JP2004318589A - Slip reading device and method, and program storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To read a mark entry frame of a non-dropout slip without making it necessary to highly precisely define the position or size of the entry frame in advance, and to reduce the conditions of the positioning of an inputted slip and preliminarily defined entry frame information. <P>SOLUTION: This slip reading device comprises a horizontal ruled line extracting means, a vertical ruled line extracting means, an entry frame segmenting means for defining a range surrounded by the horizontal ruled line and the vertical ruled line as an entry frame, an image extracting means for extracting an entry image from the range segmented by the entry frame segmenting means, and a mark deciding means for deciding that the entry of a mark has been performed when the extracted image is present, and for deciding that the entry of the mark has not been performed when any image is not present. This slip reading device is configured to automatically search the entry frame from the designated range, and to decide whether or not the entry of the mark has been performed. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a form reading apparatus, a method, and a program storage medium for reading a mark written in a frame on a form.
[0002]
[Prior art]
If you want the entrant to select the appropriate item on a form such as an application form, print out the entry frame in advance near each item, and the entrant should put a circle (丸) or check mark (チ ェ ッ ク) in the frame. I try to write.
[0003]
Conventionally, when a completed form is input to a computer or the like, the form and the like on the form are printed in dropout color, optically read by an image scanner or the like so that the form and the like are dropped out, and stored in advance. It is configured to determine whether or not there is a written mark at a read position where the mark is written. When a drop-out form is used, only the registration mark and the written mark remain after the image is read by the image scanner, so that it is easy to determine whether or not the data is filled, and the reading can be easily performed.
[0004]
In Japanese Patent Application Laid-Open No. 11-265422, an unfilled form image is stored in advance, and when a filled-out form image is input, a difference from the unfilled form image is extracted. , And that part is used as the entered part. In this case, the entry frame may be non-dropout.
[0005]
Alternatively, in Japanese Patent Application Laid-Open No. 7-28923, an area used for determination is configured to be smaller than the size of an entry frame stored in advance. In this case, even if a non-dropout is used for the entry frame, it is possible to prevent the frame line of the entry frame from being erroneously used to determine whether or not there is entry.
[0006]
[Problems to be solved by the invention]
However, the first conventional example has a drawback that a dedicated dropout form is required, and an image scanner that can handle the dropout form must be used for reading.
[0007]
Further, in the second conventional example, since an unfilled form image is stored in advance, it is necessary to store the image every time the form to be handled increases, and there is a disadvantage that the storage capacity is significantly increased. Further, in this example, since the difference between the images is obtained, there is a drawback that high accuracy is required for the alignment between the blank form and the completed form. Even if the image is slightly enlarged or reduced, the effect on the difference is large, and sufficient correction must be performed for translation and rotation.
[0008]
Alternatively, in the third conventional example, the definition information for storing the position and size of the entry frame must be accurately obtained in advance, and similarly to the second conventional example, the input form image and the There was a disadvantage that the deviation from the definition had to be corrected with high precision.
[0009]
The present invention has been made to solve the above problems, and an object of the present invention is to make it possible to determine whether or not there is a marked mark even in a non-dropout form. .
[0010]
Further, the position of the entry frame can be automatically searched out without specifying the reading range with high accuracy.
[0011]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to the present application is directed to a horizontal ruled line extracting means for extracting a horizontal ruled line from a designated range of an input image, a vertical ruled line extracting means for extracting a vertical ruled line, and a method defined by a horizontal ruled line and a vertical ruled line. Entry frame cutting means having an entry frame within the range, image extraction means for extracting an image entered from the range cut by the entry frame cutting means, if there is an extracted image, it is assumed that a mark has been entered, When there is no image, a mark determination unit that determines that there is no entry is provided.
[0012]
In the above configuration, an operation is performed to automatically search for an entry frame from within the specified range and determine whether a mark has been entered.
[0013]
Further, in order to achieve the above object, the invention according to the present application provides a row rectangle extracting means for projecting a designated range of an input image in a horizontal direction and extracting a row rectangle, and a column for projecting a vertical range and extracting a column rectangle. The image processing apparatus further includes a block selecting unit that selects the largest one of the image blocks extracted by the rectangular extracting unit, the row rectangular extracting unit, and the column rectangular extracting unit.
In the above configuration, an operation is performed to narrow the range in which the entry frame is searched for by selecting a block.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
(First embodiment)
FIG. 1 is a block diagram showing a configuration of a form reading apparatus for carrying out the present invention. Reference numeral 101 denotes a CPU that controls the entire apparatus according to a control program stored in a ROM 102; 102, a ROM that stores a control program of the apparatus such as a process shown in a flowchart to be described later which is executed by the CPU 101; A storage RAM 104 is an external storage device such as a magnetic disk, 105 is a display, 106 is a keyboard, 107 is a pointing device such as a mouse, and 108 is an image scanner for reading an image. Reference numeral 109 denotes a network interface, which communicates with a remote device (not shown) to read and write programs and data.
[0015]
The outline of the present embodiment executed by the form reading apparatus having the configuration shown in FIG. 1 will be described with reference to the flowchart of FIG.
[0016]
FIG. 2 illustrates a method of extracting an entry frame and determining whether a mark is entered in the frame. FIG. 3 shows an example of a form processed by the present embodiment, where 301 indicates an entry frame and 302 indicates a range in which processing is performed.
[0017]
In step S201, horizontal ruled lines are extracted from the processing range 302 of the image read from the photoelectric conversion device such as the image scanner 108 or the image already read and stored in the external storage device 104 or the like. In S202, it is checked whether or not there are two extracted horizontal ruled lines. If there are not two, the process ends assuming that there is no entry frame. Next, a vertical ruled line is extracted from between the two horizontal ruled lines obtained in S203, and it is checked whether there are two vertical ruled lines similarly extracted in S204.
[0018]
Here, a method of extracting ruled lines will be described with reference to FIG. 5 using vertical ruled lines as an example. The extraction of the horizontal ruled line is equivalent to the case where the coordinate axis of the vertical ruled line is rotated by 90 degrees.
In step S501, the frequency of a black pixel is calculated with respect to the horizontal projection plane. As a result, the vertical ruled lines and the written characters are projected, as shown in FIG.
[0019]
In S502, a frequency threshold Tb for obtaining a range including the vertical ruled line is determined. The threshold value Tb is an average value of the frequencies of the portions where the projection exists. Further, in S503, a frequency portion larger than the threshold value Tb is extracted, and the number is set to N. If N is 0 in S504, there is no vertical ruled line in S505, and the process ends. In step S506, a threshold value Tp based on the maximum value of each part is determined. In step S507, a part having a frequency higher than Tp is searched. Tp may be 80% of the maximum value of the maximum value of each part, or the average of the top N / 2 of the maximum values. If there is a projection whose frequency is equal to or greater than Tp in S508, that portion is set as a vertical ruled line in S509.
[0020]
In S205, the inner part surrounded by the extracted horizontal ruled line and vertical ruled line is extracted as an entry frame area. From the obtained entry frame area, an image in the area is extracted in S206. If there is no image in the entry frame in S207, the entry frame is left blank in S208. If there is an image in the entry frame in S207, the entry frame has a mark in S209. However, since there is a possibility that noise is present in the frame, when the area of the image portion is sufficiently smaller than the size of the frame, it is sufficient that no image exists in the entry frame.
[0021]
As described above, according to this embodiment, the processing range is specified to be larger than the entry frame, and the entry frame can be automatically extracted within the range. Therefore, it is necessary to define the position and size of the entry frame with high precision in advance. There is no effect.
[0022]
Further, even if a part of the character touches the entry frame, the ruled line is extracted without any problem.
[0023]
Note that there is no problem even if these are supplied by a program and configured to be executed by the CPU 101.
[0024]
FIG. 6 is a diagram illustrating a memory map of a storage medium that stores various data processing programs that can be read by the form reading apparatus according to the present invention. Although not particularly shown, data dependent on various programs is also managed in the directory. Further, a program for installing various programs on a computer may be stored.
[0025]
(Second embodiment)
Another embodiment executed by the form reading apparatus shown in FIG. 1 will be described with reference to the flowchart of FIG. FIG. 8 shows an example of a form processed by the present embodiment. Reference numerals 801, 802, and 803 denote entry frames, and reference numeral 804 denotes a processing range. In this example, three entry frames are arranged vertically, and the processing range includes a part of the character portion on the right.
[0026]
In step S701, projection is performed in the horizontal direction within the processing range. As a result, a row projection is obtained, and extraction of a row rectangle is extracted in S702. In the example of FIG. 8, upper and lower blank portions in the processing range are removed, and one row rectangle is obtained.
[0027]
Next, in step S703, projection is performed in the vertical direction on each row rectangle. As a result, in S704, an image block divided by a rectangle is extracted. In the example of FIG. 8, three entry frame portions 805 and an image block 806 of a right character portion are extracted.
[0028]
In step S705, the largest block is selected from the obtained image blocks, and the process is performed on the selected block. In the example of FIG. 8, 805 is selected.
[0029]
In step S706, a vertical ruled line is extracted. In S707, it is checked whether or not two vertical ruled lines exist. If not, the process is terminated assuming that there is no entry frame. Next, a horizontal ruled line is extracted from the obtained two ruled lines in S708, and it is checked in S709 whether two or more horizontal ruled lines exist.
[0030]
In step S710, each of the inner portions surrounded by the obtained vertical and horizontal ruled lines is extracted as an entry frame area, and in step S711, an image in the area is extracted.
[0031]
It is checked in step S712 whether or not an image exists for all entry frames. If there is an image, it is determined that there is a mark in step S714.
[0032]
As described above, according to the present embodiment, there is an effect that the processing can be performed even if the processing range includes a character, a part of another image, noise, or the like. In addition, there is an effect that the design conditions of the form are relaxed and the degree of freedom is increased.
[0033]
In the present embodiment, an example in which the entry frames are arranged vertically is used. However, even when the entry frames are arranged horizontally, it is needless to say that the processing can be performed by changing the order of the vertical and horizontal processing. Alternatively, the number of entry frames may be specified in advance, and after the processing in this embodiment, the numbers may be compared. If the numbers are the same, the processing may be completed normally.
[0034]
【The invention's effect】
As described above, according to the present invention, the reading of the mark entry frame of the non-dropout form does not require the position and size of the entry frame to be defined with high precision in advance.
[0035]
Further, according to the present invention, there is an effect that the condition for aligning the input form with the previously defined entry frame information can be reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram of a form reading apparatus according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating an example of a mark reading procedure according to the first embodiment.
FIG. 3 is a diagram showing an example of a form processed by FIG. 2 according to the first embodiment.
FIG. 4 is a diagram illustrating a state in which the frequency of a black pixel is taken with respect to a horizontal projection plane according to the first embodiment.
FIG. 5 is a flowchart illustrating an example of a procedure for extracting a vertical ruled line according to the first embodiment.
FIG. 6 is a diagram illustrating a memory map of a storage medium of the form reading apparatus according to the present invention.
FIG. 7 is a flowchart illustrating an example of a mark reading procedure according to the second embodiment.
FIG. 8 is a diagram showing an example of a form processed by FIG. 7 according to the second embodiment.
[Explanation of symbols]
101 CPU
102 ROM
103 RAM
104 external storage device 105 display 106 keyboard 107 pointing device 108 image scanner 109 network interface

Claims (5)

Horizontal ruled line extracting means for extracting horizontal ruled lines from the specified range of the input image, vertical ruled line extracting means for extracting vertical ruled lines, entry frame cutting means for defining an area enclosed by the horizontal ruled lines and vertical ruled lines, and entry frame Image extraction means for extracting an image entered from within a range cut out by the cutting means, mark determination that a mark has been entered if an extracted image exists, and that no mark has been entered if no image exists Means for automatically reading out an entry frame from a designated range and determining whether or not a mark has been entered. A row rectangle extracting means for projecting the specified range of the input image in the horizontal direction to extract a row rectangle, a column rectangle extracting means for projecting in the vertical direction to extract a column rectangle, a row rectangle extracting means, and a column rectangular extracting means. 2. The form reading apparatus according to claim 1, further comprising a block selecting means for selecting a largest one of the image blocks, and narrowing down a search range of the entry frame by selecting the block. Horizontal ruled line extraction process for extracting horizontal ruled lines from the specified range of the input image, vertical ruled line extracting process for extracting vertical ruled lines, an input frame cutout process where the area enclosed by the horizontal ruled line and the vertical ruled line is an input frame, an input frame An image extraction step of extracting an image entered from within a range cut out in the cutting step, a mark judgment that a mark has been entered if an extracted image exists, and that no mark has been entered if no image exists. A form reading method comprising a step, wherein it is possible to automatically search for an entry frame from a specified range and determine whether or not a mark has been entered. The specified range of the input image is horizontally projected to extract a row rectangle. A row rectangle extraction step is performed, and a vertical projection is performed to extract a column rectangle. A column rectangle extraction step, a row rectangle extraction step, and a column rectangle extraction step are performed. 4. The form reading method according to claim 3, further comprising a block selecting step of selecting the largest image block from the selected image blocks, and narrowing a search range of the entry frame by selecting the block. Horizontal ruled line extraction process for extracting horizontal ruled lines from the specified range of the input image, vertical ruled line extracting process for extracting vertical ruled lines, an input frame cutout process where the area enclosed by the horizontal ruled line and the vertical ruled line is an input frame, an input frame An image extraction step of extracting an image entered from within a range cut out in the cutting step, a mark judgment that a mark has been entered if an extracted image exists, and that no mark has been entered if no image exists. A storage medium storing a computer-readable program, comprising: a step of automatically searching for an entry frame from a specified range and determining whether a mark has been entered.

Images