JP2015119431A - Page image correction device, page image correction method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately and efficiently correct a cut image so as to clearly read the image even in the case where a curved paper surface is imaged.SOLUTION: A page image correction device (document camera system 1) includes: extraction means (CPU 45) for extracting a contour of a page out of an image obtained by imaging means; contour correction means (CPU 45) for applying contour correction on the basis of four corners of the extracted contour of the page; dividing means (CPU 45) for dividing the entire corrected page image in such a manner that (n) pieces of substantially rectangular divided regions are arrayed in a width direction orthogonal to a seam direction; first correction means (CPU 45) for correcting a length in the seam direction for each of the divided regions; estimation means (CPU 45) for estimating a binding amount of the divided region in a normal direction on an actual page for each corrected divided region; second correction means (CPU 45) for correcting a length in the width direction for each divided region on the basis of the estimated binding direction; and combination means (CPU 45) for combining the (n) pieces of divided regions corrected by the correction means.

Description

  The present invention relates to a page image correction apparatus, a page image correction method, and a program.

  2. Description of the Related Art Conventionally, for example, an image reading apparatus is known in which an image pickup unit is arranged above a book in a spread state, and the pages of the book are read page by page by the image pickup unit. Here, since the size differs depending on the book, the image reading device can specify the document size so that only a necessary area is cut out from the image acquired by the imaging unit in order to perform reading processing according to the size. (See, for example, Patent Document 1).

JP 2006-114993 A

By the way, there is a case where the paper surface is curved at the time of imaging. In that case, the cut image is also curved and is difficult to read.
Accordingly, an object of the present invention is to enable appropriate and efficient correction so that a cut image can be easily read even when a curved paper surface is imaged.

In order to solve the above problems, according to one aspect of the present invention,
Imaging means for imaging a page of a book;
Extraction means for extracting the outline of the page from the image acquired by the imaging means;
Contour correcting means for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted by the extracting means;
The entire image of the page corrected by the contour correcting unit is arranged in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. a dividing means for dividing the data so that n pieces are arranged;
First correcting means for correcting the length in the binding direction for each of the divided areas divided by the dividing means;
Estimating means for estimating the amount of deflection in the normal direction of the divided area on an actual page for each of the divided areas corrected by the first correcting means;
Second correction means for correcting the length in the width direction for each of the divided regions based on the amount of deflection estimated by the estimation means;
There is provided a page image correction apparatus comprising: a combining unit that combines the n number of divided regions corrected by the first correction unit and the second correction unit.

  According to the present invention, even when a curved paper surface is imaged, it is possible to appropriately and efficiently correct the cut image so that it can be easily read.

1 is a perspective view showing a schematic configuration of a document camera system according to the present embodiment. It is a block diagram which shows the main control structure of the document camera system which concerns on this embodiment. It is a flowchart which shows the flow of a process of the image processing mode which concerns on this embodiment. It is explanatory drawing which shows the outline of the left page extracted from the reference | standard image which concerns on this embodiment, and the whole image of a page. It is explanatory drawing which shows the outline and page image after trapezoid correction which concern on this embodiment. It is explanatory drawing which shows the outline and image after trapezoid correction | amendment which concern on this embodiment, and the rectangular outline which consists of four corners after trapezoid correction | amendment. It is a schematic diagram for demonstrating the conversion formula used with the vertical direction correction | amendment process which concerns on this embodiment. It is a flowchart which shows the flow of the vertical direction correction process which concerns on this embodiment. It is explanatory drawing which shows the outline and image after trapezoid correction | amendment which concern on this embodiment, and the rectangular outline which consists of four corners after trapezoid correction | amendment. It is a flowchart which shows the flow of the width direction correction process which concerns on this embodiment. It is explanatory drawing which shows the deflection amount d (z) in the division area which concerns on this embodiment. It is explanatory drawing which shows the relationship between ds (i), dz (i), and dx '(i) concerning this embodiment. It is explanatory drawing which shows the relationship between x '(i) before the width direction correction process which concerns on this embodiment, and x (i) after the width direction correction process. It is explanatory drawing which shows the example of the image after correction | amendment which changes with the curvature degree of a page.

  The best mode for carrying out the present invention will be described below with reference to the drawings. However, although various technically preferable limitations for implementing the present invention are given to the embodiments described below, the scope of the invention is not limited to the following embodiments and illustrated examples.

  FIG. 1 is a perspective view showing a schematic configuration of a document camera system as a page image correction apparatus according to the present invention. As shown in FIG. 1, the document camera system 1 can communicate with a document camera 2 as an imaging unit that images the page P1 of the book B, a mounting table 3 on which the book B is placed, and the document camera 2. And a connected personal computer 4. The book B according to the present invention includes not only a general book but also a bound document.

  The document camera 2 is provided with a stand portion 21 and a camera 22 attached to the upper end of the stand portion 21. The stand unit 21 can be tilted in the front-rear direction and the left-right direction and can be expanded and contracted in the vertical direction so that the relative positional relationship between the book B and the camera 22 can be adjusted. The lens of the camera 22 faces downward so that the book B is within the angle of view. A position adjustment mechanism is provided at a joint portion between the camera 22 and the stand portion 21, so that the direction in which the lens of the camera 22 faces can be adjusted. And the camera 22 is arrange | positioned in the position which can acquire the image of both the left and right pages P1 of the book B opened on the mounting base 3 by one imaging | photography.

Next, the main control configuration of the document camera system 1 of the present embodiment will be described. FIG. 2 is a block diagram showing a main control configuration of the document camera system 1. As shown in FIG. 2, the personal computer 4 has a recording unit 41 for recording a reference image captured by the camera 22 and a corrected image obtained by correcting the reference image, and various operation instructions. Based on operation means 42 such as a keyboard, a touch panel, and a mouse input to itself, a ROM 43 in which various programs are recorded, a RAM 44 in which the programs are expanded when the programs in the ROM 43 are executed, and instructions from the operation means 42 And a CPU 45 that expands and executes the program in the ROM 43 in the RAM 44.
When the image reading program is executed by the CPU 45 of the personal computer 4, a predetermined button of the operation means 42 is associated as a photographing button. The camera 22 shoots the book B image in conjunction with the operation of the shooting button.

Hereinafter, an image reading method by the document camera system 1 will be described.
First, the user operates the operation means 42 to cause the personal computer 4 to execute an image reading program. Thereby, first, the image capturing mode is executed. In the image shooting mode, the user first opens the first page P1 of book B and presses the page P1. In this state, the left and right pages P1 that are currently open are photographed as one image by the camera 22 by operating the photographing button.
By repeating this until the last page P1 of book B, the reference images of all the pages P1 of one book B are recorded in the recording unit 41 of the personal computer 4.
When shooting of the last page P1 is completed, the user operates the operation means 42 to prompt the personal computer 4 to end the image shooting mode. The personal computer 4 switches to the image processing mode as the image shooting mode ends.

FIG. 3 is a flowchart showing the flow of processing in the image processing mode.
In step S <b> 1, the CPU 45 reads the reference image from the recording unit 41.
In step S2, the CPU 45 extracts the contour of the page P1 in the reference image using a known image process. That is, the CPU 45 is an extraction unit according to the present invention. At this time, the left and right pages P1 in the reference image are divided, and the outline is extracted for each page P1. FIG. 4 is an explanatory diagram showing the outline R of the left page P1 and the entire image G of the page P1 extracted from the reference image.

In step S3, the CPU 45 acquires the coordinate values of the four corners r1, r2, r3, r4 from the contour R of one page P1.
In step S4, the CPU 45 performs known distortion correction on the contour R of the page P1 based on the coordinate values of the four corners r1, r2, r3, r4 acquired in step S3. That is, the CPU 45 is a contour correcting unit according to the present invention. In this distortion correction, not only the outline R of the page P1 is trapezoidal, but the entire image of the page P1 is corrected following the deformation of the outline R by the trapezoid correction.
FIG. 5 is an explanatory diagram showing the contour R1 after trapezoid correction and the image G1 of the page P1.

In step S5, the CPU 45 obtains the contour R1 of the page P1 after the keystone correction.
In step S6, the CPU 45 executes vertical direction correction processing on the image G1 after the keystone correction based on the contour R1 after the keystone correction. That is, the CPU 45 is the first correction unit according to the present invention.

FIG. 6 is an explanatory view showing the contour R1 and the image G1 after the trapezoid correction, and the rectangular contour R2 including the four corners r1, r2, r3, r4 after the trapezoid correction.
In the contour R1 and the image G1 after the trapezoid correction, the curve in the binding direction (Y direction) remains. When the page P1 is curved, the central portion of the page P1 is close to the camera 22, so that the image is taken larger than the left and right ends of the page P1. For example, the region U extending along the stitch direction from the upper end to the lower end of the image G1 in the vicinity of the approximate center in the width direction of the image G1 is larger (longer) in the stitch direction than the rectangular contour R2. That is, the length in the binding direction of the page P1 varies in the width direction of the page P1. In order to make this uniform and within the rectangular outline R2, the vertical direction correction processing is executed.

FIG. 7 is a schematic diagram for explaining a conversion formula used in the vertical direction correction processing.
In the figure, F is a focal point set for convenience. A ′ is the upper point of the arbitrary region U in the contour R1 after trapezoid correction, B ′ is the lower point of the arbitrary region U in the region U, P ′ is an arbitrary point on the A′-B ′ line, and A is a rectangular contour The upper point of the portion corresponding to the region U in R2, B is the lower point of the portion corresponding to the region U in the rectangular contour R2, and P is the point corresponding to the point P on the AB line.

The following relationship can be seen from FIG.
ΔA'FB'∝ΔAFB
∴FP: FP '= h: H
ΔAFP∝ΔA'FP '
∴AP: A'P '= h: H
∴AP = h / H · A'P '
Here, the Y coordinate value of the point A and y _a, the Y coordinate value of the point P and y _p, 'a Y coordinate value of the point y _a' A and 'a Y coordinate value of the point y _p' P and Then
A'P '= y _a' -y _{p '}
AP = y _a -y _p
It becomes.
∴y _p = h / H · (y _{p ′} −y _{a ′} ) + y _{a ′} (2)
The Y coordinate value of P after conversion of an arbitrary point P can be obtained by this equation (2).
Actually, correction is performed by applying the gradation at the point of the corresponding contour R1 to the gradation of each point in the mesh in the rectangular contour R2.

FIG. 8 is a flowchart showing the flow of the vertical direction correction process.
In step S61, the CPU 45 obtains the length h in the stitch direction of the rectangular contour R2. Specifically, the distance between the upper and lower corners on the right side or the left side of the contour R1 after the trapezoid correction is defined as a length h.
In step S62, the CPU 45 divides the contour R1 after the keystone correction into n parts in the width direction, and sets the abscissa x (i) of each divided region U1. That is, the CPU 45 is a dividing unit according to the present invention.

In step S63, the CPU 45 sets i to 1.
In step S64, the CPU 45 obtains the length H in the vertical direction at x (i) based on the contour R1 after trapezoid correction.
In step S65, the CPU 45 calculates the Y coordinate value (y _{p ′} ) corresponding to the P ′ point at the boundary between the divided areas U1 using the formula (1), and converts the image of the P ′ point.
In step S66, the CPU 45 determines whether or not the conversion process has been completed at all points in the stitch direction at x (i). If not, the process proceeds to step S65. Migrate to

In step S67, the CPU 45 determines whether i is n−1. If n−1, the CPU 45 ends the vertical direction correction process and proceeds to step S7. If not, the process proceeds to step S68. Transition.
In step S68, the CPU 45 sets i = i + 1 and proceeds to step S64. Thus, the vertical correction is performed on all the boundaries between the n−1 divided areas U1.

In step S7, the CPU 45 executes a width direction correction process on the image G1 after the keystone correction based on the image after the vertical direction correction. That is, the CPU 45 is the second correction unit according to the present invention.
FIG. 9 is an explanatory diagram showing a contour R1 and image G1 after trapezoid correction, and a rectangular contour R2 composed of four corners r1, r2, r3, r4 after trapezoid correction.
In the contour R1 and the image G1 after the trapezoidal correction, the curve in the binding direction remains, so it is necessary to perform a process of returning the curve W1 to the straight line L1 as shown in FIG. At this time, since the correction amount varies depending on the location in the width direction, the width direction correction in step S7 is executed.

FIG. 10 is a flowchart showing the flow of the width direction correction process.
In step S71, the CPU 45 acquires the length h in the stitch direction of the rectangular contour R2.
In step S72, the CPU 45 divides the trapezoid-corrected contour R1 into n portions in the width direction, and sets the abscissa x (i) of each divided region U1.

In step S73, the CPU 45 sets i to 1.
In step S74, the CPU 45 estimates the deflection amount dz (i) at x (i). That is, the CPU 45 is an estimation unit according to the present invention.
FIGS. 11A and 11B are explanatory views showing the deflection amount d (z) in the divided region U1, where FIG. 11A is a cross-sectional view of an actual page P1, and FIG. 11B is a front view showing an image after contour correction.
As shown in FIG. 11 (a), the deflection amount dz (i) in the normal direction (Z direction) of the actual page P1 is, strictly speaking, the i-th divided region U1 and the i-1th divided region. It is obtained from the difference in the Z direction from U1. However, this calculation requires a distance and angle from the camera 22 to the paper surface, and these measurements are troublesome. Here, if the width of the divided area U1 is very small, the length H (i) of the i-th divided area U1 in the binding direction and the length H (i-1) of the i-1th divided area U1. Is approximately equal to the deflection amount dz (i). Using this relationship, the deflection amount dz (i) of the divided region U1 is estimated.
Specifically, the amount of deflection dz (i) is estimated by equation (1).
dz (i) = k · {H (i) −H (i−1)} (1)
Here, k is an arbitrary coefficient. By using this coefficient k, the image after the width direction correction process can be adjusted in the width direction. Here, the calculation is temporarily performed with k = 1.

In step S75, the CPU 45 obtains the actual width dx ′ (i).
dx (i) is the width of the i-th divided area U1 after trapezoidal correction.
dx (i) = x (i) -x (i-1) = {x (n) -x (0)} / n
The actual width dx ′ (i) has been modified by the height z (i) of x (i). When z (i) is high, the width increases as H increases. That is, the actual width dx ′ (i) is small. This relationship is summarized in the following equation.
dx ′ (i) = h / H (i) · dx (i)

In step S76, the CPU 45 obtains ds (i).
ds (i) is the minute length of the actual page P1 in the i-th divided region U1.
FIG. 12 is an explanatory diagram showing the relationship between ds (i), dz (i), and dx ′ (i).
As shown in FIG. 12, ds (i) is a square based on the length dz (i) in the Z direction obtained in step S74 and the length dx '(i) in the X direction obtained in step S75. It is calculated by the theorem.
ds (i) = {dz (i) ² + dx ′ (i) ² } ^1/2

In step S77, the CPU 45 determines whether i is n. If n, the process proceeds to step S79. If not, the process proceeds to step S78.
In step S78, the CPU 45 sets i = i + 1 and proceeds to step S74. As a result, dz (i), dx ′ (i), and ds (i) are obtained for all n divided regions U1.

In step S79, the CPU 45 obtains the actual length P of the page P1 in the width direction. Specifically, the length S in the width direction of the actual page P1 is calculated by calculating the sum of all ds (i) of the n divided areas U1.
In step S80, the CPU 45 sets i to 1 again.

  In step S81, ds (i) of the divided area U1 is obtained from the 1st to the i-th, and its accumulated value S (i) is calculated.

In step S82, the image is converted with x (i) corresponding to x ′ (i).
FIG. 13 is an explanatory diagram showing the relationship between x ′ (i) before the width direction correction process and x (i) after the width direction correction process. As shown in FIG. 13, the ratio between the actual S and S (i) of page P1 is the length from x (0) to x (n) and the length from x (0) to x (i). Since it is the same value as the ratio, the relationship is as shown in the following equation.
S: S (i) = x (n) -x (0): x (i) -x (0)
Based on this relationship, an image at an arbitrary point P is converted by converting x ′ (i) into x (i). Note that the conversion formula is x (i) = f (x ′ (i)), but there is nothing clear about this formula, and x ′ (i) corresponding to S (i) is found sequentially. x (i).

  In step S83, the CPU 45 determines whether or not the conversion process has been completed for all points in the stitch direction at x (i). If not, the process proceeds to step S82. Migrate to

In step S84, the CPU 45 determines whether i is n. If i, the CPU 45 ends the width direction correction process and proceeds to step S8. If not, the CPU 45 proceeds to step S85.
In step S85, the CPU 45 sets i = i + 1 and proceeds to step S81. As a result, the width direction correction is executed for all n divided regions U1.

In step S8, the CPU 45 synthesizes the n divided areas U1 that have been subjected to the vertical direction correction process and the width direction correction process, and creates an entire image. That is, the CPU 45 is a combining unit according to the present invention. Specifically, by accumulatively adding the lengths in the width direction corrected by the width direction correction process, n divided regions U1 after the height direction correction process are combined and stored in the recording unit 41.
These processes are executed on the right page P1 included in the reference image, and further, the above process is repeated for all the reference images, thereby distorting all the pages P1 included in the plurality of photographed reference images. Can be obtained in the absence of

As described above, according to the present embodiment, before the image of the page P1 is divided into the substantially rectangular divided areas U1 extending in the binding direction from the upper end to the lower end, the entire image of the page P1 is contour-corrected. Therefore, the entire image P1 of the page P can be easily divided into the divided areas U1.
Then, after correcting the length in the binding direction for each divided region U1, the amount of deflection in the normal direction of the divided region U1 on the actual page P1 is estimated, and the length in the width direction is also calculated based on the estimated amount of deflection. Since correction is performed, correction in the binding direction and correction in the width direction can be performed efficiently.
Therefore, even when a curved paper surface is captured, it is possible to appropriately and efficiently correct the cut image so that it can be easily read.

  Further, since the amount of bending of the divided region U1 is estimated based on the length of the divided region U1 in the stitch direction, it is not necessary to actually measure the amount of bending, and the amount of bending can be efficiently obtained. .

  Further, since the n divided regions U1 after the height direction correction processing are synthesized by cumulatively adding the lengths in the width direction corrected by the width direction correction processing, the length in the width direction is set to the entire image. Can be corrected appropriately.

Further, in the present embodiment, the coefficient k (temporarily k = 1) is used for the equation (1) when dz (i) is obtained in the width direction correction processing. Although the above-described processing makes the divided regions U1 uniform in the width direction, there is a problem in the accuracy of the length of the entire width. For example, FIG. 14A shows a case where the correction is performed on the page P1 having a small degree of curvature, and FIG. 14B shows that the same page P1 as in FIG. The case where it gave is shown. As is apparent from FIG. 14, the corrected images Q1 and Q2 have different lengths in the width direction. This is because even if the actual width of the page P1 is the same, the difference between H (i) and H (i-1) differs greatly depending on the degree of curvature. In order to suppress this variation, the processing can be performed in consideration of the coefficient k.
That is, since the aspect ratio (for example, a prescribed paper size) of the page P1 is actually known, the ratio between the horizontal width of the image obtained in step S8 and the horizontal width obtained from the aspect ratio of the page P1 corresponds to the coefficient k. .
Therefore, by scaling the image obtained in step S8 in the width direction corresponding to this ratio, an image having a desired aspect ratio and no distortion can be obtained.
Specifically, it is preferable that the corrected image is enlarged or reduced in the width direction so as to have a predetermined size (for example, a prescribed paper size).
Note that the present invention is not limited to the above embodiment, and can be modified as appropriate.

Although several embodiments of the present invention have been described, the scope of the present invention is not limited to the above-described embodiments, but includes the scope of the invention described in the claims and equivalents thereof.
The invention described in the scope of claims attached to the application of this application will be added below. The item numbers of the claims described in the appendix are as set forth in the claims attached to the application of this application.

[Appendix]
<Claim 1>
Imaging means for imaging a page of a book;
Extraction means for extracting the outline of the page from the image acquired by the imaging means;
Contour correcting means for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted by the extracting means;
The entire image of the page corrected by the contour correcting unit is arranged in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. a dividing means for dividing the data so that n pieces are arranged;
First correcting means for correcting the length in the binding direction for each of the divided areas divided by the dividing means;
Estimating means for estimating the amount of deflection in the normal direction of the divided area on an actual page for each of the divided areas corrected by the first correcting means;
Second correction means for correcting the length in the width direction for each of the divided regions based on the amount of deflection estimated by the estimation means;
A page image correction apparatus comprising: a combining unit that combines the n divided regions corrected by the first correction unit and the second correction unit.
<Claim 2>
The page image correction apparatus according to claim 1,
The page image correcting apparatus, wherein the estimating means estimates the amount of deflection of the divided area based on a length of the divided area in the binding direction.
<Claim 3>
In the page image correction apparatus according to claim 2,
The bending amount of the i-th divided area is dz (i), the length of the i-th divided area in the binding direction is H (i), and the i-1th divided area is When the length in the binding direction is H (i−1) and an arbitrary coefficient is k,
The page image correcting apparatus according to claim 1, wherein the estimating means estimates the amount of deflection of the i-th divided area using equation (1).
dz (i) = k · {H (i) −H (i−1)} (1)
<Claim 4>
In the page image correction device according to any one of claims 1 to 3,
The synthesizing unit cumulatively adds the lengths in the width direction corrected by the second correction unit, so that the n pieces of the divisions in which the length in the binding direction is corrected by the first correction unit. A page image correction apparatus characterized by combining regions.
<Claim 5>
In the page image correction means device according to any one of claims 1 to 4,
A page image correction apparatus, further comprising third correction means for expanding or reducing the image synthesized by the synthesis means in the width direction so as to correspond to the aspect ratio of the page.
<Claim 6>
An imaging process for imaging a book page;
An extraction step of extracting the outline of the page from the image acquired in the imaging step;
A contour correction step for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted in the extraction step;
The entire image of the page corrected in the contour correction step is divided into n in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. A dividing step of dividing the pieces so as to be individually arranged;
A first correction step for correcting the length in the binding direction for each of the divided regions divided in the division step;
For each of the divided regions corrected in the first correction step, an estimation step for estimating the amount of deflection in the normal direction of the divided regions on the actual page;
A second correction step of correcting the length in the width direction for each of the divided regions based on the deflection amount estimated in the estimation step;
A page image correction method comprising: a combining step of combining the n divided regions corrected in the first correction step and the second correction step.
<Claim 7>
A computer that can communicate with an imaging means for imaging a page of a book,
Extraction means for extracting the outline of the page from the image acquired by the imaging means;
Contour correcting means for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted by the extracting means;
The entire image of the page corrected by the contour correcting unit is arranged in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. a dividing means for dividing the data so that n pieces are arranged;
First correcting means for correcting the length in the binding direction for each of the divided areas divided by the dividing means;
Estimating means for estimating the amount of deflection in the normal direction of the divided area on an actual page for each of the divided areas corrected by the first correcting means;
Second correction means for correcting the length in the width direction for each of the divided regions based on the amount of deflection estimated by the estimation means;
A program which functions as a combining unit that combines n divided regions corrected by the first correcting unit and the second correcting unit.

1. Document camera system (page image correction device)
2. Document camera (imaging means)
3 Mounting Table 4 Personal Computer 21 Stand Unit 22 Camera 41 Recording Unit 42 Operating Means 43 ROM
44 RAM
45 CPU (extracting means, contour correcting means, dividing means, first correcting means, estimating means, second correcting means, combining means)
B U1 divided area

Claims (7)

Imaging means for imaging a page of a book;
Extraction means for extracting the outline of the page from the image acquired by the imaging means;
Contour correcting means for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted by the extracting means;
The entire image of the page corrected by the contour correcting unit is arranged in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. a dividing means for dividing the data so that n pieces are arranged;
First correcting means for correcting the length in the binding direction for each of the divided areas divided by the dividing means;
Estimating means for estimating the amount of deflection in the normal direction of the divided area on an actual page for each of the divided areas corrected by the first correcting means;
Second correction means for correcting the length in the width direction for each of the divided regions based on the amount of deflection estimated by the estimation means;
A page image correction apparatus comprising: a combining unit that combines the n divided regions corrected by the first correction unit and the second correction unit. The page image correction apparatus according to claim 1,
The page image correcting apparatus, wherein the estimating means estimates the amount of deflection of the divided area based on a length of the divided area in the binding direction. In the page image correction apparatus according to claim 2,
The bending amount of the i-th divided area is dz (i), the length of the i-th divided area in the binding direction is H (i), and the i-1th divided area is When the length in the binding direction is H (i−1) and an arbitrary coefficient is k,
The page image correcting apparatus according to claim 1, wherein the estimating means estimates the amount of deflection of the i-th divided area using equation (1).
dz (i) = k · {H (i) −H (i−1)} (1) In the page image correction device according to any one of claims 1 to 3,
The synthesizing unit cumulatively adds the lengths in the width direction corrected by the second correction unit, so that the n pieces of the divisions in which the length in the binding direction is corrected by the first correction unit. A page image correction apparatus characterized by combining regions. In the page image correction means device according to any one of claims 1 to 4,
A page image correction apparatus, further comprising third correction means for expanding or reducing the image synthesized by the synthesis means in the width direction so as to correspond to the aspect ratio of the page. An imaging process for imaging a book page;
An extraction step of extracting the outline of the page from the image acquired in the imaging step;
A contour correction step for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted in the extraction step;
The entire image of the page corrected in the contour correction step is divided into n in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. A dividing step of dividing the pieces so as to be individually arranged;
A first correction step for correcting the length in the binding direction for each of the divided regions divided in the division step;
For each of the divided regions corrected in the first correction step, an estimation step for estimating the amount of deflection in the normal direction of the divided regions on the actual page;
A second correction step of correcting the length in the width direction for each of the divided regions based on the deflection amount estimated in the estimation step;
A page image correction method comprising: a combining step of combining the n divided regions corrected in the first correction step and the second correction step. A computer that can communicate with an imaging means for imaging a page of a book,
Extraction means for extracting the outline of the page from the image acquired by the imaging means;
Contour correcting means for correcting the entire image of the page by performing distortion correction based on the four corners of the contour of the page extracted by the extracting means;
The entire image of the page corrected by the contour correcting unit is arranged in a width direction in which a substantially rectangular divided region extending along the stitch direction from the upper end to the lower end of the image of the page is perpendicular to the stitch direction. a dividing means for dividing the data so that n pieces are arranged;
First correcting means for correcting the length in the binding direction for each of the divided areas divided by the dividing means;
Estimating means for estimating the amount of deflection in the normal direction of the divided area on an actual page for each of the divided areas corrected by the first correcting means;
Second correction means for correcting the length in the width direction for each of the divided regions based on the amount of deflection estimated by the estimation means;
A program which functions as a combining unit that combines n divided regions corrected by the first correcting unit and the second correcting unit.

Images