JP5343190B2 - Image processing apparatus and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image divider which does not require binarization of a captured image, is not affected by scars and dirt of a film, can deal with a mixture of small images difference in size, can deal with shapes of the small images other than rectangle, and can deal with an arrangement of small images even when the small images are not in a grid shape. <P>SOLUTION: A frequency distribution of image brightness constituting captured images and an average value of frequency of the brightness are obtained. The brightness having frequency exceeding the average value is detected as a background brightness to be brightness of a background part. The frequency distribution of the brightness other than the background brightness is obtained, and an arrangement of the small images in an image region of images captured on the basis of the frequency distribution is obtained. Thus, it is possible to divide small images precisely without being affected by scars or dirt of the film even when small images different in size mix, shapes of the small images are not rectangle, or the arrangement of the small images is not in the grid shape. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

The present invention is an apparatus for dividing a small image included in this image from a captured image, and in particular, a plurality of small images included inside a medical image captured from one film. And an image processing program .

  When electronically storing a plurality of small images printed on one film, it may be necessary to divide and store the small images one by one. For example, in the medical field, medical images are usually printed as a plurality of small images on a single film. When medical images are stored electronically, they can be stored separately for each small image. Required.

  Conventionally, when a plurality of medical images printed on a single film are individually divided and electronically stored, the plurality of images on the screen are viewed, and an instruction is manually given to the apparatus to perform the divided electronic storage. It was.

As an apparatus that can automatically divide a plurality of small images contained in one film, an “image processing apparatus and method” (Patent Document 1) has been proposed.
The image processing apparatus described in Patent Document 1 is characterized in that the arrangement of small images is detected based on the dispersion of pixel values constituting the captured image. As a result, a plurality of small images can be cut out accurately, easily and in a short time.
JP 2000-342539 A

  However, in the conventional image dividing apparatus described in Patent Document 1, it is necessary to binarize the captured image in order to suppress the influence of film scratches and dirt. For this reason, in an image with many intermediate gradations, there is a possibility that the binarized image varies greatly depending on the setting of the threshold value for binarization, and the division is not accurately performed after all.

In addition, the conventional image segmentation device described in Patent Document 1 cannot cope with a case where small images having different sizes are included, and cuts out a small image based on the size of the largest small image. There was a problem.
Furthermore, the conventional image segmentation device described in Patent Document 1 has a problem in that it cannot be handled unless small images are properly arranged in a grid pattern.
Furthermore, in the conventional dividing apparatus described in Patent Document 1, there is a possibility that a small image cannot be accurately divided in a shape other than a rectangle.

The present invention has been made in view of such circumstances, does not require binarization, is not affected by film scratches and dirt, and can cope with a mixture of small images of different sizes. It is an object of the present invention to provide an image dividing apparatus and an image processing program that can deal with shapes other than rectangles and can deal with small images arranged in a grid pattern.

The problems of the present invention can be solved by the following inventions.
That is, the image division apparatus of the present invention is an image processing apparatus having an image processing unit for processing the image capture unit and Installing elaborate image capturing an image,
The image processing unit includes a background detection unit that detects a background portion that is a non-image portion in the area of the captured image, and an image dividing unit that divides a plurality of small images included in the captured image, respectively. Have
The background detecting means obtains an average value of the frequency of lightness frequency distribution and front Symbol brightness of the pixels constituting the pre-Symbol captured image, a brightness of the brightness of the background portion having a frequency which exceeds the average background Detect as lightness,
The image dividing means, among the brightness of the pixels constituting the front Symbol captured image, it obtains a frequency distribution of the lightness than the background brightness, the small image in the image region of the captured image on the basis of the said frequency distribution Rutoki determine the placement,
The vertical direction of the captured image is the Y axis and the horizontal direction is the X axis,
The captured image is divided into a plurality of strip-shaped areas by straight lines parallel to the Y-axis, a frequency distribution of brightness other than the background brightness is obtained for each of the strip-shaped areas, and from a preset value of the strip-shaped areas The strip-like region having a large frequency is detected as a first strip-like region, and the captured image is divided into a plurality of strip-like regions by a straight line parallel to the X axis, and each of the strip-like regions other than the background brightness is detected. A frequency distribution of brightness is obtained, and the strip-shaped region having a frequency greater than a preset value among the strip-shaped regions is detected as a second strip-shaped region, and the first strip-shaped region and the second strip-shaped region are detected. Detecting a crossed area of strip-shaped areas as a divided image core part;
Or
The X-axis and Y-axis of the captured image are set as the X-axis and Y-axis of the image area, respectively, and a three-dimensional histogram is obtained in which the Z-axis is a frequency of lightness other than background lightness. Detecting a region on the XY plane having a frequency greater than a value as a divided image core portion;
A point obtained by moving each vertex of the rectangle forming the divided image core portion in the X-axis direction until the brightness of the image reaches the background brightness is detected as an X movement point. Detecting each vertex as a Y movement point that is moved in the Y-axis direction until the brightness of the image reaches the background brightness in the outer direction of the rectangle, including all of the X movement point and the Y movement point, A first feature is that a rectangle having the smallest area is detected as an enlarged rectangle, and the arrangement of the small images is obtained based on the arrangement of the enlarged rectangle.
As a result, the image segmentation device according to the present invention detects the background brightness based on the frequency distribution of the brightness of the pixels and detects the arrangement of the small images, so that binarization is not necessary, and it affects film scratches and dirt. In addition, the size of the divided image core portion is adjusted to the size of the corresponding small image one by one, so that it is possible to accurately divide even when small images having different sizes are mixed. Even if the small images are not arranged in a grid pattern, the size of the divided image core is adjusted to the size of the corresponding small image one by one, so that it can be divided accurately.

Further, the present invention is an image processing apparatus of the first aspect, come and overlap a plurality of the expanded rectangle, detects a rectangle area including all of the overlapping said enlarged rectangle is minimized as binding rectangular and, to that it seeks placement of the small image on the basis of arrangement of said coupling rectangular and second features.

According to the present invention, in the image processing apparatus of the second feature , at least one check point is provided for each side constituting the enlarged rectangle or the combined rectangle, and the lightness of the check point becomes background lightness. moving the side including the check point on the outside, to detect a correction rectangle is enlarged the enlarged rectangular or the binding rectangle, the third feature that you seek placement of the small image on the basis of the corrected rectangular And

Furthermore, the present invention causes a computer, an image processing program for executing the image processing steps for processing the image capture step and captured image capturing an image,
The image processing step includes a background detection step for detecting a background portion that is a non-image portion in the captured image region, and an image division step for dividing each of a plurality of small images included in the captured image. Including
Said background detecting step, an average value of the frequency of lightness frequency distribution and front Symbol brightness of the pixels constituting the pre-Symbol captured image, a brightness of the brightness of the background portion having a frequency which exceeds the average background Detect as lightness,
Wherein the image dividing step is, among the brightness of the pixels constituting the pre-Symbol captured image, it obtains a frequency distribution of the brightness other than the background brightness of the small image in the image region of the captured image on the basis of the said frequency distribution place the required Rutoki,
The vertical direction of the captured image is the Y axis and the horizontal direction is the X axis,
The captured image is divided into a plurality of strip-shaped areas by straight lines parallel to the Y-axis, a frequency distribution of brightness other than the background brightness is obtained for each of the strip-shaped areas, and from a preset value of the strip-shaped areas The strip-like region having a large frequency is detected as a first strip-like region, and the captured image is divided into a plurality of strip-like regions by a straight line parallel to the X axis, and each of the strip-like regions other than the background brightness is detected. A frequency distribution of brightness is obtained, and the strip-shaped region having a frequency greater than a preset value among the strip-shaped regions is detected as a second strip-shaped region, and the first strip-shaped region and the second strip-shaped region are detected. Detecting a crossed area of strip-shaped areas as a divided image core part;
Or
The X-axis and Y-axis of the captured image are set as the X-axis and Y-axis of the image area, respectively, and a three-dimensional histogram is obtained in which the Z-axis is a frequency of lightness other than background lightness. Detecting a region on the XY plane having a frequency greater than a value as a divided image core portion;
A point obtained by moving each vertex of the rectangle forming the divided image core portion in the X-axis direction until the brightness of the image reaches the background brightness is detected as an X movement point. Detecting each vertex as a Y movement point that is moved in the Y-axis direction until the brightness of the image reaches the background brightness in the outer direction of the rectangle, including all of the X movement point and the Y movement point, A fourth feature is that a rectangle having the smallest area is detected as an enlarged rectangle, and the arrangement of the small images is obtained based on the arrangement of the enlarged rectangle .

The present invention, in the fourth feature image processing program, can that overlap a plurality of the expanded rectangle, a rectangle area including all of the enlarged rectangle each other an overlapping is minimized as binding rectangular A fifth feature includes a step of detecting, and a step of obtaining an arrangement of the small image based on the arrangement of the combined rectangles.

According to the present invention, in the image processing program of the fifth feature, the step of providing at least one check point for each side constituting the enlarged rectangle or the combined rectangle, and the lightness of the check point as background lightness made up by moving the side including the check point on the outside, a step of detecting a correction rectangle is enlarged the enlarged rectangular or the binding rectangle, and determining the placement of the small image on the basis of the corrected rectangular It is the sixth feature that it is provided.

As a result, the image dividing apparatus and the image processing program of the present invention do not erroneously recognize the background brightness portion as the divided image core portion. In addition, since the arrangement of small images is detected based on a three-dimensional histogram, it can be accurately divided even if the small images are not arranged in a grid pattern, and when a plurality of the enlarged rectangles are overlapped, they are overlapped. A rectangle that includes all the enlarged rectangles and has the smallest area is detected as a combined rectangle, and the arrangement of the small images is obtained based on the arrangement of the combined rectangles, even if small images having different sizes are mixed. It can be accurately divided, and even if the small image has a shape other than a rectangle, it can be accurately divided.

As described above, according to the image segmentation device and the image processing program of the present invention, since the background brightness is detected based on the frequency distribution of the brightness of pixels and the arrangement of small images is detected, binarization is unnecessary. There is a feature that it is hardly affected by scratches and dirt on the film.

  Further, since the detection is performed based on the three-dimensional histogram, the background brightness portion is not erroneously recognized as the divided image core portion, and the image can be accurately divided even if a small image is not arranged in a grid pattern.

  Furthermore, since the size of the divided image core is adjusted to the size of the corresponding small image one by one, even if small images of different sizes are mixed, it can be accurately divided, and the small image is formed in a grid pattern. Even if it is not arranged, it can be accurately divided.

  Furthermore, since the size of the small image is finely adjusted one by one, even if the small image has a shape other than a rectangle, it can be accurately divided.

The best mode for carrying out an image processing apparatus and an image processing program of the present invention will be described below in detail with reference to the accompanying drawings.

<Configuration>
A configuration of an image processing apparatus that realizes an image processing program according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an image processing apparatus according to an embodiment of the present invention.

As shown in FIG. 1, the image processing apparatus of the present invention includes an image capturing unit 1 and an image processing unit 2. The image capturing unit 1 captures an image from a film. For example, the image capturing unit 1 has functions such as a film scanner and an image scanner, and can directly capture an image from a film.
Further, the image capturing unit 1 can capture an image not only from a film but also from an external storage device such as a hard disk.

  The image processing unit 2 includes a background detection unit 3 and an image division unit 4. The background detection means 3 detects a background portion in the image. Here, the background portion refers to a portion that is not an image in the image region, that is, a frame portion. For example, in the case where a plurality of small images are arranged in the image area, it means a portion of a frame that partitions each small image.

  The image dividing unit 4 generates a dividing line that divides a plurality of small images in the image using the background detected by the background detecting unit 3. Further, the image dividing unit 4 can detect the arrangement of a plurality of small images in the image using the background detected by the background detecting unit 3 and cut out these small images individually. Each can be saved as a separate file.

<Operation>
Next, the operation of the image processing apparatus according to the embodiment of the present invention will be described with reference to FIGS. FIG. 2 is an operation flowchart of the image processing apparatus of the present invention.

As shown in FIG. 2, the image capturing unit 1 captures an image from the film (S21). The background detection means 3 detects the brightness of the background portion in the image captured by the image capture unit 1 (S22).
Hereinafter, the lightness of the background portion is referred to as background lightness. Here, background lightness detection (S22) will be further described with reference to FIGS. FIG. 3 is a flowchart of background lightness detection. FIG. 4 is a diagram showing the frequency distribution of the brightness of the pixels constituting the image. FIG. 5 is also a diagram showing the frequency distribution of the brightness of the pixels constituting the image captured by the image capturing unit 1.

  As shown in FIG. 3, the background detection unit 3 obtains the frequency distribution of the brightness of the pixels constituting the image captured by the image capturing unit 1. FIG. 4 shows this frequency distribution. As shown in FIG. 4, for example, when the background lightness is dark, a peak is present in a dark portion that is the background lightness, and there are many cases in which the frequency distribution has no particular peak in other cases.

  This is because the background lightness is high because the background lightness is the same, whereas the image portion often has a wide variety of lightnesses, so the frequency is not high compared to the background lightness. is there.

  Next, the background detection means 3 calculates the average value of the frequencies (S32). A line denoted by reference numeral 41 is a line indicating an average frequency. The background detection means 3 sets the lightness having a frequency exceeding the average value of the frequencies as the background lightness. At this time, the background lightness often has a certain range. Reference numeral 42 indicates a range of background brightness.

Here, the case where the background lightness is dark has been described, but the same applies to cases where the background lightness is bright or any lightness.
As a result of inventor's diligent research, it was found that the background lightness can be detected very accurately in many cases by obtaining the lightness having a frequency exceeding the frequency average value of the image lightness. Thus, by obtaining the lightness having a frequency exceeding the lightness frequency average value, it is possible to detect the background lightness easily and accurately.

  Next, returning to FIG. 2, the background detection means 3 determines whether or not the background exists, that is, whether or not the background lightness has been detected (S23). If it cannot be detected, an error message to that effect is displayed on the screen and the process is terminated.

  The case where the background lightness cannot be detected includes, for example, the case where the image area has no frame portion and is entirely an image. In this case, the lightness frequency distribution has a frequency that exceeds the lightness frequency average value. The brightness will be scattered in various ranges from dark to bright without being fixed in one place. Thus, in the present invention, it is possible to detect the presence or absence of the background.

  The background detection means 3 determines whether or not the background lightness range is larger than a preset setting range (S24). If it is larger than the set range, S26 is executed. If not, the next step is executed. Execute S25

If the background lightness range is larger than the set range, the background detection means 3 performs a background detection retry (S26). Background detection retry (S26) will be described with reference to FIG. When the background lightness range indicated by reference numeral 52, which is a lightness having a higher frequency than the frequency average value indicated by reference numeral 51, is larger than the set value, the background detection means 3 is one of the following preset methods: The background lightness is determined by the method.
(1) The range of the darkest side among the peaks having the frequency peak is set as the background brightness.
(2) The range of the mountain on the brightest side among the peaks having the frequency peak is set as the background brightness.
(3) The range of the mountain at the center of the peaks having the frequency peak is set as the background brightness.
(4) The range of the mountain having the highest peak among the peaks having the frequency peak is defined as the brightness range.
Which of the above methods can detect the background brightness most appropriately depends on the type of image, and can be arbitrarily set depending on the type of image.

  Returning to FIG. 2, the background detection means 3 determines whether or not the background lightness has been determined (S27). If the background lightness cannot be determined by the above method, an error has occurred. If the background brightness is determined after displaying the effect on the screen or the like, S25 is executed.

  If it is determined in S24 that the background brightness range is not larger than the preset range, the image dividing unit 4 divides each of the divided images for dividing the plurality of small images in the image area into small images. The divided image core portion, which is the core portion, is detected.

  The detection of the divided image core part will be described with reference to FIG. FIG. 6 is an explanatory diagram of detection of the divided image core unit. In FIG. 6, the horizontal axis of the image area is described as the X axis, and the vertical axis is described as the Y axis. As shown in FIG. 6, the image dividing means 4 finely divides the image area by a straight line perpendicular to the X axis, that is, a straight line parallel to the Y axis, and the frequency of lightness other than the background lightness for each of the divided strip-like areas. Ask for. A histogram representing the position of the X axis on the horizontal axis and the frequency of lightness other than the background lightness on the vertical axis is denoted by reference numeral 61.

  Similarly, the image dividing means 4 finely divides the image area by a straight line perpendicular to the Y axis, that is, a straight line parallel to the X axis, and obtains the frequency of lightness other than the background lightness for each of the divided strip-like areas. The histogram thus obtained is indicated by reference numeral 62.

  The image dividing means 4 obtains a strip-like area having a frequency larger than a preset value in each obtained histogram, and detects the intersecting portion of the obtained strip-like areas as the divided image core unit 63. The image dividing means 4 examines the lightness of the pixels constituting each obtained divided image core unit 63 itself, and those in which the lightness of all the pixels is the background lightness can be excluded as not being the divided image core unit. This is because the background portion may also enter the intersecting region in the method in which the region where the strip-shaped regions whose longitudinal directions intersect the X-axis direction and the Y-axis direction are the divided image core part.

  Here, the dividing means 4 can obtain the divided image core part by obtaining a three-dimensional histogram other than the above method. Specifically, the X-axis and Y-axis of the three-dimensional histogram are the X-axis and Y-axis of the image area, respectively, the Z-axis is the frequency of brightness other than the background brightness, and X− has a frequency greater than a preset frequency. A region on the Y plane can be used as a divided image core portion. In this method, the background portion is not erroneously recognized as the divided image core portion.

Returning to FIG. 2, next, the image dividing unit 4 enlarges the obtained divided image core part, and detects a rectangle in which the divided image core part is enlarged to the same size as the small image existing in the image area. I do. A specific example will be described with reference to FIG. FIG. 7 is an enlarged explanatory diagram of the divided image core part.
In FIG. 7, the horizontal axis is referred to as the X axis, and the vertical axis is referred to as the Y axis.

  As shown in FIG. 7, the image dividing means 4 is a rectangular ABCD (hereinafter referred to as the ABCD) constituting the divided image core unit 63 for the four vertices (A, B, C, D) of the divided image core unit 63 in the image region. It is extended in the outward direction of the rectangle ABCD) until the background brightness is reached. The vertex A will be described in more detail as an example. The vertex A is extended in the Y axis direction outside the rectangle ABCD until the lightness becomes the background lightness. Let Ay be the point where the background brightness is reached. Similarly, the vertex A is extended in the Y axis direction outside the rectangle ABCD until the lightness becomes the background lightness. Let Ax be the point where the background brightness is reached. The same operation is performed for the remaining vertices B, C, and D.

  An enlarged rectangle 71 that is a rectangle that includes all of Ax, Ay, Bx, By, Cx, Cy, Dx, and Dy and has the smallest area is detected.

  Here, when all of the vertices A, B, C, and D have the background lightness, conversely, until the lightness disappears from the background lightness in the X-axis and Y-axis directions at the respective vertexes and inside the rectangle ABCD. Each vertex can be stretched. At this time, if no brightness other than the background brightness can be detected even if the image is extended until it extends beyond the rectangle ABCD, the point is extinguished. Thereby, when the divided image core part 63 is a background part, all the vertices disappear and the divided image core part 63 disappears. That is, even if the background portion is erroneously recognized as the divided image core unit 63, the divided image core unit 63 of the background portion can be erased by this operation.

This way, the slide into expanding each divided image core portion 63, there are cases where those meeting will be either in the enlarged rectangle 71 with each other is generated. The operation at this time will be described with reference to FIG. FIG. 8 is an explanatory diagram of an enlarged rectangular combination. As shown in FIG. 8, when the enlarged rectangles 71 overlap with each other, the image dividing unit 4 detects a combined rectangle 81 that includes all the overlapped rectangles and has the smallest area.

  The image dividing unit 4 corrects the finally obtained enlarged rectangle 71 and combined rectangle 81 (hereinafter referred to as the final rectangle). This correction will be described with reference to FIG. FIG. 9 is an explanatory diagram of correction of the final rectangle. As shown in FIG. 9, when the small image in the image area is not rectangular, for example, when it is a circular small image as indicated by reference numeral 91, the small image 91 is not fully contained in the final rectangle. There can be.

Therefore, the image dividing means 4 detects the lightness of each of the midpoints a, b, c, and d of the four sides of the final rectangle for all the final rectangles, and if it is not the background lightness, the midpoint becomes the background lightness. The final rectangle can be corrected by moving the side to the outside of the final rectangle. Here, not only the midpoints of the four sides, but also checkpoints that are a plurality of checkpoints are provided on each side, and each side is moved outside the final rectangle so that the background brightness is at all checkpoints. You can also. In addition, the entire four sides are checked instead of the points, and if the side includes a point that is not background lightness, the side can be moved outside the final rectangle.
As a result, even when the small image is not a rectangle, the entire small image can be reliably contained in the final rectangle by correcting the final rectangle.

  Returning to FIG. 2, the image dividing means 4 counts the number of the enlarged rectangle 71 and the combined rectangle 81 finally obtained in this way, and determines whether or not the number is one (S29). ). If it is determined that the number is one, the fact is displayed on a screen or the like and the process is terminated. Thereby, the number of the enlarged rectangle 71 and the combined rectangle 81 finally obtained is the number of small images in the captured image, so that the number of small images included in the captured image can be accurately detected. .

Returning to FIG. 2, the image dividing means 4 generates a dividing line for dividing each small image in the image region. The dividing line can be provided at the opposite side of the adjacent final rectangle and in the middle of the side. At this time, the image dividing unit 4 inspects the generated dividing line, and if the dividing line intersects with the final rectangle, deletes the portion where the dividing line intersects.
Thereby, even if the small images are not arranged in a grid pattern, it is possible to generate a dividing line for accurately dividing the small image.
As described above, the image processing apparatus of the present invention detects the background lightness and obtains the frequency distribution of the lightness other than the background lightness, thereby accurately detecting the arrangement of the small images, dividing the small images, and subdividing the small images. It is possible to easily and accurately generate a dividing line for dividing.

  Further, since background brightness detection and small image arrangement detection are performed based on the frequency distribution, it is difficult to be affected by scratches and dirt, and accurate and stable division of small images is possible.

As described above, the image processing apparatus and the image processing program of the present invention have been described by showing specific embodiments, but the present invention is not limited to these. A person skilled in the art can make various changes and improvements to the configuration and functions of the image processing apparatus according to the above-described embodiment without departing from the gist of the present invention.

The image dividing apparatus and the image processing program of the present invention are realized by an OS, an application, an image sensor, and the like built on hardware resources including a computer CPU, a memory, an auxiliary storage device, and a display. Since the process of recognizing and dividing small images included in the captured image is specifically realized using the above hardware resources, it falls under the technical idea using the laws of nature. Image data processing, such as electronic storage of medical images, processing of digital camera images, processing of image data read by an image scanner, is a field where it is necessary to divide a small image from an image including a plurality of small images. It can be used in various fields that require it.

1 is a schematic configuration diagram of an image processing apparatus according to an embodiment of the present invention. It is an operation | movement flowchart of the image processing apparatus of this invention. It is a flowchart of background lightness detection. It is the figure which showed the frequency distribution of the brightness of the pixel which comprises an image. It is the figure which showed the frequency distribution of the brightness of the pixel which comprises an image. It is detection explanatory drawing of a division | segmentation image core part. It is a divisional image core part expansion explanatory drawing. It is coupling | bonding explanatory drawing of an expansion rectangle. It is correction | amendment explanatory drawing of the last rectangle.

DESCRIPTION OF SYMBOLS 1 Image capture part 2 Image processing part 3 Background detection means 4 Image division means 41 Frequency average value 42 Background brightness 51 Frequency average value 52 Background brightness 61 Histogram showing the frequency of brightness other than background brightness 62 Brightness other than background brightness Histogram 63 representing frequency Divided image core portion 71 Enlarged rectangle 81 Combined rectangle 91 Circular small image

Claims (6)

An image processing apparatus having an image processing unit for processing the image capture unit and Installing elaborate image capturing an image,
The image processing unit includes a background detection unit that detects a background portion that is a non-image portion in the area of the captured image, and an image dividing unit that divides a plurality of small images included in the captured image, respectively. Have
The background detecting means obtains an average value of the frequency of lightness frequency distribution and front Symbol brightness of the pixels constituting the pre-Symbol captured image, a brightness of the brightness of the background portion having a frequency which exceeds the average background Detect as lightness,
The image dividing means, among the brightness of the pixels constituting the front Symbol captured image, it obtains a frequency distribution of the lightness than the background brightness, the small image in the image region of the captured image on the basis of the said frequency distribution Rutoki determine the placement,
The vertical direction of the captured image is the Y axis and the horizontal direction is the X axis,
The captured image is divided into a plurality of strip-shaped areas by straight lines parallel to the Y-axis, a frequency distribution of brightness other than the background brightness is obtained for each of the strip-shaped areas, and from a preset value of the strip-shaped areas The strip-like region having a large frequency is detected as a first strip-like region, and the captured image is divided into a plurality of strip-like regions by a straight line parallel to the X axis, and each of the strip-like regions other than the background brightness is detected. A frequency distribution of brightness is obtained, and the strip-shaped region having a frequency greater than a preset value among the strip-shaped regions is detected as a second strip-shaped region, and the first strip-shaped region and the second strip-shaped region are detected. Detecting a crossed area of strip-shaped areas as a divided image core part;
Or
The X-axis and Y-axis of the captured image are set as the X-axis and Y-axis of the image area, respectively, and a three-dimensional histogram is obtained in which the Z-axis is a frequency of lightness other than background lightness. Detecting a region on the XY plane having a frequency greater than a value as a divided image core portion;
A point obtained by moving each vertex of the rectangle forming the divided image core portion in the X-axis direction until the brightness of the image reaches the background brightness is detected as an X movement point. Detecting each vertex as a Y movement point that is moved in the Y-axis direction until the brightness of the image reaches the background brightness in the outer direction of the rectangle, including all of the X movement point and the Y movement point, An image processing apparatus , wherein a rectangle having a minimum area is detected as an enlarged rectangle, and the arrangement of the small images is obtained based on the arrangement of the enlarged rectangle . Can that overlap a plurality of the expanded rectangle, it detects a rectangle area including all of the overlapping said enlarged rectangle is minimized as binding rectangle, obtaining the arrangement of the small image on the basis of arrangement of said coupling rectangular the image processing apparatus according to claim 1, wherein the this. At least one check point is provided for each side constituting the enlarged rectangle or the combined rectangle, and the side including the check point is moved outward until the lightness of the check point reaches the background lightness. the binding rectangle detects corrected rectangle is enlarged, the image processing apparatus according to claim 2, characterized that you seek placement of the small image on the basis of the corrected rectangular. A computer, an image processing program for executing the image processing steps for processing the image capture step and captured image capturing an image,
The image processing step includes a background detection step for detecting a background portion that is a non-image portion in the captured image region, and an image division step for dividing each of a plurality of small images included in the captured image. Including
Said background detecting step, an average value of the frequency of lightness frequency distribution and front Symbol brightness of the pixels constituting the pre-Symbol captured image, a brightness of the brightness of the background portion having a frequency which exceeds the average background Detect as lightness,
Wherein the image dividing step is, among the brightness of the pixels constituting the pre-Symbol captured image, it obtains a frequency distribution of the brightness other than the background brightness of the small image in the image region of the captured image on the basis of the said frequency distribution place the required Rutoki,
The vertical direction of the captured image is the Y axis and the horizontal direction is the X axis,
The captured image is divided into a plurality of strip-shaped areas by straight lines parallel to the Y-axis, a frequency distribution of brightness other than the background brightness is obtained for each of the strip-shaped areas, and from a preset value of the strip-shaped areas The strip-like region having a large frequency is detected as a first strip-like region, and the captured image is divided into a plurality of strip-like regions by a straight line parallel to the X axis, and each of the strip-like regions other than the background brightness is detected. A frequency distribution of brightness is obtained, and the strip-shaped region having a frequency greater than a preset value among the strip-shaped regions is detected as a second strip-shaped region, and the first strip-shaped region and the second strip-shaped region are detected. Detecting a crossed area of strip-shaped areas as a divided image core part;
Or
The X-axis and Y-axis of the captured image are set as the X-axis and Y-axis of the image area, respectively, and a three-dimensional histogram is obtained in which the Z-axis is a frequency of lightness other than background lightness. Detecting a region on the XY plane having a frequency greater than a value as a divided image core portion;
A point obtained by moving each vertex of the rectangle forming the divided image core portion in the X-axis direction until the brightness of the image reaches the background brightness is detected as an X movement point. Detecting each vertex as a Y movement point that is moved in the Y-axis direction until the brightness of the image reaches the background brightness in the outer direction of the rectangle, including all of the X movement point and the Y movement point, An image processing program for detecting a rectangle having a minimum area as an enlarged rectangle and obtaining an arrangement of the small image based on the arrangement of the enlarged rectangle . When a plurality of enlarged rectangles are overlapped, a step of detecting a rectangle including all of the overlapped enlarged rectangles and having the smallest area as a combined rectangle, and determining the arrangement of the small image based on the arrangement of the combined rectangles The image processing program according to claim 4 , further comprising steps.
Providing at least one check point for each side constituting the enlarged rectangle or the combined rectangle, and moving the side including the check point outward until the lightness of the check point becomes background lightness, detecting a corrected rectangle is enlarged rectangular or the binding rectangle, the image processing program according to claim 5, characterized in that it comprises a step of obtaining the arrangement of the small image on the basis of the corrected rectangular.

Images