundle”. Further, since the original bundle BE is a removal failure original bundle, an image file corresponding to “document ID = 5” is not generated in the first reading.

Further, when the processing of steps ST07 to ST13 is performed on the document F1, for example, as shown in FIG. 10, an image file “20171212_01_006.pdf” corresponding to “document ID = 6” is generated. The image file “20171212_01_006.pdf” includes all the images F1f and F1b of the document F1. The file generation unit 104 registers the image file “20171212_01_006.pdf” in the document table in association with “document ID = 6”. In addition, since the staple SP is not detected in the original document F1, the control unit 11 sets the status corresponding to “document ID = 6” to “OK”.

The number “nnn” at the end of the file name “YYYYMMDD_mm_nnn.pdf” of the image file matches the document ID value. For this reason, the image files generated by the file generation unit 104 are sorted in ascending order of the document names by sorting in ascending order of the file names.

FIG. 11 is a diagram illustrating an example of area division according to the first embodiment, and FIG. 12 is a diagram illustrating an example of a pixel density table according to the first embodiment. This pixel density table is stored in the storage unit 101.

In acquiring the pixel density in step ST37 in the first reading, the insertion position determination unit 102 first binarizes the top surface image.

Next, as shown in FIG. 11, the insertion position determination unit 102 divides the entire area of the document into a plurality of areas. FIG. 11 shows, as an example, a case where an entire area of an original having an area of “vertical × horizontal = 4 m × 4 l” is divided into eight regions RE1 to RE8 having the same size.

Next, the insertion position determination unit 102 acquires the pixel density for each of the regions RE1 to RE8. The pixel density of each region is calculated by, for example, “the number of black pixels in each region / the area of each region”. Further, the area of each region is calculated by “m × l”.

Therefore, when the processing of steps ST31 to ST39 is performed on the document bundle BC, the insertion position determination unit 102 corresponds to “document ID = 3” in the pixel density table as shown in FIG. 12 in step ST37. Set the surface type to be “top”. Further, as shown in FIG. 12, the insertion position determination unit 102 calculates the pixel density for each of the regions RE1 to RE8 of the uppermost image (that is, the image C1f (FIG. 4)) of the document bundle BC. Each pixel density is recorded in association with “Document ID = 3”. Here, on the top surface of the document bundle BC, the pixel density of the region RE1 is “0.32”, the pixel density of the region RE2 is “0.11”, the pixel density of the region RE3 is “0.07”, and the pixel density of the region RE4 is “0.01”. Assume that the pixel density of the region RE5 is calculated as “0.23”, the pixel density of the region RE6 is “0.07”, the pixel density of the region RE7 is “0.11”, and the pixel density of the region RE8 is “0.16”.

When the processing of steps ST31 to ST39 is performed on the document bundle BE, the insertion position determining unit 102 corresponds to “document ID = 5” in the pixel density table in step ST37 as shown in FIG. Set the surface type to be “top”. Further, as shown in FIG. 12, the insertion position determination unit 102 calculates the pixel density for each of the regions RE1 to RE8 of the uppermost image (that is, the image E1f (FIG. 4)) of the document bundle BE. Each pixel density is recorded in association with “Document ID = 5”. Here, on the top surface of the original bundle BE, the pixel density of the region RE1 is “0.12”, the pixel density of the region RE2 is “0.05”, the pixel density of the region RE3 is “0.23”, and the pixel density of the region RE4 is “0.26”. Assume that the pixel density of the region RE5 is calculated as “0.17”, the pixel density of the region RE6 is “0.17”, the pixel density of the region RE7 is “0.12”, and the pixel density of the region RE8 is “0.11”.

<Second reading process>
FIG. 13 is a diagram illustrating an example of a processing flow during second reading according to the first embodiment. At the start of the processing flow shown in FIG. 13, the staple SP has already been removed from the original bundle that has been offset and discharged in step ST39 at the time of the first reading by the operator's manual work, and the original bundle after the staple SP has been removed becomes the shooter. Already put in.

First, in step ST41, one of the plurality of documents forming the bundle of documents placed on the shooter is conveyed to the document reading unit 12.

Next, in step ST43, the document reading unit 12 reads the document and stores the read image in the storage unit 101 as a second document image.

Next, in step ST45, the insertion position determination unit 102 acquires the pixel density of the image read in step ST43 in the same manner as the acquisition of the pixel density in step ST37 in the first reading.

Next, in step S47, the control unit 11 determines whether or not there is a document remaining on the shooter, that is, whether or not the document is not finished. If there is a document remaining on the shooter, that is, if the document is not finished (step ST47: No), the process returns to step ST41. On the other hand, when there is no document remaining in the shooter, that is, when the document is finished (step ST47: Yes), the process proceeds to step ST49.

In step ST49, the insertion position determination unit 102 refers to the document table (FIG. 10) stored in the storage unit 101, and acquires the document ID whose status is “bundle”.

Next, in step ST51, the insertion position determination unit 102 obtains the pixel density acquired in step ST37 (FIG. 9) at the time of the first reading (hereinafter sometimes referred to as “first reading pixel density”), the first The pixel density acquired in step ST45 during the second reading (hereinafter, sometimes referred to as “second reading pixel density”) is compared. In the comparison between the first reading pixel density and the second reading pixel density, for example, the insertion position determination unit 102 determines the difference between the first reading pixel density and the second reading pixel density over the regions RE1 to RE8. The total absolute value is calculated as a distance value between the first document image and the second document image. Calculation of the distance value between the first document image and the second document image is equivalent to calculating the degree of coincidence between the first document image and the second document image.

Next, in step ST53, the insertion position determination unit 102 specifies the minimum distance value (hereinafter, sometimes referred to as “minimum distance value”) in the plurality of second document images.

Next, in step ST55, the insertion position determination unit 102 determines whether or not the minimum distance value is less than the threshold value TH. Determining whether or not the minimum distance value is less than the threshold value TH corresponds to determining whether or not the degree of coincidence between the first document image and the second document image is greater than or equal to the threshold value. When the minimum distance value is less than the threshold value TH (that is, when the matching degree is greater than or equal to the threshold value) (step ST55: Yes), the process proceeds to step ST57.

In step ST57, the insertion position determination unit 102 determines the second document image to be stored in the image file generated in step ST59 based on the minimum distance value specified in step ST53, and the first document image of the second document image. Determine the insertion position for.

Next, in step ST59, the file generation unit 104 generates an image file including the second document image to be inserted into the first document image.

Next, in step ST61, the image insertion unit 103 inserts the image file generated in step ST59 according to the insertion position determined in step ST57. After the process of step ST61, the process proceeds to step ST63.

On the other hand, when the minimum distance value is greater than or equal to the threshold value TH (that is, when the matching degree is less than the threshold value) (step ST55: No), the process proceeds to step ST63 without performing the processes of steps ST57 to ST61.

In step ST63, the control unit 11 determines whether there is an image (hereinafter, referred to as “residual image”) that is not stored in the image file among the images read by the second reading. When there is no remaining image (step ST63: No), the control unit 11 ends the second reading.

On the other hand, if there is a remaining image (step ST63: Yes), in step ST65, the control unit 11 displays a warning that the insertion position of the second document image relative to the first document image is unknown (not shown). To display. In accordance with the warning, the operator manually designates the insertion position of the second document image with respect to the first document image to the image processing system 1.

In step ST67, the image insertion unit 103 inserts the second original image at the insertion position manually designated by the operator. After the process of step ST67, the control unit 11 ends the second reading.

Here, in the same way as described above, in the document bundles BA, BB, BC, BD, BE and the document F1 forming the document group GR, the staple SP is successfully removed from the document bundles BA, BB, BD, but the document. Assume that the removal of the staple SP from the bundles BC and BE has failed. Therefore, at the time of the second reading, the original bundle BC is first placed on the shooter among the original bundles BC and BE after the staple SP is removed by the manual operation of the operator, and the original bundle BC is read for the second reading. It becomes. Further, after the second reading with respect to the document bundle BC is completed, the document bundle BE is then placed on the shooter, and the document bundle BE becomes a reading target for the second reading.

FIG. 14 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BC of Embodiment 1, and FIG. 15 is calculated at the time of the second reading for the document bundle BC of Embodiment 1. FIG. 16 is a diagram illustrating an example of a distance value, and FIG. 16 is a diagram illustrating an example of a document table at the time of second reading of the document bundle BC according to the first embodiment. FIG. 17 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BE of the first embodiment, and FIG. 18 is a calculation at the time of the second reading for the document bundle BE of the first embodiment. FIG. 19 is a diagram illustrating an example of a document table at the time of second reading of the document bundle BE according to the first embodiment.

When the processing of steps ST41 to ST47 is performed on the document bundle BC, for example, as shown in FIG. 14, the images C1f ′, C1b ′, C2f ′, and the like of the documents C1, C2, and C3 that form the document bundle BC. In C2b ′, C3f ′, and C3b ′, the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8. In FIG. 14, for example, in the image C1f ′, the pixel density of the region RE1 is “0.31”, the pixel density of the region RE2 is “0.11”, the pixel density of the region RE3 is “0.06”, and the pixel density of the region RE4 is “0.01”. Assume that the pixel density of the region RE5 is calculated as “0.23”, the pixel density of the region RE6 is “0.08”, the pixel density of the region RE7 is “0.11”, and the pixel density of the region RE8 is “0.15”. Further, for example, in the image C1b ′, the pixel density of the region RE1 is “0.35”, the pixel density of the region RE2 is “0.52”, the pixel density of the region RE3 is “0.25”, the pixel density of the region RE4 is “0.31”, and the region RE5 Is calculated as “0.10”, the pixel density of the region RE6 is “0.20”, the pixel density of the region RE7 is “0.25”, and the pixel density of the region RE8 is “0.67”. Similarly, in the images C2f ′, C2b ′, C3f ′, and C3b ′, the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8.

In step ST51, the insertion position determination unit 102 compares the first reading pixel density (FIG. 12) with the second reading pixel density of the document bundle BC (FIG. 14), and the document acquired in step ST49. For each ID, the sum of the absolute values of the differences between the first reading pixel density over the regions RE1 to RE8 and the second reading pixel density of the document bundle BC is calculated as the second document of the first document image and the document bundle BC. It is calculated as the distance value between the images. Here, the document IDs acquired in step ST49 with reference to the document table (FIG. 10) are “3” and “5”. Therefore, for example, as illustrated in FIG. 15, the insertion position determination unit 102 associates the image C1f (that is, the first image) with the “top surface” of “document ID = 3” in accordance with the pixel density table (FIG. 12). The distance value of the image C1f ′ with respect to “the top image of“ document ID = 3 ”in one reading) is set to“ | 0.32-0.31 | + | 0.11-0.11 | + | 0.07-0.06 | + | 0.01-0.01 | + | 0.23 -0.23 | + | 0.07-0.08 | + | 0.11-0.11 | + | 0.16-0.15 | = 0.04 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image C1b ′ with the image C1f in association with the “top surface” of “document ID = 3” by “| 0.32-0.35 | + | 0.11-0.52 | + | 0.07-0.25 | + | 0.01-0.31 | + | 0.23-0.10 | + | 0.07-0.20 | + | 0.11-0.25 | + | 0.16-0.67 | = 1.83 ”. Further, for example, the insertion position determination unit 102 associates the “top ID” of “document ID = 5” with the image C1f for the image E1f (that is, the topmost image of “document ID = 5” in the first reading). The distance value of “| 0.12-0.31 | + | 0.05-0.11 | + | 0.23-0.06 | + | 0.26-0.01 | + | 0.17-0.23 | + | 0.17-0.08 | + | 0.12-0.11 | + | 0.11 -0.15 | = 0.87 ” Further, for example, the insertion position determination unit 102 associates the distance value of the image C1b ′ with the image E1f in association with the “top surface” of “document ID = 5” by “| 0.12-0.35 | + | 0.05-0.52 | + | 0.23-0.25 | + | 0.26-0.31 | + | 0.17-0.10 | + | 0.17-0.20 | + | 0.12-0.25 | + | 0.11-0.67 | = 1.56 ”. The distance values of the images C2f ', C2b', C3f ', C3b' are calculated in the same way (FIG. 15).

Further, since the document IDs acquired in step ST49 are “3” and “5”, in step ST53, the insertion position determination unit 102 determines “document ID = 3” and “document ID = 5” in FIG. The minimum distance value among the plurality of distance values of 0.04, 1.83, 1.41, 1.59, 2.88, 0.48, 0.87, 1.56, 1.36, 1.10, 2.43, 0.97 is set to “0.04” of the image C1f ′. "" (FIG. 15).

For example, if the threshold value TH used in step ST55 is “0.30”, in step ST55, the insertion position determination unit 102 determines that the minimum distance value “0.04” specified in step ST53 is less than the threshold value TH. . That is, for the document bundle BC, the insertion position determination unit 102 sets the image C1f ′ having the minimum distance value among the plurality of second document images to be equal to the image C1f that is the uppermost image of the first document images. Judge that you are doing.

In step ST57, since the minimum distance value “0.04” specified in step ST53 corresponds to “document ID = 3” (FIG. 15), the insertion position determination unit 102 determines the second original image C1f ′ of the original bundle BC. , C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′ are determined as second document images to be stored in the image file of “document ID = 3”.

Here, the “top surface” of “document ID = 3” corresponds to the top surface of the document bundle BC (that is, the reading surface of the image C1f) by the processing of steps ST33 to ST37 during the first reading. Therefore, determining the second document image to be stored in the image file generated in step ST59 based on the document ID corresponds to determining the insertion position of the second document image with respect to the first document image. For example, it is determined in step ST57 that a series of images from the image C1f ′ to the image C3b ′ is determined as the second document image to be stored in the image file with “document ID = 3” from the image C1f ′ to the image C3b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image C1f (FIG. 4) in the first document image.

Next, in step ST59, the file generation unit 104 is inserted at the position of the image C1f in the first document image as an image file corresponding to “document ID = 3” according to the determination of the insertion position determination unit 102 in step ST57. An image file “20171212_01_003.pdf” including a series of images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′ is generated. The images C1f ', C1b', C2f ', C2b', C3f ', and C3b' are all images of the originals C1, C2, and C3 that form the original bundle BC to be read in the second reading.

Next, in step ST61, the image insertion unit 103 deletes the images C1f and C3b that are the first document images stored in the storage unit 101, and then, as shown in FIG. 16, the images C1f ′, C1b ′, and C2f. The image file “20171212_01_003.pdf” including “, C2b”, C3f ′, and C3b ′ is registered in the document table in association with “document ID = 3” to update the document table. Further, the image insertion unit 103 updates the status corresponding to “document ID = 3” from “bundle” (FIG. 10) to “insertion”. As a result, the image file “20171212_01_003.pdf” is inserted between the image file “20171212_01_002.pdf” and the image file “20171212_01_004.pdf”. That is, the images C1f ′, C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′ that are the second document images of the document bundle BC are replaced with the images C1f and C3b that are the first document images of the document bundle BC. In the plurality of first document images, the images are inserted between the image B4b and the image D1f.

Next, when the processing of steps ST41 to ST47 is performed on the document bundle BE placed on the shooter after the second reading with respect to the document bundle BC is completed, for example, as shown in FIG. In each image E1f ′, E1b ′, E2f ′, E2b ′ of the originals E1, E2 to be formed, the pixel density is calculated for each region RE1, RE2, RE3, RE4, RE5, RE6, RE7, RE8. In FIG. 17, for example, in the image E1f ′, the pixel density of the region RE1 is “0.11”, the pixel density of the region RE2 is “0.05”, the pixel density of the region RE3 is “0.22”, and the pixel density of the region RE4 is “0.26”. Assume that the pixel density of the region RE5 is calculated as “0.17”, the pixel density of the region RE6 is “0.16”, the pixel density of the region RE7 is “0.12”, and the pixel density of the region RE8 is “0.11”. Further, for example, in the image E1b ′, the pixel density of the region RE1 is “0.54”, the pixel density of the region RE2 is “0.32”, the pixel density of the region RE3 is “0.10”, the pixel density of the region RE4 is “0.40”, and the region RE5 The pixel density of “0.68”, the pixel density of the region RE6 is “0.21”, the pixel density of the region RE7 is “0.55”, and the pixel density of the region RE8 is “0.77”. Similarly, in the images E2f 'and E2b', the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8.

Therefore, in step ST51, the insertion position determination unit 102 compares the first reading pixel density (FIG. 12) with the second reading pixel density of the document bundle BE (FIG. 17), and the document acquired in step ST49. For each ID, the sum of absolute values of differences between the first reading pixel density over the regions RE1 to RE8 and the second reading pixel density of the document bundle BE is calculated as the second document of the first document image and the document bundle BE. It is calculated as the distance value between the images. Here, the document IDs acquired in step ST49 with reference to the document table (FIG. 10) are “3” and “5”. Therefore, for example, as illustrated in FIG. 18, the insertion position determination unit 102 associates the image C1f (that is, the first image) in association with the “top surface” of “document ID = 3” in accordance with the pixel density table (FIG. The distance value of the image E1f ′ with respect to “the top image of“ document ID = 3 ”in one reading) is set to“ | 0.32-0.11 | + | 0.11-0.05 | + | 0.07-0.02 | + | 0.01-0.26 | + | 0.23 -0.17 | + | 0.07-0.16 | + | 0.11-0.12 | + | 0.16-0.01 | = 0.88 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image E1b ′ with respect to the image C1f “| 0.32-0.54 | + | 0.11-0.32 | + |” in association with the “top surface” of “document ID = 3”. 0.07-0.10 | + | 0.01-0.40 | + | 0.23-0.68 | + | 0.07-0.21 | + | 0.11-0.55 | + | 0.16-0.77 | = 2.49 ”. Further, for example, the insertion position determination unit 102 associates with the “uppermost surface” of “document ID = 5”, the image E1f for the image E1f (that is, the uppermost image of “document ID = 5” in the first reading). The distance value of “| 0.12-0.11 | + | 0.05-0.05 | + | 0.23-0.22 | + | 0.26-0.26 | + | 0.17-0.17 | + | 0.17-0.16 | + | 0.12-0.12 | + | 0.11 -0.11 | = 0.03 ” Further, for example, the insertion position determination unit 102 associates the distance value of the image E1b ′ with the image E1f in association with the “top surface” of “document ID = 5” “| 0.12-0.54 | + | 0.05-0.32 | + | 0.23-0.10 | + | 0.26-0.40 | + | 0.17-0.68 | + | 0.17-0.21 | + | 0.12-0.55 | + | 0.11-0.77 | = 2.60 ”. The distance value between the images E2f 'and E2b' is calculated in the same manner (FIG. 18).

Further, since the document IDs acquired in step ST49 are “3” and “5”, in step ST53, the insertion position determination unit 102 determines “document ID = 3” and “document ID = 5” in FIG. The minimum distance value is specified as “0.03” of the image E1f ′ from among a plurality of distance values 0.88, 2.49, 2.23, 1.02, 0.03, 2.60, 2.26, 0.65 (FIG. 18). .

For example, if the threshold value TH used in step ST55 is “0.30”, in step ST55, the insertion position determination unit 102 determines that the minimum distance value “0.03” specified in step ST53 is less than the threshold value TH. . That is, for the document bundle BE, the insertion position determination unit 102 sets the image E1f ′ having the minimum distance value among the plurality of second document images to be equal to the image E1f that is the uppermost image of the first document images. Judge that you are doing.

In step ST57, since the minimum distance value “0.03” specified in step ST53 corresponds to “document ID = 5” (FIG. 18), the insertion position determination unit 102 determines the second document image E1f ′ of the document bundle BE. , E1b ′, E2f ′, and E2b ′ are determined as second document images to be stored in the image file of “document ID = 5”.

Here, the “top surface” of “document ID = 5” corresponds to the top surface of the document bundle BE (that is, the reading surface of the image E1f) by the processing of steps ST33 to ST37 during the first reading. Therefore, determining the second document image to be stored in the image file generated in step ST59 based on the document ID corresponds to determining the insertion position of the second document image with respect to the first document image. For example, in step ST57, determining a series of images from the image E1f ′ to the image E2b ′ as the second document image to be stored in the image file with “document ID = 5” is from the image E1f ′ to the image E2b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image E1f (FIG. 4) in the first document image.

Next, in step ST59, the file generation unit 104 is inserted at the position of the image E1f in the first document image as an image file corresponding to “document ID = 5” according to the determination of the insertion position determination unit 102 in step ST57. An image file “20171212_01_005.pdf” including a series of images E1f ′, E1b ′, E2f ′, E2b ′ is generated. The images E1f ', E1b', E2f ', and E2b' are all images of the documents E1 and E2 that form the document bundle BE to be read in the second reading.

Next, in step ST61, the image insertion unit 103 deletes the images E1f and E2b that are the first document images stored in the storage unit 101, and then, as shown in FIG. 19, the images E1f ′, E1b ′, and E2f. The image file “20171212_01_005.pdf” including “, E2b” is registered in the document table in association with “document ID = 5”, and the document table is updated. Further, the image insertion unit 103 updates the status corresponding to “document ID = 5” from “bundle” (FIG. 10) to “insertion”. As a result, the image file “20171212_01_005.pdf” is inserted between the image file “20171212_01_004.pdf” and the image file “20171212_01_006.pdf”. That is, the images E1f ′, E1b ′, E2f ′, E2b ′, which are the second document images of the document bundle BE, are replaced with the images E1f, E2b, which are the first document images of the document bundle BE, and a plurality of first document images. In this case, the image is inserted between the image D2b and the image F1f.

As described above, in the first embodiment, the image processing apparatus 10 includes the storage unit 101, the insertion position determination unit 102, and the image insertion unit 103. The storage unit 101 stores the first original image read by the first reading on the original group GR including the original bundles BA, BB, BC, BD, BE. The insertion position determination unit 102 determines the insertion position of the second document image that is an image of each document of the document bundles BC and BE read by the second reading with respect to the first document image. The image insertion unit 103 inserts the second document image between the first document images according to the insertion position determined by the insertion position determination unit 102.

In this way, the second document image is automatically inserted at an appropriate position of the first document image, so that the operator's work efficiency can be improved.

In the first embodiment, the insertion position determination unit 102 determines the second document image based on the comparison result between the top image of the document bundle BC or BE and the second document image. The insertion position for one document image is determined.

By doing so, it is possible to accurately determine the insertion position of the second document image with respect to the first document image.

In the first embodiment, the image processing apparatus 10 includes the file generation unit 104. The file generation unit 104 generates an image file for each original bundle of the original bundles BA, BB, BC, BD, BE according to a predetermined file format such as PDF.

By doing so, it is possible to reduce the trouble of operator file management.

In the first embodiment, the image processing system 1 includes the image processing apparatus 10, the document bundle detection unit 13, and the document reading unit 12. The document bundle detection unit 13 detects document bundles BC and BE that are document bundles that have failed to be removed in the document group GR. When the document bundle detection unit 13 detects the document bundles BC and BE that are unsuccessful removal document bundles, the document reading unit 12 follows the first reading of the document group GR and the document that is the removal failure document bundle. A second reading is performed on the bundles BC and BE.

In this way, the image processing apparatus 10, the document bundle detection unit 13, and the document reading unit 12 can be linked to insert the second document image of the removal failure document bundle into an appropriate position of the first document image. it can.

In the first embodiment, the image processing system 1 includes the offset discharge unit 16. The offset discharge unit 16 offsets and discharges the document bundles BC and BE, which are unremoved document bundles, in the document group GR.

By doing so, in the document group GR, the unsuccessful removal document bundles BC, BE are offset with respect to the removal success document bundles BA, BB, BD and the document F1 and stacked on the stacker. It is possible to easily identify a bundle of documents that have failed to be removed.

[Example 2]
In the first embodiment, the comparison target of the second document image is the uppermost image of the first document image. On the other hand, in the second embodiment, the comparison target of the second document image is the lowermost image of the first document image instead of the uppermost image of the first document image. Hereinafter, differences from the first embodiment will be described.

<Processing at first reading>
When the processing of steps ST31 to ST39 (FIG. 9) is performed on the document bundle BC, the insertion position determination unit 102 sets the surface type corresponding to “document ID = 3” in the document table to “highest” in step ST37. Set to “Bottom”. Further, the insertion position determination unit 102 calculates the pixel density for each of the regions RE1 to RE8 of the lowermost image (that is, the image C3b (FIG. 4)) of the document bundle BC, and calculates the calculated pixel density as “document”. Record in association with ID = 3 ″.

When the processing of steps ST31 to ST39 is executed for the document bundle BE, the insertion position determination unit 102 sets the surface type corresponding to “document ID = 5” in the document table to “bottom surface” in step ST37. Set to. Further, the insertion position determination unit 102 calculates the pixel density for each of the regions RE1 to RE8 of the lowermost image (that is, the image E2b (FIG. 4)) of the document bundle BE, and calculates the calculated pixel density as “document”. Record in association with ID = 5 ″.

That is, in Example 2, the pixel density of the lowermost image of the document bundle BC and the pixel density of the lowermost image of the document bundle BE are the first reading pixel density.

<Second reading process>
At the time of the second reading with respect to the document bundle BC, the image having the minimum distance value among the plurality of second document images is the image C3b ′. Therefore, in step ST55 (FIG. 13), the insertion position determination unit 102, for the document bundle BC, the image C3b ′ having the minimum distance value among the plurality of second document images is the highest of the first document images. It is determined that the image matches the image C3b which is the bottom image.

In step ST57, since the minimum distance value specified in step ST53 corresponds to “document ID = 3”, the insertion position determination unit 102 determines the second document images C1f ′ and C1b ′ of the document bundle BC. , C2f ′, C2b ′, C3f ′, and C3b ′ are determined as second document images to be stored in the image file of “document ID = 3”.

Here, the “lowermost surface” of “document ID = 3” corresponds to the lowermost surface of the document bundle BC (that is, the reading surface of the image C3b) by the processing of steps ST33 to ST37 during the first reading. Therefore, in step ST57, determining a series of images from the image C1f ′ to the image C3b ′ as the second document image to be stored in the image file of “document ID = 3” is from the image C1f ′ to the image C3b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image C3b (FIG. 4) in the first document image.

In the second reading of the document bundle BE, an image having the minimum distance value among the plurality of second document images is an image E2b ′. Therefore, in step ST55 (FIG. 13), for the document bundle BE, the insertion position determination unit 102 determines that the image E2b ′ having the minimum distance value among the plurality of second document images is the highest of the first document images. It is determined that it matches the image E2b which is the lower surface image.

In step ST57, since the minimum distance value specified in step ST53 corresponds to “document ID = 5”, the insertion position determination unit 102 determines the second document images E1f ′ and E1b ′ of the document bundle BE. , E2f ′, E2b ′ are determined as second document images to be stored in the image file of “document ID = 5”.

Here, the “lowermost surface” of “document ID = 5” corresponds to the lowermost surface of the document bundle BE (that is, the reading surface of the image E2b) by the processing of steps ST33 to ST37 during the first reading. Therefore, the determination of the series of images from the image E1f ′ to the image E2b ′ as the second document image to be stored in the image file with “document ID = 5” in step ST57 is from the image E1f ′ to the image E2b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image E2b (FIG. 4) in the first document image.

As described above, in the second embodiment, the insertion position determination unit 102 determines the second original based on the comparison result between the lowermost image of the original bundle BC and BE and the second original image in the first original image. The insertion position of the image with respect to the first document image is determined.

By doing so, the insertion position of the second document image with respect to the first document image can be accurately determined as in the first embodiment.

[Example 3]
In the first embodiment, the comparison target of the second document image is the uppermost image of the first document image. In the second embodiment, the second original image is compared with the lowermost image of the first original image. However, in the document bundles BC and BE, which are unremoved document bundles generated at the time of the first reading, the uppermost image is approximated between the respective document bundles, or the lowermost image is approximated between the respective document bundles. Sometimes. Therefore, in the third embodiment, the comparison target of the second document image is both the uppermost image and the lowermost image of the first document image. Hereinafter, differences from the first and second embodiments will be described.

<Processing at first reading>
FIG. 20 is a diagram illustrating an example of a pixel density table according to the third embodiment. This pixel density table is stored in the storage unit 101.

When the processing of steps ST31 to ST39 (FIG. 9) is performed on the document bundle BC, the insertion position determination unit 102 sets “document ID = 3” in the document table as shown in FIG. 20 in step ST37. The corresponding surface type is set to “uppermost surface” and “lowermost surface”. Further, as shown in FIG. 20, the insertion position determination unit 102 includes a region RE1 of the uppermost image (that is, the image C1f (FIG. 4)) and the lowermost image (that is, the image C3b (FIG. 4)) of the document bundle BC. The pixel density is calculated for each area of RE8, and the calculated pixel density is recorded in association with “document ID = 3”. Here, on the top surface of the document bundle BC, the pixel density of the region RE1 is “0.01”, the pixel density of the region RE2 is “0.27”, the pixel density of the region RE3 is “0.24”, and the pixel density of the region RE4 is “0.12”. It is assumed that the pixel density of the region RE5 is “0.19”, the pixel density of the region RE6 is “0.07”, the pixel density of the region RE7 is “0.24”, and the pixel density of the region RE8 is “0.32”. On the lowermost surface of the document bundle BC, the pixel density of the region RE1 is “0.07”, the pixel density of the region RE2 is “0.01”, the pixel density of the region RE3 is “0.27”, and the pixel density of the region RE4 is “0.01”, Assume that the pixel density of the region RE5 is calculated as “0.25”, the pixel density of the region RE6 is “0.30”, the pixel density of the region RE7 is “0.24”, and the pixel density of the region RE8 is “0.01”.

When the processing of steps ST31 to ST39 is performed on the document bundle BE, the insertion position determination unit 102 corresponds to “document ID = 5” in the document table as shown in FIG. 18 in step ST37. Set the surface type to “Top” and “Bottom”. Further, as shown in FIG. 18, the insertion position determination unit 102 includes a region RE1 of the uppermost image (that is, the image E1f (FIG. 4)) and the lowermost image (that is, the image E2b (FIG. 4)) of the document bundle BE. The pixel density is calculated for each area of RE8, and the calculated pixel density is recorded in association with “document ID = 5”. Here, on the top surface of the document bundle BE, the pixel density of the region RE1 is “0.01”, the pixel density of the region RE2 is “0.26”, the pixel density of the region RE3 is “0.26”, and the pixel density of the region RE4 is “0.10”. It is assumed that the pixel density of the region RE5 is “0.18”, the pixel density of the region RE6 is “0.08”, the pixel density of the region RE7 is “0.24”, and the pixel density of the region RE8 is “0.33”. Further, on the lowermost surface of the document bundle BE, the pixel density of the region RE1 is “0.32”, the pixel density of the region RE2 is “0.11”, the pixel density of the region RE3 is “0.07”, and the pixel density of the region RE4 is “0.01”, Assume that the pixel density of the region RE5 is calculated as “0.20”, the pixel density of the region RE6 is “0.23”, the pixel density of the region RE7 is “0.11”, and the pixel density of the region RE8 is “0.16”.

<Second reading process>
FIG. 21 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BC of Embodiment 3, and FIG. 22 is calculated at the time of the second reading for the document bundle BC of Embodiment 3. It is a figure which shows an example of a distance value. FIG. 23 is a diagram illustrating an example of the pixel density acquired at the time of the second reading for the document bundle BE of the third embodiment, and FIG. 24 is a calculation at the time of the second reading for the document bundle BE of the third embodiment. It is a figure which shows an example of the distance value performed.

When the processing of steps ST41 to ST47 (FIG. 13) is performed on the document bundle BC, for example, as shown in FIG. 21, the images C1f ′ and C1b ′ of the documents C1, C2, and C3 that form the document bundle BC. , C2f ′, C2b ′, C3f ′, and C3b ′, the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8. In FIG. 21, for example, in the image C1f ′, the pixel density of the region RE1 is “0.01”, the pixel density of the region RE2 is “0.26”, the pixel density of the region RE3 is “0.25”, and the pixel density of the region RE4 is “0.01”. It is assumed that the pixel density of the region RE5 is “0.18”, the pixel density of the region RE6 is “0.07”, the pixel density of the region RE7 is “0.24”, and the pixel density of the region RE8 is “0.32”. Further, for example, in the image C1b ′, the pixel density of the region RE1 is “0.25”, the pixel density of the region RE2 is “0.02”, the pixel density of the region RE3 is “0.47”, the pixel density of the region RE4 is “0.28”, and the region RE5 , The pixel density of the region RE6 is calculated to be “0.24”, the pixel density of the region RE7 is “0.09”, and the pixel density of the region RE8 is “0.12”. Similarly, in the images C2f ′, C2b ′, C3f ′, and C3b ′, the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8.

Therefore, in step ST51, the insertion position determination unit 102 compares the first reading pixel density (FIG. 20) with the second reading pixel density of the document bundle BC (FIG. 21), and the document acquired in step ST49. For each ID, on both the uppermost surface and the lowermost surface, the sum of absolute values of differences between the first reading pixel density over the regions RE1 to RE8 and the second reading pixel density of the document bundle BC is calculated as the first document. This is calculated as a distance value between the image and the second original image of the original bundle BC. Here, the document IDs acquired in step ST49 with reference to the document table (FIG. 10) are “3” and “5”.

Therefore, for example, as illustrated in FIG. 22, the insertion position determination unit 102 associates the image C1f (that is, the first image) in association with the “top surface” of “document ID = 3” in accordance with the pixel density table (FIG. 20). The distance value of the image C1f ′ with respect to “the uppermost image of“ document ID = 3 ”in one reading) is set to“ | 0.01-0.01 | + | 0.27-0.26 | + | 0.24-0.25 | + | 0.12-0.11 | + | 0.19 -0.18 | + | 0.07-0.07 | + | 0.24-0.24 | + | 0.32-0.32 | = 0.04 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image C1b ′ with the image C1f in association with the “top surface” of “document ID = 3” by “| 0.01-0.25 | + | 0.27-0.02 | + | 0.24-0.47 | + | 0.12-0.28 | + | 0.19-0.18 | + | 0.07-0.24 | + | 0.24-0.09 | + | 0.32-0.12 | = 1.41 ”. Further, for example, the insertion position determination unit 102 associates the “lowermost surface” of “document ID = 3” with the image C1b for the image C3b (that is, the lowermost image of “document ID = 3” in the first reading). The distance value of “| 0.07-0.01 | + | 0.01-0.26 | + | 0.27-0.25 | + | 0.01-0.11 | + | 0.25-0.18 | + | 0.30-0.07 | + | 0.24-0.24 | + | 0.01 -0.32 | = 1.04 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image C1b ′ with respect to the image C3b “| 0.07−0.25 | + | 0.01−0.02 | + |” in association with “bottom surface” of “document ID = 3”. 0.27-0.47 | + | 0.01-0.28 | + | 0.25-0.18 | + | 0.30-0.24 | + | 0.24-0.09 | + | 0.01-0.12 | = 1.05 ”.

Further, for example, the insertion position determination unit 102 associates the “top ID” of “document ID = 5” with the image C1f for the image E1f (that is, the topmost image of “document ID = 5” in the first reading). The distance value of “| 0.01-0.01 | + | 0.26-0.26 | + | 0.26-0.25 | + | 0.10-0.11 | + | 0.18-0.18 | + | 0.08-0.07 | + | 0.24-0.24 | + | 0.33 -0.32 | = 0.04 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image C1b ′ with the image E1f in association with the “top surface” of “document ID = 5” by “| 0.01-0.25 | + | 0.26-0.02 | + | 0.26-0.47 | + | 0.10-0.28 | + | 0.18-0.18 | + | 0.08-0.24 | + | 0.24-0.09 | + | 0.33-0.12 | = 1.39 ”. Further, for example, the insertion position determination unit 102 associates with the “lowermost surface” of “document ID = 5”, the image C1f for the image E2b (that is, the lowermost image of “document ID = 5” in the first reading). The distance value of “| 0.32-0.01 | + | 0.11-0.26 | + | 0.07-0.25 | + | 0.01-0.11 | + | 0.20-0.18 | + | 0.23-0.07 | + | 0.11-0.24 | + | 0.16 -0.32 | = 1.21 ” Further, for example, the insertion position determination unit 102 sets the distance value of the image C1b ′ with respect to the image E2b to “| 0.32-0.25 | + | 0.11-0.02 | + |” in association with “bottom surface” of “document ID = 5”. 0.07-0.47 | + | 0.01-0.28 | + | 0.20-0.18 | + | 0.23-0.24 | + | 0.11-0.09 | + | 0.16-0.12 | = 0.92 ”.

The distance values of the images C2f ′, C2b ′, C3f ′, C3b ′ are calculated in the same manner (FIG. 22).

Further, since the document IDs acquired in step ST49 are “3” and “5”, in step ST53, the insertion position determination unit 102 determines “document ID = 3” and “document ID = 5” in FIG. ", The minimum distance value is specified for each surface type. For example, for the “top surface” of “document ID = 3”, the insertion position determination unit 102 sets the minimum distance value from among a plurality of distance values 0.04, 1.41, 1.22, 0.57, 0.79, and 1.07 to the image C1f. Specify “0.04” of '. In addition, for example, the insertion position determination unit 102 sets the minimum distance value from the plurality of distance values 1.04, 1.05, 1.12, 0.99, 0.75, and 0.05 for the “bottom surface” of “document ID = 3”. It is specified as “0.05” of C3b ′. For example, the insertion position determination unit 102 sets the minimum distance value for the “top surface” of “document ID = 5” from the plurality of distance values 0.04, 1.39, 1.24, 0.57, 0.79, and 1.03. C1f ′ is specified as “0.04”. For example, the insertion position determination unit 102 sets the minimum distance value from the plurality of distance values 1.21, 0.92, 0.75, 1.20, 0.60, and 0.98 for “document ID = 5” as the image. It is specified as “0.60” of C3f ′.

Further, in step ST53, the insertion position determination unit 102 adds the minimum distance value of the uppermost surface and the minimum distance value of the lowermost surface in each of “document ID = 3” and “document ID = 5” in FIG. To do. Hereinafter, the total value of the minimum distance value on the top surface and the minimum distance value on the bottom surface may be referred to as a “total distance value”. Therefore, for example, in FIG. 22, the total distance value of “document ID = 3” is calculated as “0.04 + 0.05 = 0.09”, and the total distance value of “document ID = 5” is calculated as “0.04 + 0.60 = 0.64”. Is done.

For example, if the threshold value TH used in step ST55 is “0.50”, in step ST55, the insertion position determination unit 102 calculates the total distance among the total distance values “0.09” and “0.64” calculated in step ST53. It is determined that the value “0.09” is less than the threshold value TH.

In step ST57, since the total distance value “0.09” that is less than the threshold TH corresponds to “document ID = 3” (FIG. 22), the insertion position determination unit 102 determines the second document image C1f ′ of the document bundle BC. , C1b ′, C2f ′, C2b ′, C3f ′, and C3b ′ are determined as second document images to be stored in the image file of “document ID = 3”.

Here, by the processing of steps ST33 to ST37 at the time of the first reading, the “uppermost surface” of “document ID = 3” corresponds to the uppermost surface of the document bundle BC (that is, the reading surface of the image C1f). The “lowermost surface” of the document ID = 3 ”corresponds to the lowermost surface of the document bundle BC (that is, the reading surface of the image C3b). Therefore, in step ST57, determining a series of images from the image C1f ′ to the image C3b ′ as the second document image to be stored in the image file of “document ID = 3” is from the image C1f ′ to the image C3b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image C1f and the image C3b (FIG. 4) in the first document image.

When the processing of steps ST41 to ST47 (FIG. 13) is performed on the document bundle BE, as shown in FIG. 23, for example, the images E1f ′ and E1b ′ of the documents E1 and E2 forming the document bundle BE. , E2f ′, E2b ′, the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8. In FIG. 23, for example, in the image E1f ′, the pixel density of the region RE1 is “0.01”, the pixel density of the region RE2 is “0.25”, the pixel density of the region RE3 is “0.27”, and the pixel density of the region RE4 is “0.09”. Assume that the pixel density of the region RE5 is calculated as “0.17”, the pixel density of the region RE6 is “0.08”, the pixel density of the region RE7 is “0.24”, and the pixel density of the region RE8 is “0.33”. Further, for example, in the image E1b ′, the pixel density of the region RE1 is “0.54”, the pixel density of the region RE2 is “0.32”, the pixel density of the region RE3 is “0.10”, the pixel density of the region RE4 is “0.40”, and the region RE5 The pixel density of “0.68”, the pixel density of the region RE6 is “0.21”, the pixel density of the region RE7 is “0.55”, and the pixel density of the region RE8 is “0.77”. Similarly, in the images E2f 'and E2b', the pixel density is calculated for each of the regions RE1, RE2, RE3, RE4, RE5, RE6, RE7, and RE8.

Therefore, in step ST51, the insertion position determination unit 102 compares the first reading pixel density (FIG. 20) with the second reading pixel density of the document bundle BE (FIG. 23), and the document acquired in step ST49. For each ID, on both the uppermost surface and the lowermost surface, the sum of the absolute values of the differences between the first reading pixel density and the second reading pixel density of the document bundle BE over the regions RE1 to RE8 is calculated as the first document. A distance value between the image and the second original image of the original bundle BE is calculated. Here, the document IDs acquired in step ST49 with reference to the document table (FIG. 10) are “3” and “5”.

Therefore, for example, as illustrated in FIG. 24, the insertion position determination unit 102 associates with the “top surface” of “document ID = 3” in accordance with the pixel density table (FIG. 20), so The distance value of the image E1f ′ with respect to “the top image of“ document ID = 3 ”in one reading) is set to“ | 0.01-0.01 | + | 0.27-0.25 | + | 0.24-0.27 | + | 0.12-0.09 | + | 0.19 -0.17 | + | 0.07-0.08 | + | 0.24-0.24 | + | 0.32-0.33 | = 0.12 ”. For example, the insertion position determination unit 102 associates the distance value of the image E1b 'with the image C1f in association with the "top surface" of "document ID = 3" as "| 0.01-0.54 | + | 0.27-0.32 | + | 0.24-0.10 | + | 0.12-0.40 | + | 0.19-0.68 | + | 0.07-0.21 | + | 0.24-0.55 | + | 0.32-0.77 | = 2.39 ”. In addition, for example, the insertion position determination unit 102 associates with the “lowermost surface” of “document ID = 3”, the image E1f for the image C3b (that is, the lowermost image of “document ID = 3” in the first reading). The distance value of “| 0.07-0.01 | + | 0.01-0.25 | + | 0.27-0.27 | + | 0.01-0.09 | + | 0.25-0.17 | + | 0.30-0.08 | + | 0.24-0.24 | + | 0.01 -0.33 | = 1.00 ” Also, for example, the insertion position determination unit 102 associates the distance value of the image E1b ′ with respect to the image C3b “| 0.07−0.54 | + | 0.01−0.32 | + | in association with the“ bottom surface ”of“ document ID = 3 ”. 0.27-0.10 | + | 0.01-0.40 | + | 0.25-0.68 | + | 0.30-0.21 | + | 0.24-0.55 | + | 0.01-0.77 | = 2.93 ”.

Further, for example, the insertion position determination unit 102 associates with the “uppermost surface” of “document ID = 5”, the image E1f for the image E1f (that is, the uppermost image of “document ID = 5” in the first reading). The distance value of “| 0.01-0.01 | + | 0.26-0.25 | + | 0.26-0.27 | + | 0.10-0.09 | + | 0.18-0.17 | + | 0.08-0.08 | + | 0.24-0.24 | + | 0.33 -0.33 | = 0.04 ”. Further, for example, the insertion position determination unit 102 associates the distance value of the image E1b ′ with the image E1f in association with the “top surface” of “document ID = 5” by “| 0.01-0.54 | + | 0.26-0.32 | + | 0.26-0.10 | + | 0.10-0.40 | + | 0.18-0.68 | + | 0.08-0.21 | + | 0.24-0.55 | + | 0.33-0.77 | = 2.43 ". Further, for example, the insertion position determination unit 102 associates with the “lowermost surface” of “document ID = 5”, and the image E1f for the image E2b (that is, the lowermost image of “document ID = 5” in the first reading). The distance value of “| 0.32-0.01 | + | 0.11-0.25 | + | 0.07-0.27 | + | 0.01-0.09 | + | 0.20-0.17 | + | 0.23-0.08 | + | 0.11-0.24 | + | 0.16 -0.33 | = 1.21 ” Further, for example, the insertion position determination unit 102 associates the distance value of the image E1b ′ with respect to the image E2b “| 0.32-0.54 | + | 0.11-0.32 | + |” in association with “bottom surface” of “document ID = 5”. 0.07-0.10 | + | 0.01-0.40 | + | 0.20-0.68 | + | 0.23-0.21 | + | 0.11-0.55 | + | 0.16-0.77 | = 2.40 ”.

The distance value between the images E2f ′ and E2b ′ is calculated in the same manner (FIG. 24).

Further, since the document IDs acquired in step ST49 are “3” and “5”, in step ST53, the insertion position determination unit 102 determines “document ID = 3” and “document ID = 5” in FIG. ", The minimum distance value is specified for each surface type. For example, for the “top surface” of “document ID = 3”, the insertion position determination unit 102 sets the minimum distance value from “0.12, 2, 39, 2.07, and 1.20” that are the plurality of distance values to “0.12” of the image E1f ′. " Further, for example, the insertion position determination unit 102 sets the minimum distance value among the plurality of distance values 1.00, 2.93, 2.11 and 0.94 for the “bottom surface” of “document ID = 3” to “ Specify 0.94 ”. Further, for example, the insertion position determination unit 102 sets the minimum distance value among the plurality of distance values 0.04, 2.43, 2.09, and 1.20 for the “top surface” of “document ID = 5” to “ Specify 0.04 ”. In addition, for example, the insertion position determination unit 102 sets the minimum distance value for “bottom surface” of “document ID = 5” from the plurality of distance values 1.21, 2.40, 2.10, and 0.05 in the image E2b ′. Specify 0.05 ”.

Further, in step ST53, the insertion position determination unit 102 adds the minimum distance value on the uppermost surface and the minimum distance value on the lowermost surface in each of “document ID = 3” and “document ID = 5” in FIG. To do. Therefore, for example, in FIG. 24, the total distance value of “document ID = 3” is calculated as “0.12 + 0.94 = 1.06”, and the total distance value of “document ID = 5” is calculated as “0.04 + 0.05 = 0.09”. Is done.

For example, if the threshold value TH used in step ST55 is “0.50”, in step ST55, the insertion position determination unit 102 calculates the total distance among the total distance values “1.06” and “0.09” calculated in step ST53. It is determined that the value “0.09” is less than the threshold value TH.

In step ST57, since the total distance value “0.09” that is less than the threshold TH corresponds to “document ID = 5” (FIG. 24), the insertion position determination unit 102 determines the second original image E1f ′ of the original bundle BE. , E1b ′, E2f ′, and E2b ′ are determined as second document images to be stored in the image file of “document ID = 5”.

Here, by the processing of steps ST33 to ST37 at the time of the first reading, the “uppermost surface” of “document ID = 5” corresponds to the uppermost surface of the document bundle BE (that is, the reading surface of the image E1f). The “lowermost surface” of the document ID = 5 ”corresponds to the lowermost surface of the document bundle BE (that is, the reading surface of the image E2b). Therefore, the determination of the series of images from the image E1f ′ to the image E2b ′ as the second document image to be stored in the image file with “document ID = 5” in step ST57 is from the image E1f ′ to the image E2b for the first document image. This corresponds to determining that the insertion position of the series of second document images up to 'is the position of the image E1f and the image E2b (FIG. 4) in the first document image.

As described above, in the third embodiment, the insertion position determination unit 102 compares the top image of the document bundle BC or BE with the second document image in the first document image, and the first document image. The insertion position of the second original image with respect to the first original image is determined based on the comparison result of both the lowermost image of the original bundle BC and BE and the comparison result of each of the second original image.

In this way, the insertion position of the second document image with respect to the first document image can be determined more accurately than in the first and second embodiments.

[Example 4]
The storage unit 101 is realized by hardware, for example, as a memory. Examples of the memory include a RAM (Random Access Memory) such as SDRAM (Synchronous Dynamic Random Access Memory), a ROM (Read Only Memory), a flash memory, and the like.

The insertion position determination unit 102, the image insertion unit 103, the file generation unit 104, and the control unit 11 can be realized as hardware by, for example, a processor. Examples of the processor include a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and an FPGA (Field Programmable Gate Array). The insertion position determination unit 102, the image insertion unit 103, the file generation unit 104, and the control unit 11 may be realized by an LSI (Large Scale Integrated circuit) including a processor and peripheral circuits. Furthermore, the insertion position determination unit 102, the image insertion unit 103, the file generation unit 104, and the control unit 11 may be realized using a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), or the like.

All or a part of each process in the above description in the image processing system 1 may be realized by causing a processor included in the image processing system 1 to execute a program corresponding to each process. For example, a program corresponding to each process in the above description may be stored in the memory, and the program may be read from the memory by the processor and executed. The program is stored in a program server connected to the image processing system 1 via an arbitrary network, downloaded from the program server to the image processing system 1 and executed, or recorded by the image processing system 1. The program may be stored in a medium, read from the recording medium, and executed. Examples of the recording medium readable by the image processing system 1 include a memory card, USB memory, SD card, flexible disk, magneto-optical disk, CD-ROM, DVD, and Blu-ray (registered trademark) disk. A portable storage medium is included. The program is a data processing method described in an arbitrary language or an arbitrary description method, and may be in any format such as a source code or a binary code. In addition, the program is not necessarily limited to a single configuration, and the functions are achieved in cooperation with a plurality of modules, a configuration that is distributed as a plurality of libraries, and a separate program represented by the OS. Including things.

The specific form of distribution / integration of the image processing system 1 is not limited to the one shown in the figure, and the whole or a part of the three-dimensional shape measuring apparatus 2 may be arbitrarily selected according to various additions or according to functional loads. It can be configured to be functionally or physically distributed and integrated in units.

[Other embodiments]
[1] In the above embodiment, the pixel density is adopted as an example of the image feature amount. However, the feature amount applicable to the disclosed technology is not limited to the pixel density. For example, an average luminance value, a pixel spatial dispersion value, a luminance dispersion value, or the like in each of the regions RE1 to RE8 may be used as the image feature amount.

[2] In the above embodiment, as an example, the case where the entire area of the document is divided into the areas RE1 to RE8 has been described. However, the number of area divisions is not limited to eight. For example, the entire area of the document may be divided into 4, 16, or 32.

[3] The predetermined format of the image file generated by the file generation unit 104 is not limited to PDF, and may be, for example, JPEG (Joint Photographic Experts Group), TIFF (Tagged Image File Format), or the like.

[4] When the file generation unit 104 generates an image file, if the image to be included in the image file includes a blank image (hereinafter sometimes referred to as “blank image”), the file The generation unit 104 may create an image file after removing the blank image.

[5] In the above embodiment, the case where the original bundle to be read in the second reading is a removal failure original bundle has been described as an example. However, the original bundle to be read in the second reading is not limited to the unremoved original bundle. For example, a document bundle to be read in the second reading is a document bundle that has been double-fed (multifeed) in the first reading although it is not bound by the staple SP at the time of the first reading. It may be.

DESCRIPTION OF SYMBOLS 1 Image processing system 10 Image processing apparatus 11 Control part 12 Document reading part 13 Document bundle detection part 14 Staple detection part 15 Staple removal part 16 Offset discharge part 101 Storage part 102 Insertion position judgment part 103 Image insertion part 104 File generation part

Claims (8)

1. A storage for storing a plurality of first images read by a first reading of a group of documents formed of a plurality of documents, the document group including a bundle of documents as a bundle of a plurality of documents. And
   Inserting a plurality of second images, which are images of each document of the document bundle, read by a second reading subsequent to the first reading into the plurality of first images A determination unit for determining a position;
   An insertion unit for inserting the plurality of second images between the plurality of first images according to the determined insertion position;
   An image processing apparatus comprising:
2. The determination unit includes an uppermost image of the bundle of documents among the plurality of first images read by the first reading, and each of the plurality of second images read by the second reading. Each of the plurality of first images read by the first reading and the plurality of second images read by the second reading. The insertion position is determined based on one of the comparison results of
   The image processing apparatus according to claim 1.
3. The determination unit includes an uppermost image of the bundle of documents among the plurality of first images read by the first reading, and each of the plurality of second images read by the second reading. And the lowermost image of the bundle of documents among the plurality of first images read by the first reading and the plurality of second images read by the second reading. And determining the insertion position based on both comparison results of
   The image processing apparatus according to claim 1.
4. A generating unit that generates an image file for each bundle of documents according to a predetermined file format;
   The image processing apparatus according to claim 1, further comprising:
5. A detection unit for detecting the document bundle in the document group;
   A reading unit that performs the second reading on the document bundle subsequent to the first reading on the document group when the document bundle is detected in the document group;
   An image processing apparatus according to claim 1;
   An image processing system comprising:
6. A discharge unit that offsets and discharges the bundle of documents in the document group;
   The image processing system according to claim 5, further comprising:
7. Storing a plurality of first images read by a first reading of a group of documents formed of a plurality of documents, the document group including a bundle of documents as a bundle of a plurality of documents;
   Inserting a plurality of second images, which are images of each document of the document bundle, read by a second reading subsequent to the first reading into the plurality of first images Determine the position,
   According to the determined insertion position, the plurality of second images are inserted between the plurality of first images.
   Image processing method.
8. A plurality of first images read by a first reading on a group of documents formed of a plurality of documents, each of which includes a bundle of documents that is a bundle of a plurality of documents, are stored in a storage unit. Remember,
   Inserting a plurality of second images, which are images of each document of the document bundle, read by a second reading subsequent to the first reading into the plurality of first images Determine the position,
   According to the determined insertion position, the plurality of second images are inserted between the plurality of first images.
   An image processing program for causing a processor to execute processing.

Images