JP3636447B2 - Other event description detection apparatus and other event description detection method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus for detecting other event descriptions and a method for detecting other event descriptions, and more particularly to an apparatus for detecting other event descriptions and a method for detecting other event descriptions for detecting other event descriptions such as improper entries other than marks made on a mark sheet. .
[0002]
[Prior art]
In recent years, from the viewpoint of efficiently and reliably performing data aggregation work, mark-type entry sheets (mark sheets) provided with mark entry locations have been widely used, and are widely used in various fields. In general, such a mark sheet is printed with a timing mark associated with a mark entry position for each row or each column so that the entered mark can be read by a mark sheet reader. . The mark sheet reading device reads the timing mark and determines the position of the mark written in the mark entry position corresponding to the timing mark.
[0003]
In such a mark sheet, there are cases where so-called other matters such as improper description or dirt written other than a predetermined mark entry location are generated at the time of mark entry. If there are other matters on the mark sheet, it may be necessary to treat the mark sheet as invalid depending on the field of use.
[0004]
For example, in recent years, an electronic voting system that digitizes a series of operations in election voting has been proposed. In this electronic voting system, a candidate for voting is selected by marking a ballot on a mark sheet printed with a number assigned to a candidate name or a party name. Then, a predetermined reading device image-recognizes the filled-in voting form, and the voted election candidate is determined based on the position of the read mark.
[0005]
[Problems to be solved by the invention]
However, when other matters are detected, for example, the entire mark sheet after entry is read by an image scanner or the like, and the read mark sheet image data and the original mark sheet image data before entry are each read in the number of dots. However, in this case, since the image is composed entirely of black dots, it will be filled in later to see if the entry other than the specified part is the content originally printed on the mark sheet. However, there is a problem that it is difficult to automatically identify whether it is a description of other matters.
[0006]
In the electronic balloting system, if it is not possible to properly identify other matters on the ballot on the mark sheet, it may harm the election fairness and affect the election results. There is also a risk of an increase.
[0007]
This invention is made in view of such a point, and provides the other-thing description detection apparatus and the other-thing description detection method which can identify and accurately detect the other-thing description which exists in a mark sheet. Objective.
[0008]
[Means for Solving the Problems]
In the present invention, in order to solve the above-described problem, in the event description detection apparatus for detecting the event description of the mark sheet, a mark sheet reading unit that reads the dot image before and after the mark entry of the mark sheet, and the read dot image before the mark entry Using the comparison information, a comparison information storage unit that stores the obtained print content and the position of the mark entry location as comparison information, a division setting unit that divides the read dot image after the mark entry into a predetermined division, and division using the comparison information The number of black dots in the post-marked dot image for each of the divided sections, and the number of black dots in the section obtained by subtracting the number of mark black dots when a mark black dot is detected at the mark entry location A ratio calculation unit that calculates a ratio to the total number of dots, and the calculated ratio is less than a certain value A determination unit that determines that there is no other description in the section, and determines that there is an entry other than the print content or mark when the ratio exceeds the certain value. The other matter description detection device is provided.
[0009]
According to such an other matter description detection apparatus, the ratio calculation unit calculates the ratio of the number of black dots in a section to the total number of dots in the section for each divided section. At this time, if the ratio calculation unit detects the mark black dot at the mark entry location in this section using the comparison information, it is regarded as a valid entry, and the number of mark black dots, which is the number of black dots, is subtracted. The ratio is calculated as the number of black dots in the section. Then, based on the ratio, the determination unit determines whether or not an entry has been made in places other than the mark entry area, and if the calculated ratio is less than a certain value, there is another description in that section. Judge that not.
[0010]
Further, in the present invention, in the other event detection method for detecting the other event description of the mark sheet, the dot image before the mark entry on the mark sheet is read, and the printed content and the mark entry location obtained from the read dot image before the mark entry are read. Stores the position as comparison information, reads the dot image after entering the mark on the mark sheet, divides the read dot image after entering the mark into certain sections, and using the comparison information, for each of the divided sections, The number of black dots in the area of the dot image after the mark entry, and when the mark black dot is detected at the mark entry position, the ratio of the number of black dots in the area obtained by subtracting the number of mark black dots to the total number of dots in the area If the calculated ratio is less than a certain value, other matters are described in the section. It determined that no standing, and other events described detection method and judging that there is complete except for the printed contents or mark if the proportion exceeds the predetermined value is provided.
[0011]
According to such an other matter description detection method, first, the print contents obtained by reading the dot image before mark entry and the position of the mark entry location are stored as comparison information. After the mark is entered, the dot image is read and divided into fixed sections, and the ratio of the number of black dots in the section to the total number of dots in the section is calculated for each divided section. At this time, using the comparison information, if a mark black dot at the mark entry location is detected in this section, this is regarded as a valid entry, and the number of black dots, which is the number of black dots, is subtracted. The ratio is calculated as the number of black dots. Then, the existence of other matters in the section is determined depending on whether the ratio is equal to or less than a certain value.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram illustrating a configuration example of the other-affairs description detection device.
[0013]
The other event description detection apparatus 10 includes a mark sheet reading unit 11, a comparison information storage unit 12, and an image analysis unit 13.
The mark sheet reading unit 11 reads the dot images of the mark sheet 20 before and after the mark entry. In the following, the dot image of the mark sheet 20 before entering the mark is referred to as “dot image before entering the mark”, and the dot image of the mark sheet 20 after entering the mark is referred to as “dot image after entering the mark”.
[0014]
The print information on the mark sheet 20 included in the pre-mark dot image read by the mark sheet reading unit 11 and information on the position of the mark entry location where the mark is written are stored in the comparison information storage unit 12 as comparison information. The
[0015]
In addition, the image analysis unit 13 of the other matter description detection apparatus 10 includes a section setting unit 13a, a ratio calculation unit 13b, an exclusive logic processing unit 13c, a pattern matching unit 13d, a determination unit 13e, and a determination result output unit 13f. .
[0016]
The section setting unit 13a divides the dot image after the mark entry read by the mark sheet reading unit 11 into fixed sections.
First, the ratio calculation unit 13b obtains the number of black dots in a section, which is the number of black dots in the dot image after the mark entry in the section actually read for each divided section. At this time, the ratio calculation unit 13b uses the comparison information to detect a mark black dot that is a black dot that is generated when a mark entry location exists in this section and a mark is entered in the mark entry location. In this case, a value obtained by subtracting the number of mark black dots, which is the number of black dots, is set as the number of black dots in the section. Then, the ratio calculation unit 13b calculates the ratio of the number of black dots in the partition to the total number of dots occupying this partition (hereinafter referred to as “total number of dots in the partition”) using the number of black dots in the partition thus obtained. To do.
[0017]
The exclusive logic processing unit 13c performs exclusive logic processing on the dot image after the mark entry read by the mark sheet reading unit 11 and the dot image before the mark entry stored in the comparison information storage unit 12 as comparison information. In this exclusive logic processing, the dot image after the mark entry and the dot image before the mark entry are overlapped to extract a non-matching portion.
[0018]
The pattern matching unit 13d performs collation with a character pattern stored in advance in order to determine whether or not the portion formed later on the mark sheet 20 before the mark has been formed forms a character.
[0019]
Based on the calculation result of the ratio calculation unit 13b and the processing result of the exclusive logic processing unit 13c and the collation result of the pattern matching unit 13d, the determination unit 13e may enter an inappropriate entry or the like on the mark sheet 20 afterwards. It is determined whether or not it is a so-called other matter description.
[0020]
The determination result output unit 13f outputs the determination result of the determination unit 13e by an external output device or the like.
When the other event description is detected by the other event description detecting apparatus 10 as described above, first, the dot image before mark entry is read by the mark sheet reading unit 11, and the printed contents and the position of the mark entry location are compared with the comparison information. Is stored in the comparison information storage unit 12.
[0021]
In the other event description detection device 10, the dot image after the mark entry is read by the mark sheet reading unit 11 in order to detect the other event description. The dot image after the mark entry is subjected to image analysis by the image analysis unit 13 using the comparison information in the comparison information storage unit 12.
[0022]
2A and 2B are explanatory diagrams of image analysis in the other event description detection device, in which FIG. 2A is a diagram showing a state in which the mark sheet is partitioned and FIG. 2B is an enlarged view of the partition A. FIG.
A plurality of mark entry portions 21 are provided on the mark sheet 20, and are arranged in a state of being aligned vertically and horizontally on the mark sheet 20. In this mark sheet 20, there is an other description 22 which is an entry other than the mark entered in the mark entry location 21.
[0023]
In the detection of the other matter description 22, first, the section setting unit 13a of the image analysis unit 13 divides the dot image into a predetermined section after the mark is entered.
Next, the ratio calculation unit 13b of the image analysis unit 13 first obtains the number of black dots in the division of the dot image after the mark entry in the division for each divided division. At this time, if the mark entry location 21 exists in this section and the mark black dot is detected at the mark entry location using the comparison information, the ratio calculation unit 13b regards this as a valid entry. The number of black dots, which is the number of black dots, is subtracted from the number of black dots in the section. Then, the ratio calculation unit 13b calculates a ratio X of the number of black dots in the section obtained in this way to the total number of dots in the section for each section. That is, the ratio X is calculated by the following formula: X = {(number of black dots in section) − (number of mark black dots)} / (total number of dots in section).
[0024]
In calculating the ratio X in this way, when an inter-compartment dot image that is a dot image extending between a plurality of sections is detected, the image analysis unit 13 performs a new operation centering on the inter-compartment dot image. A new partition is set, and the ratio X is calculated for each new partition.
[0025]
FIG. 3 is an explanatory diagram of image analysis of an inter-compartment dot image, in which (a) shows an inter-compartment dot image that spans between sections, and (b) shows a newly set section.
[0026]
As shown in FIG. 3A, when the image analysis unit 13 detects an inter-compartment dot image 30 that straddles between the initially set divisions A and B, the division setting unit 13a performs the processing shown in FIG. ), A new section C is set with the inter-partition dot image 30 as the center. Then, the ratio X is calculated for the newly set section C.
[0027]
As a result, the ratio X of the inter-compartment dot image 30 to the section C is larger than the ratio X when a part of the inter-compartment dot image 30 occupies the section A or the section B. 30 can be detected more accurately.
[0028]
After calculating the ratio X, the determination unit 13e of the image analysis unit 13 determines that there is no entry other than the mark in the section when the ratio X is a certain value or less, that is, there is no other matter description 22. To do. On the other hand, when the ratio X exceeds a certain value, it is determined that there is some entry other than the mark in the section.
[0029]
When the ratio X is calculated by the above image analysis, the number of black dots in the section to be read is larger and the ratio X is higher in the section including the print contents together with the mark entry portion 21 than in the case where the print content is not included. In addition, in the case where only the print contents are included in the section, the set constant value is always exceeded. Therefore, when the determination unit 13e determines that there is some entry other than the mark entry location 21 in this section, it is necessary to determine whether the entry is a print content or an entry other than a mark. is there.
[0030]
For this reason, the exclusive logic processing unit 13c performs exclusive logic processing on the dot image after mark entry and the dot image before mark entry to determine whether or not the print contents are included. That is, the dot image after the mark entry and the dot image before the mark entry are overlapped to extract a non-matching portion. If both match, the determination unit 13e determines that it is caused by the print content even if the ratio X exceeds a certain value. On the other hand, if a non-matching part occurs, it is determined that the ratio X exceeding a certain value is due to the entry other than the mark.
[0031]
By the way, as the entry other than the mark, in addition to the above-mentioned other matter description 22, for example, the name entered in the name entry column provided on the mark sheet 20 is a character entered later in a predetermined place. In some cases. In such a case, the pattern matching unit 13d performs pattern matching with a character pattern stored in advance for a non-matching portion generated during processing in the exclusive logic processing unit 13c. Thereby, it is determined whether or not the entry other than the mark forms a character.
[0032]
The determination unit 13e determines whether or not there is the other event description 22 in the section based on the value of the ratio X and the exclusive logic processing result and the pattern matching result. This determination result is externally output by the determination result output unit 13 f of the image analysis unit 13.
[0033]
In the other event description detection apparatus 10, the above processing is performed for all the set sections, and the other event description 22 is detected for each section.
Hereinafter, the embodiment of the present invention will be described in detail with reference to the drawings, taking as an example a case where the electronic balloting system is applied to the detection of other matters in a ballot paper.
[0034]
FIG. 4 is a diagram illustrating a configuration example of an electronic voting system using the other-affairs description detection device.
The electronic voting system 40 includes image scanners 41a, 41b, 41c, and 41d for reading dot images before and after filling in a ballot paper. The image scanners 41a, 41b, 41c, and 41d function as the mark sheet reading unit 11 shown in FIG.
[0035]
Computer terminals 42a, 42b, 42c, and 42d are connected to the image scanners 41a, 41b, 41c, and 41d, respectively, for performing image analysis on the read dot images before and after the entry. The computer terminals 42a, 42b, 42c, and 42d function as the comparison information storage unit 12 and the image analysis unit 13 shown in FIG.
[0036]
Each of these computer terminals 42 a, 42 b, 42 c, 42 d is connected to a server computer 44 for counting up votes via a network 43. The server computer 44 is connected to an output device 45 such as a printer or a display board so that the voting results are printed or displayed.
[0037]
FIG. 5 is a diagram showing an example of voting paper used in the electronic voting system.
The ballot 50 is provided with a mark entry location 52 on each candidate name 51, and a voter selects a candidate to vote by entering a mark in the mark entry location 52. It is like that.
[0038]
In addition, the ballot form 60 is provided with mark entry portions 61a, 61b, and 61c for marking each digit in a vertical line in order to represent a hundred digit, a tenth digit, and a first digit. Yes. A voter enters a mark in a mark entry location 61a, 61b, 61c with a number previously assigned to the candidate or political party, and the candidate or political party to vote is specified.
[0039]
Next, in the electronic voting system 40 having the above-described configuration, the flow of processing will be described by taking as an example the case of detecting other description of the voting paper 50.
FIG. 6 is a diagram showing a flow of processing before voting.
[0040]
When detecting other information on the ballot paper, first, the dot images before marking on the ballot paper 50 are read by the image scanners 41a, 41b, 41c, and 41d before voting (step S10).
[0041]
Then, the print contents described on the ballot 50 and the position of the mark entry location 52 are stored in the computer terminals 42a, 42b, 42c, and 42d as comparison information (step S11).
[0042]
It should be noted that reading of the voting paper 50 before the mark entry may be performed only by, for example, one computer terminal 42a among the plurality of image scanners 41a, 41b, 41c, 41d. In this case, the obtained comparison information is transmitted to the other computer terminals 42b, 42c, and 42d via the network 43 and stored.
[0043]
FIG. 7 is a diagram showing the flow of image analysis processing.
In the voting, first, the computer terminals 42a, 42b, 42c, and 42d read the dot images after the marks on the voting paper 60 with the image scanners 41a, 41b, 41c, and 41d (step S20).
[0044]
Next, the read dot image after mark entry is divided into predetermined sections (step S21).
For each section, the ratio X of the number of black dots in the section to the total number of dots in the section is calculated (step S22). At that time, based on the comparison information stored in step S11 shown in FIG. 6, the ratio X is obtained by subtracting the number of mark black dots of the mark entered at the position of the mark entry location 52 as the number of black dots in the section. Is calculated.
[0045]
It is determined whether or not the calculated ratio X is equal to or less than a certain value (step S23).
Here, when it is determined that the ratio X exceeds a certain value, exclusive logic processing is performed on the dot image before the mark entry and the dot image after the mark entry (step S24). It is determined whether or not (step S25).
[0046]
If a portion that does not match is generated by this exclusive logic processing, it is determined that there is an entry other than a mark in this section (step S26). On the other hand, if there is no unmatched portion, it is determined that the print content is in this section (step S27).
[0047]
If it is determined in step S23 that the ratio X is equal to or less than a certain value, it is determined that there is no entry other than the mark in this section (step S28).
Furthermore, when it is determined that there is an entry other than the mark, if it is determined whether or not it is a character entered at a predetermined location, for the non-matching part generated by the exclusive logic process, Perform pattern matching.
[0048]
FIG. 8 is a diagram showing a flow of pattern matching processing.
In pattern matching, a non-matching part generated by exclusive logic processing is extracted (step S30), and the non-matching part is compared with a character pattern candidate stored in advance to determine whether or not they match. Judgment is made (step S31).
[0049]
Here, if the candidate matches, the character pattern is determined (step S32). If the character pattern candidate does not match in step S31, it is determined whether or not the next candidate matches. This process is repeated until the character patterns match.
[0050]
Thereby, it is determined whether or not the entry other than the mark forms a character.
The computer terminals 42a, 42b, 42c, and 42d display, on the screen, the presence / absence of other matters determined in consideration of the exclusive logic processing result and the pattern matching result together with the value of the ratio X for each partition. This information is transmitted to the server computer 44.
[0051]
The server computer 44 that has received information indicating the presence or absence of other matters stores information regarding whether the voting by the ballot is valid or invalid, and reflects it in the voting result.
[0052]
In the above description, the case where the other description of the ballot 50 is detected has been described. Of course, the same processing can be performed even in the case of the ballot 60.
[0053]
【The invention's effect】
As described above, in the present invention, the comparison information obtained from the dot image before the mark entry is stored, the dot image after the mark entry is read and divided into fixed sections, and the total number of dots in the section is determined for each section. The ratio of the number of black dots in the compartment after subtracting the number of mark black dots was calculated. Thereby, the presence or absence of other matters in the section can be determined depending on whether the ratio is equal to or less than a certain value.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration example of an other-event description detection apparatus.
2A and 2B are explanatory diagrams of image analysis in the other-affairs description detection apparatus, wherein FIG. 2A is a diagram showing a state in which a mark sheet is partitioned, and FIG.
FIG. 3 is an explanatory diagram of image analysis of an inter-compartment dot image, in which (a) shows a dot image that straddles between sections, and (b) shows a newly set section.
FIG. 4 is a diagram illustrating a configuration example of an electronic voting system using an other-affairs description detection device.
FIG. 5 is a diagram showing an example of voting paper used in the electronic voting system.
FIG. 6 is a diagram showing a flow of processing before voting.
FIG. 7 is a diagram illustrating a flow of image analysis processing.
FIG. 8 is a diagram illustrating a flow of pattern matching processing.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 Other matter description detection apparatus 11 Mark sheet reading part 12 Comparison information storage part 13 Image analysis part 13a Section setting part 13b Ratio calculation part 13c Exclusive logic processing part 13d Pattern matching part 13e Determination part 13f Determination result output part 20 Mark sheets 21, 52, 61a, 61b, 61c Mark entry location 22 Other information 30 Inter-compartment dot image 40 Electronic voting system 41a, 41b, 41c, 41d Image scanner 42a, 42b, 42c, 42d Computer terminal 43 Network 44 Server computer 45 Output device 50, 60 Ballot 51 Each candidate name

Claims (7)

In the other event description detection device that detects the other event description of the mark sheet,
A mark sheet reading unit that reads dot images before and after filling in a mark on the mark sheet;
A comparison information storage unit that stores the print content obtained from the read dot image before mark entry and the position of the mark entry part as comparison information;
A section setting unit that divides the dot image into a certain section after writing the read mark;
Using the comparison information, for each of the divided sections, the number of black dots in the section of the dot image after the mark entry, and when the mark black dot is detected at the mark entry position, the number of mark black dots is subtracted. A ratio calculation unit that calculates a ratio of the number of black dots in a section to the total number of dots in the section;
When the calculated ratio is equal to or less than a certain value, it is determined that there is no other description in the section, and when the ratio exceeds the certain value, it is determined that there is an entry other than the printed content or mark. A determination unit to perform,
A device for detecting other description, comprising: An exclusive logic processing unit that performs an exclusive logic process on the dot image before the mark entry and the dot image after the mark entry;
In the case where the calculated ratio exceeds the certain value, and the dot image before mark entry and the dot image after mark entry coincide with each other by the exclusive logic processing, the determination unit includes the section. The apparatus according to claim 1, wherein it is determined that the print content is included. An exclusive logic processing unit that performs an exclusive logic process on the dot image before the mark entry and the dot image after the mark entry;
If the calculated ratio exceeds the certain value, and the dot image before the mark entry and the dot image after the mark entry do not match by the exclusive logic process, the coincidence occurred in the exclusive logic process The apparatus according to claim 1, further comprising a pattern matching unit that performs pattern matching with a character pattern stored in advance for a portion that is not to be stored. In the other event description detection method for detecting the other event description of the mark sheet,
Read the dot image before marking on the mark sheet,
Store the printed content obtained from the read dot image before mark entry and the position of the mark entry location as comparison information,
Read the dot image after entering the mark on the mark sheet,
Divide the dot image after writing the read mark into certain sections,
Using the comparison information, for each of the divided sections, the number of black dots in the section of the dot image after the mark entry, and when the mark black dot is detected at the mark entry position, the number of mark black dots is subtracted. Calculate the ratio of the number of black dots in the block to the total number of dots in the block,
When the calculated ratio is equal to or less than a certain value, it is determined that there is no other description in the section, and when the ratio exceeds the certain value, it is determined that there is an entry other than the printed content or mark. A method for detecting other matters described in the above. When the read dot image after marking the dot image is divided into the predetermined partitions, if an inter-compartment dot image extending between the plurality of partitions is detected, the dot image after marking is converted to the inter-compartment dot image. The method according to claim 4, further comprising: dividing the image into new sections centered on an image and calculating the ratio for each new section. When it is determined that there is an entry other than the print content or the mark when the ratio exceeds the certain value, an exclusive logic process is performed on the dot image before the mark entry and the dot image after the mark entry,
5. The printing apparatus according to claim 4, wherein when the dot image before the mark entry matches the dot image after the mark entry by the exclusive logic processing, it is determined that the print content is included in the section. Detection method of other matters. When it is determined that there is an entry other than the print content or the mark when the ratio exceeds the certain value, an exclusive logic process is performed on the dot image before the mark entry and the dot image after the mark entry,
If the dot image before the mark entry and the dot image after the mark entry do not match by the exclusive logic process, pattern matching with a pre-stored character pattern is performed on the mismatched part generated by the exclusive logic process. 5. The method for detecting a foreign matter according to claim 4, wherein:

Images