JP2005275950A - Image processor, image processing method, program, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor, image processing method, program and storage medium suitably calculating the inclination of an input image even when characters of a larger size than that of a window for detecting projection exists, or a document image whose character's density is low, and whose character lines do not uniformly exist in a document is input. <P>SOLUTION: The projection of character lines is detected by setting two partial image regions for a document image, and the inclination of the document image is decided on the basis of the projection. At that time, the first partial image region is changed on the basis of the projection, and the second partial image region is set on the basis of the changed first partial image region. Then, the projection of the character lines is detected from the first and second partial image regions, and the inclination of the document image is decided on the basis of the projection. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus, an image processing method, a program, and a storage medium that automatically detect and correct a tilt of a document image or the like.

  By automatically correcting the inclination of the document image read from the scanner or the like, it is possible to improve the accuracy of document image analysis processing such as area division and character recognition. For example, the document direction is automatically detected, the projections in the row direction are obtained in two search windows arranged along the row direction, one window is fixed, and the other window is perpendicular to the row direction. A technique is disclosed in which the tilt amount of a document image is obtained from the amount of shift in the vertical direction of two windows and the distance in the row direction when the correlation between projections is maximized while shifting in the direction (see, for example, Patent Document 1). ).

  According to the invention described in Patent Document 1, the projection indicates the position of the character line in the window, and the case where the correlation is maximized is the state where the line position matches. Then, the inclination of the document image can be determined from the inclination of the character line.

Further, a plurality of sets are set so that the set of windows extends over the entire document image, and the inclination as the document image is obtained using the magnitude of the inclination obtained from the plurality of positions. As described above, according to the invention described in Patent Document 1, by using the projection in the row direction in the set window on the character image, the character image can be processed at a high processing speed and with a small storage area. A tilt detection method is disclosed.
Japanese Patent Laid-Open No. 9-6914

  However, the invention described in the above-mentioned Patent Document 1 has a drawback in that the inclination extraction accuracy is lowered when the projection in the line direction of the characters cannot be sufficiently extracted in the set window. For example, if the size of the character is larger than the size of the window, or if the text line is not uniformly present in the document and there are many windows that cover only part of the line, etc. The extraction accuracy of is reduced.

  The present invention has been made in consideration of such circumstances, and there are characters of a size larger than that in the window for detecting projections, the character density is low, and character lines are not contained in the document. An object of the present invention is to provide an image processing apparatus, an image processing method, a program, and a storage medium that can suitably determine the inclination of the input image even when a document image that does not exist is input.

In order to solve the above problems, the present invention provides:
Setting means for setting two partial image areas for the input document image, detection means for detecting projection of a character line from each of the two partial image areas, and the document image based on the detected projection An image processing apparatus comprising a determining means for determining the inclination of
Changing means for changing the first partial image area based on the projection detected from the first partial image area;
The setting means sets a second partial image area based on the first partial image area changed by the changing means;
The detecting means detects a projection of a character line from each of the first partial image region and the second partial image region.
The image processing method according to the present invention includes:
A first setting step of setting a first partial image region for the document image;
A first detection step of detecting a projection of a character line from the first partial image region;
A changing step of changing the first partial image area based on the projection detected from the first partial image area;
A second setting step of setting a second partial image region for the document image based on the first partial image region changed by the changing step;
A detection step of detecting a projection of a character line from each of the first partial image region and the second partial image region;
And a determining step for determining the inclination of the document image based on the detected projection.

  Furthermore, the present invention is a program for causing a computer to execute the image processing method.

  Furthermore, the present invention is a computer-readable storage medium storing the above program.

  According to the present invention, a document image in which characters having a size larger than that in a window for detecting projections or a character density is low and character lines are not uniformly present in the document is input. Even in this case, the inclination of the input image can be obtained suitably.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

<First Embodiment>
FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to the first embodiment of the present invention. In FIG. 1, reference numeral 101 denotes a CPU that controls the entire image processing apparatus according to a control program stored in the ROM 102. Reference numeral 102 denotes a ROM that stores a control program of the image processing apparatus executed by the CPU 101 in accordance with a processing procedure shown in a flowchart to be described later. Reference numeral 103 denotes a RAM for storing document images and the like, and 104 denotes an external storage device such as a magnetic disk. Furthermore, 105 is a display, 106 is a keyboard, 107 is a pointing device such as a mouse, and 108 is a scanner for reading an image from a paper document or the like.

  Furthermore, the CPU 101, ROM 102, RAM 103, external storage device 104, display 105, and scanner 108 described above are connected to each other via a bus 110.

  Note that the image processing performed by the image processing apparatus according to the present embodiment can also be performed by a general-purpose computer. In this case, a control program provided on a storage medium or the like is stored in the external storage device 104 and the operator's It may be configured to be executed by the CPU 101 according to an instruction or the like. Reference numeral 109 denotes a network interface (I / F) that communicates with an information processing apparatus or the like that exists in a remote place, and that reads and writes programs and data, and is connected to the bus 110. Has been.

  The devices such as the scanner 108 and the display 105 may not be directly connected to the bus 110 but may be connected via the network I / F 109.

  The image processed by the image processing apparatus is an image optically read from a paper document or the like by the scanner 108, an image stored in the external storage device 104, or a network from another image input device or the like. What was received via I / F109 can be used. Further, in the present embodiment, the inspection area for detecting the tilt and the document direction in the inspection area are determined for the image obtained in this manner using a known technique. For example, the inspection area can be determined by removing margins of the input image, or can be determined by extracting a text area from document layout information.

  FIG. 2 is a flowchart for explaining a processing procedure for detecting the tilt amount of the input image from within the inspection region by the image processing apparatus according to the first embodiment. First, a detection position for detecting a local inclination of a character line or the like existing in the inspection area is determined for the inspection area in the input image determined using a known technique (step S201). For example, the detection positions are distributed at equal intervals over the entire inspection region, or are varied using random numbers.

  Next, a detection window having a predetermined size is set at the detection position determined in step S201 (step S202). In this embodiment, as an example, a rectangular partial image region is used as a detection window. The direction of the detection window to be set is determined according to the document direction.

  Then, the projection in the row direction within the detection window set in the inspection area is acquired (step S203). Here, the projection means that each pixel is scanned in the row direction, and if there is a black pixel (a pixel constituting the character portion), it is determined that the row is projected. As a result, the height of the character line existing in the detection window is obtained as a projection. And the character size which exists in a detection window is estimated from the obtained projection (step S204).

  FIG. 4 is a diagram for explaining a method for estimating the character size from the projection in the detection window and a method for changing the size of the detection window. In FIG. 4, 401 indicates the initial state of the detection window set in the inspection area in step S202. Here, the projection in the initial state 401 of the detection window is cut off at the upper end because it does not completely include the line. Therefore, by extending the detection window to a position where the projection disappears, a projection corresponding to the height of the row can be obtained with certainty. For example, in order to ensure that a character line is completely included, when a pixel line without projection is found, the detection window is further extended in that direction by a predetermined number of pixels from that position (for example, FIG. 4 is extended upward). When the projection is cut at the lower end, the detection window is extended downward.

  Since the character size can be estimated from the height of the character line obtained from the projection, the size of the detection window is changed according to the estimated character size (step S205). FIG. 4 shows that the height and width of the detection window are changed from the initial state 401 to 402 of the detection window. As described above, the height and width of the detection window are determined based on the estimated character size, and another detection that is paired with the detection window 401 to obtain the inclination of the input document image at this time. In consideration of the inclination with respect to the window 403, it is important that the size of the character line is sufficient.

  Then, after the size of the detection window 401 is changed to the detection window 402 in step 205, a detection window 403 that is paired with the detection window 402 is set at a position separated by a preset distance in the row direction (step S206). ). Next, one detection window (for example, detection window 402) is fixed, and the other detection window (for example, detection window 403) is shifted in the row direction and the vertical direction to obtain a correlation between the projections of the two. Then, a deviation amount that maximizes the correlation is obtained (step S207). As a result, the local inclination angle is obtained from the amount of deviation at the time of the maximum correlation and the interval between the two detection windows (step S208). Then, the obtained angle is stored (step S209).

  FIG. 6 is a diagram for explaining the correlation of the projection of the detection window in the image processing apparatus according to the first embodiment. In FIG. 6A, 601a is a set first detection window, and 602a is a projection in the detection window 601a. Similarly, 601b is a second detection window set as a pair of detection windows 601a, and 602b is a projection in the detection window 601b. FIG. 6B is a graph 605 showing the detection window 604 shifted up and down with respect to the fixed detection window 607 and the magnitude of the correlation. Here, the inclination angle θ has a relationship of tan θ = t / d using the shift amount t when the correlation is maximum and the distance d of the detection window. Note that if the shift upward to t is positive and the shift downward is negative, it is possible to determine which is inclined by the sign.

  Then, it is determined whether or not the inclination angles at all the detection positions have been obtained (step S210). As a result, when the detection position for which the inclination angle has not yet been obtained remains (No), the process returns to step S201 again, and the processes of steps S201 to S209 described above are repeated for the next detection position. On the other hand, when all are detected (Yes), the inclination angle of the input image is determined from the stored local inclination angle (step S211). For example, they can be determined by using the average value or median value of all inclination angles.

  In the process from step S203 to step S208, if the tilt angle cannot be obtained, for example, if an appropriate projection is not obtained or the correlation is insufficient, the detection window The local detection angle is not obtained, and the next detection position may be used. If the local tilt angle stored in step S211 is small, it may be determined that the accuracy cannot be expected and the tilt angle of the input image has not been obtained.

  As described above, even if a document image with an unknown character size is input by image processing by the image processing apparatus according to the present embodiment, the character size is estimated from the projection within the set detection window. Therefore, for example, the inclination of the input image can be obtained with high accuracy even for a document image including a plurality of character sizes.

  In the above embodiment, the description is given with the minimum configuration requirements for carrying out the technical idea of the present invention. However, for example, an image processing program for performing processing for implementing the present invention on a general-purpose computer, etc. The data may be provided from the outside, or may be stored in advance in the external storage device 107 and stored in the RAM 103. Further, an input image stored in advance in the external storage device 107 may be processed.

<Second Embodiment>
Next, another embodiment executed by an image processing apparatus similar to the image processing apparatus shown in FIG. 1 will be described. FIG. 3 is a flowchart for explaining a processing procedure for detecting the tilt amount of the input image from within the inspection region by the image processing apparatus according to the second embodiment.

  First, detection for detecting a local inclination of a character line or the like existing in an inspection area in an input image determined using a known technique, as in the first embodiment. The position is determined (step S301). For example, the detection positions are distributed at equal intervals over the entire inspection region, or are varied using random numbers.

  Next, a detection window having a predetermined size is set at the detection position determined in step S301 (step S302). In this embodiment, as an example, a rectangular partial image region is used as a detection window. Then, similarly to the first embodiment, the projection in the row direction in the detection window is acquired (step S303). As a result, in the present embodiment, the position of the detection window is shifted from the obtained projection, and the position of the detection window is changed so that the character line is within the detection window (step S304).

  FIG. 5 is a diagram for explaining a method of changing the position of the detection window in the second embodiment. In FIG. 5, reference numeral 501 denotes the initial state of the detection window set in step S302. In the detection window 501, there is only one projection of a complete character line, and the upper portion of the detection window is not a projection of a complete character line. That is, since there is no projection at the lower part of the detection window and the upper part is incomplete, the position of the detection window is changed and shifted upward in step S304 so that as many character line projections as possible are included in the detection window. As a result, as shown in FIG. 5, two lines of character line projections are included in the shifted detection window 502.

  Then, after the position of the detection window 501 is changed to the detection window 502 in step 304, the detection window 503 that is paired with the detection window 502 is set at a position that is a predetermined distance in the row direction (step S305). . Next, one detection window (for example, detection window 502) is fixed, and the other detection window (for example, detection window 503) is shifted in the row direction and the vertical direction. Get the correlation of the projection. Then, a deviation amount that maximizes the correlation is obtained (step S306). As a result, a local inclination angle is obtained from the amount of deviation at the time of the maximum correlation and the interval between the two detection windows (step S307). Then, the obtained angle is stored (step S308).

  Then, it is determined whether or not the inclination angles at all detection positions have been obtained (step S309). As a result, when there remains a detection position for which the inclination angle has not yet been obtained (No), the process returns to step S301 again, and the processes of steps S301 to S308 described above are repeated for the next detection position. On the other hand, when all are detected (Yes), the inclination angle of the input image is determined from the stored local inclination angle (step S310). For example, they can be determined by using the average value or median value of all inclination angles.

  As described above, according to the image processing apparatus according to the second embodiment, the position where the projection is extracted is corrected to an appropriate position. Therefore, the inclination angle can be extracted efficiently and at the same time the accuracy can be improved. The effect that it can be obtained.

<Third Embodiment>
You may perform both the change of the magnitude | size of the detection window demonstrated in 1st Embodiment mentioned above, and the change of the position of the detection window demonstrated in 2nd Embodiment. In that case, the detection window becomes a position and size more suitable for the situation of the input image, and it is not necessary to extract a useless part by projective extraction of the pair of detection windows, and the detection window is inclined from the pair of detection windows. There is an effect of increasing the possibility of obtaining a corner. Further, the image processing based on the first embodiment or the image processing based on the second embodiment may be selectively executed according to a user instruction. The configuration of the image processing apparatus according to the third embodiment is the same as that of the image processing apparatus shown in FIG. 1 described in the first and second embodiments. Here, FIG. 7 is a diagram for explaining an example of a memory map of a storage medium that stores various data processing programs that can be read out by the image processing apparatus according to the third embodiment.

  In the description of the present embodiment, the size and the position of one of the detection windows are determined, but may be determined using the other detection window that is a pair. The paired detection windows may be set again according to the obtained tilt angle, and the tilt angle may be extracted again.

<Other embodiments>
As described above, the embodiment has been described in detail. However, the present invention can take an embodiment as a system, apparatus, method, program, storage medium, or the like, and specifically includes a plurality of devices. The present invention may be applied to a system that is configured, or may be applied to an apparatus that includes a single device.

  In the present invention, a software program (in the embodiment, a program corresponding to the flowchart shown in the figure) that realizes the functions of the above-described embodiment is directly or remotely supplied to the system or apparatus, and the computer of the system or apparatus Is also achieved by reading and executing the supplied program code.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, or the like.

  As a recording medium for supplying the program, for example, floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card ROM, DVD (DVD-ROM, DVD-R) and the like.

  As another program supply method, a client computer browser is used to connect to an Internet homepage, and the computer program of the present invention itself or a compressed file including an automatic installation function is downloaded from the homepage to a recording medium such as a hard disk. Can also be supplied. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS running on the computer based on the instruction of the program is a part of the actual processing. Alternatively, the functions of the above-described embodiment can be realized by performing all of them and performing the processing.

  Furthermore, after the program read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion board or The CPU or the like provided in the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

1 is a block diagram illustrating a configuration of an image processing apparatus according to a first embodiment of the present invention. It is a flowchart for demonstrating the process sequence for detecting the inclination amount of an input image from the test | inspection area | region by the image processing apparatus which concerns on 1st Embodiment. It is a flowchart for demonstrating the process sequence for detecting the inclination amount of an input image from the test | inspection area | region by the image processing apparatus which concerns on 2nd Embodiment. It is a figure for demonstrating the method to estimate the character size from the projection in a detection window, and the method to change the magnitude | size of a detection window. It is a figure for demonstrating the change method of the position of the detection window in 2nd Embodiment. It is a figure for demonstrating the correlation of the projection of the detection window in the image processing apparatus which concerns on 1st Embodiment. It is a figure for demonstrating an example of the memory map of the storage medium which stores the various data processing program which can be read with the image processing apparatus which concerns on 3rd Embodiment.

Explanation of symbols

101 CPU
102 ROM
103 RAM
104 External storage device 105 Display 106 Keyboard 107 Pointing device 108 Scanner 109 Network interface 110 Bus

Claims (14)

Setting means for setting two partial image areas for the input document image, detection means for detecting projection of a character line from each of the two partial image areas, and the document image based on the detected projection An image processing apparatus comprising a determining means for determining the inclination of
Changing means for changing the first partial image area based on the projection detected from the first partial image area;
The setting means sets a second partial image area based on the first partial image area changed by the changing means;
The image processing apparatus, wherein the detection unit detects a projection of a character line from each of the first partial image region and the second partial image region.   The changing means estimates a character size of a character included in the first partial image region based on the projection acquired from the first partial image region, and based on the estimated character size, The image processing apparatus according to claim 1, wherein the size of the first partial image area is changed.   The changing means expands the first partial image region in a direction in which the projection of the character line of the character included in the first partial image region is interrupted until the projection is not interrupted. The image processing apparatus according to claim 2.   4. The apparatus according to claim 1, wherein the changing unit moves the position of the first partial image area based on a projection acquired from the first partial image area. 5. The image processing apparatus described.   The changing means moves the first partial image area in a direction in which the projection of the character line of the character included in the first partial image area is interrupted until the projection is not interrupted. The image processing apparatus according to claim 4.   The determination unit fixes one of the first partial image region and the second partial image region and vertically moves the other to obtain a correlation of projections detected from each partial image region, and the correlation 6. The image processing apparatus according to claim 1, wherein the inclination is determined from a positional relationship between the partial image regions when the value of the partial image region becomes maximum. A first setting step of setting a first partial image region for the document image;
A first detection step of detecting a projection of a character line from the first partial image region;
A changing step of changing the first partial image area based on the projection detected from the first partial image area;
A second setting step of setting a second partial image region for the document image based on the first partial image region changed by the changing step;
A detection step of detecting a projection of a character line from each of the first partial image region and the second partial image region;
A determination step of determining an inclination of the document image based on the detected projection.   The changing step estimates a character size of a character included in the first partial image region based on the projection acquired from the first partial image region, and based on the estimated character size, The image processing method according to claim 7, wherein the size of the first partial image region is changed.   The changing step is to enlarge the first partial image area in a direction in which the projection of the character line of the character included in the first partial image area is interrupted until the projection is not interrupted. The image processing method according to claim 8.   10. The method according to claim 7, wherein the changing step moves the position of the first partial image area based on a projection acquired from the first partial image area. The image processing method as described.   The changing step moves the first partial image area in a direction in which the projection of the character line of the character included in the first partial image area is interrupted until the projection is not interrupted. The image processing method according to claim 10.   In the determining step, one of the first partial image region or the second partial image region is fixed and the other is vertically moved to obtain a correlation of projections detected from the respective partial image regions, and the correlation 12. The image processing method according to claim 7, wherein the inclination is determined from a positional relationship between the partial image regions when the value of the partial image region becomes maximum.   A program for causing a computer to execute the image processing method according to any one of claims 7 to 12.   A computer-readable storage medium storing the program according to claim 13.

Images