JP2011139371A - Image processor, image processing method, and program by which computer executes - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor that detects with high accuracy a pleat line of the read image obtained by optically reading a reading medium, and also to provide an image processing method, and a program by which a computer executes. <P>SOLUTION: A detection portion 14 detects each of the pleat lines of first and second surfaces of the read images of the reading medium, and a correction portion 15 compares the pleat line detection result of the first surface of the read image with that of the second surface of the read image by using the geometrical feature of each pleat line so as to correct the pleat line detection results of the first and second surfaces of the read images. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image processing method, and a computer-executable program, and in particular, causes of a fold in the reading medium from image information obtained by optically reading a reading medium such as a paper document. Thus, the present invention relates to an image processing apparatus, an image processing method, and a computer-executable program that execute processing for detecting noise superimposed on a read image.

  When a creased paper document is read using a reading device such as a scanner, the fold portion may be read as a thin line (hereinafter referred to as a crease line). Since these thin lines are not drawn on the manuscript, they should be unnecessary information for most users. In addition, if there is an application for image recognition such as OCR or image area separation, the crease line may cause a decrease in recognition accuracy. For this reason, it is desirable that the crease line is erased or not noticeable.

  For example, an image processing apparatus disclosed in Patent Document 1 and a crease noise removing apparatus disclosed in Patent Document 2 have been proposed as techniques for erasing or making the crease line inconspicuous. In Japanese Patent Laid-Open No. 2004-260688, “locations where the brightness difference from the surroundings is large but the color difference is small” are searched from the read image, and are detected as a crease line when they are connected in a line. Then, the crease line is made inconspicuous by making the portion detected as the crease line resemble the surrounding lightness.

  Further, in Patent Document 2, a read image is converted into a black and white image, and “a portion having a run length of a predetermined length or more in a direction parallel to an arbitrary side of a black and white image and a similar run length continuing in a vertical direction” Is detected as a crease line. Then, the fold line is made inconspicuous by making the portion detected as the fold line similar to the surrounding density.

  In both patent documents, the crease line is detected from the read image by paying attention to the appearance such as the color and shape of the crease line. Certainly, the crease lines have characteristic colors and shapes, so it is considered appropriate to use them as detection conditions. However, since various objects such as characters, lines, and figures are drawn on the paper document, if attention is paid only to the appearance, there is a possibility that an object that looks exactly like the crease line is erroneously detected as a crease line.

  The present invention has been made in view of the above, and is an image processing apparatus, an image processing method, and an image processing apparatus that can detect a crease line of a read image obtained by optically reading a reading medium with high accuracy. An object is to provide a computer-executable program.

  In order to solve the above-described problems and achieve the object, the present invention provides an image processing apparatus for detecting a fold line of a read image obtained by optically reading a read medium, wherein A crease line detection result of the read image of the surface and a crease line detection result of the read image of the second surface of the reading medium are compared using a geometric feature of the crease line, and the first surface and the first surface are compared. The image forming apparatus includes a correcting unit that corrects the crease line detection result of the two-side read image.

  In order to solve the above-described problems and achieve the object, the present invention provides an image processing method for detecting a crease line of a read image obtained by optically reading a read medium, A fold line detection result of the read image of the first surface and a fold line detection result of the read image of the second surface of the reading medium are compared using a geometric feature of the fold line, and the first surface and A correction step of correcting a crease line detection result of the read image on the second surface is included.

  According to the image processing apparatus of the present invention, the image processing apparatus detects a crease line of a read image obtained by optically reading the read medium, and the crease line of the read image on the first surface of the read medium. The detection result and the fold line detection result of the read image on the second surface of the reading medium are compared using the geometric feature of the fold line, and the folding of the read images on the first surface and the second surface is compared. Since the correction means for correcting the line-of-sight detection result is provided, it is possible to provide an image processing apparatus capable of detecting the crease line of the read image obtained by optically reading the reading medium with high accuracy. The effect of becoming.

FIG. 1 is a diagram for explaining the principle of the present invention. FIG. 2 is a diagram showing a configuration example of the image processing apparatus of the present invention. FIG. 3 is a diagram showing an example of an operation sequence of the image processing apparatus of the present invention. FIG. 4 is a diagram illustrating a display example of the operation unit. FIG. 5 is a diagram illustrating an example of the crease line detection result. FIG. 6 is a diagram illustrating an example of data recorded in the recording unit. FIG. 7-1 is a diagram illustrating an example of a fold line detection result. FIG. 7B is a diagram illustrating an example of a fold line detection result. FIG. 8 is a diagram illustrating a flowchart example of the correction operation. FIG. 9 is a diagram illustrating an example of group processing. FIG. 10A is a diagram illustrating an example of a geometric feature of a crease line. FIG. 10B is a diagram illustrating an example of a geometric feature of the crease line. FIG. 11A is a diagram illustrating an example of a geometric feature of a crease line. FIG. 11B is a diagram illustrating an example of a geometric feature of a crease line. FIG. 12 is a diagram illustrating an example of mirror surface processing. FIG. 13 is a diagram illustrating an example of the contraction process. FIG. 14 is a diagram illustrating an example of the expansion process.

  Embodiments of an image processing apparatus, an image processing method, and a computer-executable program according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments. In addition, constituent elements in the following embodiments include those that can be easily assumed by those skilled in the art or that are substantially the same.

(Principle of the present invention)
According to the present invention, noise superimposed on a read image due to a fold of the read medium is detected with high accuracy from image information obtained by optically reading a read medium such as a paper document. FIG. 1 is a diagram for explaining the principle of the present invention.

  In FIG. 1, a crease in a paper document or the like is a scar caused by folding the paper (see 101). The fold line of the paper document enters both sides. When a paper document is optically read, the crease line superimposed on the read images on the first surface and the second surface has mirror coordinates. The figure shows a state where a paper document (first side: 101) with an expanded fold is turned over (reversed) in the horizontal and vertical directions (second side: 102, 103). The crease line between the surface and the second surface has specular coordinates. In the present invention, the crease line detection result of the read image on the first surface of the reading medium and the fold line detection result of the read image on the second surface of the reading medium are represented by the geometric characteristics of the crease line, that is, the same reading. By using the feature that the crease line between the first surface and the second surface of the medium has a mirror surface relationship, the crease line detection result of the read images of the first surface and the second surface is corrected, thereby making the crease line Is improved.

(Embodiment)
FIG. 2 is a diagram schematically illustrating a configuration example of the image processing apparatus according to the embodiment of the present invention. As shown in FIG. 2, the image processing apparatus 1 includes parts such as a reading unit 11, an output unit 12, an operation unit 13, a detection unit 14, a correction unit 15, a recording unit 16, and a control unit 17. It has.

  The reading unit 11 is a part that optically reads a reading medium and generates a read image. The reading unit 11 includes a mechanism that inverts the reading medium in the horizontal direction and / or the vertical direction, and is configured to be able to read both surfaces (first surface and second surface) of the reading medium. When a creased paper document is read, a crease line is superimposed on the generated read image.

  The detection unit 14 is a part for detecting a crease line from the read image. Specifically, the detection unit 14 determines whether or not each pixel constituting the read image is a crease line, and generates binary data having the same size as the read image as a detection result.

  The recording unit 16 is a part for recording the read image and the crease line detection result in association with each other. The recording unit 16 can use, for example, a RAM (Random Access Memory) as a recording medium, and can partially read and write recorded information. Further, the recording capacity of the recording unit 16 is a size capable of recording at least the read image and the crease line detection result for each two sides.

  The correction unit 15 is a part that corrects the crease line detection result of the detection unit 14. Specifically, the correction unit 15 reads the crease line detection result from the recording unit 16, corrects the detection result as necessary, and writes it back to the recording unit 16.

  The output unit 12 is a part for outputting the read image and the crease line detection result. For example, the output unit 12 is connected to a device that performs image processing (for example, processing for making a fold line superimposed on a read image inconspicuous).

  The operation unit 13 is a part for the user of the image processing apparatus 1 to perform various settings and to provide various information to the user of the image processing apparatus 1. The operation unit 13 can be configured by a display device that can display soft keys such as a touch panel, for example.

  The control unit 17 includes a CPU, a ROM, a RAM, an A / D converter, a port, and the like, and is a part for controlling the operation of each part of the image processing apparatus 1. The control unit 17 controls each part during a period from the start to the end of a basic operation described later.

  The basic operation of the image processing apparatus 1 in FIG. 2 will be described with reference to FIGS. FIG. 3 shows an example of an operation sequence of the image processing apparatus 1 of FIG. In particular, FIG. 3 shows the operation of each part and the flow of information in time series, and can be classified into four types: a scan operation, a detection operation, a correction operation which is a characteristic part of the present invention, and an output operation. In the correction operation, the crease line detection result on the front and back sides of the same document is required. Therefore, before the correction operation is started, the scan operation and the detection operation on the front and back sides of the same document are executed. FIG. 4 shows an example of the operation screen of the operation unit 13. FIG. 5 is a diagram illustrating an example of the crease line detection result. FIG. 6 is a diagram illustrating an example of data recorded in the recording unit 16.

  In FIG. 3, first, the image processing apparatus 1 performs a Scan operation for optically reading a paper document and recording the read image in the recording unit 16. Specifically, the user installs a paper document in the reading unit 11 and uses the operation unit 13 to set information (at least (1) the size of the paper document, (2) the spread direction of the paper document, (3) (3 types of reading surface) are set, and when the confirm button is pressed, the setting information is confirmed and the scanning operation is started. For example, as shown in FIG. 4, the operation screen of the operation unit 13 includes (1) buttons (A3, B4, Letter) for selecting the size of a paper document, and (2) buttons for selecting a facing direction of the paper document ( (3) A button for selecting a reading surface (front surface, back surface) and a confirmation button are provided.

  The confirmed setting information is once transmitted from the operation unit 13 to the control unit 17 (step S1). Based on the received setting information, the control unit 17 transfers a control parameter for realizing an operation according to the setting of the user to each part (step S2). Each part that has received the control information returns a status signal 1 to the control unit 17 (step S3).

  When receiving the status signal 1 from all the parts, the control unit 17 instructs the reading unit 11 to start the reading operation (step S4). When the reading unit 11 receives an instruction to start the reading operation, the reading unit 11 performs a reading operation for the size of the paper document to generate a read image, and transfers the read image to the recording unit 16 (step S5). The recording unit 16 stores three types of information, that is, the size of the paper document, the spread direction, and the reading surface, in association with the read image. When the recording of the read image is completed, the recording unit 16 returns the status signal 2 to the control unit (step S6), and when the control unit 17 receives the status signal 2, the scanning operation is terminated.

  Next, in the image processing apparatus 1, a detection operation for detecting a crease line from the read image stored in the recording unit 16 is performed. Specifically, first, when the control unit 17 instructs the detection unit 14 to start a detection operation (step S7), the detection unit 14 reads a read image from the recording unit 16 (step S8). It is determined whether or not the pixel is a crease line. The determination algorithm can be the same as that of the prior art, for example, and thus detailed description thereof is omitted. When the determination is completed, the detection unit 14 generates binary bitmap data having the same size as the read image as a crease line detection result, and stores it in the recording unit 16 in association with the read image (step S9).

  Since binary bitmap data having the same size as the read image can represent the state of “0” or “1” for each pixel of the read image, for example, as shown in FIG. If it is “1” and not a crease line, “0” can be assigned. Further, as shown in FIG. 6, the recording unit 16 records the size of the reading medium, the spread direction, the reading surface, the crease line detection result, and the read image in association with each other.

  When the accumulation of the crease line detection results is completed, the recording unit 16 returns the status signal 3 to the control unit 17 (step S10). When the control unit 17 receives the status signal 3, the detection operation ends.

  The Scan operation and the Detection operation are executed for the front and back surfaces of the same document.

  The Correction operation for correcting the crease line detection result recorded in the recording unit 16 will be described. When the control unit 17 instructs the correction unit 15 to start the correction operation (step S11), the correction unit 15 uses the detection result of the crease lines on the front and back surfaces of the same document recorded in the recording unit 16 in a timely manner. Reading (step S12), the crease line detection result of both is corrected using the geometric features of the crease line. Details of the algorithm for correcting the crease line detection result will be described later (see FIGS. 8 to 13). The correction unit 15 stores the new result in the recording unit 16 for the portion covered by the correction of the crease line detection result, that is, the portion where “0” is corrected to “1” or “1” is corrected to “0”. Overwriting is performed (step S13). When the correction of all the crease line detection results is completed, the correction unit 15 returns a status signal 4 to the control unit 17 (step S14). When receiving the status signal 4, the control unit 17 ends the correction operation.

  Finally, in the image processing apparatus 1, an output operation for outputting the read image and the crease line detection result stored in the recording unit 16 from the output unit 12 to the external device is performed. Specifically, first, when the control unit 17 instructs the output unit 12 to start an output operation (step S15), the output unit 12 reads the read image and the crease line detection result from the recording unit 16 (step S15). S16), after converting into a format interpretable by the external device, output to the external device (step S17). When all of the read image and the crease line detection result are output, the output unit 12 returns a status signal 5 to the control unit 17 (step S18). When the control unit 17 receives the status signal 5, the output operation ends.

  With reference to FIGS. 7A to 14, an algorithm for correcting the crease line detection result of the correction operation will be described in detail. 7A and 7B are diagrams illustrating an example of the crease line detection result. FIG. 8 is a flowchart for explaining an example of the correction operation. FIG. 9 is a diagram illustrating an example of group processing. 10-1, FIG. 10-2, FIG. 11-1, and FIG. 11-2 are diagrams showing examples of geometric features of crease lines. FIG. 12 is a diagram for explaining an example of the mirror surface processing. FIG. 13 is a diagram for explaining an example of the contraction process. FIG. 14 is a diagram for explaining an example of the expansion process.

  Here, as an example, as a result of performing the Scan operation and the Detection operation on the left and right open paper document as shown in FIG. 7A, the crease line detection result as shown in FIG. The case of being recorded in No. 16 will be described.

  As shown in FIG. 8, the correction operation includes four steps of group processing (step S21), mirror surface processing (step S22), contraction processing (step S23), and expansion processing (step S24). .

  In the group processing (step S21), the crease line detection results are grouped so that the area continuously determined as the crease line is one unit. For example, in the case of the crease line detection result as shown in FIG. 7-2, as shown in FIG. 9, the crease line detection result on the front surface is grouped into groups A to E, and the crease line detection result on the back surface is grouped into groups F to L. Can be

  In the mirror surface processing (step S22), a crease line detection result on the back surface is made a mirror surface in the horizontal direction or the vertical direction, and is superimposed on the crease line detection result on the front surface. In this case, the mirroring direction is switched depending on the spread direction of the reading medium. This is because the top and bottom of the reading medium and the read image are switched depending on the spread direction. For example, in the case of a left-and-right reading medium (see FIG. 10-1), the crease line superimposed on the front and back read images has a mirror-surface relationship in the horizontal direction (see FIG. 10-2). In the case of vertical opening (see FIG. 11-1), it has a mirror surface relationship (see FIG. 11-2) in the vertical direction. In the case of FIG. 9, the original paper document (see FIG. 7-1) opens left and right. Therefore, if the mirrored ones are overlapped in the horizontal direction, the result shown in FIG. 12 is obtained. Note that FIG. 12 is a diagram in which the overlapping of the crease line detection results is slightly shifted so as to be easily seen.

  In the contraction process (step S23), the first correction is performed on a group basis for the mirror-finished process. Specifically, the group where the fold line is detected on both the front and back surfaces is the same as the detection result (that is, the determination result that it is a crease line). The detection result is corrected for a group in which the fold line and the detected part do not overlap each other (that is, the determination result is changed to a determination result indicating that the fold line is not present).

  For example, if the mirror-finished surface is FIG. 12, the front surface group A is the back surface groups F and H, the front surface group B is the back surface groups I, J, and K, the front surface group C is the back surface group J, Since the front surface group E overlaps the rear surface group L, the determination result that these groups are crease lines is not changed. On the other hand, the groups D and G are changed to a determination result that they are not crease lines because no overlapping group is found. That is, the result shown in FIG. 12 is corrected to the result shown in FIG. 13 by the contraction process.

  In the expansion process (step S24), the second correction is performed on a pixel-by-pixel basis for the processed contraction process. Specifically, if there is a part that has been detected as a crease line on either the front surface or the back surface, it is corrected that the crease line has been detected on both sides. For example, when the contraction process is completed as shown in FIG. 13, the expansion process corrects as shown in FIG.

  As described above, according to the present embodiment, the detection unit 14 detects the crease lines of the read images on the first surface and the second surface of the reading medium, and the correction unit 15 reads the first surface. The crease line detection result of the first surface and the second surface is compared by comparing the crease line detection result of the image and the crease line detection result of the read image of the second surface using the geometric characteristics of the crease line. Therefore, it is possible to prevent erroneous detection of a crease line-like object as a crease line, and to improve the detection result of the crease line of the read image.

  Further, when the correction unit 15 compares the crease line detection result of the read image on the first surface with the crease line detection result of the read image on the second surface, the result of the position having a mirror surface relationship in the horizontal or vertical direction. Since the comparison is made, it is possible to make a comparison using the geometric features of the crease line.

  In addition, since the direction having a mirror-like relationship in the horizontal or vertical direction is determined by the top and bottom state of the reading medium and the read image, the same crease line can be correctly compared.

  Further, when comparing the crease line detection result of the read image on the first surface and the crease line detection result of the read image on the second surface, the correction unit 15 minimizes an area where the portions detected as crease lines are continuous. Since the unit is used, it is possible to improve the accuracy of comparison using the geometric features of the crease line.

  Further, the correction unit 15 detects the crease line on both sides when the crease line detection result on the first surface and the crease line detection result on the second surface read image are not detected on both sides. Since the contraction process is performed to correct that there is no crease, over-detection of the crease line can be corrected.

  Further, the correction unit 15 folds both sides when a crease line is detected on at least one of the crease line detection result of the read image on the first surface and the crease line detection result on the read image of the second surface. Since the expansion process is performed to correct the detection of the line of sight, it is possible to correct the insufficient detection of the fold line.

  Further, since the correction unit 15 executes the expansion process after executing the contraction process, it is possible to correct more overdetection and underdetection.

(program)
Note that the image processing apparatus according to the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a scanner, a printer, and the like). ) May be applied.

  Another object of the present invention is to supply a recording medium recording a program code of software for realizing the functions of the above-described image processing apparatus to the system or apparatus, and the computer of the system or apparatus (or CPU, MPU, It can also be achieved by the DSP) executing the program code stored in the recording medium. In this case, the program code read from the recording medium itself realizes the functions of the image processing apparatus described above, and the program code or the recording medium storing the program constitutes the present invention. Examples of the recording medium for supplying the program code include an FD, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory, an optical recording medium such as a ROM, and a magnetic recording medium. A magneto-optical recording medium or a semiconductor recording medium can be used.

  Further, by executing the program code read by the computer, not only the functions of the image processing apparatus described above are realized, but also an OS (operating system) operating on the computer based on the instruction of the program code. However, it goes without saying that a case where the function of the image processing apparatus described above is realized by performing part or all of the actual processing.

  In addition, after the program code read from the recording medium is written in a memory provided in a function expansion board inserted in the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the above-described functions of the image processing apparatus are realized by the processing.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 11 Reading part 12 Output part 13 Operation part 14 Detection part 15 Correction part 16 Recording part 17 Control part

JP 2002-77625 A Japanese Patent No. 3925851

Claims (9)

An image processing apparatus for detecting a crease line of a read image obtained by optically reading a reading medium,
The crease line detection result of the read image on the first surface of the read medium is compared with the crease line detection result of the read image on the second surface of the read medium using a geometric feature of the crease line, and An image processing apparatus comprising correction means for correcting a crease line detection result of a read image on the first surface and the second surface.   When the correction means compares the crease line detection result of the read image of the first surface and the crease line detection result of the read image of the second surface, the result of the position having a mirror surface relationship in the horizontal or vertical direction. The image processing apparatus according to claim 1, wherein the image processing apparatuses compare each other.   The image processing apparatus according to claim 2, wherein an orientation having a mirror-like relationship in the horizontal or vertical direction is determined by a top and bottom state of the reading medium and the read image.   When the crease line detection result of the read image of the first surface and the crease line detection result of the read image of the second surface are compared, the correction unit minimizes an area where the portions detected as crease lines are continuous. The image processing apparatus according to claim 1, wherein the image processing apparatus is a unit.   The correction means detects a crease line on both sides when the crease line detection result of the read image of the first surface and the crease line detection result of the read image of the second surface are not detected on both sides. The image processing apparatus according to claim 1, wherein a contraction process for correcting that there is no image is executed.   The correcting means is configured to fold lines on both sides when a crease line is detected on at least one of the crease line detection result of the read image on the first surface and the crease line detection result on the read image on the second surface. The image processing apparatus according to claim 1, wherein an expansion process is performed to correct the detected image.   The correction means detects a crease line on both sides when the crease line detection result of the read image of the first surface and the crease line detection result of the read image of the second surface are not detected on both sides. After performing the contraction process for correcting that there is no crease line, if a crease line is detected on at least one surface, an expansion process is performed to correct that the crease line is detected on both sides. The image processing apparatus according to claim 1. An image processing method for detecting a crease line of a read image obtained by optically reading a reading medium,
The crease line detection result of the read image on the first surface of the read medium is compared with the crease line detection result of the read image on the second surface of the read medium using a crease line geometric feature, An image processing method comprising: a correction step of correcting a crease line detection result of a read image on the first surface and the second surface.   A computer-executable program for causing a computer to function as each unit of the image processing apparatus according to any one of claims 1 to 7.

Images