JP2007328652A - Image processing device and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that a constitution for applying a Hough transform and histogram generation to an entire image results in the erroneous detection of the inclination of lines of paragraphs as the skew angle of the entire image if, for example, vertically or horizontally adjacent paragraphs included in the image have evenly shifted lines. <P>SOLUTION: A white area search part 202 searches an input image for white areas, and an area division part 203 divides the input image into a plurality of areas on the basis of such white areas. Every divided area is Hough-transformed by a Hough transform part 204 and angle is etected by an angle calculation part 205. An overall angle calculation part 206 calculates the skew angle of the entire image from the angles detected on the divided area basis, and a skew correction part 207 corrects the skew of the input image according to the skew angle. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing program, and more particularly to an image processing apparatus and an image processing program for detecting an inclination angle of a document image read in an inclined manner when reading an image and correcting the inclination.

  Analyzes the image data obtained when the document image is read by the image scanner and the image data received by the facsimile machine, extracts the region from the document image, and determines the type and attribute of the image included in the document. A technique is known in which character recognition processing is performed on an area determined as a character, and a character or a sentence is extracted from a document read as an image.

  In the field of this type of image processing, in order to correctly execute processing such as segmentation and character recognition, it is premised that the document image read by the image scanner or the facsimile machine on the transmission side has no inclination. Therefore, when the read image has a tilt, as a pre-process for processing such as segmentation and character recognition, the tilt angle of the image (hereinafter referred to as “skew angle”) is detected and the tilt is detected. It is necessary to perform so-called skew correction.

  Conventionally, a technique using Hough transformation is known as a technique for detecting a skew angle of a document image and performing skew correction based on the detection result (see, for example, Patent Document 1). In the prior art described in Patent Document 1, the following processing is performed when detecting the skew angle.

  First, a filtering process is performed on an input image, and a binarization process is performed on an image whose density difference is enhanced by the filtering process to generate a binary image. Next, a Hough transform is performed on each pixel of the generated binary image, a histogram is created in the Hough space, and coordinates whose frequency is equal to or higher than a predetermined threshold in the Hough space are extracted. Then, after the extracted coordinates are grouped, representative point coordinates are extracted for each group, and the skew angle of the image is estimated from the extracted coordinates.

  Further, in Patent Document 1, after extracting coordinates having a frequency equal to or higher than a predetermined threshold in the Hough space, the number of extracted coordinates is integrated for each angle to generate a histogram, and the angle at which the frequency is maximum is obtained. A technique for estimating the skew angle of an image is also described.

JP 11-328408 A

  However, since the conventional technique described above employs a configuration in which the Hough transform process and the histogram generation process are performed on the entire document image, for example, as shown in FIG. If the lines are evenly inclined (shifted) between adjacent paragraphs, the line inclination between the paragraphs may be mistakenly detected as the inclination angle (skew angle) of the entire document image, or In order to determine one inclination angle for the entire document image, for example, as shown in FIG. 8, when the left and right pages have different inclination angles, as shown in FIG. On the other hand, skew correction may not be performed correctly.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an image processing apparatus and an image processing program capable of detecting a skew angle of an image and correcting the image with high accuracy. There is.

  In order to achieve the above object, the present invention adopts the following configuration. That is, a white area is searched from the input image, the input image is divided into a plurality of areas based on the searched white area, and angle detection is performed by Hough transform for each divided area. Then, a skew angle is obtained based on the detected angle of the region unit, and skew correction is performed on the input image based on the obtained skew angle.

  In the image processing for detecting and correcting the skew angle, a white area is found from the image, and angle detection is performed by Hough transform for each divided area divided by the white area. In this manner, by performing angle detection in units of divided areas, it is possible to detect the angles of individual divided areas while preventing interference with other areas. Then, a skew angle for the entire image or for each region is calculated from the angle detected for each divided region, and skew correction is performed based on the skew angle.

  According to the present invention, by detecting the angle of each divided region while preventing interference with other regions, it is possible to more accurately detect the skew angle of the entire image or of the region unit. Detection and correction thereof can be realized with high accuracy.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration example of an image processing system including an image processing apparatus according to the present invention. As shown in FIG. 1, the image processing system includes an image data input unit 10, an image processing unit 20, and an image data output unit 30.

  The image data input unit 10 includes an image reading device such as an image scanner, reads an image of a document original, and inputs the image data to the image processing unit 20. In the case of adopting a system configuration in which image data read by an image reading device such as an image scanner is temporarily stored in a storage device such as a server, the server functions as the image data input unit 10. In the facsimile apparatus, the receiving facsimile apparatus functions as the image data input unit 10.

  The image processing unit 20 corresponds to the image processing apparatus according to the present invention, detects a skew angle (tilt angle) of an image (input image) based on the image data input from the image data input unit 10, and based on the detected angle. In addition, skew correction is performed on the image data.

  The image processing unit 20 includes a CPU (Central Processing Unit) 21, an I / O circuit 22, a ROM 23, a RAM 24, an HDD (Hard Disk Drive) device 25, and the like, and these components are connected via a bus line 26. It is the structure connected so that mutual communication was possible.

  The CPU 21 controls processing of the entire image processing unit 20 including calculation processing. The I / O circuit 22 manages input / output with peripheral devices including the image data input unit 10 and the image data output unit 30. The ROM 23 stores processing programs for various processes executed under the control of the CPU 21. The RAM 24 is a primary storage device used when executing various processes. The HDD 25 stores image data processed under the control of the CPU 21 and image data captured from the outside.

  The image data output unit 30 outputs the image data whose skew has been corrected by the image processing unit 20 to an image processing device (not shown) such as a character recognition device in the next stage. The image data output unit 30 includes an output device such as a printer or a display and its control means, and prints out an image based on the image data skew-corrected by the image processing unit 20 on printing (recording) paper, or It is also possible to display on the display screen.

[First Embodiment]
FIG. 2 is a block diagram illustrating a functional configuration example of the image processing apparatus (image processing unit) 20A according to the first embodiment of the present invention.

  As shown in FIG. 2, the image processing apparatus 20A according to the present embodiment includes a binarization unit 201, a white region search unit 202, a region division unit 203, a Hough conversion unit 204, an angle calculation unit 205, and an overall angle. The calculation unit 206 and the skew correction unit 207 are included.

  The binarization unit 201 converts the image data input from the image data input unit 10 in FIG. 1 and representing the shading with a plurality of bits into foregrounds such as characters, lines, patterns, and photographs included in the input image. Each pixel value is 1 when the pixel belonging to is an ON pixel (for example, a black pixel) whose pixel value is logic “1” and the pixel belonging to the background area is an OFF pixel (for example, a white pixel) whose pixel value is logic “0” The binarized binary image data is supplied to the white area searching unit 202 and the area dividing unit 203.

  Based on the binary image data supplied from the binarization unit 201, the white region search unit 202 searches for a white region (white band) from the binary image based on the image data. The white area search unit 202 performs logical sum reduction on a binary image when performing a white area search, specifically m × n pixels (m and n are arbitrary integers of 2 or more), for example, 4 ×. If four pixels are used as a unit, and even one pixel in the unit is an ON pixel, the image is reduced by setting the unit as one ON pixel.

  As for this image reduction, as shown in FIG. 3, the original image (A) may be reduced in two steps, for example, reduced image (B) → reduced image (C). A predetermined magnification, for example, the original image (A) may be reduced to a reduced image (C).

  The area dividing unit 203 divides the image binarized by the binarizing unit 201 into a plurality of areas based on the white area searched by the white area searching unit 202. Here, when the image is divided into a plurality of areas, it is not necessary to completely divide the image into white areas. It suffices that the area is roughly divided on the basis of the white area, and the character area or the like may be divided to some extent.

The Hough conversion unit 204 performs a well-known Hough conversion process for each region divided (divided) by the region dividing unit 203. Here, the Hough conversion process will be described. When the position of a pixel is represented by an x coordinate and ay coordinate, if the straight line distance from the origin to the pixel is ρ, and the angle between the straight line from the origin to the pixel and the x axis is θ, the coordinate on the xy coordinate All straight lines passing through the pixel located at (x, y) are
ρ = x cos θ + ysin θ (0 ≦ θ <π)
It can be expressed by the following formula.

Then, for the pixel located at the position of the coordinate (x, y), the angle θ of the above equation is sequentially changed from 0 to π, and the distance ρ obtained corresponding to the change of the angle θ is put on the ρ−θ coordinate. When plotted, all straight lines passing through a certain pixel can be represented as curves on the ρ-θ coordinates (on the polar coordinates). This curve is a Hough curve, and the process for obtaining this Hough curve is the Hough transform.

  The angle calculation unit 205 calculates the inclination angle of the image in units of divided regions based on the result of the Hough conversion in units of regions in the Hough conversion unit 204. Specifically, the angle calculation unit 205 calculates the inclination angle of the straight line in the xy plane subjected to the Hough transform by the Hough transform unit 204 as the skew angle of the input image for each divided region. The angle calculation unit 205, together with the Hough conversion unit 204, constitutes the angle detection means in the claims.

  The overall angle calculation unit 206 corresponds to the skew angle calculation means in the claims, and calculates the skew angle of the entire image before division based on the inclination angle of the image for each divided region calculated by the angle calculation unit 205. .

  As a method for calculating the skew angle of the entire image in the overall angle calculation unit 206, the following method can be considered. However, these are merely examples, and the present invention is not limited to these.

Method 1: The average of the calculated angles is set as the skew angle of the entire image.
Method 2: Of the calculated angles, the angle with the highest calculation frequency is set as the skew angle of the entire image.
Method 3: With respect to the calculated angle, an average of weights using the size of the divided area as a coefficient is set as the skew angle of the entire image.
Method 4: With respect to the calculated angle, an average of weighting using the number of ON pixels in the divided area as a coefficient is set as the skew angle of the entire image.

  Based on the skew angle of the entire image calculated by the overall angle calculation unit 206, the skew correction unit 207 performs skew correction by rotating the input image so that the skew angle becomes zero. As a method for rotating the image, for example, a known method such as affine transformation can be used.

  Each component of the image processing apparatus 20A having the above configuration, that is, a binarizing unit 201, a white region searching unit 202, a region dividing unit 203, a Hough transform unit 204, an angle calculating unit 205, an overall angle calculating unit 206, and a skew correcting unit 207. As for PC, it is conceivable to realize a software configuration using a computer device that executes functions such as information storage processing, image processing, and arithmetic processing by executing a predetermined program, such as a PC (personal computer). .

  However, the present invention is not limited to the implementation by software configuration, and can also be implemented by a hardware configuration or a combined configuration of hardware and software. When realized by a software configuration, a program that causes a computer to function as the binarization unit 201, the white region search unit 202, the region division unit 203, the Hough transform unit 204, the angle calculation unit 205, the overall angle calculation unit 206, and the skew correction unit 207 Is an image processing program according to the present invention.

  It can also be said that the program for executing the processing of each step in the image processing sequence described below is the image processing program according to the present invention. These image processing programs may be installed in the computer in advance. However, it may be provided by being stored in a computer-readable storage medium instead of being installed in advance, or distributed via wired or wireless communication means. .

(Skew angle detection / correction processing)
Next, an example of a procedure of skew angle detection / correction processing according to the first embodiment in which a skew angle of an input image is detected and skew correction is performed based on the detected angle will be described with reference to the flowchart of FIG.

  First, input image data is captured (step S11), and then the captured image data is binarized to generate binary image data (step S12). Next, based on the binary image data, a white area (white band) is searched from the binary image based on the image data (step S13), and then a plurality of binarized images are searched based on the searched white area. (Step S14).

  Here, when the number of divided areas is n, the divided area number i is set to 1 (step S15), the Hough transform process is performed on the i-th divided area (step S16), and then the Hough transform is performed. Based on the result, the inclination angle of the i-th divided region is calculated (step S17), and then the number i of the divided region is incremented (step S18).

  Next, it is determined whether or not the number i of the divided areas exceeds the number n of divided areas (step S19). If the number i does not exceed the number n of divided areas, the process returns to step S16 to reach the nth divided area. The processing from step S16 to step S18 is repeatedly executed for each region.

  When the calculation process of the inclination angle is completed for all the divided areas, the skew angle of the entire image before the division is calculated based on the calculated inclination angle for each divided area (step S20). Then, the skew correction is performed by rotating the input image based on the calculated skew angle of the entire image (step S21), and the series of skew angle detection / correction processing ends.

  As described above, a white area (white band) is found in the image, angle detection is performed by Hough transform for each area divided by the white area, and the skew angle of the entire image is calculated from the angle detected for each divided area. By doing so, it is possible to detect the angle of each divided area while preventing interference with other areas, and as a result, the skew angle of the entire image can be detected more accurately. Can be realized with high accuracy.

  For example, as shown in FIG. 7, even if the lines included in the image are evenly inclined (shifted) between adjacent paragraphs in the vertical and horizontal directions, for each divided region, that is, for each paragraph By detecting the tilt angle, it is possible to detect the angle of each paragraph while preventing interference with other paragraphs. Therefore, the tilt of the line between paragraphs is erroneously detected as the skew angle of the entire image. Can be avoided. Thereby, since the skew angle of the whole image can be detected more accurately, the detection of the skew angle and the skew correction can be realized with high accuracy.

[Second Embodiment]
FIG. 5 is a block diagram illustrating a functional configuration example of an image processing apparatus (image processing unit) 20B according to the second embodiment of the present invention.

  As shown in FIG. 5, the image processing apparatus 20B according to the present embodiment includes a binarization unit 211, a white region search unit 212, a region division unit 213, a Hough transform unit 214, an angle calculation unit 215, a region integration unit 216, The image division unit 217, the skew correction unit 218, and the image composition unit 219 are included.

  Based on the image data input from the image data input unit 10 in FIG. 1, the binarizing unit 211 converts pixels belonging to the foreground such as characters, lines, patterns, photographs, and the like included in the image to ON pixels (for example, Black pixel) and 1-bit binary image data in which pixels belonging to the background region are OFF pixels (for example, white pixels) are generated and supplied to the white region search unit 212 and the region division unit 213.

  Based on the binary image data supplied from the binarization unit 211, the white region search unit 212 searches for a white region (white band) from the binary image based on the image data. When the white area search unit 212 performs a white area search, the logical area reduction is performed on the binary image, specifically, m × n pixels (m and n are arbitrary integers of 2 or more), for example, 4 ×. If four pixels are used as a unit, and even one pixel in the unit is an ON pixel, the image is reduced by setting the unit as one ON pixel. At this time, as in the case of the first embodiment, the image may be reduced in stages, or may be reduced to a predetermined magnification at a time.

  The area dividing unit 213 divides the image binarized by the binarizing unit 211 into a plurality of areas based on the white area searched by the white area searching unit 212.

  The Hough transform unit 214 performs a well-known Hough transform process for each region divided (divided) by the region dividing unit 213. The angle calculation unit 215 calculates the inclination angle of the image in units of divided regions based on the result of the Hough conversion in units of regions in the Hough conversion unit 214. The angle calculation unit 215, together with the Hough conversion unit 214, constitutes an angle detection unit in the scope of claims, and detects (calculates) an angle in the same manner as in the first embodiment.

  The region integration unit 216 corresponds to the skew angle calculation means in the claims, compares the inclination angles calculated for each divided region by the angle calculation unit 215 between the divided regions, and the comparison result is the same (substantially the same). In addition, the divided areas are integrated for each of the same inclination angles, and the skew angle of the integrated area is set to the same inclination angle.

  The image dividing unit 217 divides the input image in units of the integrated regions based on the integration result in the region integration unit 216.

  The skew correction unit 218 divides each image divided by the image division unit 217 so that the skew angle becomes zero based on the skew angle of the region (corresponding to the divided image) integrated by the region integration unit 216. Skew correction is performed by rotating the image. As a method for rotating the image, for example, a known method such as affine transformation can be used.

  The image synthesis unit 219 generates a skew-corrected image corresponding to the input image by combining the divided images that have been skew-corrected by the skew correction unit 218, and outputs the generated image.

  Each component of the image processing apparatus 20B having the above configuration, that is, a binarizing unit 211, a white region searching unit 212, a region dividing unit 213, a Hough transform unit 214, an angle calculating unit 215, a region integrating unit 216, an image dividing unit 217, The skew correction unit 218 and the image composition unit 219 are realized by a software configuration using a computer device that executes functions such as information storage processing, image processing, and arithmetic processing by executing a predetermined program like a PC. It is possible to do.

  However, the present invention is not limited to the implementation by software configuration, and can also be implemented by a hardware configuration or a combined configuration of hardware and software. When realized by a software configuration, the binarization unit 211, the white region search unit 212, the region division unit 213, the Hough transform unit 214, the angle calculation unit 215, the region integration unit 216, the image division unit 217, the skew correction unit 218, and the image A program that causes a computer to function as the combining unit 219 is an image processing program according to the present invention.

  It can also be said that the program for executing the processing of each step in the image processing sequence described below is the image processing program according to the present invention. These image processing programs may be installed in the computer in advance. However, it may be provided by being stored in a computer-readable storage medium instead of being installed in advance, or distributed via wired or wireless communication means. .

(Skew angle detection / correction processing)
Next, an example of a procedure of skew angle detection / correction processing according to the second embodiment in which a skew angle of an input image is detected and skew correction is performed based on the detected angle will be described with reference to a flowchart of FIG.

  First, input image data is captured (step S31), and then the captured image data is binarized to generate binary image data (step S32). Next, based on the binary image data, a white area (white band) is searched from the binary image based on the image data (step S33), and then a plurality of binarized images are searched based on the searched white area. (Step S34).

  Here, when the number of divided areas is n, the divided area number i is set to 1 (step S35), the Hough transform process is performed on the i-th divided area (step S36), and then the Hough transform is performed. Based on the result, the inclination angle of the i-th divided region is calculated (step S37), and then the number i of the divided region is incremented (step S38).

  Next, it is determined whether or not the number i of the divided areas exceeds the number n of divided areas (step S39). If the number i does not exceed the number n of divided areas, the process returns to step S36 to the nth divided area. The processing from step S36 to step S38 is repeated for each region.

  After the calculation of the tilt angle is completed for all the divided areas, the tilt angles calculated for each divided area are compared between the divided areas. If the comparison results are the same (including almost the same), the same. The divided areas are integrated for each inclination angle, and the skew angle of the integrated area is set to the same inclination angle (step S40).

  Next, based on the integration result for the divided areas, the input image is divided in units of the integrated areas (step S41), and then, for each divided image, the divided images are rotated based on the skew angle of the integrated area. Thus, the skew correction is performed (step S42), and then the divided images subjected to the skew correction are synthesized (step S43), and the series of skew angle detection / correction processing is completed.

  As described above, by finding a white area (white band) from the image, performing angle detection by Hough transform for each area segmented by the white area, and calculating a skew angle from the angle detected for each divided area , It is possible to detect the angle of each of the divided areas while preventing interference with other areas, furthermore, by integrating the areas where the same angle is detected, the detection angle can be detected as the skew angle of the integrated area, As a result, since the skew angle of the entire image can be detected more accurately, the detection of the skew angle and the skew correction can be realized with high accuracy.

  For example, as shown in FIG. 8, even when the image tilt angle is different between the left and right pages as in a spread document, the skew angle is obtained for each integrated region, that is, for both the left and right pages, By dividing the input image based on the integrated area and performing skew correction for each divided image, it is possible to detect the angle of each page and prevent skew correction while preventing interference with other pages. In addition, skew angle detection and skew correction can be realized with high accuracy for both the left and right page images.

It is a block diagram which shows the structural example of the image processing system containing the image processing apparatus by this invention. It is a block diagram which shows the function structural example of the image processing apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing about image reduction in the case of a white area search. It is a flowchart which shows an example of the procedure of the skew angle detection and correction process which concerns on 1st Embodiment. It is a block diagram which shows the function structural example of the image processing apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart which shows an example of the procedure of the skew angle detection and correction process which concerns on 2nd Embodiment. It is a figure which shows the state in which the line inclines equally between the paragraphs adjacent to the upper and lower sides or right and left in the paragraph contained in an image. It is a figure which shows the state from which the inclination angle of an image differs in a right-and-left page.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Image data input part, 20, 20A, 20B ... Image processing apparatus (image processing part), 30 ... Image data output part, 21 ... CPU, 22 ... I / O circuit, 23 ... ROM, 24 ... RAM, 25 ... HDD (Hard Disk Drive) device, 26 ... bus line, 201, 211 ... binarization unit, 202, 212 ... white area search unit, 203, 213 ... area division unit, 204, 214 ... Hough conversion unit, 205, 215 ... Angle calculation unit, 206 ... Overall angle calculation unit, 207, 218 ... Skew correction unit, 216 ... Area integration unit, 217 ... Image separation unit, 219 ... Image composition unit

Claims (4)

Search means for searching for a white area in the input image;
Area dividing means for dividing the input image into a plurality of areas based on the white area searched by the search means;
Angle detection means for performing angle detection by Hough transform for each area divided by the area dividing means;
A skew angle calculating means for obtaining a skew angle based on an angle of a region unit detected by the angle detecting means;
An image processing apparatus comprising: a skew correction unit that performs skew correction on the input image based on the skew angle obtained by the skew angle calculation unit. The image processing according to claim 1, wherein the skew angle calculation unit calculates a skew angle of the entire image before the division by the region dividing unit based on the angle of the region unit detected by the angle detection unit. apparatus. The skew angle calculation means compares the angle of the area unit detected by the angle detection means between areas, and in the case of the same angle, the areas are integrated for each same angle and the same angle is integrated. The image processing apparatus according to claim 1, wherein a skew angle of the region is used. A search step for searching for a white region in the input image;
A region dividing step of dividing the input image into a plurality of regions based on the white region searched in the searching step;
An angle detection step of performing angle detection by Hough transform on the region units divided in the region division step;
A skew angle calculating step for obtaining a skew angle based on the angle of the region unit detected in the angle detecting step;
An image processing program that causes a computer to execute each processing of a skew correction step of performing skew correction on the input image based on the skew angle obtained in the skew angle calculation step.

Images