JP2002247348A - Image distortion correcting apparatus, storage medium, program, image distortion correcting method, image reader and image forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct a scanned image of bound originals even when the back of the originals is perpendicular to the direction of horizontal scanning at the time of scanning. SOLUTION: The outer rectangle containing each character is extracted from a binary image obtained by thresholding an scanned image (step S1). The amount of distortion of the scanned image is estimated by using the change in the shape of the extracted rectangle of the character and the change in the background gray level of the scanned image (steps S2-S9). The scanned image or the binary image is subject to magnification on the basis of the estimated amount of distortion (step S10). Thus, if the back of the bound original is perpendicular to the direction of horizontal scanning at the time of scanning, since the shape of the rectangle of each character changes, the scanned image of the bound originals can be corrected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image distortion correcting device, a storage medium, a program, an image distortion correcting method, an image reading device, and an image forming device.

[0002]

2. Description of the Related Art Most originals read using a floodbed scanner are sheet-like originals. An openable and closable pressure plate is provided on a contact glass, and after the original is placed on the contact glass, the pressure plate is closed to close the original. I try to scan. However, manuscripts are not limited to sheet-like ones, and book manuscripts (books, booklets, etc.) may be treated as manuscripts. In such a case, place the book manuscript on the contact glass and scan the manuscript. Will do.

However, when a book document is used as the document, the page binding portion 101 of the book document 100 rises from the contact glass 102 as shown in FIG. , The density of the background becomes distorted, and characters and ruled lines are blurred.

Techniques for correcting the distortion, density abnormality, and blur of a scanned image when a book original is used as described above are disclosed in, for example, JP-A-11-41455 and JP-A-11-414456. It is disclosed in Japanese Unexamined Patent Publication No. Hei 11-414457.

According to the technology disclosed in Japanese Patent Application Laid-Open No. 11-41455, a bookbinding document image reading unit, a bookbinding document shape recognizing unit, and a unit for separating a double-page spread binding area of a bookbinding document based on read data. Means for correcting the distortion of the binding portion image of the spread pages of the bookbinding document from the shape of the bookbinding document, and the binding portion region separating means separates the binding portion region in the image of the facing pages of the binding document. And a means for limiting a range for detecting a position at which a binding region is separated by this means, thereby accurately correcting binding portion distortion.

According to the technique disclosed in Japanese Patent Application Laid-Open No. H11-414456, the shape of a bound page of a bound manuscript is recognized, the shape of a spread book is recognized based on the shape of the page boundary, and the spread book is recognized. Means for correcting the distortion of the double-bound bookbinding portion image from the document shape, and the page boundary portion shape recognizing means is provided in the image density data of the reading line in a direction perpendicular to the line above or below the page boundary of the bookbinding document. By determining an adaptive threshold value for each line from pixel data on the high density side and pixel data on the low density side, and calculating the position of the page boundary from the read data of the page, the distortion of the binding portion is accurately corrected. ing.

Further, according to the technique disclosed in Japanese Patent Application Laid-Open No. 11-414457, there is provided a means for recognizing the shape of a bound document from the read data and correcting the image distortion of the binding portion of the bound document based on the shape of the bound document. This unit calculates Mm = (Ak−A) / (Ak) from the detected page boundary position A, the optical axis position Ak of the reading unit, and the page plane position Ka of the bookbinding original.
By calculating the projection magnification Mm in the direction orthogonal to the page boundaries of the bound manuscript by the formula of -Ka), and performing expansion processing of the read data by the reciprocal magnification of the projection magnification Mm, the binding portion distortion is accurately corrected. Like that.

[0008]

However, the techniques disclosed in JP-A-11-41455, JP-A-11-414456, and JP-A-11-414457 described above,
As shown in FIG. 22, when the page binding unit 101 and the main scanning direction of scanning are parallel, correction is performed using distortion of the outer shape of the page of the scanned image. That is, using the distortion amount indicated by d in FIG. 22 as a clue, based on information on the focal plane distance of the scanner and the position of the optical axis,
This is a technique of estimating the rising shape of the page binding section 101 of the book document 100 and correcting the distortion.

However, according to such a method,
As shown in FIG. 23, this cannot be applied when the page binding unit 101 is perpendicular to the main scanning direction of scanning. Because, in this case, as shown in FIG. 23, the outer shape of the page of the scanned image is not deformed, and the image is merely clogged (characters become thinner) in the direction toward the page binding unit 101. This is because the correction based on the distortion amount d as shown in FIG. 22 cannot be performed.

An object of the present invention is to make it possible to correct a scanned image of a book document even when the page binding portion of the book document is perpendicular to the main scanning direction of scanning.

[0011]

According to a first aspect of the present invention, there is provided an image distortion correcting apparatus for extracting a circumscribed rectangle of each character by cutting out a character from a scan image read by an image reading means for reading an original image. A rectangle extracting means, and a distortion amount estimating means for estimating a distortion amount of the scanned image using a change in the shape of the character circumscribed rectangle extracted by the character circumscribed rectangle extracting means and a change in the background density of the scanned image. Image correcting means for performing an enlargement process on the scanned image based on the distortion amount estimated by the distortion amount estimating means.

Therefore, a circumscribed rectangle of a character is extracted from the scanned image read by the image reading means, and distortion of the scanned image is corrected based on a change in the shape of the circumscribed rectangle of the character. Thereby, even when the page binding portion of the book document is perpendicular to the main scanning direction of the scan,
Since a change appears in the shape of the character circumscribed rectangle, the scan image of the book document can be corrected.

According to a second aspect of the present invention, in the image distortion correcting apparatus of the first aspect, the distortion amount estimating means extends the scan image in a direction perpendicular to a main scanning direction of reading by the image reading means. The scan image is divided into a plurality of strip-shaped areas serving as sides, and the distortion amount of the scan image is estimated based on a statistical feature quantity relating to a set of the character circumscribed rectangles having a center in each strip area.

Therefore, it is possible to reliably detect a change in the shape of the character circumscribed rectangle.

According to a third aspect of the present invention, in the image distortion correction apparatus according to the second aspect, the statistical feature amount is: feature amount = length of a horizontal side of a character circumscribed rectangle / length of a vertical side of a character circumscribed rectangle. It is determined based on the following equation.

Therefore, it is possible to easily catch the change in the shape of the character circumscribed rectangle.

According to a fourth aspect of the present invention, in the image distortion correcting apparatus according to the third aspect, an average value of the respective characteristic amounts of each of the character circumscribed rectangles having a center in the strip area is calculated by using the statistical value in the strip area. It is regarded as a characteristic feature.

Therefore, it is possible to more accurately grasp the change in the shape of the character circumscribed rectangle.

According to a fifth aspect of the present invention, in the image distortion correcting apparatus according to the fourth aspect, the length of the vertical side of the character circumscribed rectangle is predetermined when the statistical feature in the strip region is obtained. If it is smaller than the value, it is excluded in advance.

Therefore, it is possible to eliminate a characteristic amount of a character whose size is originally small, such as a punctuation mark or a symbol in a mathematical expression.

The invention according to claim 6 is the invention according to claim 4 or 5.
In the image distortion correction apparatus described above, when the feature amount of the character circumscribed rectangle is larger than a predetermined value, the statistical feature amount in the strip region is excluded in advance.

Therefore, for example, it is possible to eliminate a feature amount of a character circumscribed rectangle having a very long lateral side of a character circumscribed rectangle, such as a case where adjacent characters are stuck together at the time of rectangle extraction. Will be possible.

The invention according to claim 7 is the invention according to claims 2 to 6
In the image distortion correction device according to any one of the above, the background density of each of the strip regions is measured, and the strip region having the highest background density is regarded as the strip region including the page binding portion of the book document.

Accordingly, the strip area including the page binding portion of the book document is obtained by using the density change of the background, so that the distortion amount of the image near the page binding portion can be more appropriately estimated.

According to an eighth aspect of the present invention, there is provided the image distortion correcting apparatus according to the seventh aspect, wherein the character circumscribed rectangle which can be a calculation target of the statistical feature amount exists in the strip region including the page binding portion. Calculating the feature value of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical feature value in the strip region including the page binding portion. Does not exist, the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists, and the feature amount of the strip region located closest to the strip region including the page binding portion is determined in advance. A value multiplied by the determined constant value is regarded as the statistical feature amount in the strip region including the page binding section.

Therefore, even if there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

According to a ninth aspect of the present invention, in the image distortion correction apparatus according to the eighth aspect, a moving average in a direction in which the position of the strip region changes is obtained, and a filtering process for the change in the characteristic amount is performed based on the moving average. Is applied.

Therefore, it is possible to make the change in the characteristic amount (in the main scanning direction) gentle with the change in the position of the strip region.

According to a tenth aspect of the present invention, in the image distortion correcting apparatus according to the ninth aspect, when performing the filtering process, the window of the filter is positioned on both sides of the strip region including the page binding portion. The window length is adjusted in the vicinity of the page binding section so as not to straddle the area.

Therefore, a change in the characteristic amount in the vicinity of the page binding portion can be appropriately expressed, so that good image correction can be realized.

According to an eleventh aspect of the present invention, in the image distortion correcting apparatus according to any one of the second to tenth aspects, the distortion amount estimating means is configured to calculate the distortion amount of the rectangular region. And the feature amount of the reference strip region is set as a reference feature amount of the entire scan image, and the estimated distortion amount of each strip region is calculated by the following equation: estimated distortion amount = feature amount of each strip region / reference feature amount. Ask based on.

Therefore, it is possible to easily calculate the estimated distortion amount.

According to a twelfth aspect of the present invention, in the image distortion correcting apparatus according to the eleventh aspect, the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value.

Therefore, it is possible to accurately calculate the estimated distortion amount.

According to a thirteenth aspect of the present invention, in the image distortion correcting apparatus according to any one of the first to twelfth aspects, the image correcting means sets a reciprocal of the estimated distortion amount as a correction magnification for each of the strip regions. Enlargement processing is performed in the short side direction of the strip region.

Therefore, it is possible to easily correct the scanned image of the book document.

According to a fourteenth aspect of the present invention, in the image distortion correcting apparatus according to the thirteenth aspect, the correction enlargement ratio at the boundary between the strip regions continuously changes.
The correction magnification at the boundary between the strip regions is corrected in advance.

Therefore, a natural corrected image can be obtained.

According to a fifteenth aspect of the present invention, in the image distortion correction apparatus according to any one of the eleventh to fourteenth aspects, the estimated distortion amount is set to “1” for the strip region where the background density is lower than a predetermined density. ".

Therefore, since the strip region distant from the vicinity of the page binding portion is originally likely to be a region without image distortion, the image of the strip region distant from the vicinity of the page binding portion is excluded from the correction target. And it is possible to prevent unnatural image quality deterioration.

According to a sixteenth aspect of the present invention, in the image distortion correction apparatus according to any one of the eleventh to fourteenth aspects, the reference strip area is obtained only from the strip area whose background density is higher than a predetermined density.

Therefore, the search range of the reference strip region can be limited in advance to the vicinity of the page binding portion.
It is possible to more accurately calculate the estimated distortion amount.

According to a seventeenth aspect of the present invention, in the image distortion correcting apparatus according to any one of the eleventh to sixteenth aspects, the reference strip region is obtained independently for each of left and right pages of the book document, and the estimated distortion amount is calculated. Is also calculated independently for the left and right pages of the book document.

Therefore, it is possible to more accurately calculate the estimated distortion amount.

According to an eighteenth aspect of the present invention, in the image distortion correcting apparatus according to the eighth aspect, the strip including the character circumscribed rectangle that can be a target of calculation of the statistical feature amount and including the page binding portion is provided. Regarding a process in which a value obtained by multiplying the feature amount of the strip region located nearest to the region by a predetermined constant value is regarded as the statistical feature amount in the strip region including the page binding unit, Are performed independently for each page, and the feature amount on the left page side of the book document and the feature value on the right page side of the book document are respectively associated with the strip region including the page binding portion.

Therefore, since the page distortion is greatly different between the left and right sides of the page binding portion of the book original, such as when a page close to the front or back cover of the book is opened, the character circumscribed rectangles on the left and right sides of the page binding portion are different. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

According to a nineteenth aspect of the present invention, in the image distortion correcting apparatus according to the eighteenth aspect, the character circumscribed rectangle that can be a calculation target of the statistical feature is present in the strip region including the page binding portion. Calculating the feature amount of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical circumscription of the character region in the strip region including the page binding portion. If there is no rectangle, the statistical feature amount in the strip region including the page binding unit is the average value of the feature amount on the left page side of the book document and the feature amount on the right page side of the book document. I reckon.

Therefore, even if there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

According to a twentieth aspect of the present invention, in the image distortion correcting apparatus according to the eighteenth aspect, the character circumscribed rectangle which can be a calculation target of the statistical feature amount exists in the strip region including the page binding portion. Calculating the feature amount of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical circumscription of the character region in the strip region including the page binding portion. If there is no rectangle, the statistical feature amount in the strip region including the page binding portion is undecided.

Therefore, when there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature amount of the strip area including the page binding section is avoided.

According to a twenty-first aspect of the present invention, in the image distortion correcting apparatus according to the twentieth aspect, when the statistical feature amount in the strip region including the page binding portion is undecided,
The estimated distortion amount is “1”.

Therefore, since there is no character circumscribed rectangle in the strip region including the page binding portion, the calculation of the statistical feature of the strip region including the page binding portion is avoided. Since there is a high possibility that the region is absent, it is possible to exclude the image of the strip region including the page binding portion having no character circumscribed rectangle from the correction target.

According to a twenty-second aspect of the present invention, in the image distortion correction apparatus according to any one of the eighteenth to twenty-first aspects, when performing the filtering process, the left page of the book document is filtered for the left page. The feature value is used, and the right page feature value is used for filtering the right page of the book document.

Therefore, since the page distortion is greatly different between the left and right sides of the page binding portion of the book original, such as when a page close to the front cover or back cover of the book is opened, the character circumscribed rectangles on the left and right sides of the page binding portion. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

A storage medium according to a twenty-third aspect of the present invention is a storage medium readable by a computer for storing a program for causing a computer to execute image distortion correction of a scan image read by an image reading means for reading a document image. The computer has a character circumscribing rectangle extracting function of extracting characters from the scanned image read by the image reading means and extracting a circumscribing rectangle of each character, and a shape of the character circumscribing rectangle extracted by the character circumscribing rectangle extracting function. And a distortion amount estimating function for estimating the distortion amount of the scanned image using the change in the background density of the scanned image and the distortion amount estimated by the distortion amount estimating function. And a program for executing an image correction function of performing an enlargement process.

Therefore, the circumscribed rectangle of the character is extracted from the scanned image read by the image reading means, and the distortion of the scanned image is corrected based on the change in the shape of the circumscribed rectangle of the character. Thereby, even when the page binding portion of the book document is perpendicular to the main scanning direction of the scan,
Since a change appears in the shape of the character circumscribed rectangle, the scan image of the book document can be corrected.

According to a twenty-fourth aspect of the present invention, in the storage medium according to the twenty-third aspect, a plurality of strip-shaped areas having a long side in a direction perpendicular to a main scanning direction of reading the scanned image by the image reading means. And stores a program for executing the distortion amount estimating function of estimating the distortion amount of the scan image based on the statistical feature amount relating to the set of the character circumscribed rectangles having a center in each strip region.

Therefore, a change in the shape of the character circumscribed rectangle can be reliably detected.

According to a twenty-fifth aspect of the present invention, in the storage medium according to the twenty-fourth aspect, the statistical feature amount is defined as: feature amount = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. Stores the program obtained based on the formula.

Therefore, a change in the shape of the character circumscribed rectangle can be easily grasped.

According to a twenty-sixth aspect of the present invention, in the storage medium according to the twenty-fifth aspect, an average value of the feature amounts of the respective character circumscribed rectangles having a center in the strip region is calculated by using the statistical feature in the strip region. Stores programs that are considered as quantities.

Therefore, it is possible to more accurately catch the change in the shape of the character circumscribed rectangle.

According to a twenty-seventh aspect of the present invention, in the storage medium according to the twenty-sixth aspect, the length of the vertical side of the character circumscribed rectangle is set to a predetermined value when obtaining the statistical feature in the strip region. If it is smaller, a program for eliminating this in advance is stored.

Therefore, it is possible to eliminate a feature amount of a character whose size is originally small, such as a punctuation mark or a symbol in a mathematical expression.

According to a twenty-eighth aspect of the present invention, in the storage medium according to the twenty-sixth aspect or the twenty-seventh aspect, when the statistical feature quantity in the strip area is obtained, the feature quantity of the character circumscribed rectangle is larger than a predetermined value. If it is larger, a program for eliminating this in advance is stored.

Therefore, for example, it is possible to exclude a feature amount of a character circumscribed rectangle having a very long side of the character circumscribed rectangle, such as a case where adjacent characters are attached to each other at the time of rectangle extraction. Will be possible.

According to a twenty-ninth aspect of the present invention, in the storage medium according to any one of the twenty-fourth to twenty-eighth aspects, the background density of each of the strip areas is measured, and the strip area having the highest background density is determined based on the book document. A program that is regarded as the strip area including a page binding section is stored.

Accordingly, the strip area including the page binding portion of the book document is obtained by using the density change of the background, so that the distortion amount of the image near the page binding portion can be more appropriately estimated.

According to a thirtieth aspect of the present invention, in the storage medium according to the twenty-ninth aspect, when the character circumscribed rectangle which can be a target of calculation of the statistical feature amount exists in the strip region including the page binding portion, The feature amount of the strip region including the page binding portion is calculated using the character circumscribed rectangle, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount exists in the strip region including the page binding portion. If not, the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists, and the feature amount of the strip region located closest to the strip region including the page binding portion is determined in advance. A program for storing a value obtained by multiplying a value multiplied by a constant value as the statistical feature amount in the strip region including the page binding portion is stored.

Therefore, even if there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

According to a thirty-first aspect of the present invention, in the storage medium according to the thirty-third aspect, a moving average in a direction in which the position of the strip region changes is obtained, and a filtering process is performed on the change in the characteristic amount based on the moving average. Store the program.

Therefore, it is possible to make the change of the characteristic amount (in the main scanning direction) with respect to the change of the position of the strip region gentle.

According to a thirty-second aspect of the present invention, in the storage medium according to the thirty-first aspect, when the filtering process is performed, a window of a filter is provided in the strip region located on both sides of the strip region including the page binding portion. A program for adjusting the window length is stored in the vicinity of the page binding section so as not to be straddled.

Therefore, it is possible to appropriately express the change in the characteristic amount near the page binding portion, and it is possible to realize good image correction.

According to a thirty-third aspect of the present invention, in the storage medium according to any one of the twenty-fourth to thirty-second aspects, the strip area serving as a reference for calculating a distortion amount of the strip area is defined,
The feature amount of the reference strip region is used as a reference feature amount of the entire scanned image, and the estimated distortion amount of each strip region is obtained based on the following equation: Estimated distortion amount = feature amount of each strip region / reference feature amount. A program for executing the distortion amount estimation function is stored.

Therefore, it is possible to easily calculate the estimated distortion amount.

According to a thirty-fourth aspect of the present invention, in the storage medium according to the thirty-third aspect, a program is stored in which the feature amount having the largest value among the feature amounts of all the strip regions is set as the reference feature amount. .

Therefore, it is possible to accurately calculate the estimated distortion amount.

According to a thirty-fifth aspect of the present invention, in the storage medium according to any one of the twenty-third to thirty-fourth aspects, a reciprocal of the estimated distortion amount for each of the strip regions is used as a correction enlargement factor in a short side direction of the strip region. A program for executing the image correction function for performing enlargement processing is stored.

Therefore, it is possible to easily correct the scanned image of the book document.

According to a thirty-sixth aspect of the present invention, in the storage medium according to the thirty-fifth aspect, the correction magnification at the boundary portion of each of the strip regions changes continuously so that the correction magnification at the boundary portion of each of the strip regions changes continuously. A program for correcting the correction magnification in advance is stored.

Therefore, a natural corrected image can be obtained.

According to a thirty-seventh aspect of the present invention, in the storage medium according to any one of the thirty-third to thirty-sixth aspects, the estimated distortion amount is set to “1” for the strip region having a background density lower than a predetermined density. Store the program to be executed.

Therefore, since the strip region distant from the vicinity of the page binding portion is originally likely to be a region without distortion of the image, the image of the strip region distant from the vicinity of the page binding portion is excluded from the correction target. And it is possible to prevent unnatural image quality deterioration.

According to a thirty-eighth aspect of the present invention, in the storage medium according to any one of the thirty-third to thirty-sixth aspects, a program for obtaining the reference strip area from only the strip areas whose background density is higher than a predetermined density is stored. I do.

Accordingly, it is possible to limit the search range of the reference strip region to the vicinity of the page binding portion in advance,
It is possible to more accurately calculate the estimated distortion amount.

According to a thirty-ninth aspect of the present invention, in the storage medium according to any one of the thirty-third to thirty-third aspects, the reference strip regions are independently obtained for left and right pages of the book document, and the estimated distortion amount is calculated. Also stores a program to be performed independently on the left and right pages of the book document.

Therefore, it is possible to more accurately calculate the estimated distortion amount.

The invention according to claim 40 is the storage medium according to claim 30, wherein the character circumscribed rectangle which can be a target of calculation of the statistical feature amount is present, and the rectangular region including the page binding portion is included. Regarding a process in which a value obtained by multiplying the feature amount of the strip region located in the nearest neighborhood by a predetermined constant value is regarded as the statistical feature amount in the strip region including the page binding unit, the left and right pages of the book document And stores a program for associating the feature value on the left page side of the book document and the feature value on the right page side of the book document with the strip region including the page binding unit.

Therefore, since the page distortion is greatly different between the left and right sides of the page binding portion of the book original, such as when a page close to the front or back cover of the book is opened, the character circumscribed rectangles on the left and right sides of the page binding portion. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

According to a 41st aspect of the present invention, in the storage medium according to the 40th aspect, when the character circumscribed rectangle which can be a calculation target of the statistical feature amount exists in the strip region including the page binding portion, Using the character circumscribed rectangle, the feature amount of the strip region including the page binding portion is calculated, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount is calculated in the strip region including the page binding portion. If not, a program that regards the average value of the feature value on the left page side of the book document and the feature value on the right page side of the book document as the statistical feature value in the strip region including the page binding unit. Is stored.

Therefore, even when the character circumscribed rectangle does not exist in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

[0093] The invention according to claim 42 is the storage medium according to claim 40, wherein the character circumscribed rectangle that can be the calculation target of the statistical feature is present in the strip region including the page binding portion. Using the character circumscribed rectangle, the feature amount of the strip region including the page binding portion is calculated, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount is calculated in the strip region including the page binding portion. If not, a program is stored in which the statistical feature amount in the strip region including the page binding section is undecided.

Therefore, when there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature amount of the strip area including the page binding section is avoided.

The invention according to claim 43 is the storage medium according to claim 40, wherein the estimated distortion amount is set to "1" when the statistical feature amount in the strip region including the page binding portion is undecided. Store the program.

Therefore, since there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature amount of the strip area including the page binding section is avoided. Since there is a high possibility that the region is absent, it is possible to exclude the image of the strip region including the page binding portion having no character circumscribed rectangle from the correction target.

According to a forty-fourth aspect of the present invention, in the storage medium according to any one of the forty-fourth to thirty-fourth aspects, when performing the filtering process, the left page of the book document is filtered for the left page. The amount is used, and a program that uses the feature amount for the right page is stored for filtering the right page of the book document.

Therefore, since the page distortion is greatly different between the left and right sides of the page binding portion of the book original, such as when a page close to the front or back cover of the book is opened, the character circumscribed rectangles on the left and right sides of the page binding portion are different. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

The invention according to claim 45 is a program for causing a computer to execute image distortion correction of a scan image read by an image reading means for reading a document image, wherein a character is converted from the scan image read by the image reading means. A character circumscribing rectangle extracting function of extracting a circumscribing rectangle of each character by extracting The computer causes the computer to execute a distortion amount estimation function for estimating the distortion amount of the scanned image and an image correction function for performing an enlargement process on the scanned image based on the distortion amount estimated by the distortion amount estimation function.

Accordingly, a circumscribed rectangle of a character is extracted from the scanned image read by the image reading means, and distortion of the scanned image is corrected based on a change in the shape of the circumscribed rectangle of the character. Thereby, even when the page binding portion of the book document is perpendicular to the main scanning direction of the scan,
Since a change appears in the shape of the character circumscribed rectangle, the scan image of the book document can be corrected.

The invention according to claim 46 is the program according to claim 45, wherein the scanned image is divided into a plurality of strip-shaped areas whose long sides are perpendicular to the main scanning direction of reading by the image reading means. The computer causes the computer to execute the distortion amount estimation function of estimating the distortion amount of the scanned image based on the statistical feature amount regarding the set of the character circumscribed rectangles having the center in each of the strip regions.

Therefore, a change in the shape of the character circumscribed rectangle can be reliably detected.

The invention according to claim 47 is the program according to claim 46, wherein the statistical feature quantity is calculated by the following formula: feature quantity = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. Based on

Therefore, a change in the shape of the character circumscribed rectangle can be easily grasped.

[0105] The invention according to claim 48 is the program according to claim 47, wherein the average value of the characteristic amounts of the respective character circumscribed rectangles having a center in the strip region is used as the statistical feature amount in the strip region. Consider

Therefore, a change in the shape of the character circumscribed rectangle can be grasped more accurately.

The invention according to claim 49 is the program according to claim 48, wherein the length of the vertical side of the character circumscribed rectangle is smaller than a predetermined value when the statistical feature in the strip region is obtained. In this case, this is excluded in advance.

Therefore, it is possible to eliminate a feature amount of a character whose size is originally small, such as a punctuation mark or a symbol in a mathematical expression.

The invention according to claim 50 is the program according to claim 48 or 49, wherein, when obtaining the statistical feature in the strip region, the feature of the character circumscribed rectangle is larger than a predetermined value. In this case, this is excluded in advance.

Therefore, for example, it is possible to eliminate a feature amount of a character circumscribed rectangle having a very long lateral side of a character circumscribed rectangle, such as a case where adjacent characters are attached to each other at the time of rectangle extraction. Will be possible.

The invention according to claim 51 is the program according to any one of claims 46 to 50, wherein the background density of each of the strip regions is measured, and the strip region having the highest background density is defined as a page of a book document. It is regarded as the strip area including the binding portion.

Therefore, the strip area including the page binding portion of the book document is obtained by using the density change of the background, so that the distortion amount of the image near the page binding portion can be more appropriately estimated.

[0113] The invention according to claim 52 is the program according to claim 51, wherein the character circumscribed rectangle that can be a target of calculation of the statistical feature amount is present in the strip region including the page binding portion. The feature amount of the strip region including the page binding portion is calculated using a character circumscribed rectangle, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion. In the case, the character circumscribing rectangle which can be a target of calculation of the statistical feature amount exists, and a predetermined value is set in the feature amount of the strip region located closest to the strip region including the page binding portion. The value obtained by multiplying the numerical value is regarded as the statistical feature amount in the strip region including the page binding portion.

Therefore, even when there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

The invention according to claim 53 is the program according to claim 52, wherein a moving average in a change direction of the position of the strip region is obtained, and a filtering process for the change in the feature amount is performed based on the moving average.

Therefore, it is possible to make the change of the characteristic amount (in the main scanning direction) gentle with respect to the change of the position of the strip region.

The invention according to claim 54 is the program according to claim 53, wherein, when the filtering process is performed, a window of a filter straddles the strip region located on both sides of the strip region including the page binding portion. The window length is adjusted near the page binding section so as not to be distorted.

Therefore, a change in the characteristic amount in the vicinity of the page binding portion can be appropriately expressed, so that good image correction can be realized.

According to a fifty-fifth aspect of the present invention, in the program according to any one of the thirty-sixth to thirty-fourth aspects, the strip area serving as a reference for calculating the distortion amount of the strip area is determined, and the characteristic of the reference strip area is determined. The amount of distortion is used as a reference feature amount of the entire scanned image, and the estimated distortion amount of each of the strip regions is calculated based on the following equation: Estimated distortion amount = Feature amount of each strip region / Reference feature amount. Let the computer run.

Therefore, it is possible to easily calculate the estimated distortion amount.

The invention according to claim 56 is the program according to claim 55, wherein the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value.

Therefore, it is possible to accurately calculate the estimated distortion amount.

[0123] The invention according to claim 57 is the program according to any one of claims 45 to 56, wherein the reciprocal of the estimated distortion amount for each of the strip areas is expanded in the short side direction of the strip area as a correction expansion rate. And causing the computer to execute the image correction function for performing the processing.

Therefore, it is possible to easily correct the scanned image of the book document.

The invention according to claim 58 is the program according to claim 57, wherein the correction at the boundary between the strip regions is changed so that the correction magnification at the boundary between the strip regions changes continuously. The magnification is corrected in advance.

Therefore, a natural corrected image can be obtained.

The invention according to claim 59 is the program according to any one of claims 55 to 58, wherein the estimated distortion amount is set to "1" for the strip region in which the background density is lower than a predetermined density. .

[0128] Therefore, since the strip region distant from the vicinity of the page binding portion is originally likely to be a region without image distortion, the image of the strip region distant from the vicinity of the page binding portion is excluded from the correction target. And it is possible to prevent unnatural image quality deterioration.

The invention according to claim 60 is the program according to any one of claims 55 to 58, wherein the reference strip area is obtained only from the strip area whose background density is higher than a predetermined density.

Therefore, it is possible to limit the search range of the reference strip region to the vicinity of the page binding portion in advance,
It is possible to more accurately calculate the estimated distortion amount.

The invention according to claim 61 is the program according to any one of claims 55 to 60, wherein the reference strip regions are independently obtained for left and right pages of the book document, and the estimated distortion amount is calculated. This operation is performed independently on the left and right pages of the book document.

Therefore, it is possible to more accurately calculate the estimated distortion amount.

[0133] The invention according to claim 62 is the program according to claim 52, wherein the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists, and the latest information of the strip region including the page binding portion exists. Regarding a process in which a value obtained by multiplying the feature amount of the strip region located by the side by a predetermined constant value is regarded as the statistical feature amount in the strip region including the page binding unit, regarding the left and right pages of the book document Independently, the feature value on the left page side of the book document and the feature value on the right page side of the book document are associated with the strip region including the page binding unit.

[0134] Therefore, the page skew differs greatly between the left and right sides of the page binding portion of the book original, such as when a page close to the front or back cover of a book is opened. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

[0135] The invention according to claim 63 is the program according to claim 62, wherein the character circumscribed rectangle which can be a target of calculation of the statistical feature is present in the strip region including the page binding portion. The feature amount of the strip region including the page binding portion is calculated using a character circumscribed rectangle, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount exists in the strip region including the page binding portion. If not, the average value of the feature value on the left page side of the book document and the feature value on the right page side of the book document is regarded as the statistical feature value in the strip region including the page binding section.

Therefore, even if there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

[0137] The invention according to claim 64 is the program according to claim 62, wherein the character circumscribed rectangle that can be a target of calculation of the statistical feature is present in the strip region including the page binding portion. The feature amount of the strip region including the page binding portion is calculated using a character circumscribed rectangle, and the character circumscribed rectangle that can be the calculation target of the statistical feature amount exists in the strip region including the page binding portion. If not, the statistical feature amount in the strip region including the page binding section is undecided.

Therefore, when there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature amount of the strip area including the page binding section is avoided.

[0139] The invention according to claim 65 is the program according to claim 62, wherein the estimated distortion amount is set to "1" when the statistical feature amount in the strip region including the page binding portion is undecided.

Accordingly, since there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature of the strip area including the page binding section is avoided. Since there is a high possibility that the region is absent, it is possible to exclude the image of the strip region including the page binding portion having no character circumscribed rectangle from the correction target.

The invention according to claim 66 is the program according to any one of claims 62 to 65, wherein in performing the filtering process, the filtering of the left page of the book document is performed using the left-page feature value. Using
The feature value for the right page is used for filtering the right page of the book document.

Accordingly, the page skew differs greatly between the left and right sides of the page binding portion of the book original, such as when a page close to the front or back cover of a book is opened. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

The invention according to claim 67 is an image distortion correction method for correcting distortion of a scan image read by image reading means for reading an original image, wherein characters are cut out from the scan image and a circumscribed rectangle of each character is extracted. And extracting a distortion amount of the scanned image by using a change in the shape of the character circumscribed rectangle extracted in the character circumscribed rectangle extraction process and a change in the background density of the scanned image. A distortion amount estimating step; and an image correcting step of performing an enlargement process on the scan image based on the distortion amount estimated in the distortion amount estimating step.

Accordingly, a circumscribed rectangle of a character is extracted from the scanned image read by the image reading means, and distortion of the scanned image is corrected based on a change in the shape of the circumscribed rectangle of the character. Thereby, even when the page binding portion of the book document is perpendicular to the main scanning direction of the scan,
Since a change appears in the shape of the character circumscribed rectangle, the scan image of the book document can be corrected.

The invention according to claim 68 is the image distortion correction method according to claim 67, wherein the distortion amount estimating step is
The scan image is divided into a plurality of strip-shaped areas each having a long side in a direction perpendicular to the main scanning direction of reading by the image reading unit, and statistics relating to a set of the character circumscribed rectangles each having a center in each strip area. The amount of distortion of the scanned image is estimated based on the characteristic amount.

Therefore, it is possible to reliably detect a change in the shape of the character circumscribed rectangle.

[0147] The invention according to claim 69 is the image distortion correction method according to claim 68, wherein the statistical feature amount is defined as: feature amount = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. It is calculated based on the following equation.

Therefore, a change in the shape of the character circumscribed rectangle can be easily grasped.

[0149] The invention according to claim 70 is the image distortion correction method according to claim 69, wherein the average value of the respective characteristic amounts of each of the character circumscribed rectangles having a center in the strip area is used as the statistical value in the strip area. It is regarded as a characteristic feature.

Therefore, it is possible to more accurately grasp the change in the shape of the character circumscribed rectangle.

The invention according to claim 71 is the image distortion correction method according to claim 70, wherein the length of the vertical side of the character circumscribed rectangle is predetermined when the statistical feature in the strip region is obtained. If it is smaller than the value, it is excluded in advance.

Therefore, it is possible to eliminate a feature amount of a character whose size is originally small, such as a punctuation mark or a symbol in a mathematical expression.

The invention according to claim 72 is the image distortion correction method according to claim 70 or 71, wherein the characteristic amount of the character circumscribed rectangle is predetermined when the statistical characteristic amount in the strip region is obtained. If it is larger than the value, it is excluded in advance.

Therefore, for example, it is possible to exclude a feature amount of a character circumscribed rectangle having an extremely long lateral side of a character circumscribed rectangle, such as a case where adjacent characters are stuck together at the time of rectangle extraction. Will be possible.

The invention according to claim 73 is the image distortion correction method according to any one of claims 68 to 72, wherein the background density of each of the strip regions is measured, and the strip region having the highest background density is determined as a book. This is regarded as the strip region including the page binding portion of the document.

Therefore, the strip area including the page binding portion of the book document is obtained by utilizing the density change of the background, so that the distortion amount of the image near the page binding portion can be more appropriately estimated.

[0157] The invention according to claim 74 is the image distortion correction method according to claim 73, wherein the character circumscribed rectangle that can be a calculation target of the statistical feature is present in the strip region including the page binding portion. Calculating the feature value of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical feature value in the strip region including the page binding portion. Does not exist, the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists, and the feature amount of the strip region located closest to the strip region including the page binding portion is determined in advance. A value multiplied by the determined constant value is regarded as the statistical feature amount in the strip region including the page binding section.

Therefore, even if there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section can be calculated.

The invention according to claim 75 is the image distortion correction method according to claim 74, wherein a moving average in a direction in which the position of the strip region changes is obtained, and a filtering process for the change in the characteristic amount is performed based on the moving average. Is applied.

Therefore, it is possible to make the change in the characteristic amount (in the main scanning direction) gentle with respect to the change in the position of the strip region.

The invention according to claim 76 is the image distortion correction method according to claim 75, wherein the window of the filter is positioned on both sides of the strip region including the page binding portion when performing the filtering process. The window length is adjusted in the vicinity of the page binding section so as not to straddle the area.

Therefore, it is possible to appropriately express the change in the characteristic amount near the page binding portion, and it is possible to realize good image correction.

According to a 77th aspect of the present invention, in the image distortion correction method according to any one of the 68th to 76th aspects, the distortion amount estimating step comprises the step of calculating the distortion amount of the strip region. And the feature amount of the reference strip region is set as the reference feature amount of the entire scan image, and the estimated distortion amount of each strip region is calculated by the following equation: estimated distortion amount = feature amount of each strip region / reference feature amount. Ask based on.

Therefore, it is possible to easily calculate the estimated distortion amount.

The invention according to claim 78 is the image distortion correction method according to claim 77, wherein the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value.

Therefore, it is possible to accurately calculate the estimated distortion amount.

The invention according to claim 79 is the image distortion correction method according to any one of claims 67 to 78, wherein the image correction step uses a reciprocal of the estimated distortion amount as a correction enlargement factor for each of the strip regions. Enlargement processing is performed in the short side direction of the strip region.

Therefore, it is possible to easily correct the scanned image of the book document.

The invention according to claim 80 is the image distortion correction method according to claim 79, wherein the correction enlargement ratio at a boundary portion of each of the strip regions changes continuously.
The correction magnification at the boundary between the strip regions is corrected in advance.

Therefore, a natural corrected image can be obtained.

The invention according to claim 81 is the image distortion correction method according to any one of claims 77 to 80, wherein the estimated distortion amount is set to "1" for the strip region whose background density is lower than a predetermined density. ".

Therefore, since it is highly probable that the strip region distant from the vicinity of the page binding portion is originally a region without image distortion, the image of the strip region distant from the vicinity of the page binding portion is excluded from the correction target. And it is possible to prevent unnatural image quality deterioration.

The invention according to claim 82 is the image distortion correction method according to any one of claims 77 to 80, wherein the reference strip area is obtained only from the strip area whose background density is higher than a predetermined density.

Therefore, it is possible to limit the search range of the reference strip region to the vicinity of the page binding portion in advance,
It is possible to more accurately calculate the estimated distortion amount.

The invention according to claim 83 is the image distortion correction method according to any one of claims 77 to 82, wherein the reference strip regions are independently obtained for left and right pages of the book document, and the estimated distortion amount is calculated. Is also calculated independently for the left and right pages of the book document.

Therefore, it is possible to calculate the estimated distortion amount more accurately.

The invention according to claim 84 is based on the image distortion correction method according to claim 74, wherein the character circumscribed rectangle which can be a target of calculation of the statistical feature amount exists, and the strip including the page binding portion is provided. Regarding a process in which a value obtained by multiplying the feature amount of the strip region located nearest to the region by a predetermined constant value is regarded as the statistical feature amount in the strip region including the page binding unit, Are performed independently for each page, and the feature amount on the left page side of the book document and the feature value on the right page side of the book document are respectively associated with the strip region including the page binding portion.

[0178] Therefore, the page skew differs greatly between the left and right sides of the page binding portion of the book original, such as when a page near the front or back cover of a book is opened. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

[0179] The invention according to claim 85 is the image distortion correction method according to claim 84, wherein the character circumscribed rectangle which can be a calculation target of the statistical feature is present in the strip region including the page binding portion. Calculating the feature amount of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical circumscription of the character region in the strip region including the page binding portion. If there is no rectangle, the statistical feature amount in the strip region including the page binding unit is the average value of the feature amount on the left page side of the book document and the feature amount on the right page side of the book document. I reckon.

Therefore, even if there is no character circumscribed rectangle in the strip region including the page binding portion, the statistical feature amount of the strip region including the page binding portion can be calculated.

The invention according to claim 86 is the image distortion correction method according to claim 84, wherein the character circumscribed rectangle which can be a target of calculation of the statistical feature is present in the strip region including the page binding portion. Calculating the feature amount of the strip region including the page binding portion using the character circumscribed rectangle, and calculating the statistical circumscription of the character region in the strip region including the page binding portion. If there is no rectangle, the statistical feature amount in the strip region including the page binding portion is undecided.

Therefore, when there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature amount of the strip area including the page binding section is avoided.

The invention according to claim 87 is based on the image distortion correction method according to claim 86, wherein the statistical feature amount in the strip region including the page binding portion is undecided.
The estimated distortion amount is “1”.

Accordingly, since there is no character circumscribed rectangle in the strip area including the page binding section, the calculation of the statistical feature of the strip area including the page binding section is avoided. Since there is a high possibility that the region is absent, it is possible to exclude the image of the strip region including the page binding portion having no character circumscribed rectangle from the correction target.

The invention according to claim 88 is the image distortion correction method according to any one of claims 84 to 87, wherein the filtering of the left page of the book document is performed using a left page filter when performing the filtering process. The feature value is used, and the right page feature value is used for filtering the right page of the book document.

Therefore, as in the case where a page close to the front or back cover of a book is opened, the page distortion is greatly different between the left and right sides of the page binding section of the book document. Even when the difference between the distortion amounts becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing.

[0187] The image reading apparatus according to the 89th aspect of the present invention,
2. An image reading means for reading a document image, and correcting a scanned image read by the image reading means.
23. The image distortion correction device according to any one of claims 22 to 22.

Therefore, an image reading apparatus having the same effect as the invention according to any one of claims 1 to 22 is obtained.

The image forming apparatus according to the ninety aspect of the present invention
2. An image reading means for reading a document image, and correcting a scanned image read by the image reading means.
23. An image distortion correction apparatus according to any one of claims 22 to 22, and an image printing apparatus that prints an image on paper based on image data output from the image distortion correction apparatus.

Therefore, an image forming apparatus having the same effects as the invention according to any one of claims 1 to 22 is obtained.

[0191]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described with reference to FIGS. The image distortion correcting apparatus according to the present embodiment is provided in a digital copying machine which is an image forming apparatus, and a scanner unit of the digital copying machine is applied as an image reading apparatus.

Here, FIG. 1 is a vertical sectional front view showing the structure of the scanner section 1. As shown in FIG. As shown in FIG.
Is a contact glass 2 on which a document is placed, a first traveling body 5 including an exposure lamp 3 and a first reflection mirror 4 for exposing the document, a second reflection mirror 6 and a third reflection mirror 7
A second traveling body 8 comprising a CCD, a CCD (Charge Coupled Device) 9 as an image sensor for reading an image of a document, and a lens unit 10 for forming an image on the CCD 9.
It becomes the standard for placing the manuscript and the contact glass 2
An original scale 11 for preventing displacement and separation of the original, a white reference plate 12 for shading correction and a frame 14 installed below the original scale 11 are provided. CC
D9 is formed on the sensor board 13.

When scanning a document, the first traveling body 5 and the second traveling body 8 are moved in the sub-scanning direction by the stepping motor 24 (see FIG. 3). That is, the first traveling body 5 and the second traveling body 8 travel under the contact glass 2 to expose and scan the original with the exposure lamp 3, and reflect the reflected light to the first reflecting mirror 4, the second reflecting mirror 6, and The light is reflected by the third reflection mirror 7 and passes through the lens unit 10 to the CCD 9.
Image. Here, an image reading unit is realized.

Such a scanner unit 1 is a printer unit (an image printing apparatus) that forms an image on paper in an electrophotographic manner, for example, in accordance with image data based on an image of a document read by the scanner unit 1. (Not shown). FIG. 2 is a perspective view showing an upper portion of the digital copying machine 16 on which the scanner unit 1 is mounted. As shown in FIG. 2, the scanner unit 1 is provided with a pressure plate 17 that can be opened and closed with respect to the contact glass 2, and an open / close sensor 18 that detects whether the pressure plate 17 is opened or closed. In addition, as a printer provided in the digital copying machine 16, in addition to an electrophotographic system, an inkjet system,
Sublimation type thermal transfer method, silver halide photographic method, direct thermal recording method,
Various printing methods such as a fusion type thermal transfer method can be applied. Since the specific configuration is well known, detailed description is omitted.

FIG. 3 is a block diagram showing the electrical connection of the control system of the scanner section 1. As shown in FIG. As shown in FIG. 3, the control system includes a main control unit 19 that controls the entire scanner unit 1, an image processing unit 20 that is a circuit that performs various types of image processing on image data read by the CCD 9, Running body 5
And a traveling body control unit 21 which is a circuit for controlling the second traveling body 8 and various operations to the digital copying machine 16 are received,
An operation panel 22 for displaying various messages;
A memory 23 for storing image data read by the CD 9, predetermined data, and the like is connected. The operation panel 22 is provided with a copy start key for declaring the start of copying. In addition, the traveling body control unit 21 includes the exposure lamp 3, the first traveling body 5 and the second traveling body 8.
Home position sensor (HP sensor) for detecting whether or not the first traveling body 5 and the second traveling body 8 are at the home position.
25 and the open / close sensor 18 are connected.

FIG. 4 is a block diagram showing the basic internal configuration of the image processing section 20. As shown in FIG.
The image processing unit 20 includes an analog video processing unit 26 that performs an amplification process and a digital conversion process of an analog image signal obtained by reading a document by the CCD 9, a shading correction processing unit 27 that performs a shading correction process, and a digital image signal after the shading correction process. An image data processing unit 28 that performs various image data processing such as MTF correction, scaling processing, and γ correction to generate a scan image, and realizes a scan image distortion correction function that is a characteristic function of the present embodiment. The image distortion correcting section 29 is provided. The digital image signal after the image processing as described above is transmitted to the printer unit via the main control unit 19, and is provided for image formation.

The main control unit 19, as shown in FIG.
CPU (Central Processing U) that controls each part centrally
nit) 31. The CPU 31 has a BIO
ROM (Read-only memory) that stores
Only Memory) 32 and a RAM (Ra) that stores various data in a rewritable manner and functions as a work area for the CPU 31.
ndom Access Memory) 33 is connected by a bus 34 to constitute a microcomputer. Further, the bus 34 has an HDD 35 storing a control program,
CD (Compact Disc)-CD-R that reads ROM37
The OM drive 36 is connected to an interface (I / F) 38 that controls communication with a printer unit and the like.

A CD-ROM 37 shown in FIG. 5 embodies the storage medium of the present invention, and stores a predetermined control program. The CPU 31 has a CD-ROM 37
Is read by the CD-ROM drive 36 and installed in the HDD 35. Accordingly, the main control unit 19 is in a state where various processes described below can be performed.

The storage medium is a CD-ROM 3
In addition to the media 7, various types of media such as various types of optical disks such as DVDs, various types of magneto-optical disks, various types of magnetic disks such as floppy (registered trademark) disks, and semiconductor memories can be used. You can also download programs from networks such as the Internet,
35 may be installed. In this case, a storage device storing a program in the server on the transmission side is also a storage medium of the present invention. Note that the program may operate on a predetermined OS (Operating System), and in that case, the OS may substitute a part of execution of various processes to be described later, Predetermined application software such as software and OS
And the like may be included as a part of a group of program files constituting the above.

Next, the CP provided in the main control unit 19
The contents of various processes executed by U31 based on the control program will be described. Here, of the processing executed by the CPU 31, only the distortion correction processing of the scan image in the image distortion correction unit 29 that realizes the distortion correction function of the scan image, which is a characteristic function of the scanner unit 1 of the present embodiment, is described. explain.

FIG. 6 is a flowchart schematically showing the flow of the scan image distortion correction processing. Here, as shown in FIG. 7, the book document 40 is placed on the contact glass 2 so that the page binding section 41 and the main scanning direction of the scanning of the scanner section 1 are perpendicular to each other. Correction of the distortion of the scanned image in the case will be described.

As shown in FIG. 6, in step S1, the scanned image output from the image data processing section 28 is binarized, and based on the binarized image data (binary image). Then, a circumscribed rectangle A (see FIG. 8) of the character is extracted. Here, the function of the character circumscribed rectangle extracting means is executed. Here, the character recognition processing is a well-known technique, and a description thereof will be omitted. The reason why the circumscribed rectangle A of the character is extracted is to correct the distortion of the scanned image based on the change in the shape of the circumscribed rectangle A of the character. Here, as shown in FIG. 8, the length w of the horizontal side, the length h of the vertical side, and the center B of the character of the character circumscribed rectangle A are defined. Here, the center B of the character is the intersection of the diagonal lines of the circumscribed rectangle A. Here, the image data processing unit 28
The circumscribed rectangle A of the character is extracted after the scan image output from is binarized, but the present invention is not limited to this.
A circumscribed rectangle A of a character may be extracted from a scanned image (multi-valued image).

Subsequently, as shown in FIG. 9, after the binarized image is divided into a plurality of strip-shaped areas C in a direction parallel to the page binding section 41 of the book document 40 (step S2), each strip area With respect to C, a feature amount related to the character circumscribed rectangle A included therein is obtained (step S3). Here, the character circumscribed rectangle A included in a certain strip area C is a circumscribed rectangle A whose center is included in the strip area C. For example, the circumscribed rectangle A included in the strip area C1 in FIG.
Is a shaded rectangle in the figure.

The feature value of the character circumscribing rectangle A is obtained based on (length of horizontal side of character) / (length of vertical side of character) = w / h. That is, for each strip area C,
The average value of the w / h values of all the character circumscribed rectangles A included therein is used as the feature value of the strip region C.

However, it is sometimes inappropriate to simply calculate the average value of w / h. Some characters are originally small in size, such as punctuation marks and symbols in mathematical formulas, w / h
Some values are unstable. In addition, in extracting a rectangle, adjacent characters may be stuck together and extracted, resulting in a character circumscribed rectangle A having an extremely large w. When a feature amount is obtained, it is necessary to eliminate such special characters and those having an extremely large w in advance. Therefore, in the subsequent step S4, a threshold value is determined in advance, the character circumscribed rectangle A having a value of h smaller than the threshold value is excluded in advance, and a threshold value for the w / h ratio is determined in advance. Circumscribed rectangle A whose value is larger than the threshold
Are also eliminated in advance. For example, the shaded character circumscribed rectangle A shown in FIG. 10 is removed in advance.

In the following step S5, the average value of w / h of the character circumscribed rectangle A in each strip region C is obtained after excluding the extreme character circumscribed rectangle A as described above. Figure 1
1 shows an example of the average value of w / h of the circumscribed rectangle A in each strip region C. Note that the strip area C2 in FIG. 11 is a strip area including the page binding section 41 of the book document 40.

Subsequently, the page binding section 4 of the book document 40
It is determined whether or not the character circumscribed rectangle A exists in the strip area C2 including 1 (step S6). This is because, as shown in FIG. 10, the character circumscribed rectangle A generally does not exist near the page binding portion 41 of the book document 40 in many cases. When the character circumscribed rectangle A exists in the strip area C2 including the page binding section 41 of the book document 40 (step S6).
Y), since the characteristic amount has been calculated using the character circumscribed rectangle A, the process directly proceeds to step S8.

On the other hand, the page binding section 41 of the book document 40
If the character circumscribed rectangle A does not exist in the strip area C2 including the character (N in step S6), the process proceeds to step S7, and the feature amount of the strip area C2 including the page binding unit 41 of the book document 40 is obtained. The page binding unit 41 of the book document 40
Of the strip region C2 including, for example, is realized by obtaining a background density change of a scanned image (for example, a monochrome multivalued image) for each strip region C, and obtaining a pixel value of the lightest density in the strip region C. Is done. FIG. 12 shows an example in which the background density change is obtained. The strip region having the highest background density is regarded as the strip region C2 including the page binding portion 41 of the book document 40.

When the scan image is a color multi-valued image, the identification of the strip region C2 including the page binding portion 41 of the book document 40 focuses on, for example, any one of the RGB components (for example, the G component). The identification may be performed using the background density of the G component. Alternatively, the RGB area may be converted into a luminance component and a color difference component by color conversion, and the strip region C2 including the page binding portion 41 of the book document 40 may be identified using the luminance component.

The feature amount of the strip area C2 including the page binding section 41 of the book document 40 is determined as follows.
Here, there is a character circumscribing rectangle A that can be a statistical feature amount calculation target, and a predetermined constant value for the feature amount of the strip region C that is the closest to the strip region C2 including the page binding unit 41. Is calculated as the characteristic amount in the strip region C2 including the page binding portion 41 of the book document 40. That is, in the example shown in FIG. 11, the character circumscribed rectangle A exists in the left and right strip regions C3 and C4 of the strip region C2 including the page binding unit 41 of the book document 40, so that whichever is more appropriate. An amount is selected (here, the circle on the right side) is multiplied by a predetermined constant value (here, 0.5), and this is multiplied by the characteristic amount of the strip region C2 including the page binding portion 41 of the book document 40. And

In the following step S8, an appropriate filtering process for the characteristic amount of each strip region C, for example, a process of obtaining a moving average in the direction of change in the position of the strip region C (ie, the main scanning direction) is performed. The change in the feature amount (in the main scanning direction) with respect to the change in the position of the strip region C is made gentle. However, also in this case, special processing is required in the vicinity of the page binding section 41 of the book document 40. This is because if filtering is performed using windows whose lengths are all equal in the main scanning direction, the sharpness of the change in the feature amount near the page binding portion 41 of the book document 40 is lost.

Here, FIG. 13 shows a result obtained by performing a filtering process on the feature amount of each strip region C shown in FIG. 11 using a window whose length is all five. FIG.
As shown in the figure, when the filtering process is performed using a window whose length is all 5, the book original 40
The characteristic amount (w / h) in the vicinity of the page binding unit 41 changes too slowly. In such a case, appropriate image correction near the page binding section 41 of the book document 40 becomes impossible.

Therefore, in the present embodiment, at the time of the filtering process, the window of the filter is set so as not to straddle the strip areas C3 and C4 on both sides of the strip area C2 including the page binding section 41 of the book document 40. The window length is adjusted near the page binding section 41. Here, FIG.
Is a graph showing the result of performing the filtering process by adjusting the window length near the page binding unit 41. Figure 1
As shown in FIG. 4, when the window length is adjusted in the vicinity of the page binding section 41, a change in the characteristic amount (w / h) in the vicinity of the page binding section 41 can be appropriately expressed, so that good image correction can be realized. .

In the following step S9, the estimated distortion amount of each strip region C is calculated. The calculation method of the estimated distortion amount of each strip region C is as follows.

First, a strip area (reference strip area) serving as a reference for calculating the distortion amount of the strip area is determined. Here, a strip area C that is considered to have the smallest distortion, for example, a strip area C having the largest characteristic amount (w / h) is set as a reference strip area. This processing may be performed in common for the left and right pages, but the reference strip area may be determined independently for the left and right pages.
FIG. 14 shows an example in which the reference strip areas are defined independently for the left and right sides. The strip area C marked with a circle is the reference strip area, the left reference feature amount is “Lw0 / Lh0”, and the right reference The feature amount is indicated by “Rw0 / Rh0”, respectively.

Next, the feature value w0 / h0 of the reference strip region is used as the reference feature value of the entire scanned image, and (the feature value of each strip region) / (reference feature value) = (w / h)
The value of / (w0 / h0) is calculated as the estimated distortion amount of each strip region.

That is, in steps S2 to S9, the function of the distortion amount estimating means is executed.

Note that the page binding section 41 of the book document 40
If the strip area C near the outside of the page deviating from the page is set as the reference strip area, there may be a case where it is not possible to calculate an appropriate estimated distortion amount due to a large difference in font and print size from the vicinity of the page binding section 41 of the book document 40. Can be When such an image is targeted, it is effective to limit the search range of the reference strip region to the vicinity of the page binding section 41 of the book document 40 in advance. In order to realize this, the reference strip region may be obtained only from the strip region C whose background density is higher than a predetermined density.

Finally, the scan image is enlarged in the short side direction (main scanning direction) of the strip region C to correct distortion near the page binding section 41 of the book document 40 (step S10: image correction). means). In that case, the magnification is
The reciprocal of the estimated distortion amount calculated in step S9, that is, (reference characteristic amount) / (characteristic amount of each strip region) = (w0 / h)
0) / (w / h). Here, when the above-mentioned reference strip region is defined for both the left and right sides, this enlargement ratio is also calculated based on the reference feature amounts common to the left and right. To do it. FIG.
9 shows the corrected magnification calculated based on the feature amounts shown in FIG.

[0220] In this case, too, the strip region C of the book document 40 which is remote from the vicinity of the page binding portion 41 is likely to be a region where the image is not originally distorted. is there. This is because there is a possibility that unnatural distortion may occur due to the enlargement processing. In order to prevent this, the estimated distortion amount is set to “1” for the strip region C whose background density is lower than a predetermined density.

Further, when a common correction magnification rate is applied in the strip area C, the correction image becomes unnatural because the correction magnification rate at the boundary between the adjacent strip areas C becomes discontinuous. Therefore, the correction magnification is corrected in advance so that the correction magnification at the boundary between the adjacent strip regions C changes continuously. This is done, for example, by plotting the correction enlargement ratio of the central portion of the strip region C shown in FIG. 15 as points indicating the reciprocal of the estimated distortion amount, connecting these points with line segments, and linearly complementing the other portions. This can be realized by setting the correction magnification. Through the above processing, the correction magnification rate of the scanned image in the main scanning direction is determined.

Note that the image enlargement processing is executed by using, for example, a convolution method of a cubic function often used as a copy scaling function.

Next, as shown in FIG. 16, the book original 4 is displayed as if a page close to the front or back cover of the book was opened.
A description will be given of a process in the case where the difference in the amount of distortion of the character circumscribed rectangle A between the left and right sides of the page binding unit 41 is large because the left and right sides of the page binding unit 41 have significantly different pages.

FIG. 17 shows an example of the characteristic amount (w / h) of each strip region C in a case where the page distortion is greatly different between the left and right sides of the page binding section 41 of the book document 40. FIG.
As shown in FIG. 7, when the page distortion is greatly different between the left and right sides of the page binding section 41 of the book document 40, the characteristic amounts ( A difference occurs in (circle mark in FIG. 17). That is, in this case, the difference in the distortion amount of the character circumscribed rectangle A in the strip areas C on both sides adjacent to the strip area C2 including the page binding section 41 increases.

In order to cope with such a case, in step S7 described above, a character circumscribed rectangle A that can be a target of calculation of a statistical feature exists, and the page binding section 41
The value calculated by multiplying the feature value of the strip area C closest to the strip area C2 including a predetermined constant value by a predetermined constant value is used as the statistical value in the strip area C2 including the page binding unit 41 of the book document 40. It is effective to perform the processing regarded as the feature amount independently for the left and right pages, and to associate the two feature amounts of the left page side feature amount and the right page side feature amount. That is, in the example shown in FIG. 17, the characteristic amount of each of the left and right strip regions C3 and C4 of the page binding unit 41 is multiplied by a predetermined constant value (here, 0.5).
Both of them are associated with the strip region C2 including the page binding section 41.

Then, when performing the filtering process on the feature value in the subsequent step S8, the feature value for the left page is used for filtering the left page, and the feature value for the right page is used for filtering the right page. . Here, FIG. 18 shows a result obtained by performing a filtering process on the feature amount of each strip region C shown in FIG.

In this case, how to set the characteristic amount of the strip area C2 including the page binding section 41 becomes a problem. For example, the strip area C2 including the page binding section 41 is associated as described above. There is a method of setting the average value of the feature values of the left and right pages to this, and FIG. 18 shows this case. Also, at this stage, the page binding section 41
, The feature amount of the strip region C2 including “.” Is set to be undetermined, and when the estimated distortion amount is set later, it is set to “1”.

The following steps S9 and S10 are not different from those described above, and will not be described.

Here, FIG. 19 is a plan view showing an image before correcting the distortion near the page binding section 41 of the book document 40, and FIG. 20 is an image after correcting the distortion near the page binding section 41 of the book document 40. FIG. According to the above processing, the page binding unit 4 of the book document 40 shown in FIG.
The distortion of the scanned image in the case where 1 is perpendicular to the main scanning direction of the scan is corrected as shown in FIG.

In the present embodiment, the image distortion correcting device is provided in the digital copying machine 16 as an image forming apparatus, and the image distortion correcting process is performed on the scanned image read by the scanner unit 1 of the digital copying machine 16. , But is not limited to this. For example, an image scanner having image reading means for reading an original image is connected to a personal computer, and a storage medium such as a CD-ROM 3 is stored in an HDD of the personal computer.
Even if the image distortion correction apparatus is configured by installing the program stored in the storage device 7, the same operation and effect as those described above can be obtained. Also, an image distortion correction apparatus is configured by installing a program stored in a CD-ROM 37 as a storage medium into an HDD of a personal computer, and performs distortion correction processing on a scanned image read in advance by image reading means. You may do it.

[0231]

According to the present invention, a circumscribed rectangle of a character is extracted from the scanned image read by the image reading means, and scanning is performed based on a change in the shape of the character circumscribed rectangle. By correcting the image distortion, even if the page binding portion of the book document is perpendicular to the main scanning direction of the scan, the shape of the character circumscribed rectangle changes, so the scanned image of the book document is Can be corrected.

According to the invention described in claims 2, 24, 46, and 68, a change in the shape of the character circumscribed rectangle can be reliably detected.

According to the invention described in claims 3, 25, 47 and 69, a change in the shape of the character circumscribed rectangle can be easily grasped.

According to the invention as set forth in claims 4, 26, 48, and 70, a change in the shape of the character circumscribed rectangle can be grasped more accurately.

According to the inventions described in claims 5, 27, 49, and 71, it is possible to eliminate a feature amount of a character whose size is originally small, such as a punctuation mark or a symbol in a mathematical expression.

According to the inventions described in claims 6, 28, 50, and 72, the length of the horizontal side of the character circumscribed rectangle is extremely large, for example, when adjacent characters are attached to each other when the rectangle is extracted. Can be excluded from the character circumscribed rectangle having a long character.

According to the seventh, 29, 51, and 73 aspects of the present invention, a strip region including a page binding portion of a book document is obtained by utilizing the density change of the background, thereby distorting the image near the page binding portion. The quantity can be estimated more appropriately.

According to the eighth, thirty-second, and seventy-fourth aspects of the present invention, the statistical characteristics of the strip area including the page binding part can be obtained even when the character circumscribed rectangle does not exist in the strip area including the page binding part. The amount can be calculated.

According to the ninth, third, fifth and ninth aspects of the present invention, the change in the characteristic amount (in the main scanning direction) with respect to the change in the position of the strip region can be made gentle.

According to the inventions described in claims 10, 32, 54, and 76, a change in the characteristic amount in the vicinity of the page binding portion can be appropriately expressed, so that good image correction can be realized.

According to the inventions described in claims 11, 33, 55, and 77, the estimated distortion amount can be easily calculated.

According to the twelfth aspect, the estimated distortion amount can be accurately calculated.

According to the thirteenth, thirty-fifth, fifty-seventh and seventy-ninth aspects of the present invention, a scanned image of a book document can be easily corrected.

According to the invention described in claims 14, 36, 58 and 80, a natural corrected image can be obtained.

According to the inventions described in claims 15, 37, 59, and 81, the strip region far from the vicinity of the page binding portion is likely to be a region in which the image is not originally distorted. The image of the strip region distant from the vicinity of the portion can be excluded from the correction target, and unnatural image quality degradation can be prevented.

According to the inventions of claims 16, 38, 60, and 82, the search range of the reference strip region can be limited to the vicinity of the page binding portion in advance, and the estimated distortion amount can be calculated more accurately. Can be.

According to the present invention, the estimated distortion amount can be calculated more accurately.

According to the inventions described in claims 18, 40, 62, and 84, the distortion of the page on the left and right sides of the page binding portion of the book document is large, such as when a page close to the front or back cover of a book is opened. Even if the difference between the distortion amounts of the character circumscribed rectangles on the left and right sides of the page binding portion is large, it is possible to obtain a better correction result by performing the distortion amount estimation processing independently of each other. it can.

According to the inventions set forth in claims 19, 41, 63, and 85, the statistical characteristics of the strip area including the page binding section even when no character circumscribed rectangle exists in the strip area including the page binding section. The amount can be calculated.

According to the twentieth, twenty-second, sixty-fourth, and eighty-sixth aspects, when there is no character circumscribed rectangle in the strip area including the page binding part, the statistical feature amount of the strip area including the page binding part is determined. Calculation can be avoided.

According to the inventions described in claims 21, 43, 65 and 87, since there is no character circumscribed rectangle in the strip area including the page binding section, the statistical feature amount of the strip area including the page binding section is calculated. Has been avoided, and since this area is likely to be an area without image distortion,
An image of a strip region including a page binding portion having no character circumscribed rectangle can be excluded from the correction target.

According to the inventions of claims 22, 44, 66, and 88, the distortion of the page on the left and right sides of the page binding portion of the book document is large, such as when a page close to the front or back cover of the book is opened. Due to the difference, even when the difference between the distortion amounts of the character circumscribed rectangles on the left and right sides of the page binding portion becomes large, it is possible to obtain a better correction result by independently performing the distortion amount estimation processing. it can.

According to the invention of claim 89, claim 1
According to the present invention, an image reading apparatus having the same effects as the invention described in any one of Items 22 to 22 can be obtained.

According to the nineteenth aspect, the first aspect is provided.
An image forming apparatus having the same effects as the invention according to any one of the above-described aspects can be obtained.

[Brief description of the drawings]

FIG. 1 is a vertical sectional front view showing a configuration of a scanner unit according to an embodiment of the present invention.

FIG. 2 is a perspective view showing an upper part of a digital copier equipped with a scanner unit.

FIG. 3 is a block diagram illustrating an electrical connection of a control system of the scanner unit.

FIG. 4 is a block diagram illustrating a basic internal configuration of an image processing unit.

FIG. 5 is a block diagram showing an electrical connection of a main control unit.

FIG. 6 is a flowchart schematically showing a flow of a scan image distortion correction process.

FIG. 7 is a perspective view showing a state in which a book document is placed on a contact glass of a scanner unit.

FIG. 8 is an explanatory diagram showing extracted character circumscribed rectangles;

FIG. 9 is an explanatory diagram illustrating a state in which the binarized image is divided into a plurality of strip-shaped areas in a direction parallel to a page binding portion of a book document.

FIG. 10 is an explanatory diagram illustrating a character circumscribed rectangle that is excluded in advance.

FIG. 11 is a graph showing an example of an average value of feature values of a circumscribed rectangle in each strip region.

FIG. 12 is a graph showing an example of a change in background density in each strip region.

13 is a graph showing a result of performing a filtering process on a feature amount of each strip region shown in FIG. 11 using a window whose length is all 5;

14 is a graph showing a result obtained by performing a filtering process by adjusting a window length near a page binding portion with respect to a feature amount of each strip region shown in FIG. 11;

FIG. 15 is a graph showing a corrected magnification calculated based on the feature amounts shown in FIG. 14;

FIG. 16 is a front view showing a state in which a book document is placed on a contact glass with its first page opened.

FIG. 17 is a graph showing an example of a feature amount of each strip region in a case where page distortion is greatly different between left and right of a page binding portion of a book document.

18 is a graph showing a result of performing a filtering process on the feature amount of each strip region shown in FIG.

FIG. 19 is a plan view showing an image before correcting distortion near a page binding portion of a book document.

FIG. 20 is a plan view showing an image in which distortion near a page binding portion of a book document has been corrected.

FIG. 21 is a front view showing a state in which a book document is placed on a contact glass.

FIG. 22 is an explanatory diagram showing distortion of the outer shape of the page of the scanned image when the page binding section and the main scanning direction of scanning are parallel.

FIG. 23 is an explanatory diagram illustrating a scanned image in a case where the page binding section and the main scanning direction of scanning are perpendicular to each other.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image reading device 16 Image forming device 37 Storage medium 40 Book original 41 Page binding part A Character circumscribed rectangle B Center of character circumscribed rectangle C Strip area C2 Strip area including page binding part

Claims (90)

[Claims] 1. A character circumscribing rectangle extracting means for extracting characters from a scanned image read by an image reading means for reading a document image and extracting a circumscribing rectangle of each character, and a character circumscribing rectangle extracted by the character circumscribing rectangle extracting means. Distortion amount estimating means for estimating a distortion amount of the scan image using a change in the shape of a rectangle and a change in the background density of the scan image; and the scan based on the distortion amount estimated by the distortion amount estimator. An image distortion correction device comprising: an image correction unit that performs an enlargement process on an image. 2. The distortion amount estimating means divides the scanned image into a plurality of strip-shaped areas each having a long side in a direction perpendicular to the main scanning direction of reading by the image reading means, and 2. The image distortion correction device according to claim 1, wherein the distortion amount of the scan image is estimated based on a statistical feature amount regarding a set of the character circumscribed rectangles having a center at the center. 3. The image distortion correction apparatus according to claim 2, wherein the statistical feature amount is obtained based on an equation: feature amount = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. . 4. The image distortion correction apparatus according to claim 3, wherein an average value of the characteristic amounts of the respective character circumscribed rectangles having a center in the strip region is regarded as the statistical feature amount in the strip region. 5. The image distortion according to claim 4, wherein when calculating the statistical feature amount in the strip region, if the length of the vertical side of the character circumscribed rectangle is smaller than a predetermined value, the image distortion is eliminated in advance. Correction device. 6. The method according to claim 4, wherein when calculating the statistical feature in the strip region, if the feature of the character circumscribed rectangle is larger than a predetermined value, the feature is excluded in advance.
The image distortion correction apparatus according to the above. 7. The strip area having the highest background density is measured as the strip area including a page binding portion of a book document, and the background density of each of the strip areas is measured. The image distortion correction apparatus according to the above. 8. When the character circumscribed rectangle that can be a target of calculating the statistical feature quantity exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, and when the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the statistical feature amount can be calculated. The strip including the page binding section is a value in which a character circumscribed rectangle is present, and a value obtained by multiplying the feature amount of the strip area located closest to the strip area including the page binding section by a predetermined constant value. The image distortion correction device according to claim 7, wherein the image distortion correction device regards the statistical feature amount in a region. 9. The image distortion correction apparatus according to claim 8, wherein a moving average in a direction in which the position of the strip region changes is obtained, and filtering processing for the change in the feature amount is performed based on the moving average. 10. When performing the filtering process,
The image distortion correction device according to claim 9, wherein a window length is adjusted near the page binding portion so that a window of the filter does not straddle the strip regions located on both sides of the strip region including the page binding portion. 11. The distortion amount estimating means determines the strip region as a reference for calculating the distortion amount of the strip region, sets the feature amount of the reference strip region as a reference feature amount of the entire scan image, The image distortion correction device according to any one of claims 2 to 10, wherein the estimated distortion amount of each strip region is obtained based on an equation: estimated distortion amount = feature amount of each strip region / reference feature amount. 12. The image distortion correction device according to claim 11, wherein the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value. 13. The apparatus according to claim 1, wherein said image correcting means performs an enlarging process in a short side direction of said strip region with respect to each of said strip regions as a correction enlargement factor using a reciprocal of said estimated distortion amount.
3. The image distortion correction device according to any one of 2. 14. The image distortion correction according to claim 13, wherein the correction magnification at the boundary between the strip regions is corrected in advance so that the correction magnification at the boundary between the strip regions continuously changes. apparatus. 15. The image distortion correction apparatus according to claim 11, wherein the estimated distortion amount is set to “1” for the strip region whose background density is lower than a predetermined density. 16. The image distortion correction device according to claim 11, wherein the reference strip region is obtained only from the strip region whose background density is higher than a predetermined density. 17. The book strip document according to claim 11, wherein the reference strip regions are independently determined for left and right pages of the book document, and the estimated distortion amount is calculated independently for the left and right pages of the book document. An image distortion correction device according to one of the claims. 18. A method according to claim 1, wherein the character circumscribed rectangle which can be a target of calculation of the statistical feature amount exists, and the feature amount of the strip region located closest to the strip region including the page binding portion is predetermined. The process of regarding the value multiplied by the constant value as the statistical feature amount in the strip region including the page binding portion is performed independently for each of the left and right pages of the book document, and the left page side of the book document is 9. The image distortion correction apparatus according to claim 8, wherein a characteristic amount and the characteristic amount on the right page side of the book document are respectively associated with the strip region including the page binding unit. 19. When the character circumscribed rectangle that can be a target of calculating the statistical feature quantity exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the feature on the left page side of the book document 19. The image distortion correction apparatus according to claim 18, wherein an amount and an average value of the feature amount on the right page side of the book document are regarded as the statistical feature amount in the strip region including the page binding unit. 20. When the character circumscribed rectangle that can be the calculation target of the statistical feature amount exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, in the strip region including the page binding portion 19. The image distortion correction device according to claim 18, wherein the statistical feature is undecided. 21. The image distortion correction device according to claim 20, wherein the estimated distortion amount is set to “1” when the statistical feature amount in the strip region including the page binding portion is undecided. 22. When performing the filtering processing,
22. The filtering method of the left page of the book document using the feature value for the left page, and the filtering of the right page of the book document using the feature value for the right page. Image distortion correction device. 23. A computer-readable storage medium storing a program for causing a computer to execute image distortion correction of a scan image read by an image reading unit that reads a document image, wherein the computer has the image reading unit. A character circumscribing rectangle extracting function for extracting characters circumscribing rectangles of each character by extracting characters from the scanned image read by A distortion amount estimating function for estimating the distortion amount of the scanned image using the change of the image, and an image correcting function of performing an enlargement process on the scanned image based on the distortion amount estimated by the distortion amount estimating function. , A storage medium for storing a program for executing the program. 24. The scanned image is divided into a plurality of strip-shaped areas whose long sides extend in a direction perpendicular to the main scanning direction of reading by the image reading means, and the character circumscription having a center in each strip area. 24. The storage medium according to claim 23, wherein the storage medium stores a program for executing the distortion amount estimation function of estimating the distortion amount of the scan image based on a statistical feature amount regarding a set of rectangles. 25. A program for calculating the statistical feature quantity based on the following formula: feature quantity = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle.
4. The storage medium according to 4. 26. The storage medium according to claim 25, wherein a program for storing an average value of the characteristic amounts of the respective character circumscribed rectangles having a center in the strip region as the statistical feature amount in the strip region is stored. 27. A program for storing the statistical feature value in the strip area, wherein a program for excluding the length of a vertical side of the character circumscribed rectangle if the length is smaller than a predetermined value is stored. The storage medium according to the above. 28. A program for storing the statistical feature value in the strip area, wherein a program for eliminating the feature value of the circumscribed rectangle of the character if the feature value is larger than a predetermined value is stored. The storage medium according to the above. 29. A background density of each of the strip regions is measured,
29. The storage medium according to claim 24, storing a program for regarding the strip region having the highest background density as the strip region including a page binding portion of a book document. 30. When the character circumscribed rectangle that can be a target of calculation of the statistical feature quantity exists in the strip region including the page binding part, the character circumscribed rectangle is used to include the page binding part. Calculating the feature amount of the strip region, and when the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the statistical feature amount can be calculated. The strip including the page binding section is a value in which a character circumscribed rectangle is present, and a value obtained by multiplying the feature amount of the strip area located closest to the strip area including the page binding section by a predetermined constant value. 30. The storage medium according to claim 29, which stores a program regarded as the statistical feature amount in an area. 31. The storage medium according to claim 30, wherein a program for calculating a moving average in a direction in which the position of the strip region is changed and performing a filtering process for a change in the feature amount based on the moving average is stored. 32. When performing the filtering process,
32. The storage medium according to claim 31, wherein a program for adjusting a window length near the page binding portion is stored such that a window of the filter does not straddle the strip regions located on both sides of the strip region including the page binding portion. 33. The method according to claim 33, further comprising: determining the strip region as a reference for calculating the distortion amount of the strip region; setting the feature amount of the reference strip region as a reference feature amount of the entire scan image; 33. The storage medium according to claim 24, wherein the storage medium stores a program for executing the distortion amount estimating function for obtaining the amount based on an equation: estimated distortion amount = characteristic amount of each strip region / reference characteristic amount. 34. The storage medium according to claim 33, wherein a program for storing the feature value having the largest value among the feature values of all the strip regions as the reference feature value is stored. 35. A program for executing the image correction function of performing an enlarging process in a short side direction of the strip region using a reciprocal of the estimated distortion amount as a correction expansion ratio for each of the strip regions. The storage medium according to any one of the above. 36. A program for preliminarily correcting the correction enlargement ratio at the boundary between the strip regions so that the correction enlargement ratio at the boundary portion between the strip regions changes continuously. Storage media. 37. The storage medium according to claim 33, wherein a program for setting the estimated distortion amount to “1” is stored for the strip region whose background density is lower than a predetermined density. 38. The storage medium according to claim 33, wherein a program for obtaining the reference strip region only from the strip region whose background density is higher than a predetermined density is stored. 39. A program for storing the reference strip region independently for the left and right pages of the book document and calculating the estimated distortion amount independently for the left and right pages of the book document. 38
The storage medium according to any one of the above. 40. The feature amount of the strip region located in the vicinity of the strip region including the page binding portion, in which the character circumscribed rectangle that can be the calculation target of the statistical feature amount is present. The process of regarding the value multiplied by the constant value as the statistical feature amount in the strip region including the page binding portion is performed independently for each of the left and right pages of the book document, and the left page side of the book document is 31. The storage medium according to claim 30, storing a program for associating a feature amount and the feature amount on the right page side of the book document with the strip region including the page binding unit. 41. When the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists in the strip region including the page binding portion, the character circumscribed rectangle is used to include the page binding portion using the character circumscribed rectangle. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the feature on the left page side of the book document 41. The storage medium according to claim 40, wherein a program for storing an amount and an average value of the characteristic amount on the right page side of the book document as the statistical characteristic amount in the strip region including the page binding unit is stored. 42. When the character circumscribed rectangle that can be a target of calculation of the statistical feature quantity exists in the strip region including the page binding part, the character circumscribed rectangle is used to include the page binding part. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, in the strip region including the page binding portion 41. The storage medium according to claim 40, wherein a program is stored for which the statistical feature is undecided. 43. The storage medium according to claim 40, wherein a program for setting the estimated distortion amount to “1” is stored when the statistical feature amount in the strip region including the page binding section is undecided. 44. When performing the filtering processing,
44. A program that uses the feature value for the left page in filtering the left page of the book document and stores the program using the feature value for the right page in filtering the right page of the book document. The storage medium according to any one of the preceding claims. 45. A program for causing a computer to execute image distortion correction of a scanned image read by image reading means for reading a document image, the program comprising the steps of: extracting characters from the scanned image read by the image reading means; A character circumscribing rectangle extraction function for extracting a circumscribing rectangle; and a distortion amount of the scan image using a change in the shape of the character circumscribing rectangle extracted by the character circumscribing rectangle extraction function and a change in the background density of the scan image. A program for causing the computer to execute a distortion amount estimating function for estimating, and an image correcting function of performing an enlargement process on the scan image based on the distortion amount estimated by the distortion amount estimating function. 46. The character circumscribing method, wherein the scanned image is divided into a plurality of strip-shaped areas whose long sides extend in a direction perpendicular to the main scanning direction of reading by the image reading means, and each of the strip areas has a center. The program according to claim 45, wherein the program causes the computer to execute the distortion amount estimation function of estimating a distortion amount of the scan image based on a statistical feature amount regarding a set of rectangles. 47. The program according to claim 46, wherein the statistical feature amount is obtained based on an expression: feature amount = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. 48. The program according to claim 47, wherein an average value of each of the feature amounts of each of the character circumscribed rectangles having a center in the strip region is regarded as the statistical feature amount in the strip region. 49. The program according to claim 48, wherein when calculating the statistical feature amount in the strip region, if the length of the vertical side of the character circumscribed rectangle is smaller than a predetermined value, the program is excluded. 50. The program according to claim 48, wherein when calculating the statistical feature in the strip region, if the feature of the character circumscribed rectangle is larger than a predetermined value, the program is excluded. 51. The background density of each of the strip regions is measured,
5. The strip area having the highest background density is regarded as the strip area including a page binding portion of a book document.
The program according to any one of 6 to 50. 52. When the character circumscribed rectangle that can be the calculation target of the statistical feature amount exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, and when the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the statistical feature amount can be calculated. A character circumscribed rectangle is present, and the strip including the page binding unit is obtained by multiplying a value obtained by multiplying the feature amount of the strip region located closest to the strip region including the page binding unit by a predetermined constant value. 52. The program according to claim 51, wherein the program is regarded as the statistical feature in a region. 53. The non-transitory storage medium according to claim 52, wherein a moving average in a direction in which the position of the strip region changes is obtained, and a filtering process for the change in the feature amount is performed based on the moving average. 54. When performing the filtering processing,
54. The program according to claim 53, wherein a window length is adjusted near the page binding section so that a window of the filter does not straddle the strip areas located on both sides of the strip area including the page binding section. 55. The strip area as a reference for calculating the distortion amount of the strip area, the feature amount of the reference strip area is used as a reference feature quantity of the entire scan image, and the estimated distortion of each strip area is determined. The program according to any one of claims 46 to 54, wherein the computer executes the distortion amount estimating function of obtaining the amount based on an equation: estimated distortion amount = feature amount of each strip region / reference feature amount. 56. The program according to claim 55, wherein the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value. 57. The computer according to claim 45, wherein the computer executes the image correction function of performing an enlarging process in a short side direction of the strip region using a reciprocal of the estimated distortion amount as a correction expansion ratio for each of the strip regions. The program according to (1). 58. The program according to claim 57, wherein the correction magnification at the boundary between the strip regions is corrected in advance so that the correction magnification at the boundary between the strip regions continuously changes. 59. The program according to claim 55, wherein the estimated distortion amount is set to "1" with respect to the strip region whose background density is lower than a predetermined density. 60. The program according to claim 55, wherein the reference strip area is obtained only from the strip area whose background density is higher than a predetermined density. 61. The book strip document according to claim 55, wherein the reference strip regions are obtained independently for left and right pages of the book document, and the estimated distortion amount is calculated independently for the left and right pages of the book document. A program according to one of the above. 62. The feature amount of the strip region located in the vicinity of the strip region including the page binding portion, in which the character circumscribed rectangle that can be the calculation target of the statistical feature amount is present. The process of regarding the value multiplied by the constant value as the statistical feature amount in the strip region including the page binding portion is performed independently for each of the left and right pages of the book document, and the left page side of the book document is 53. The program according to claim 52, wherein a characteristic amount and the characteristic amount on the right page side of the book document are respectively associated with the strip region including the page binding unit. 63. When the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the feature on the left page side of the book document 63. The program according to claim 62, wherein an average value of the amount and the feature value on the right page side of the book document is regarded as the statistical feature value in the strip region including the page binding portion. 64. When the character circumscribed rectangle that can be a target of calculating the statistical feature quantity exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, in the strip region including the page binding portion 63. The program according to claim 62, wherein the statistical feature amount is undecided. 65. The program according to claim 62, wherein the estimated distortion amount is set to “1” when the statistical feature amount in the strip region including the page binding portion is undecided. 66. When performing the filtering processing,
66. The filtering method of the left page of the book document using the feature value for the left page, and the filtering of the right page of the book document using the feature value for the right page. program. 67. An image distortion correction method for correcting distortion of a scan image read by image reading means for reading a document image, the method comprising: extracting characters from the scan image to extract a circumscribed rectangle of each character; An extracting step, a distortion amount estimating step of estimating a distortion amount of the scan image using a change in the shape of the character circumscribed rectangle extracted in the character circumscribed rectangle extraction step and a change in the background density of the scan image; An image correction step of performing an enlargement process on the scanned image based on the distortion amount estimated in the distortion amount estimation step. 68. The distortion amount estimating step includes dividing the scanned image into a plurality of strip-shaped areas each having a longer side in a direction perpendicular to the main scanning direction of reading by the image reading means, and 68. The image distortion correction method according to claim 67, wherein the distortion amount of the scan image is estimated based on a statistical feature amount relating to the set of the character circumscribed rectangles having a center at the center. 69. The image distortion correction method according to claim 68, wherein the statistical feature amount is calculated based on an expression: feature amount = length of horizontal side of character circumscribed rectangle / length of vertical side of character circumscribed rectangle. 70. The image distortion correction method according to claim 69, wherein an average value of the respective characteristic amounts of the respective character circumscribed rectangles having a center in the strip region is regarded as the statistical feature amount in the strip region. 71. The image distortion according to claim 70, wherein when calculating the statistical feature amount in the strip region, if the length of the vertical side of the character circumscribed rectangle is smaller than a predetermined value, the image distortion is eliminated in advance. Correction method. 72. The image distortion according to claim 70, wherein, when obtaining the statistical feature amount in the strip region, if the feature amount of the character circumscribed rectangle is larger than a predetermined value, the feature amount is excluded in advance. Correction method. 73. A background density of each of the strip regions is measured,
7. The strip area having the highest background density is regarded as the strip area including a page binding portion of a book document.
73. The image distortion correction method according to any one of items 8 to 72. 74. When the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists in the strip region including the page binding portion, the character circumscribed rectangle is used to include the page binding portion using the character circumscribed rectangle. Calculating the feature amount of the strip region, and when the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the statistical feature amount can be calculated. The strip including the page binding section is a value in which a character circumscribed rectangle is present, and a value obtained by multiplying the feature amount of the strip area located closest to the strip area including the page binding section by a predetermined constant value. 74. The image distortion correction method according to claim 73, wherein the method is regarded as the statistical feature amount in a region. 75. The image distortion correction method according to claim 74, wherein a moving average in a direction in which the position of the strip region changes is obtained, and a filtering process for the change in the feature amount is performed based on the moving average. 76. When performing the filtering process,
76. The image distortion correction method according to claim 75, wherein a window length is adjusted near the page binding portion so that a window of the filter does not straddle the strip regions located on both sides of the strip region including the page binding portion. 77. The distortion amount estimation step defines the strip region as a reference for calculating the distortion amount of the strip region, sets the feature amount of the reference strip region as a reference feature amount of the entire scan image, 77. The image distortion correction method according to any one of claims 68 to 76, wherein the estimated distortion amount of each strip region is obtained based on an equation: estimated distortion amount = feature amount of each strip region / reference feature amount. 78. The image distortion correction method according to claim 77, wherein the feature value having the largest value among the feature values of all the strip regions is set as the reference feature value. 79. The image correction method according to claim 67, wherein, in the image correction step, an enlarging process is performed in a short side direction of the strip region with respect to each of the strip regions, using a reciprocal of the estimated distortion amount as a correction expansion ratio. Image distortion correction method. 80. The image distortion correction according to claim 79, wherein the correction magnification at the boundary between the strip regions is corrected in advance so that the correction magnification at the boundary between the strip regions continuously changes. Method. 81. The image distortion correction method according to claim 77, wherein the estimated distortion amount is “1” with respect to the strip region whose background density is lower than a predetermined density. 82. The image distortion correction method according to claim 77, wherein the reference strip region is obtained only from the strip region whose background density is higher than a predetermined density. 83. The reference strip region is obtained independently for left and right pages of the book document, and the estimated distortion amount is calculated independently for left and right pages of the book document. An image distortion correction method according to claim 1. 84. The character circumscribing rectangle that can be a target of calculation of the statistical feature is present, and is predetermined in the feature of the strip region located closest to the strip region including the page binding portion. The process of regarding the value multiplied by the constant value as the statistical feature amount in the strip region including the page binding portion is performed independently for each of the left and right pages of the book document, and the left page side of the book document is 75. The image distortion correction method according to claim 74, wherein a characteristic amount and the characteristic amount on the right page side of the book document are respectively associated with the strip region including the page binding unit. 85. When the character circumscribed rectangle that can be a target of calculation of the statistical feature amount exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, the feature on the left page side of the book document 85. The image distortion correction method according to claim 84, wherein an average value of the amount and the feature value on the right page side of the book document is regarded as the statistical feature value in the strip region including the page binding portion. 86. When the character circumscribed rectangle that can be the calculation target of the statistical feature quantity exists in the strip region including the page binding unit, the character circumscribed rectangle is used to include the page binding unit using the character circumscribed rectangle. Calculating the feature amount of the strip region, if the character circumscribed rectangle that can be the calculation target of the statistical feature amount does not exist in the strip region including the page binding portion, in the strip region including the page binding portion The image distortion correction method according to claim 84, wherein the statistical feature amount is undecided. 87. The image distortion correction method according to claim 86, wherein the estimated distortion amount is set to “1” when the statistical feature amount in the strip region including the page binding section is undecided. 88. When performing the filtering processing,
89. The filtering method for the left page of the book document using the feature value for the left page, and the filtering for the right page of the book document using the feature value for the right page. Image distortion correction method. 89. An image reading apparatus comprising: an image reading means for reading a document image; and the image distortion correcting apparatus according to claim 1, which corrects a scanned image read by the image reading means. . 90. An image reading device for reading a document image, an image distortion correcting device according to any one of claims 1 to 22 for correcting a scanned image read by said image reading device, and said image distortion correcting device. And an image printing device for printing an image on paper based on image data output from the image forming apparatus.