JP2016163142A - Program and image processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a program and image processing apparatus for automatically executing image processing, in accordance with a shape of a document to be scanned while being folded into two and held between carrier sheets, upon image data corresponding to the document.SOLUTION: First image data and second image data corresponding to a first region and a second region of the document are acquired. A first size of a first side of the document with characters recorded thereon is longer than a second size of a second side and the document is folded into two and held between the carrier sheets. The first region of the document is a region to be a side of a first face of the carrier sheet, and the second region of the document is a region to be a side of a second face of the carrier sheet. A short-side size of the first image data corresponding to the second size is acquired. If the short-side size is smaller than a reference size, a rotation amount is determined in accordance with at least a recognition result of OCR processing upon the first image data and rotation processing is executed. If the short-side size is not smaller than the reference size, the processing is not executed.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a program and an image processing apparatus that perform image processing on image data corresponding to a document that is sandwiched between carrier sheets and scanned by a scanner device.

  Scanning using a carrier sheet and a double-sided simultaneous scanning scanner device has been put into practical use. Techniques relating to image processing for image data generated by scanning using a carrier sheet and a double-sided simultaneous reading scanner device have been proposed. For example, Patent Document 1 discloses a technique for scanning both front and back sides of a document, generating front surface image data and back surface image data, and synthesizing both image data. When combining the front surface image data and the back surface image data, the document is folded in half and sandwiched between carrier sheets. The compositing instruction mark is drawn on the carrier sheet at a position other than the position where the folded document is sandwiched. When the compositing instruction mark is detected in the image processing unit, the front cut-out image and the back cut-out image cut out from the predetermined positions of the front image data and the back image data are combined into one piece of image data. When the compositing instruction mark is not detected, the front surface image data and the back surface image data are independent image data. According to the technique of Patent Document 1, a setting operation for distinguishing between a normal double-sided original reading mode and a mode in which an image is read and synthesized by folding in half is unnecessary, and there is no need to switch modes.

  Patent Document 2 discloses a reading device. In the reading device, an image on the document sheet sandwiched between the carrier sheets being conveyed is read. The document sheet is a long document such as a receipt. In the reading device, image data is generated from the images on the front and back surfaces of the carrier sheet. In the reading device, a part corresponding to the receipt sandwiched between the carrier sheets is detected as partial data from the image data of the front and back surfaces. It is determined whether or not the receipt is folded. When the receipt is folded, the partial data on the front surface and the partial data on the back surface are combined into one image data. At this time, the composition is performed such that the position of the crease in the receipt corresponding to the partial data on the front surface matches the position of the crease in the receipt corresponding to the partial data on the back surface. When the receipt is not folded, the above-described synthesis is not executed.

JP 2005-260387 A JP 2014-216998 A

  Various documents may be sandwiched between carrier sheets, and both sides of the carrier sheet may be scanned. An original document sandwiched between carrier sheets may be folded in half. For example, a long document such as a receipt may be folded in two, or an A3 size document may be folded in two. Therefore, in the scan using the carrier sheet, it is necessary to provide an image processing mode corresponding to the document sandwiched between the carrier sheets. It is desirable that the image processing mode corresponding to the document sandwiched between the carrier sheets is automatically set in accordance with the scan. In other words, the processing procedure in which the user manually sets the image processing mode is not user-friendly.

  The present invention provides a program and an image processing apparatus in which image processing for image data corresponding to a document to be scanned by being sandwiched between carrier sheets in a folded state is automatically executed according to the shape of the document. The purpose is to do.

  One aspect of the present invention is a program that can be executed by a computer that controls an image processing apparatus that performs image processing on image data corresponding to a document scanned by a scanner apparatus. The long original document whose first side size is longer than the second size of the second side is the side of the first surface of the carrier sheet that is folded and sandwiched with the fold line along the second side. First image data corresponding to the first region of the document, and second image data corresponding to the second region of the document on the second surface side opposite to the first surface of the carrier sheet; , Second acquisition means for acquiring a short side size corresponding to the second size of the document and corresponding to a short side of the first image data, and the second acquisition means. Acquired by First determination means for determining whether the short side size is smaller than a reference size, and if the short side size is smaller than the reference size, at least according to the recognition result of the OCR processing for the first image data, Performing a determination process for determining a rotation amount of the first image data, performing a rotation process according to the rotation amount determined by the determination process on at least the first image data, and the short side size Is a program that functions as a deforming means that does not execute the determination process and the rotation process when it is not smaller than the reference size.

  According to this, execution or non-execution of the determination process and the rotation process can be controlled according to the short side size corresponding to the short side of the first image data. When the short side size corresponding to the short side of the first image data is smaller than the reference size, the rotation amount can be determined, and the rotation process according to the determined rotation amount can be executed. In the determination process, the rotation amount can be determined according to the recognition result of the OCR process. When the short side size corresponding to the short side of the first image data is larger than the reference size, the rotation amount is not determined. Therefore, the rotation process is not executed.

  When the short side size is smaller than the reference size, the program causes the first image on each short side of the first image data and the second image data in a state corresponding to the original. When the first composite data is generated by combining the data and the second image data, and the short side size is not smaller than the reference size, the first image data and the second image data in a state corresponding to the document. You may make it function as a synthetic | combination means which produces | generates the 2nd synthetic | combination data which synthesize | combined said 1st image data and said 2nd image data in each long side of image data. According to this, the first composite data corresponding to the original can be generated for the folded document whose short side size corresponding to the short side of the first image data is smaller than the reference size. For a document folded in half with a short side size larger than the reference size corresponding to the short side of the first image data, second composite data corresponding to this document can be generated.

  In this program, the deformation means includes the first image data and the second image data according to the recognition result of the OCR process for the first composite data including the first image data and the second image data. A function of executing the determination process for determining the rotation amount of the first composite data, and executing the rotation process on the first composite data including the first image data and the second image data; May be included. According to this, it is possible to determine the rotation amount for the first composite data corresponding to the document that is not folded in half, and to execute the rotation process according to the determined rotation amount.

  In this program, the combining means does not combine the first image data and the second image data when the second image data is data corresponding to a region not including predetermined information including characters. And when the short side size is smaller than the reference size and the first image data and the second image data are not combined, the deformation means follows the recognition result of the OCR process for the first image data. A function of executing the determination process for determining the rotation amount of the first image data and executing the rotation process on the first image data may be included. According to this, it is possible to prevent generation of combined data in which the second image data corresponding to the area not including characters is combined with the first image data. A rotation amount can be determined for the first image data, and a rotation process according to the determined rotation amount can be executed.

  In this program, when the short side size is smaller than the reference size, the short side size acquired by the second acquisition unit and used as the determination target of the determination unit is scanned in the scanner device. And the second determining means for determining whether the size of the first image data in the horizontal direction coincides with the main scanning direction of the first image data. In the case of the size of the direction, in the determination process, as the amount of rotation, at least an amount in which the first image data is vertically reversed in the vertical direction and unrotated are determined, and the short side size is When the size is not the horizontal size of the first image data, in the determination process, as the rotation amount, at least the horizontal direction of the first image data Determining an amount of the vertical direction, including functional, may be. According to this, the rotation amount can be appropriately determined according to the direction of the short side size corresponding to the short side of the first image data.

  In this program, the deformation means has a vertical direction of the first image data in which the short side size is a horizontal size of the first image data and coincides with a sub-scanning direction orthogonal to the main scanning direction. The first parameter value indicating the reliability of the recognition result of the OCR process from the lower side to the upper side of the second indicates the reliability of the recognition result of the OCR process from the upper side to the lower side in the vertical direction of the first image data. When larger than a parameter value, as the amount of rotation, at least determine the amount that the vertical direction of the first image data is upside down, the short side size is the horizontal size of the first image data, When the first parameter value is smaller than the second parameter value, non-rotation is determined as the rotation amount, and the short side size is not the horizontal size of the first image data. The third parameter value indicating the reliability of the recognition result of the first image data from the left side to the right side in the horizontal direction is the recognition result of the first image data from the right side to the left side in the horizontal direction. When the rotation amount is larger than a fourth parameter value indicating a reliability of the first image data, at least a clockwise amount in which the horizontal direction of the first image data is a vertical direction is determined, and the short side size is the first image. If the third parameter value is smaller than the fourth parameter value instead of the horizontal size of the data, the amount of rotation is determined as a counterclockwise amount in which at least the horizontal direction of the first image data is the vertical direction. You may do it. According to this, when the short side size is the horizontal size, the rotation amount can be appropriately determined according to the magnitude relationship between the first parameter value and the second parameter value. When the short side size is not the horizontal size, the rotation amount can be appropriately determined according to the magnitude relationship between the third parameter value and the fourth parameter value. For example, the number of recognized characters can be used as the parameter indicating the reliability of the recognition result of the OCR processing in order to compare the levels.

  Another aspect of the present invention is an image processing device that performs image processing on image data corresponding to a document scanned by a scanner device, wherein characters are recorded and the first size of the first side is the second of the second side. The first document corresponding to the first region of the document, which is on the first surface side of the carrier sheet sandwiched by being folded in half with the fold line being along the second side, is longer than the size of the document. A first acquisition unit configured to acquire image data and second image data corresponding to a second region of the document on a second surface side opposite to the first surface of the carrier sheet; Second acquisition means for acquiring a short side size corresponding to the second size of the first image data and the short side size acquired by the second acquisition means is a reference. To determine whether it is smaller than the standard size When the short side size is smaller than the reference size, the determination unit executes a determination process for determining at least the rotation amount of the first image data according to at least a recognition result of the OCR process for the first image data, A rotation process according to the rotation amount determined by the determination process is performed on the first image data, and if the short side size is not smaller than the reference size, the determination process and the rotation process are performed. The image processing apparatus includes a deformation unit.

  According to this, the same operation as the corresponding program is realized. This image processing apparatus can also be specified as an image processing apparatus that realizes other functions in the same manner as the above-described program.

  According to the present invention, a program and an image processing apparatus that automatically perform image processing on image data corresponding to a document that is sandwiched between carrier sheets and scanned in a folded state according to the shape of the document Can be obtained.

1 is a diagram illustrating an example of an image processing system. It is a figure regarding the receipt and carrier sheet as an example of a reading medium. The upper stage shows a state in which the receipt is sandwiched between carrier sheets. The lower left side shows the first surface of the carrier sheet in which the receipt is sandwiched. The lower right side shows the second surface of the carrier sheet with the receipt sandwiched therebetween. It is a figure which shows the state by which a receipt is folded in half. It is a flowchart of the 1st part of image processing. It is a flowchart of the 2nd part of image processing. It is a flowchart of the 3rd part of image processing.

  Embodiments for carrying out the present invention will be described with reference to the drawings. The present invention is not limited to the configurations described below, and various configurations can be employed in the same technical idea. For example, some of the configurations shown below may be omitted or replaced with other configurations. Other configurations may be included.

<Image processing system>
The image processing system 10 will be described with reference to FIGS. The image processing system 10 includes an image processing device 20 and a scanner device 30. In the image processing system 10, the image processing device 20 and the scanner device 30 are connected by a communication cable 11. Data communication is performed between the image processing apparatus 20 and the scanner apparatus 30 via the communication cable 11. For example, image data generated by the scanner device 30 is transmitted to the image processing device 20 via the communication cable 11. The image data is data corresponding to the first surface or the second surface of the reading medium. In the image processing device 20, the above-described image data transmitted from the scanner device 30 is received, and the image processing device 20 acquires this.

  In the image processing device 20, image processing is executed on the acquired image data. The image processing apparatus 20 is an information processing apparatus such as a personal computer, for example. Image processing executed by the image processing apparatus 20 will be described later. In the scanner device 30, the reading medium is scanned, and image data corresponding to the scanned reading medium is generated. The scanner device 30 can collectively scan the first surface and the second surface of the reading medium by one scanning operation. In this case, the scanner device 30 generates image data corresponding to the first surface of the reading medium and image data corresponding to the second surface of the reading medium.

  In the scanner device 30, a scanning operation using the carrier sheet 40 can be performed. As shown in FIG. 2, the document is sandwiched between the carrier sheets 40. Various information is recorded on the manuscript. The information recorded on the document includes at least characters. For example, examples of the manuscript include A3 size materials (not shown) and receipts. The receipt is delivered from a store clerk at the time of payment when a predetermined article is purchased or a predetermined service is received. As shown in FIG. 3, on the receipt, for example, characters such as a purchased article name and its price are recorded. In the receipt, the characters are recorded along the short side. The document sandwiched between the carrier sheets 40 may be folded in half (see FIGS. 2 and 3). In the embodiment, it is assumed that the reading medium is the carrier sheet 40 in which one or a plurality of documents are sandwiched. The carrier sheet 40 shown in FIG. 2 is simply illustrated. The connection between the two transparent sheets constituting the carrier sheet 40 may be different from that shown in FIG. As the carrier sheet 40, a known carrier sheet can be adopted.

<Image processing device>
As shown in FIG. 1, the image processing apparatus 20 includes a CPU 21, a storage device 22, a RAM 23, a display unit 24, an operation unit 25, and a communication unit 26. These units 21 to 26 are communicably connected. CPU21 performs arithmetic processing. The storage device 22 is configured by a computer-readable storage medium. For example, the storage device 22 is configured by a hard disk and / or a flash memory. In addition, the storage device 22 may include a ROM. Various programs are stored in the storage device 22. For example, the storage device 22 stores an OS (Operating System) and various applications. The application stored in the storage device 22 includes a program for image processing (see FIGS. 4 to 6) described later, a labeling process (see S15 in FIG. 4), and an OCR process (see FIG. 5). , S43 of FIG. 6 or S51 of FIG. 6), rotation processing (see S59 of FIG. 6), and inclination correction processing (see S61 of FIG. 6).

  For example, the application is installed in the storage device 22 in advance. The prior installation is performed, for example, by reading a program stored in a computer-readable storage medium such as a semiconductor memory by a reading unit (not shown) of the image processing apparatus 20. For example, when the image processing apparatus 20 includes an optical drive (not shown), the prior installation may be performed by reading a program stored in the optical medium by the optical drive. In addition, when the image processing apparatus 20 is connected to a network (not shown) such as the Internet, the prior installation is stored in a computer-readable storage medium such as a hard disk of a server apparatus connected to the network. The programmed program may be received by the communication unit 26 as a transmission signal. Which form is used is appropriately determined in consideration of various circumstances. The computer-readable storage medium may be a non-transitory storage medium that does not include a temporary storage medium (for example, a transmission signal). The non-temporary storage medium may be any storage medium that can store information regardless of the period in which the information is stored.

  The RAM 23 is a storage area used when the CPU 21 executes various programs. In the RAM 23, predetermined data and information used in the process are stored in a predetermined storage area during the execution of the process. In the image processing apparatus 20, the CPU 21 controls the image processing apparatus 20 by executing the OS stored in the storage device 22 and the above-described programs (see FIGS. 4 to 6). Thereby, in the image processing apparatus 20, various functional units are realized.

  The display unit 24 is, for example, a liquid crystal display. Various information is displayed on the display unit 24. For example, the display unit 24 displays an image corresponding to the first synthesized data or the second synthesized data synthesized by the image processing shown in FIGS. The operation unit 25 receives inputs such as various instructions for the image processing apparatus 20. For example, the operation unit 25 receives an input of a scan operation start instruction. The operation unit 25 includes, for example, a keyboard and a mouse. Although details are omitted, the process of generating operation information corresponding to each operation on the keyboard and mouse is, for example, a technique adopted in a known personal computer, and is also adopted in the image processing apparatus 20.

  The communication cable 11 is connected to the communication unit 26. The communication unit 26 performs data communication with the scanner device 30 via the communication cable 11. For example, the image processing device 20 receives image data transmitted from the scanner device 30 via the communication unit 26. The communication unit 26 is an interface circuit that conforms to, for example, the USB (Universal Serial Bus) standard or the Ethernet (registered trademark) standard. It is assumed that the communication unit 26 is an interface circuit that conforms to the Ethernet (registered trademark) standard. In this case, the connection with the scanner device 30 may be either a wired connection or a wireless connection.

  The image processing apparatus 20 is different from a known image processing apparatus in that the image processing program shown in FIGS. 4 to 6 is stored in the storage device 22. However, the image processing apparatus 20 is an information processing apparatus similar to a known image processing apparatus in hardware. Therefore, although the description is omitted above, the image processing apparatus 20 includes a configuration included in a known image processing apparatus in addition to the above-described units 21 to 27.

<Scanner device>
As shown in FIG. 1, the scanner device 30 includes a CPU 31, a storage device 32, a RAM 33, a transport unit 34, a first image sensor 35, a second image sensor 36, and a communication unit 37. These units 31 to 37 are communicably connected. CPU31 performs arithmetic processing. The storage device 32 is configured by a computer-readable storage medium. For example, the storage device 32 is configured by a flash memory. Various programs are stored in the storage device 32. For example, the program stored in the storage device 32 includes a program that generates image data in accordance with a scanning operation.

  The RAM 33 is a storage area used when the CPU 31 executes various programs. In the RAM 33, predetermined data used in the processing is stored in a predetermined storage area during the execution of the processing. For example, the generated image data is stored in the RAM 33. In the scanner device 30, the CPU 31 controls the scanner device 30 by executing a program stored in the storage device 32. Thereby, various functional means are realized in the scanner device 30.

  The transport unit 34 transports the reading medium set in the scanner device 30. As described above, the reading medium includes, for example, the carrier sheet 40 in which one or more originals are sandwiched. The reading medium may be a document itself. The first image sensor 35 and the second image sensor 36 are CIS type image sensors. However, the first image sensor 35 and the second image sensor 36 may be CCD image sensors. In the scanner device 30, the two first image sensors 35 and the second image sensor 36 facing each other realize double-sided scanning in one scanning operation. In the embodiment, the surface of the carrier sheet 40 scanned by the first image sensor 35 is referred to as a “first surface”. The surface of the carrier sheet 40 scanned by the second image sensor 36 is referred to as a “second surface”.

  A case where the size that can be scanned by the scanner device 30 is set to A4 vertical size will be described as an example. The A4 vertical size is 297 mm. In this case, in the first image sensor 35 and the second image sensor 36, the maximum scan size in the main scanning direction is 210 mm. In the scanner device 30, a carrier sheet 40 having a size corresponding to the A4 size is used. In the scanner device 30, the carrier sheet 40 is conveyed along the longitudinal direction of the carrier sheet 40. The longitudinal direction of the carrier sheet 40 coincides with the vertical direction of A4 size. The communication unit 37 is connected to the communication cable 11. The communication unit 37 performs data communication with the image processing apparatus 20 via the communication cable 11. For example, in the scanner device 30, image data is transmitted from the communication unit 37 to the image processing device 20. The communication unit 37 has the same configuration as that of the communication unit 26. Therefore, the other description regarding the communication part 37 is abbreviate | omitted.

  The scanner device 30 is a known scanner device in terms of hardware. Therefore, although the description is omitted above, the scanner device 30 includes a configuration included in a known scanner device in addition to the above-described units 31 to 37.

<Image processing>
Image processing executed by the image processing apparatus 20 will be described with reference to FIGS. In the description of the image processing, it is assumed that the document sandwiched between the carrier sheets 40 is a receipt, a postcard, and an A3 size document folded in half. The receipt is a long document in which the first size on the long side is longer than the second size on the short side. Receipts are intended for those whose long side is longer than twice the size of the short side (see FIG. 3). The receipt may be sandwiched between the carrier sheets 40 without being folded in half. The image processing is started when an instruction to start the process is received by the operation unit 25 and the CPU 21 acquires the start instruction from the operation unit 25. When inputting the start instruction, the user sets the carrier sheet 40 with the document sandwiched in the scanner device 30. At this time, the number of documents sandwiched between the carrier sheets 40 is one or more. It is assumed that the number of carrier sheets 40 set in the scanner device 30 is one.

  The CPU 21 that has acquired the start instruction initializes the counter to “0” (S10). The counter counts the number of times processing of S19 to S63 (see FIG. 6) has been executed. The count number counted by the counter is stored in the RAM 23. Subsequently, the CPU 21 controls transmission of a start instruction (S11). That is, the CPU 21 outputs a start instruction transmission command to the communication unit 26. The transmission destination of the start instruction is the scanner device 30. Accordingly, a start instruction is transmitted from the communication unit 26 to the scanner device 30.

  In the scanner device 30, the CPU 31 acquires a start instruction via the communication unit 37 and starts a scanning operation. Accordingly, the conveyance unit 34 is driven, and the carrier sheet 40 set in the scanner device 30 is conveyed. In the scanner device 30, image data is generated according to the scanning operation. In one scan operation, the image data generated by the scanner device 30 is first surface data and second surface data. The first surface data is image data corresponding to the first surface of the carrier sheet 40 scanned by the first image sensor 35. The second surface data is image data corresponding to the second surface of the carrier sheet 40 scanned by the second image sensor 36. In the scanner device 30, the generated image data is transmitted from the communication unit 37 to the scanner device 30.

  In the image processing device 20, the first surface data and the second surface data from the scanner device 30 are received by the communication unit 26. CPU21 acquires 1st surface data and 2nd surface data via the communication part 26 (S13). The acquired first surface data and second surface data are stored in the RAM 23. Subsequently, the CPU 21 executes a labeling process (S15). The labeling process is executed on the first surface data and the second surface data. In the labeling process, the first image data, the second image data, and the number of labels are acquired. The number of labels corresponds to the number of documents sandwiched between the carrier sheets 40.

  In the labeling process for the first surface data, the first image data is acquired. The first image data is image data corresponding to the first region of the document on the first surface side of the carrier sheet 40 in each document sandwiched between the carrier sheets 40. For example, it is assumed that one receipt is sandwiched between carrier sheets 40 in a folded state as shown in FIG. 3 (see FIG. 2). In this case, as shown in the lower part of FIG. 2, the first image data is the entire area of the print surface of the receipt sandwiched between the carrier sheets 40 (see “<Before Bending (Whole Receipt)>” in FIG. 3). The image data corresponding to the first region of the receipt on the first surface side of the carrier sheet 40 (see “<First region>” in FIG. 3).

  In the labeling process for the second surface data, second image data is acquired. The second image data is image data corresponding to the second region of the document on the second surface side of the carrier sheet 40 in each document sandwiched between the carrier sheets 40. For example, it is assumed that one receipt is sandwiched between carrier sheets 40 in a folded state as shown in FIG. 3 (see FIG. 2). In this case, the second image data is the second region of the receipt on the second surface side of the carrier sheet 40 as shown in the lower part of FIG. 2 among the entire region of the printing surface of the receipt sandwiched between the carrier sheets 40. (Refer to “<second region>” in FIG. 3). The second region of the receipt shown in FIG. 3 is illustrated with reference to a state in which the receipt is folded in half at a crease L along the short side.

  The first image data and the second image data acquired by the labeling process are stored in the RAM 23. The first image data and the second image data are rectangular image data in which the horizontal and vertical sizes have predetermined values. In the labeling process, when the process is appropriately executed, a set number of first image data and second image data corresponding to the number of documents sandwiched between the carrier sheets 40 are acquired. For example, assume that one receipt is sandwiched between carrier sheets 40 (see FIG. 2). In the labeling process, a set of first image data and second image data is acquired. In this case, the CPU 21 acquires the number of labels “1”.

  Unlike the example described above, it is assumed that two receipts are sandwiched between carrier sheets 40. In the labeling process, two sets of first image data and second image data are acquired. In this case, the CPU 21 acquires the label number “2”. Furthermore, it is assumed that one receipt and one postcard are sandwiched between carrier sheets 40. In the labeling process, two sets of first image data and second image data are acquired. In this case, the CPU 21 acquires the label number “2”. A postcard is a document in which the second size on the short side is larger than the second size on the short side of the receipt. For example, the maximum size of the short side of the receipt is 80 mm, and the short side of the postcard is 100 mm. Furthermore, it is assumed that two postcards are sandwiched between the carrier sheets 40. In the labeling process, two sets of first image data and second image data are acquired. In this case, the CPU 21 acquires the label number “2”.

  In addition, it is assumed that the A3 size material is sandwiched between the carrier sheets 40 in a folded state. In the labeling process, a set of first image data and second image data is acquired. In this case, the CPU 21 acquires the number of labels “1”. That is, the CPU 21 sets the number that matches the number of sets of the first image data and the second image data acquired by the label processing as the number of labels. It is assumed that the labeling process is not properly executed and only the first image data is acquired. In this case, the CPU 21 sets the number of acquired first image data as the number of labels. The labeling process executed in S15 is a known process. In S15, a known labeling processing technique can be employed. Therefore, the details of the labeling process are omitted.

  After executing S15, the CPU 21 sets the acquired number of labels as the number of repetitions (S17). The number of repetitions is a number for repeating the processing of S19 to S67 (see FIG. 6). In image processing, a set of first image data and second image data corresponding to the same document are handled together. However, either the labeling process for the first surface data or the labeling process for the second surface data may not be executed properly. In such a case, in the image processing, the image data acquired by the appropriately performed labeling processing is handled alone. In the embodiment, it is assumed that the labeling process that may not be performed properly is a labeling process for the second surface data.

  CPU21 selects unprocessed 1st image data as a process target among the 1st image data acquired by the labeling process in S15 (S19). At this time, the CPU 21 increases the counter by one. Subsequently, the CPU 21 acquires the short side size of the short side of the first image data selected as the processing target (S21). It is assumed that a set of first image data and second image data has been acquired in the labeling process. One set of first image data and second image data is image data corresponding to one original. Therefore, the short side size of each short side is the same in one set of first image data and second image data.

  Next, the CPU 21 determines whether or not the short side size acquired in S19 is equal to or smaller than the reference size (S23). S23 is a process for specifying that the document sandwiched between the carrier sheets 40 is a receipt. Accordingly, the reference size is set in advance in consideration of the size of the short side of a known receipt that is commercially available. Here, in the known receipt, the size of the short side is standardized. Therefore, as the reference size, for example, the maximum dimension of the width standardized in a known receipt is set. For example, the reference size is set to 80 mm. The reference size may be that the carrier sheet 40 may be transported in an inclined state with respect to the main scanning direction and / or the receipt is inclined to the carrier sheet 40 (see “angle θ” in the lower part of FIG. 2). Considering the possibility of being caught, the dimension may be set larger than the maximum dimension described above. However, in this case, the reference size is, for example, a size that can be distinguished from a postcard or the like and can specify a receipt. The reference size is registered in the image processing program. When the short side size is equal to or smaller than the reference size (S23: Yes), the CPU 21 proceeds to S31 in FIG. In the embodiment, the document on which S23 is affirmed (see S23: Yes) is a receipt.

  When the short side size is not equal to or smaller than the reference size (S23: No), the CPU 21 determines whether the number of repetitions is set to “1” (S25). For example, assume that two postcards are sandwiched between carrier sheets 40. In this case, the number of labels is “2” corresponding to two postcards, and the number of repetitions “2” is set in S17. In addition, it is assumed that one receipt and one postcard are sandwiched in the carrier sheet 40, and the first image data corresponding to the postcard is selected as a processing target in S19. In this case, the number of labels is “2”, and the number of repetitions “2” is set in S17. If the number of repetitions is not “1” (S25: No), the CPU 21 proceeds to S63 in FIG.

  Suppose that the reading medium is a carrier sheet 40 in which an A3 size original folded in two is sandwiched. In this case, the number of labels is “1”, and the number of repetitions “1” is set in S17. When the number of repetitions is set to “1” (S25: Yes), the CPU 21 determines the long sides of the first image data and the second image data as the synthesis position (S27). The long sides of the first image data and the second image data determined as the synthesis position coincide with the fold when the A3 document is folded in half.

  Subsequently, the CPU 21 generates second composite data (S29). The second composite data is image data obtained by combining the first image data and the second image data in a state where the long sides of the first image data and the second image data determined in S27 are matched. The second composite data is stored in the RAM 23. In S27 and S29, a composition processing technique that has already been put into practical use in scanning using the carrier sheet 40 can be employed. Therefore, the other description regarding S27 and S29 is abbreviate | omitted. When the second image data is blank paper data, if S33 in FIG. 5 described later is negative (see S33: No), S29 in FIG. 5 is omitted as in the processing procedure in which S37 is not executed. May be. The blank data will be described later. Thereafter, the CPU 21 shifts the processing to S63 in FIG.

  In S <b> 31 of FIG. 5, the CPU 21 determines whether the second image data is stored in the RAM 23. The second image data is image data corresponding to the first image data selected as a processing target in S19 of FIG. This determination is a process corresponding to the case where the labeling process for the second surface data is not properly executed and the second image data is not stored in the RAM 23 in S15. When the second image data is not stored (S31: No), the CPU 21 proceeds to S39. When the second image data is stored (S33: Yes), the CPU 21 determines whether the second image data is blank paper data (S33). The blank page data is data that is determined to be a blank page area that does not include predetermined information including characters based on the color information of each pixel. For example, in the receipt sandwiched between the carrier sheets 40, it is assumed that nothing is recorded in the second area of the receipt on the second surface side of the carrier sheet 40. In such a case, the second image data is blank paper data. Even when the receipt is not folded in half and the side of the receipt on the second side of the carrier sheet 40 is a blank sheet, the second image data is blank sheet data. If the second image data is blank paper data (S33: Yes), the CPU 21 proceeds to S39.

  When the second image data is not blank paper data (S33: No), the CPU 21 determines each short side of the first image data and the second image data as the synthesis position (S35). Each short side of the first image data and the second image data determined as the synthesis position coincides with the crease L when the receipt is folded in half. The broken line with the symbol “P” shown in FIG. 3 clearly indicates the position of the receipt fold L, not the information recorded on the receipt. Subsequently, the CPU 21 generates first composite data (S37). The first combined data is image data obtained by combining the first image data and the second image data in a state where the short sides of the first image data and the second image data determined in S35 are matched. The first composite data is stored in the RAM 23. In S37, a known synthesis processing technique can be employed. Therefore, the details of the synthesis process are omitted. Thereafter, the CPU 21 shifts the processing to S39.

  In S39, the CPU 21 acquires the short side size of the first image data or the first composite data. When S31 is denied (see S31: No) or when S33 is affirmed (see S33: Yes), the CPU 21 acquires the short side size of the first image data. When S37 is executed, the CPU 21 acquires the short side size of the first composite data. The short side size of the first composite data matches the short side size of the first image data before composition. Therefore, the short side size of the first composite data can be substantially referred to as the short side size of the first image data before the composition. Subsequently, the CPU 21 determines whether the acquired short side size is the horizontal size of the first image data or the first composite data (S41). The horizontal direction of the first image data or the first composite data coincides with the main scanning direction of scanning in the scanner device 30 (see FIG. 2).

  If the short side size is not the horizontal size (S41: No), the CPU 21 proceeds to S51 in FIG. When the short side size is the horizontal size (S41: Yes), the CPU 21 executes the OCR process in which the reading pattern is set in the vertical direction (S43). In the OCR process in which the reading pattern is set in the vertical direction, the first image data or the first composite data is subjected to the OCR process from the lower side to the upper side in the vertical direction, and further, the OCR process is performed from the upper side to the lower side in the vertical direction. . Performing OCR processing from the lower side to the upper side in the vertical direction is equivalent to performing OCR processing from the upper side to the lower side in the vertical direction after rotating the first image data or the first composite data by 180 °. The vertical direction of the first image data or the first composite data matches the sub-scanning direction. The sub-scanning direction is a direction orthogonal to the main scanning direction and corresponds to the direction in which the carrier sheet 40 is conveyed by the conveyance unit 34. The CPU 21 acquires the first parameter value and the second parameter value from the OCR process in S43. The first parameter value is a parameter value indicating the reliability of the recognition result when the reading pattern is from the lower side to the upper side in the vertical direction. The second parameter value is a parameter value indicating the reliability of the recognition result when the reading pattern is changed from the upper side to the lower side in the vertical direction. The OCR process is a known process, and a known OCR process technique is employed in S43 and S51 in FIG. Therefore, the details of the OCR process are omitted.

  Subsequently, the CPU 21 determines whether the first parameter value is larger than the second parameter value (S45). When the first parameter value is larger than the second parameter value (S45: Yes), the CPU 21 determines vertical rotation as the rotation amount (S47). When the first parameter value is not larger than the second parameter value, that is, when the first parameter value is equal to or smaller than the second parameter value (S45: No), the CPU 21 determines non-rotation as the rotation amount (S49). After executing S47 or S49, the CPU 21 shifts the processing to S59 of FIG.

  In S51 of FIG. 6, the CPU 21 executes an OCR process in which the reading pattern is set in the horizontal direction. In the OCR process in which the reading pattern is set in the horizontal direction, the first image data or the first composite data is subjected to the OCR process from the left side in the horizontal direction to the right side, and further, the OCR process is performed from the right side in the horizontal direction to the left side. Performing OCR processing from the left side to the right side in the horizontal direction is equivalent to performing OCR processing from the upper side to the lower side in the vertical direction after the first image data or the first composite data is rotated 90 ° clockwise. Performing OCR processing from the right side to the left side in the horizontal direction is equivalent to performing OCR processing from the upper side to the lower side in the vertical direction after rotating the first image data or the first composite data by 90 ° counterclockwise. The CPU 21 acquires the third parameter value and the fourth parameter value from the OCR process in S51. The third parameter value is a parameter value indicating the reliability of the recognition result when the reading pattern is changed from the left side to the right side in the horizontal direction. The fourth parameter value is a parameter value indicating the reliability of the recognition result when the reading pattern is changed from the right side to the left side in the horizontal direction.

  Subsequently, the CPU 21 determines whether the third parameter value is larger than the fourth parameter value (S53). When the third parameter value is larger than the fourth parameter value (S53: Yes), the CPU 21 determines a clockwise amount in which the horizontal direction is the vertical direction as the rotation amount (S55). When the third parameter value is not larger than the fourth parameter value, that is, when the third parameter value is equal to or smaller than the fourth parameter (S53: No), the CPU 21 sets the rotation amount in the counterclockwise direction in which the horizontal direction is the vertical direction. The amount is determined (S57). After executing S55 or S57, the CPU 21 shifts the processing to S59.

  In S59, the CPU 21 executes a rotation process for the first image data or the first composite data. It is assumed that vertical rotation is determined as the rotation amount in S47 of FIG. In this case, the CPU 21 newly generates image data obtained by rotating the top and bottom of the first image data or the first composite data (for example, 180 ° rotation). Assume that non-rotation is determined as the rotation amount in S49 of FIG. In this case, the CPU 21 does not execute the rotation process. It is assumed that a clockwise amount in which the horizontal direction is the vertical direction is determined as the rotation amount in S55. In this case, the CPU 21 rotates the first image data or the first combined data clockwise by an amount such that the horizontal direction of the first image data or the first combined data is the vertical direction (for example, rotated 90 ° clockwise). The generated image data is newly generated. It is assumed that a counterclockwise amount in which the horizontal direction is the vertical direction is determined as the rotation amount in S57. In this case, the CPU 21 rotates the first image data or the first combined data counterclockwise by an amount in which the horizontal direction of the first image data or the first combined data is the vertical direction (for example, rotated 90 ° counterclockwise). The generated image data is newly generated. The rotation process executed in S59 is a known process. In S59, a known rotation processing technique can be employed. Therefore, the details of the rotation process are omitted. Thereafter, the CPU 21 shifts the processing to S61. When the rotation amount is determined to be non-rotated in S49 of FIG. 5, the CPU 21 substantially does not execute S59, or applies the rotation process to the first image data or the first composite data. Instead, image data copied as it is is newly generated, and the process proceeds to S61. That is, after executing S49 of FIG. 5, the CPU 21 shifts the process to S61.

  In S61, the CPU 21 executes an inclination correction process. The tilt that is the target of the tilt correction process is generated when the scanner device 30 scans. In other words, this inclination occurs, for example, when the carrier sheet 40 is conveyed while being inclined with respect to the main scanning direction. In addition, this occurs when the document is sandwiched between the carrier sheets 40 while being inclined with respect to the main scanning direction. The lower part of FIG. 2 shows a state in which a receipt as a document is sandwiched between carrier sheets 40 in an inclined state with respect to the main scanning direction by an angle θ.

  If both S41 and S45 in FIG. 5 are affirmed (see S41, S45: Yes), the inclination correction process is executed on the first image data or the first composite data whose upper and lower sides are rotated in S59. If S41 in FIG. 5 is negative (see S41: No), the inclination correction process is executed on the first image data or the first composite data rotated by a predetermined amount clockwise or counterclockwise in S59. The For example, it is assumed that characters included in the receipt image corresponding to the first composite data are inclined with respect to the horizontal direction. According to the inclination correction process, the first composite data can be corrected to image data in a state where the above-described characters are arranged in the horizontal direction.

  The first image data or the first composite data may be in a state where no tilt occurs, or the tilt generated in the first image data or the first composite data may be a predetermined value or less. In this case, there is no change in the image data before and after the inclination correction process, or the change amount of the image data is very small. Therefore, when it is specified that the first image data or the first combined data is in a state where no inclination occurs, or the inclination generated in the first image data or the first combined data is equal to or less than a predetermined value. , S61 may be omitted. The tilt correction process executed in S61 is a known process. In S61, a known tilt correction processing technique can be employed. Therefore, the details of the inclination correction process are omitted.

  In the embodiment, the name of the image data after the rotation process is executed in S59 and the inclination correction process is executed in S61 or the image data after the inclination correction process is executed in S61 is the target of each process. The name of the processed image data is used. That is, for the first image data, the rotation process is executed in S59 and the inclination correction process is executed in S61, or the rotation correction process is not executed in S59 and the inclination correction process is executed in S61. . In this case, the image data after the inclination correction process is referred to as “first image data” as before the process. It is assumed that the rotation process is executed in S59 and the inclination correction process is executed in S61 or the inclination correction process is executed in S61 without executing the rotation process in S59 for the first composite data. In this case, the image data after the inclination correction process is referred to as “first composite data” as before the process.

  Next, the CPU 21 controls display of the second composite data, the first image data, or the first composite data (S63). That is, the CPU 21 outputs a display command for the second composite data, the first image data, or the first composite data to the display unit 24. Accordingly, the display unit 24 displays an image corresponding to the second composite data, the first image data, or the first composite data. For example, it is assumed that the carrier sheet 40 in the state shown in the lower part of FIG. In this case, the display unit 24 displays the entire receipt as shown in FIG. 3 (refer to “<Before folding (the entire receipt)>” in FIG. 3).

  Subsequently, the CPU 21 determines whether or not the count number of the counter matches the repetition number set in S17 of FIG. 4 (S65). When the count number does not match the repeat number (S65: No), the CPU 21 returns the process to S19 of FIG. Thereafter, the CPU 21 repeatedly executes the processes after S19. When the count number and the repetition number match (S65: Yes), the CPU 21 ends the image processing.

<Effect of embodiment>
According to the embodiment described above, the following effects can be obtained.

  (1) In the image processing shown in FIGS. 4 to 6, a labeling process is executed (see S15 in FIG. 4). In the labeling process, first image data is acquired from the first surface data corresponding to the first surface of the carrier sheet 40, and second image data is acquired from the second surface data corresponding to the second surface of the carrier sheet 40. . Thereafter, it is determined whether the short side size of the first image data is equal to or smaller than the reference size (see S23 of FIG. 4). When the short side size of the first image data is equal to or smaller than the reference size (see S23 in FIG. 4: Yes), the first image data and the second image data are combined at the short side of each image data, and the first combined data is Is generated (see S37 in FIG. 5). The first composite data is subjected to OCR processing (see S43 in FIG. 5 or S51 in FIG. 6). In the image processing, the rotation amount of the first composite data is appropriately determined according to the parameter value indicating the reliability of the recognition result of the OCR processing (see S47 in FIG. 5 or S55 or S57 in FIG. 6), and the rotation processing is performed. This is executed (see S59 in FIG. 6). On the other hand, when the short side size of the first image data is larger than the reference size (see S23 in FIG. 4: Yes), the first image data and the second image data are combined on the long side of each image data, Two-composition data is generated (see S29 in FIG. 4). In this case, the second composite data is not subjected to OCR processing, and the rotation amount is not determined (see S31 in FIG. 5 to S57 not executed in FIG. 6). As a result, the rotation process is not executed (see S59 not executed in FIG. 6).

  For this reason, the image processing apparatus 20 can automatically determine the type of the document that is folded in half and sandwiched between the carrier sheets 40, and can perform composition according to the type of each document. For example, it is assumed that there is a receipt and an A3 size document, and one of the documents is sandwiched between the carrier sheet 40 in a folded state. In such a case, the image processing apparatus 20 automatically determines whether the original document sandwiched between the carrier sheets 40 is a receipt or an A3 size document. Synthesis according to the size data can be automatically executed. By determining the rotation amount according to the parameter value indicating the reliability of the recognition result of the OCR process, the character direction can be set to the correct direction in the first composite data corresponding to the receipt. In the first composite data, the horizontal direction and the vertical direction can be matched with the correct direction of the actual receipt. The operability of the image processing apparatus 20 can be improved. In the image processing apparatus 20, a user-friendly image processing procedure is realized.

  (2) In the image processing, even if the short side size of the first image data is equal to or less than the reference size (see S23 in FIG. 4), the second image data is blank paper data (S33 in FIG. 5: Yes). The first image data and the second image data are not combined (see S37 not executed in FIG. 5). In this case, the first image data is the target of OCR processing (see S43 in FIG. 5 or S51 in FIG. 6). In the image processing, the rotation amount of the first image data is appropriately determined according to the parameter value indicating the reliability of the recognition result of the OCR processing (see S47 in FIG. 5 or S55 or S57 in FIG. 6), and the rotation processing is performed. This is executed (see S59 in FIG. 6). For this reason, it is possible to suppress the generation of the first composite data including blank paper data.

<Modification>
The embodiment described above can also be performed as follows. Some configurations of the modifications shown below can be appropriately combined and employed. Hereinafter, points different from the above will be described, and description of similar points will be omitted as appropriate.

  (1) In the above, the scanner device 30 includes the first image sensor 35 and the second image sensor 36. In the scanner device 30, the first surface of the carrier sheet 40 is scanned by the first image sensor 35, and the second surface of the carrier sheet 40 is scanned by the second image sensor 36. The scanner device may include one image sensor. In this case, the transport unit of the scanner device includes a reversing mechanism provided in a known scanner device. After the first surface is scanned, the carrier sheet 40 is appropriately transported along the transport path in the transport unit and reversed. Thereafter, in the scanner device, the second surface of the carrier sheet 40 is scanned again by the image sensor that scanned the first surface. In the scanner device, the first surface data and the second surface data described above are generated in this way.

  (2) In the above, the image processing shown in FIGS. 4 to 6 is executed by the image processing device 20 connected to the scanner device 30 by the communication cable 11. The image processing may be executed by the scanner device 30. In this case, the user sets the carrier sheet 40 in the scanner device 30 as described above. Thereafter, the user operates an operation unit (not shown) provided in the scanner device 30 and inputs a start instruction. The operation unit of the scanner device 30 receives a start instruction, and the CPU 31 acquires it. The CPU 31 having acquired the start instruction starts the scanning operation and acquires the first surface data and the second surface data. The first surface data and the second surface data are stored in the RAM 33. Thereafter, the CPU 31 executes processes corresponding to the processes after S15 in FIG. 4 in the same manner as described above. The CPU 31 initializes the counter to “0” at a predetermined timing before the execution of the process corresponding to S19 of FIG.

  The CPU 31 controls the display of the second composite data, the first image data, or the first composite data in the process corresponding to S63 in FIG. Accordingly, in the scanner device 30, the second composite data, the first image data, or an image corresponding to the first composite data is displayed on a display unit (not shown) provided in the scanner device 30. When the scanner device 30 has a function of writing to a memory card, the second composite data, the first image data, or the first composite data may be stored in the memory card.

  (3) In the above description, the blank page data is defined as image data corresponding to a blank page area that does not include predetermined information including characters. In this case, the blank area that does not include the predetermined information including characters includes a plain area that does not include the predetermined information including characters. The plain is an area of one color as a whole. The plain includes a pattern.

  (4) In the above, in S35 and S37 (see FIG. 5) of the image processing, the short side is determined as the combining position, the first image data and the second image data are combined, and the first combined data is generated. Each process corresponding to S35 and S37 may be executed after the inclination correction process of S61 in FIG. 6 is executed, for example. In this case, S39 in FIG. 5 to S61 in FIG. 6 are executed for each of the first image data and the second image data. For the first image data and the second image data after the inclination correction process is executed, each short side as the synthesis position is determined as described above. Subsequently, the first image data and the second image data are synthesized in a state where the determined short sides of the first image data and the second image data coincide with each other, thereby generating first synthesized data.

  (5) In the above description, the determination condition in S45 of image processing (see FIG. 5) is “first parameter value> second parameter value”. This determination condition may be “first parameter value <second parameter value”. If this determination is negative, the CPU 21 proceeds to S47, and if this determination is affirmed, the CPU 21 proceeds to S49. The judgment condition of S53 (see FIG. 6) of the image processing is “third parameter value> fourth parameter value”. This determination condition may be “third parameter value <fourth parameter value”. If this determination is negative, the CPU 21 proceeds to S55, and if this determination is affirmed, the CPU 21 proceeds to S57. In S45 of FIG. 5 or S53 of FIG. 6, when both parameters to be compared are equal, each determination may be affirmed.

DESCRIPTION OF SYMBOLS 10 Image processing system 11 Communication cable 20 Image processing apparatus 21 CPU
22 storage device 23 RAM
24 display unit 25 operation unit 26 communication unit 30 scanner device 31 CPU
32 storage device 33 RAM
34 Conveying section 35 First image sensor 36 Second image sensor 37 Communication section 40 Carrier sheet L Fold

Claims (7)

A computer-executable program for controlling an image processing device that performs image processing on image data corresponding to a document scanned by a scanner device,
The computer,
A carrier sheet in which characters are recorded, and the original having a long first length longer than the second size of the second side is folded and sandwiched with the fold line along the second side. The first image data corresponding to the first area of the document on the first surface side and the second area of the document on the second surface side opposite to the first surface of the carrier sheet. Corresponding first image data; first acquisition means for acquiring;
Second acquisition means for acquiring a short side size corresponding to the second size of the document and corresponding to the short side of the first image data;
First determination means for determining whether the short side size acquired by the second acquisition means is smaller than a reference size serving as a reference;
When the short side size is smaller than the reference size, at least a determination process for determining a rotation amount of the first image data is executed according to at least a recognition result of the OCR process for the first image data, and at least the first image data On the other hand, a deforming unit that executes a rotation process according to the rotation amount determined by the determination process and does not execute the determination process and the rotation process when the short side size is not smaller than the reference size. And a program to make it function.   When the short side size is smaller than the reference size, the computer is configured to correspond to the original, and the first image data and the first image data on each short side of the first image data and the second image data. When first composite data is generated by combining two image data, and the short side size is not smaller than the reference size, each of the first image data and the second image data in a state corresponding to the original. The program according to claim 1, wherein the program functions as a synthesizing unit that generates second synthesized data obtained by synthesizing the first image data and the second image data on a long side.   The deformation means includes the first composition including the first image data and the second image data in accordance with a recognition result of the OCR process on the first composition data including the first image data and the second image data. A function of executing the determination process for determining the rotation amount of data and executing the rotation process on the first composite data including the first image data and the second image data. 2. The program according to 2. The synthesizing means includes a function of not synthesizing the first image data and the second image data when the second image data is data corresponding to an area not including predetermined information including characters,
In the case where the short side size is smaller than the reference size and the first image data and the second image data are not combined, the deforming unit is configured to perform the first image data according to the recognition result of the OCR process on the first image data. The program according to claim 2 or 3, further comprising a function of executing the determination process for determining the rotation amount of image data and executing the rotation process on the first image data. When the short side size is smaller than the reference size, the short side size acquired by the second acquisition unit and used as the determination target of the determination unit is the main scanning direction of scanning in the scanner device. Which functions as a second determination means for determining whether the size of the first image data is in the horizontal direction,
The deformation means includes
When the short side size is the size of the first image data in the horizontal direction, in the determination process, as the rotation amount, at least an amount that is vertically reversed in the vertical direction of the first image data and unrotated And decide
When the short side size is not the horizontal size of the first image data, the determining process determines, as the rotation amount, at least an amount in which the horizontal direction of the first image data is a vertical direction. The program according to any one of claims 1 to 4, further comprising: The deformation means includes
The OCR process from the lower side to the upper side of the vertical direction of the first image data, wherein the short side size is the horizontal size of the first image data and coincides with the sub-scanning direction orthogonal to the main scanning direction. If the first parameter value indicating the reliability of the recognition result of the first image data is greater than the second parameter value indicating the reliability of the recognition result of the OCR process from the upper side to the lower side in the vertical direction of the first image data, the rotation amount And determining at least the amount that the first image data is upside down in the vertical direction,
When the short side size is the horizontal size of the first image data and the first parameter value is smaller than the second parameter value, the rotation amount is determined as non-rotated,
The short side size is not the horizontal size of the first image data, but the third parameter value indicating the reliability of the recognition result of the OCR processing from the left side to the right side of the first image data is When the fourth parameter value indicating the reliability of the recognition result of the OCR process from the right side to the left side in the horizontal direction of the first image data is larger than the fourth parameter value, at least the horizontal direction of the first image data is the vertical direction. Determine the amount of clockwise rotation
When the short side size is not the horizontal size of the first image data and the third parameter value is smaller than the fourth parameter value, at least the horizontal direction of the first image data is the rotation amount. The program according to claim 5, wherein the amount of counterclockwise rotation in the vertical direction is determined. An image processing device that performs image processing on image data corresponding to a document scanned by a scanner device,
A carrier sheet in which characters are recorded, and the original having a long first length longer than the second size of the second side is folded and sandwiched with the fold line along the second side. The first image data corresponding to the first area of the document on the first surface side and the second area of the document on the second surface side opposite to the first surface of the carrier sheet. Corresponding first image data; first acquisition means for acquiring;
Second acquisition means for acquiring a short side size corresponding to the second size of the document and corresponding to the short side of the first image data;
First determination means for determining whether the short side size acquired by the second acquisition means is smaller than a reference size serving as a reference;
When the short side size is smaller than the reference size, at least a determination process for determining a rotation amount of the first image data is executed according to at least a recognition result of the OCR process for the first image data, and at least the first image data On the other hand, a deforming unit that executes a rotation process according to the rotation amount determined by the determination process and does not execute the determination process and the rotation process when the short side size is not smaller than the reference size. And an image processing apparatus.

Images