JPH08202818A - Format information processor - Google Patents

Abstract

PURPOSE: To easily judge whether an error exists in registered format information or not. CONSTITUTION: A picture read means reading a slip, a storage device 22 storing format information, a controller 31 and a display device are provided. The controller 31 is provided with an image memory 23 storing a slip image corresponding to the slip which the picture read means reads, an image display means 32 displaying the slip image on the display device and a dummy data write means 33 overlapping dummy data corresponding to format information with the slip image stored in the image memory 23 and writing them. The controller 31 reads the slip by the picture read means and stores the slip image into the image memory 23. Format information on the slip is read from the storage device 22 and dummy data corresponding to format information is overlapped with the slip image so as to write them.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a format information processing device in an optical character reader (hereinafter referred to as "OCR").

[0002]

2. Description of the Related Art Conventionally, in OCR, characters written on a form are imaged on an image sensor by an optical system,
The image signal obtained by the image sensor is converted into an image image, and the image image is written in the image memory. Then, the character is recognized by scanning the image memory and extracting the characteristics of the character image cut out from the image image.

FIG. 2 is a conceptual diagram of a conventional OCR. In the figure, 11 is a form in which characters are entered. The form 1
The character written in 1 is scanned by an optical system (not shown) and imaged on the image sensor. The image signal obtained by the image sensor is converted into an image image, and the image image is written in the image memory 13 provided as a working memory.

The OCR has a storage device 12, and the storage device 12 stores format information of a plurality of types of forms 11 in advance. Therefore, a personal computer or the like that constitutes the control unit of the OCR refers to the format information stored in the storage device 12 to perform the character reading process. That is, the format information is made up of the positions of the entry frames 16 and 17 on the form 11, the attributes of the characters entered in the entry frames 16 and 17 (types such as printed characters, handwritten characters, ANK, and Chinese characters). It is stored for each of the entry frames 16 and 17.

In this embodiment, entry frames 16 and 17 are formed in the form 11, and the entry frame 16 is set at a position at a distance X ₁ from the left side of the form 11 and at a position Y ₁ from the top side. The entry frame 17 is set at a position at a distance X ₂ from the left side of the form 11 and at a position Y ₂ from the upper side. The number "1234" is displayed in the entry frame 16.
Is at pitch P ₁ , and the above-mentioned entry frame 17 has the letters “ABCD”.
"E" is entered at pitch P ₂ .

On the other hand, entry frame images 19 and 20 are formed in the image memory 13, and the entry frame image 19 is located at a bit distance a ₁ from the upper side of the form image 14,
Further, it is set at a position of a bit distance b ₁ from the left side, and the entry frame image 20 is set at a position of a bit distance a ₂ from the upper side of the form image 14 and a position of a bit distance b ₂ from the left side. The numeral image "1234" is recorded at the pitch Q ₁ in the entry frame image 19, and the alphabet image "ABCDE" is recorded at the pitch Q ₂ in the entry frame image 20.

By the way, the OCR is usually adapted to read a plurality of types of forms 11, but the positions of the entry frames 16 and 17 are different for each form 11, or the entry frames 16 and 17 are different. The attributes of the characters written in 17 are different. Therefore, in the storage device 12, the O
Format information about all the forms 11 to be read by CR is stored, and each form 11 has a sheet ID.
The format information corresponding to the form 11 can be read out by being identified by 15.

Therefore, the sheet ID 15 is usually made up of a 2 to 5 digit number unique to the form 11, is given to each form 11 having the same format information, and is printed at a common position of each form 11. Therefore, a different sheet ID 15 is assigned to each form 11 having different format information. Then, in the storage device 12, the format information of each form 11 is stored with the sheet ID 15 added.

In this embodiment, the sheet I is added to the form 11.
"123" is given as D15, the position of the entry frame 16 is X ₁ , Y ₁ , the attribute of the characters entered in the entry frame 16 is "number", the pitch of each character is P ₁ , and the first item Is set as. Also, the position of the entry frame 17 is X ₂ , Y ₂ .
The character of the attribute is entered in the entry box 17 is "letter", the pitch of each character in P _2, is set as the second item.

Therefore, the character image can be cut out from the image image stored in the image memory 13 by specifying the position and the attribute of the character based on the format information. And the OCR
First, the sheet ID image 18 is stored in the image memory 1
3, the format information of the form 11 based on the sheet ID image 18 is read from the storage device 1.
2 is read out, and the character image in each entry frame image 19, 20 is cut out based on the format information.

By the way, when the format information is registered in the storage device 12, each entry frame 16 and 1 of each form 11 is registered.
The position of 7 is measured manually using a ruler, and the form 11
The value of the distance from the left side of the document, the value of the distance from the upper side of the form 11, and the attributes of the characters to be written in the writing frames 16 and 17 are manually input using a personal computer or the like. However, not only the number of entry frames 16 and 17 per one form 11 is large, but also the number of types of forms 11 is large, so that format information is often erroneously registered due to a measurement error during registration, an input error, or the like. . If the format information is erroneously registered, the character will be read based on the erroneously registered format information, and thus the character cannot be accurately recognized.

Therefore, the characters actually recognized are inspected,
It is determined whether the registered format information is incorrect.

[0013]

However, in the above-mentioned conventional OCR, when it is judged whether or not the registered format information has an error, all the forms 11
Since it is necessary to visually compare the characters entered in each of the input boxes 16 and 17 with the characters actually read and recognized and output by, for example, a printer, the work is not only troublesome, but the determination A mistake will occur.

Further, when the format information of the similar form 11 is already registered, the registered format information of the form 11 is diverted and only the different part is changed, and the work for registering the format information is performed. However, it is difficult to find a similar form 11. The present invention solves the problems of the conventional format information processing apparatus, makes it possible to easily determine whether or not there is an error in the registered format information, and a format in which it is easy to find a similar form. An object is to provide an information processing device.

[0015]

To this end, the format information processing apparatus of the present invention comprises image reading means for reading a form, a storage device storing format information, a control device, and a display device. Further, the control device includes an image memory for storing a form image corresponding to the form read by the image reading means, an image display means for displaying the form image stored in the image memory on the display device, and The form image stored in the image memory is provided with dummy data writing means for writing dummy data corresponding to the format information read from the storage device in an overlapping manner.

In another format information processing apparatus of the present invention, a storage device storing format information,
It has a control device and an input device. Further, the control device includes an image memory that stores the provisional format information input by the input device, a format information comparison unit that compares the format information read from the storage device with the provisional format information, Form selection means for selecting a similar form based on the comparison result by the format information comparison means.

[0017]

According to the present invention, as described above, the format information processing apparatus has the image reading means for reading the form, the storage device storing the format information, the control device, and the display device. Further, the control device includes an image memory for storing a form image corresponding to the form read by the image reading means, an image display means for displaying the form image stored in the image memory on the display device, and The form image stored in the image memory is provided with dummy data writing means for writing dummy data corresponding to the format information read from the storage device in an overlapping manner.

In this case, the control device normally reads the image of the form by the image reading means, develops and stores the image image corresponding to the read image in the image memory, scans the image memory, and the image image is read. A character is recognized by extracting the characteristics of the character image cut out from the character. When it is determined whether the registered format information has an error or when a similar form is searched for, the control device reads the form by the image reading means and stores the form image in the image memory. Next, the format information of the form is read from the storage device, and the dummy data corresponding to the format information is written over the form image. Therefore, the display device displays the dummy data written over the form image.

In another format information processing apparatus of the present invention, a storage device storing format information,
It has a control device and an input device. Further, the control device includes an image memory that stores the provisional format information input by the input device, a format information comparison unit that compares the format information read from the storage device with the provisional format information, Form selection means for selecting a similar form based on the comparison result by the format information comparison means.

When searching for a similar form, the operator first operates the input device to store the provisional format information in the image memory. Next, the format information comparison means of the control device reads the format information from the storage device and compares the read format information with the provisional format information. Then, the form selection means of the control device selects a similar form based on the comparison result by the format information comparison means.

[0021]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a control block diagram in the first embodiment of the present invention, FIG. 3 is a schematic diagram of a format information processing apparatus in the first embodiment of the present invention, and FIG. 4 is a display device in the first embodiment of the present invention. FIG. 5 is a diagram showing a first example of the display screen of FIG. 5, and FIG. 5 is a diagram showing a second example of the display screen of the display device according to the first embodiment of the present invention.

In the figure, 11 is a form in which characters are entered, 21 is a personal computer for controlling, registering, and storing format information while controlling the entire OCR, and 22 is format information. A storage device, 23 is an image memory provided as a working memory of the personal computer 21, 24 is an image reader as an image reading means, 25 is a CRT or the like as a display device for displaying the image image stored in the image memory 23. Display device. The image reader 24 has an optical system for scanning the image of the form 11 and an image sensor for forming the scanned image. Further, the image reader 24 can be provided separately or integrally with the OCR.

The image signal obtained by the image sensor is converted into an image image, and the image image is
It is written in the image memory 23. The OCR has a storage device 22, and the storage device 22 stores format information of a plurality of types of forms 11 in advance. Therefore, the personal computer 21 refers to the format information stored in the storage device 22 to read the character.

Therefore, the storage device 22 stores the format information about all the forms 11 to be read by the OCR. The format information includes the positions of the entry frames 16 and 17 on the form 11, the entry frame 1
Attributes of characters written in 6 and 17 (printed characters, handwritten characters,
ANK, type of kanji, etc.), character pitch, etc. In this case, the characters entered in each of the entry boxes 16 and 17 are "first
It is set as "item", "second item", .... Also,
When the attribute of the character entered in the entry boxes 16 and 17 is an alphabetic character, "A" is selected, when the character is kana, "K" is selected, when it is a number, "N" is selected, and when it is a symbol, "S" is selected. When "" is a kanji, "Kan" is added as attribute data.

The format information is stored with the sheet ID 15 assigned to each form 11. Therefore, the form 11 can be identified by the sheet ID 15. The sheet ID 15 is usually composed of 2 to 5 digit numbers unique to the form 11, is given to each form 11 having the same format information, and is printed at a common position of each form 11. Therefore, a different sheet ID 15 is assigned to each form 11 having different format information. Then, in the storage device 12, the format information of each form 11 is stored with the sheet ID 15 added.

In this embodiment, the entry frame 16 is located at a distance X ₁ from the left side of the form 11 and a distance Y ₁ from the upper side thereof, like the conventional form 11 and format information shown in FIG.
The entry frame 17 is set at a position of a distance X ₂ from the left side of the form 11 and a position of a distance Y ₂ from the upper side of the form 11. The numeral "1234" is provided in the entry frame 16 at the pitch P ₁ , and the alphabet "ABCDE" is provided in the entry frame 17.
Are entered at pitch P ₂ . In addition, the sheet I is on the form 11.
“123” is added as D15.

On the other hand, the entry frame images 19 and 20 are formed in the image memory 23, and the entry frame image 1
9 is set at a bit distance a ₁ from the upper side of the form image 14 and at a bit distance b ₁ from the left side,
The entry frame image 20 is at a bit distance a ₂ from the top side of the form image 14 and is a bit distance b from the left side.
Set to position ₂ . And the entry frame image 1
The numeral image “1234” is recorded at 9 with a pitch Q ₁ and the alphabet image “ABCDE” is recorded at a pitch Q ₂ on the entry frame image 20.

The bit distance a₁, A₂, B₁, B₂Over
And pitch Q₁, Q₂Is the distance X ₁, X₂, Y₁, Y
₂And pitch P₁, P₂Calculated based on. The a
If the resolution of the image reader 24 is different, the distance between bits is different.
Therefore, conversion is necessary. For example, 300 [DP
I] when using the image reader 24
The separation is 1/300 [inch] = 25.4 / 300 [mm] =
It becomes 0.0847 [mm]. Therefore, for example, 10 [mm] on the form 11
Becomes 10 / 0.0847 = 118 on the image memory 23.

In this case, the position of the entry frame 16 is X ₁ , Y _1.
, The attribute of the character entered in the entry frame 16 is "number",
The pitch of each character is set to P ₁ as the first item.
Further, the position of the entry frame 17 is set to X ₂ and Y ₂ , the attribute of the character entered in the entry frame 17 is set to “English character”, and the pitch of each character is set to P ₂ . By the way, in this embodiment, it is possible to judge whether or not the registered format information has an error.

For this purpose, first, the control device 31 of the personal computer 21 uses the image reader 24 to read the image of the form 11 in which no characters are entered in the entry frames 16 and 17. As a result, in the obtained image signal, the one corresponding to the white portion on the form 11 is “0”,
Those corresponding to images such as ruled lines (characters) and characters that form the entry frames 16 and 17 are converted into "1" bit by bit, developed and stored in the image memory 23.

The form image 14 stored in the image memory 23 includes a sheet ID image 18 corresponding to the sheet ID 15 of the form 11 and entry frame images 19 and 20. The image display means 3 of the control device 31
2 displays the form image 14 on the display device 25. Therefore, the form image 14 of the image read by the image reader 24 can be visually confirmed.

Next, the dummy data writing means 33 of the control device 31 reads out the registered format information corresponding to the sheet ID 15 from the storage device 22 and expands it in the image memory 23 based on the format information. Dummy data is overlaid and written on the form image 14. For that purpose, first, from the storage device 22, the entry frame 1
Format information 6 is read. Then, the dummy data writing means 33 writes the dummy data in the entry frame images 19 and 20 of the image memory 23. In this case, the attribute data indicating the attribute of the character of the format information is used as the dummy data. That is, when the characters entered in the entry frames 16 and 17 are, for example, alphabetic characters, “A”, kana, “K”, numerical characters, “N”, and symbols. Is written in the entry box images 19, 20 at the character pitch set for each entry box 16, 17.

In this way, all entry frame images 1
When the dummy data is input to 9 and 20, the form image 14 of the image actually read by the image reader 24 and the dummy data are displayed on the display device 25.
It is displayed in an overlapping manner as shown in FIGS. Then, when the registered format information has no error, the dummy data as set in each of the entry frame images 19 and 20 is displayed as shown in FIG. On the other hand, if there is an error in the registered format information, the dummy data may deviate from the entry frame images 19 and 20, or the dummy data having an incorrect attribute may be displayed, as shown in FIG.

Therefore, by visually comparing the form image 14 of the image actually read by the image reader 24 and the dummy data on the display device 25, it is possible to easily determine whether or not the registered format information has an error. You can judge. When new format information is registered for a plurality of forms 11, the form images 14 of each form 11 are sequentially displayed on the display device 25, dummy data corresponding to the entry frame images 19 and 20 are written, and the form image 14 is displayed. And the dummy data.

Next, a procedure for finding a similar form 11 when format information is already registered for the similar form 11 will be described. In this case, since the format information is set for each sheet ID 15, the images of the form 11 are sequentially read for each sheet ID 15, and the form image 14 and the dummy data of the read images are visually checked on the display device 25. Compare.

Then, as shown in FIG. 5, the form 11 having the smallest number of items in which the dummy data deviates from the entry frame images 19 and 20 and the dummy data having an incorrect attribute is displayed is selected. Next, when the format information is already registered for the similar form 11,
A second embodiment of the present invention will be described in which the task of finding a similar form 11 is automated.

FIG. 6 is a schematic diagram of a format information processing apparatus according to the second embodiment of the present invention, and FIG. 7 is a block diagram of the format information processing apparatus according to the second embodiment of the present invention. In the figure, 11 is a form in which letters are entered, 2
1 is a personal computer that controls, registers, and stores the format information while controlling the entire OCR, 22 is a storage device that stores the format information, 2
3 is an image memory provided as a working memory of the personal computer 21, 25 is a display device such as a CRT for displaying the image image stored in the image memory 23, 41 is a control device of the personal computer 21, and 42 is the personal computer 2
1 is an input device. The control device 41 comprises an image memory 23, a form selecting means 44 and a format information comparing means 45.

The storage device 22 stores the format information of a plurality of types of forms 11 in advance.
Therefore, the personal computer 21 refers to the format information stored in the storage device 22 to read the character. Therefore, in the storage device 22, the OC
Format information about all the forms 11 to be read by R is stored. The format information is
Positions of the entry frames 16 and 17 on the form 11, the entry frames 16 and
Attributes of characters written in 17 (print type, handwritten character, AN
K, type of kanji, etc.), character pitch, entry boxes 16, 17
The number of items to be set corresponding to the number of items (hereinafter referred to as "number of items") and the like.

In this case, the characters to be entered in the entry boxes 16 and 17 are set as "first item", "second item", .... Next, the operation of the format information processing apparatus having the above configuration will be described. FIG. 8 is a flow chart showing the operation of the format information processing apparatus in the second embodiment of the present invention. Step S1 The input device 42 (FIG. 7) is operated to provisionally register the format information of the form 11 (FIG. 6) and store it in the image memory 23 as provisional format information. In this case, the number of items in the form 11 and the respective entry frames 16 and 17
The positions, the attributes of the characters to be written in the entry frames 16 and 17, the number of characters, and the pitch are registered, but the number of items is registered with a correct value, and other values are registered with provisional values.

For example, regarding the positions of the entry frames 16 and 17, the temporary positions x ₁ , y ₁ , x ₂ and y ₂ are registered, respectively, while allowing an error of about ± 5 [mm]. When the attribute cannot be determined, the symbol "Φ" is registered. In this way, by temporarily registering the format information, the work can be performed more quickly than the normal registration. In step S2, the format information comparison unit 45 of the control device 41 performs the format information comparison process, sequentially reads the registered format information from the storage device 22,
The format information is compared with the provisional format information. Step S3 The form selecting means 44 of the control device 41 selects a similar form 11.

Next, the format of step S2 in FIG.
A subroutine of information comparison processing will be described. Figure 9
Format information comparison processing in the second embodiment of the present invention
It is a flow chart which shows a subroutine of processing. Step S2-1 The first from the storage device 22 (FIG. 6)
Format information is read for the form 11. Step S2-2 Item of read format information
Number n_AAnd the number of provisional format information items n_BCompare with
I do. Number of read format information items n _AAnd tentatively
Number of fixed format information items n_BIf is equal to
Step S2-3, and if they are not equal, step S2-
Return to 1. In step S2-3, the position is checked. Ie read
Positions of the entry boxes 16 and 17 for the format information
And the positions of the entry frames 16 and 17 for the provisional format information.
To match.

That is, with respect to the entry frame 16, it is determined whether or not the positional condition of | X ₁ −x ₁ | ≦ ΔX | Y ₁ −y ₁ | ≦ ΔY is satisfied. Further, with respect to the entry frame 17, it is determined whether or not the positional condition of | X ₂ −x ₂ | ≦ ΔX | Y ₂ −y ₂ | ≦ ΔY is satisfied. In this case, the deviations ΔX and ΔY are set to 5 [mm], for example.

Next, a character attribute check is performed. That is, the character attribute of the read format information is collated with the character attribute of the provisional format information. This collation is performed for the characters entered in each entry frame 16 and 17. When the symbol “Φ” is registered in the provisional format information, it is considered that the character attribute of the read format information and the character attribute of the provisional format information match unconditionally.

When the position condition is not satisfied, or when the character attribute of the read format information and the character attribute of the provisional format information do not match, the count value C of the counter is incremented by one. Increment. Step S2-4 The count value C of the counter is the set value Δ
Judge whether it is C or less. If the count value C is less than or equal to the set value ΔC, the process proceeds to step S2-5, and if the count value C is greater than the set value ΔC, step S2-.
Return to 1. Note that ΔC is 3, for example. Step S2-5 The sheet ID 15 of the form 11 is sequentially displayed on the display device 25.

In this way, when the read format information and the provisional format information are compared for all the forms 11, the format information comparison processing subroutine is ended. At this time, the sheet ID 15 having the count value C equal to or less than the set value ΔC is displayed on the display device 25. Here, the form 11 corresponding to the sheet ID 15 displayed on the display device 25 is, for example, 9 if there are 100 items.
The format information of the seven items can be used almost as it is. Then, it is only necessary to newly register the format information for only three items.

In this way, since the similar forms 11 can be easily narrowed down, the work for registering the format information can be simplified. In this embodiment, the deviations ΔX, ΔY and the set value ΔC are fixed values, but they can be changed by operating the input device 42 of the personal computer 21. Also, in the format information comparison process, the sheet ID is displayed on the display device 25.
Although 15 is displayed, the item when the count value C is incremented can also be displayed. Further, the form image 14 corresponding to the sheet ID 15 can be displayed.

The present invention is not limited to the above-mentioned embodiments, but can be variously modified within the scope of the present invention, and these are not excluded from the scope of the present invention.

[0048]

As described in detail above, according to the present invention, in the format information processing apparatus, the image reading means for reading the form, the storage device storing the format information, the control device, and the display device. Have and. Also,
The control device includes an image memory that stores a form image corresponding to the form read by the image reading unit, an image display unit that displays the form image stored in the image memory on the display device, and the image memory. And dummy data writing means for writing dummy data corresponding to the format information read from the storage device in an overlapping manner on the form image stored in.

Then, when it is determined whether the registered format information has an error or when a similar form is searched for, the control device reads the form by the image reading means and stores the form image in the image memory. . Next, the format information of the form is read from the storage device, and the dummy data corresponding to the format information is written over the form image. Then, the display device displays the dummy data written over the form image.

Therefore, by visual comparison on the display device, it is possible to easily determine whether or not the registered format information has an error. Also,
By displaying the form images of a plurality of forms on the display device, it is possible to easily find similar forms.
In another format information processing device of the present invention, a storage device storing format information, a control device,
And an input device.

Further, the control device compares the image memory for storing the provisional format information input by the input device with the format information comparing means for comparing the format information read from the storage device with the provisional format information. And a form selecting unit that selects a similar form based on the comparison result by the format information comparing unit.

When searching for a similar form, the operator first operates the input device to store the provisional format information in the image memory. Next, the format information comparison means of the control device reads the format information from the storage device and compares the read format information with the provisional format information. Then, the form selection means of the control device selects a similar form based on the comparison result by the format information comparison means.

Therefore, since it is possible to easily narrow down similar forms, the work for registering the format information can be simplified.

[Brief description of drawings]

FIG. 1 is a control block diagram according to a first embodiment of the present invention.

FIG. 2 is a conceptual diagram of a conventional OCR.

FIG. 3 is a schematic diagram of a format information processing apparatus according to the first embodiment of the present invention.

FIG. 4 is a diagram showing a first example of a display screen of the display device according to the first embodiment of the present invention.

FIG. 5 is a diagram showing a second example of a display screen of the display device according to the first embodiment of the present invention.

FIG. 6 is a schematic diagram of a format information processing apparatus according to a second embodiment of the present invention.

FIG. 7 is a block diagram of a format information processing device according to a second embodiment of the present invention.

FIG. 8 is a flowchart showing an operation of the format information processing apparatus according to the second embodiment of the present invention.

FIG. 9 is a flowchart showing a subroutine of format information comparison processing in the second embodiment of the present invention.

[Explanation of symbols]

 11 form 14 form image 22 storage device 23 image memory 24 image reader 25 display device 31, 41 control device 32 image display means 33 dummy data writing means 42 input device 44 form selection means 45 format information comparison means

Claims (2)

[Claims] 1. (a) Image reading means for reading a form,
(B) a storage device in which format information is stored,
(C) A control device and (d) a display device are provided, and (e) the control device stores an image memory for storing a form image corresponding to the form read by the image reading means, and an image memory in the image memory. Image display means for displaying the stored form image on the display device;
A format information processing apparatus, comprising: dummy data writing means for writing dummy data corresponding to the format information read from the storage device in an overlapping manner on the form image stored in the image memory. 2. A storage device storing (a) format information, (b) a control device, and (c) an input device, and (d) the control device is input by the input device. An image memory for storing the provisional format information, a format information comparison unit for comparing the format information read from the storage device with the provisional format information, and a similar form based on the comparison result by the format information comparison unit. A format information processing apparatus comprising: a form selecting unit for selecting.