JPH08335245A - Document reader - Google Patents

Abstract

PURPOSE: To vary the size of an image screen through easy operation when a recognition result is confirmed and corrected. CONSTITUTION: A control part 12 displays recognition data in a recognition memory 6 on the text screen of a display part 10 by a recognition data display means 12a and also displays image data in an image memory 2 on the image screen by an image data display means 12b. When an operator specifies characters in a desired range of the image screen on the text screen side, a display range calculating means of the control part 12 calculates the display range of image data corresponding to the specified characters and the image data display means 12b varies the size of the image screen according to the calculated display range and displays the image data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document reading device for reading characters and images (graphics other than characters, pictures, photographs, ruled lines, etc.) recorded on a document. In particular, it relates to a method of displaying a character image when confirming / correcting the recognition result.

[0002]

2. Description of the Related Art In a document reading apparatus, since recognition includes an error and is not perfect, it is necessary to confirm the recognition result by an operator and correct it based on the confirmation result. FIG. 7 is an explanatory diagram showing an example of a screen display at the time of confirmation / correction of the recognition result. In general, in this confirmation / correction processing, the recognition result is displayed in characters (font) on the text screen 200 as shown in FIG. A character image corresponding to the character designated by the operator is displayed on the image screen 210 (if the designated character is changed on the text screen, the character image on the image screen is also changed in conjunction), and these are compared. The characters on the text screen 200 were corrected using the text editing function and the candidate replacement function.

FIG. 8 is an explanatory view showing an example of a conventional text screen and an image screen. FIG. 8A shows a case where the size and position of the image screen are fixed, and FIG. 8B shows a case of the image screen. When the size is variable and the position is linked to the designated character, FIG. 8C shows the case where the size of the image screen is variable and the position is also variable. Reference numerals 312, 322 and 332 are image screens. The image screen 334 is
Image screen 332 to make it easier to see when writing vertically
Is placed horizontally on the text screen 200 to make it vertically long.

Image screen 322 of FIGS. 8B and 8C
The sizes of 332 and 332 (334) can be changed by an operator's operation. That is, when the display range of the character image is too large or too small, or when the display range is not appropriate such as not suitable for vertical writing, the size of the image screen is changed to change the display range. FIG. 9 is an explanatory diagram showing an example of a conventional operation for changing the size of the image screen. For example, by moving the mouse cursor 450 to the lower right of the image screen 410 and enlarging it by dragging the mouse, the image screen 410 can be changed to the image screen 420. The image screen 410 was reduced and changed to the image screen 430.

[0005]

In the conventional document reading apparatus, when the size of the image screen is changed by the confirmation / correction process, the operator moves the mouse from the text screen to be operated to the image screen and the size is increased. Change the
It is necessary to return the mouse to the original text screen to select the character to be corrected. Therefore, the operation amount of the operator increases.

Further, in order to determine an appropriate size of the image screen, if the image screen cannot be displayed immediately after the size is changed, trial and error such as changing the size or displaying the image and confirming the size is necessary. Therefore, even if the image screen can be displayed immediately after the size is changed, it is necessary to adjust the size while looking at the image screen. Therefore, the operation is difficult,
The operation amount of the operator increases and the work load increases.

[0007]

In order to solve the above-mentioned problems, the present invention displays recognition data obtained by character recognition of image data on a screen as characters and displays image data corresponding to this character display. Display means for displaying an image according to the layout data created from the image data, and the image data in conjunction with the display range designated by the operating means by the operator's operation on the screen on the side displaying the characters of the recognition data. And a display range calculating unit for calculating the display range of the recognition range, and the display unit displays the image data of the display range calculated by the display range calculating unit to display the recognition data in a state where the recognition data and the image data can be compared. It is a document reading device that can be confirmed and modified.

[0008]

According to the present invention having the above-described configuration, when the recognition data is confirmed and corrected by comparing the character display of the recognition data and the image display of the image data, the screen on the side displaying the character of the recognition data is displayed. When the display range of the image data is designated, the size of the image display of the image data is changed in association with the designated range.

[0009]

1 is a block diagram showing the configuration of a document reading apparatus according to a first embodiment of the present invention. The image input unit 1 optically scans a document to be read and creates image data. Also, image data can be obtained from another system (not shown) via an image file or a communication line.

The image memory 2 stores the image data output from the image input section 1. The layout analysis unit 3 extracts a region from the image data in the image memory 2, identifies each region as a character and an image, further cuts out a line from the character region, and cuts out a character from the line. The layout data is composed of the processing results of the above-mentioned areas, lines, and characters.

The layout memory 4 stores the layout data output from the layout analysis unit 3. The character recognition unit 5 recognizes the image data of each character from the layout data in the layout memory 4 and the image data in the image memory 2 and converts it into a character code. This character code is used as recognition data. Furthermore, the character recognition unit 5 can also perform knowledge processing on this recognition data.

The recognition memory 6 stores the recognition data output from the character recognition unit 5. The result output unit 7 creates document data using at least one of the image data in the image memory 2, the layout data in the layout memory 4, and the recognition data in the recognition memory 6, and the document data is printed by the printer 8 Or to an output device such as the output memory 9. Then, the document data is passed to another document data processing system (word processor, DTP system, document management system, etc.) via the output memory 9. Alternatively, the document data can be passed to another document data processing system by communication.

A display unit 10 such as a CRT and an operation unit 11 such as a keyboard and a mouse are used to instruct start and end of processing and processing results (image data, layout data, and recognition data).
The interface between the operator and the document reading device is used for displaying and checking / correcting. The control unit 12 controls the overall operation of each unit and each memory described above.

Here, in the control unit 12, the recognition data display means 12a for displaying the recognition data in the recognition memory 6 on the display unit 10 and the image data in the image memory 2 are displayed on the display unit 10 together with the recognition data. The image data display means 12b to be displayed and the display range calculation means 12c for calculating the display range of the image data in accordance with the range designated by the operator on the operation section 11 on the screen where the recognition data of the display section 10 is displayed. And are provided. In addition, these recognition data display means 12a and image data display means 12
b, the display range calculation means 12c may be incorporated in the control unit 12 as hardware, or as means realized as a software program of the control unit 12,
It may be stored in the memory storing the program of the control unit 12 and read and executed as needed.

FIG. 2 is a process flowchart showing the operation of the document reading apparatus of the first embodiment described above.
The operation of the document reading apparatus according to the embodiment will be described. First, an image is input by the image input unit 1 (S1). For this image input, the image input unit 1 optically scans an input document to be read, converts characters and images recorded on the document into image signals by photoelectric conversion, and further converts the image data into digital binary data. Image data. This image data is stored in the image memory 2.

When the image input is completed, the layout analysis unit 3
The layout is analyzed by (S2). In the layout analysis, first, the layout analysis unit 3 extracts an area from the image data in the image memory 2 by using a method of using a peripheral distribution histogram of black pixels or the like, and characterizes the area according to the geometric feature of each area. And identify to the image. Further, in the character area, a line is cut out using a method using a peripheral distribution histogram of black pixels, and a character is cut out from each line. The layout data consists of the above areas, lines, and character data. This layout data is stored in the layout memory 4. The control unit 12 superimposes and displays the layout data (frame display) and the image data on the display unit 10, and the operator confirms / corrects the layout data by using the operation unit 11. Then, the layout data after confirmation / correction is stored in the layout memory 4.

When the layout analysis is completed, the character recognition unit 5
The character is recognized by (S3). According to the layout data in the layout memory 4, the character recognition unit 5 uses the recognition image in which the character image of the image data in the image memory 2 is stored in the character recognition unit 5 to store the standard character recognition features. It is converted to a character code including candidate characters by matching. Further, this character code is knowledge-processed and corrected using the word dictionary and grammar rules in the character recognition unit 5. This character code is used as recognition data. And
This recognition data is stored in the recognition memory 5.

When the character recognition is completed, the operator confirms / corrects the recognition data (S4). In this confirmation / correction processing, first, the recognition data display means 12a of the control section 12 displays the recognition data in the recognition memory 6 on the text screen of the display section 10, and the image data display means 12b displays the image in the image memory 2. Display the data on the image screen. The text screen displaying the recognition data is compared with the image screen displaying the character image, and the operator confirms / corrects the recognition data using the operation unit 11.
This correction operation is performed by replacement with candidate characters / words or by text editing with a keyboard as in a general word processor or the like. Then, the recognition data after confirmation / correction is stored in the recognition memory 6. It should be noted that changing the size of the image screen at the time of confirming / correcting the recognition data will be described later.

When the confirmation / correction of the recognition data is completed, the result output section 7 outputs the read result (S5). In this result output processing, the result output unit 7 creates document data from at least one of the image data in the image memory 2, the layout data in the layout memory 4 and the recognition data in the recognition memory 6, and the document data is generated. Printer 8
Is printed or stored in the output memory 9.

Next, the operation of changing the size of the image screen in the first embodiment in the "confirmation / correction of recognition data" processing in S4 described above will be described. FIG. 3 is a processing flowchart for changing the size of the image screen, FIG. 4 is an explanatory diagram of a method of calculating the display range, and FIG. 5 is an explanatory diagram of an example of an operation for changing the size of the image screen. Here, the text screen 2 of FIG.
In the text display of 00, the display character position in the character area is aligned with the character image by using a space, and thereby the layout (line, character) in the character area and the text display can be made close to each other.

In the confirmation / correction processing of the recognition data in the above-described flowchart S4 of FIG. 2, the operator confirms / corrects the recognition data, so that the mouse of the operation unit 11 is displayed on the text screen 200 displayed on the display unit 10. For example, the character "minute" is designated with the mouse cursor 190 using. At this time, the recognition data display means 1 of the control unit 12
In 2a, "minute" is highlighted.

In conjunction with this, the control section 12 displays an image centered on "minutes" on the image screen 100 displayed on the display section 10 by the image data display means 12b. At this time, a character frame 192 is displayed in "minute".
The display on the display unit 10 before the above-described image screen size change is shown in FIG. Next, an example of a size changing operation is shown in FIG.
Describing with reference to the flow chart of No. 1, first, the operator designates the characters in the desired range on the text screen 200 (S6). That is, the operator operates the image screen 10
To change the size of 0, move the mouse to the text screen 20
Double-click at the 0 minute position. By this double-clicking, the control unit 12 is notified that the process of checking / correcting the recognition data is shifted to the process of changing the size of the image screen.

Next, the operator drags the desired range, for example, the characters "ni" to "to" with the mouse. When the mouse drag is released, the control unit 12 is notified of the end of the range designation, and the recognition data display means 12a of the control unit 12 makes the rectangle from "ni" to "to" a shaded display 110. When the range designation is completed, the display range calculation means 12 of the control unit 12
The display range of the image data corresponding to the characters in the range specified by c is calculated (S7). An example of calculating the display range will be described with reference to FIG.

The display range calculation means 12c of the control unit 12 causes the character data 300 of the layout data in the layout memory 4 to the character data 310 of the specified range.
(The shaded portion in FIG. 4) is extracted. The character data of the layout data in the layout memory 4 represents the position of each character by the number of dots with the upper left of the document image as the origin, and corresponds to the character in the range specified by the display range calculation means 12c. When you calculate the rectangle that surrounds the image,
(220,200), (450,390) (width W
= 230, height H = 190).

When the calculation of the display range is completed, the image data display means 12b of the control unit 12 changes the size of the image screen 100 according to the display range (S8). That is,
The image data display means 12b controls the display unit 10 to change the size of the image screen so that the image data in the display range calculated in S7 can be displayed. For example, when displaying one dot of image data as one dot of image data, the image screen 120 is changed to have a width W = 230 and a height H = 190.

When the size of the image screen is changed,
The control unit 12 returns the display of the display unit 10 to the same state as before the size change of the image screen (S9). That is, the display unit 10 is displayed by the recognition data display means 12a of the control unit 12.
Control is performed to return the shaded display 110 of S7 on the text screen 200 to the normal display, the character "minute" is highlighted, the mouse cursor 190 is brought into alignment, and the text screen is returned to the state before the size change of the image screen. The image screen 120 whose size has been changed displays an image centered on the character "minute" by the image data display means 12b. The above results are shown in FIG.

An example of reducing the image screen by similar processing is shown in FIG. Also in the case of reduction, by double-clicking the mouse at the position of "minute" on the text screen 200 and designating the character of the range to be viewed, the operator then mouses over only the range to be viewed, for example, the character "minute". Drag. As a result, the end of the range designation is notified to the control unit 12, and the recognition data display means 12a of the control unit 12 makes the rectangle of "minutes" a shaded display 130.

When the range designation is completed, the display range calculation means 12c of the control unit 12 calculates the display range of the image data corresponding to the characters in the designated range. When the display range calculation means 12c calculates a rectangle surrounding the image corresponding to the characters in the specified range with reference to FIG. 4, (280, 270),
(330, 320) (width W = 50, height H = 5)
0).

When the calculation of the display range is completed, the size of the image screen 100 is changed according to the display range by the image data display means 12b of the control unit 12. That is, the display unit 10 is controlled by the image data display means 12b so that the width W
= 50 and height H = 50, change to the image screen 140. After changing the size of the image screen, the control unit 1
By 2, the display on the display unit 10 is returned to the same state as before the size change of the image screen. That is, the display unit 10 of the control unit 12 is controlled by the recognition data display unit 12a to return the shaded display 130 on the text screen 200 to the normal display, the character "minute" is highlighted and the mouse cursor 190 is placed, and the text screen is displayed. To the state before changing the size of the image screen. FIG. 5C shows the result of the above processing.

After the image screen size changing process is completed, if the mouse cursor 190 is moved to change the character to be checked / corrected, the image data display means 12b does not change the size of the image screen. , Scroll the image so that the target character is displayed in the center. First explained above
According to this embodiment, the following effects can be obtained.

In other words, by confirming / correcting the recognition data, the size of the image screen can be changed without directly operating the image screen, mainly by the operator operating the text screen to be operated. Can be reduced. Also, since the size of the image screen can be changed by specifying the characters in the range you want to see on the text screen, there is little trial and error or adjustment operation to display the range you want to see on the image screen, and you can easily select the appropriate image screen. You can decide the size. Therefore, the operation amount can be reduced and the work load can be reduced.

Next, a second embodiment of the present invention will be described. The control block of the document reading apparatus in the second embodiment is the same as that in the first embodiment described with reference to FIG. The flow of processing for changing the size of the image screen is shown in FIG.
The process is the same as the process described in the flowchart. In the second embodiment, the layout analysis unit 3 calculates the average line pitch and the average character pitch for each character area at the time of layout analysis, and stores them in the layout memory 4 as layout data.

Then, the display range is calculated in S7 of the flowchart of FIG. 3 as follows. That is, the display range calculation means 12c of the control unit 12 detects the number of lines (for example, the number of lines: GN = 3) and the number of characters (for example, the number of characters: MN = 4) from the range specified in S6 of FIG. The display range is calculated by multiplying the average line pitch and the character pitch of the character area including the specified number and the specified number of characters. In the case of the layout data of FIG. 4, assuming that the average line pitch GP = 70 and the average character pitch MP = 60, the width and height of the display range are as follows. Width W = MP × MN = 60 × 4 = 240 Height H = GP × GN = 70 × 3 = 210 The image data display means 12b of the controller 12 changes the size of the image screen according to the calculated display range. .

In addition, in the range designation of the second embodiment, it is not possible to designate over a plurality of character areas. Figure 6
6A is a display example of an image screen and a text screen in the second embodiment. FIG. 6A shows a case where the character pitch (size of the character) is changed, and the mouse double-clicks with "C" on the text screen. In that case, in the first embodiment, when the range of “OCR” on the text screen is designated and the change process described in FIG. 3 of the first embodiment is performed, “OCR” appears on the image screen.
“R” is displayed. After that, when the mouse cursor was moved to "special", only "special" was displayed on the image screen. On the other hand, in the second embodiment, when "OCR" on the text screen is designated as the range and the change processing described in FIG. 3 of the first embodiment is performed, "writing" is displayed on the image screen.
OCR DR ”is displayed. After that, when the mouse cursor is moved to "special", "is characteristic" is displayed on the image screen.

FIG. 6B shows the case where the positions of the characters are deviated between the image screen and the text screen. When the mouse is double-clicked with "minute" on the text screen, in the first embodiment, " To specify the range from "to"
First Embodiment When the change process described with reference to FIG. 3 is performed, 3 lines × 4 characters from “ni” to “character” are displayed on the image screen. On the other hand, in the second embodiment, when the change processing described in FIG. 3 of the first embodiment is performed by designating the range from "ni" to "collection" on the text screen, "ni" appears on the image screen. ~ 3 lines x 3 characters up to "special" are displayed.

According to the second embodiment described above, the fact that the line pitch and the character pitch are substantially constant within one character area is utilized to calculate the range by calculating the average line pitch and the average character of the character area. By using the pitch, Fig. 6 (a)
Even if there is a local change in the character pitch, such as when the size of the image screen is changed and the character to be checked / corrected by moving the mouse cursor is changed, the change in the image screen is suppressed. it can.

Even if the character display on the text screen and the character display on the image screen are deviated from each other as shown in FIG. 6B, the average line pitch and the average character pitch of the character area are used to calculate the range. The deviation can be suppressed. As a result, the image screen can be changed to an appropriate size desired by the operator. Next, a third embodiment of the present invention will be described. The control block of the document reading apparatus in the third embodiment is the same as that in the first embodiment described with reference to FIG. The flow of processing for changing the size of the image screen is the same as the processing described in the flowchart of FIG.

In the first embodiment, the size of the image screen is made constant after the completion of the image screen size changing process (during recognition data confirmation / correction process). Every time a designated character on the screen (character designated by the mouse cursor 190 in FIG. 5) moves, the control unit 1
The display range calculation means 12c of 2 calculates the display range surrounding the number of lines and the number of characters designated in S6 of FIG. 3, and changes the size of the image screen. Alternatively, by utilizing the fact that the line pitch and the character pitch are substantially constant in one character area, when the designated character moves to another character area, the number of lines and the number of characters designated in S6 of FIG. Calculate the display range surrounding the characters and change the size of the image screen.

Similarly to the above, in the second embodiment, when the designated character moves to another character area, the display range calculation means 12c of the control unit 12 determines the average line pitch of the character area to which the new designated character belongs. The size of the image screen is changed by calculating the display range by multiplying the average character pitch. As described above, according to the third embodiment, the size of the image screen can be changed in association with the movement of the designated character, so that the image around the designated character is fixed (the designated number of lines and the number of characters). Can be displayed in the range. For this reason, the confirmation / correction of the recognition data becomes easy, and the work load on the operator can be reduced.

Next, a fourth embodiment of the present invention will be described. The control block of the document reading apparatus in the fourth embodiment is the same as that in the first embodiment described with reference to FIG. The flow of processing for changing the size of the image screen is shown in FIG.
The process is the same as the process described in the flowchart. In the third embodiment, the size of the image screen is changed along with the movement of the designated character on the text screen, but in the fourth embodiment, the size of the image screen is fixed to the size changed in S8 of FIG. If the designated range is expanded or reduced, the control unit 1
The second image data display unit 12b enlarges or reduces the image data and displays the range calculated in S7 so as to fit within the image screen.

As described above, according to the fourth embodiment, since the size of the image screen is constant, the screen configuration of the display unit 10 does not change, and the operator's eye strain can be reduced.
Therefore, the work load on the operator can be reduced. In addition,
In the first embodiment, the transition from the confirmation / correction process to the image screen size change process is performed by double-clicking the mouse, but any operation such as pressing a function key or pressing a button on the screen may be performed. It is possible.

Although the size and position of the image screen are variable as a display example of the image screen, the size of the image screen is fixed even if the position is fixed or the position is linked to the designated character. Can be changed. Furthermore, in the third embodiment, the size of the image screen is changed in accordance with the movement of the designated character, but at this time, the image screen becomes too large or too small, and the recognition data is confirmed or corrected. Make sure that the maximum and minimum image screen sizes are set so that it does not become difficult (it becomes too large to protrude from the display screen, too small to check), and the image screen size is changed within this maximum / minimum limit. You can also do it. By combining this with the fourth embodiment, when the maximum / minimum is exceeded, the display image can be fit on the image screen by reducing / enlarging the image data.

[0043]

As described above, the present invention compares the character display of the recognition data with the image display of the image data and confirms / corrects the recognition data. Since the display range of the image data is specified by, and the size of the image display of the image data is changed in conjunction with the specified range, the display screen of the recognition data, which is the operation target of the operator, is mainly displayed. The size of the screen of the image data can be changed without directly operating the screen of the image data by only the operation of, so that the operation amount of the operator can be reduced.

Further, since the size of the image data screen can be changed by designating the characters in the desired range on the recognition data screen, trial and error or adjustment operation can be performed to display the desired range on the image data screen. The size of the screen of the appropriate image data can be easily determined. Therefore, the operation amount can be reduced and the work load can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a document reading device according to a first embodiment of the present invention.

FIG. 2 is a processing flowchart showing the operation of the document reading apparatus according to the first embodiment.

FIG. 3 is a processing flowchart of changing the size of an image screen.

FIG. 4 is an explanatory diagram of a display range calculation method.

FIG. 5 is an explanatory diagram of an example of an operation of changing the size of an image screen.

FIG. 6 is a display example of an image screen and a text screen in the second embodiment.

FIG. 7 is an explanatory diagram illustrating an example of a screen display when confirming / correcting a recognition result.

FIG. 8 is an explanatory diagram showing an example of a conventional text screen and image screen.

FIG. 9 is an explanatory diagram showing an example of a conventional operation of changing the size of an image screen.

[Explanation of symbols]

 1 Image Input Section 2 Image Memory 3 Layout Analysis Section 4 Layout Memory 5 Character Recognition Section 6 Recognition Memory 7 Result Output Section 10 Display Section 11 Operation Section 12 Control Section

Claims (3)

[Claims] 1. An image input means for operating a character or an image recorded on a recording medium to create image data, an image memory for storing the image data, and layout data from the image data in the image memory. A layout analysis unit for creating the layout data; a layout memory for storing the layout data; and a character recognition unit for recognizing a character image of the image data in the image memory and creating recognition data according to the layout data in the layout memory, A recognition memory that stores this recognition data and the recognition data in this recognition memory are displayed as characters on the screen, and the image data in the image memory corresponding to this character display is displayed as an image according to the layout data in the layout memory. Display means and an image on the side displaying the recognition data in characters. And a display range calculation unit that calculates a display range of the image data in association with a display range specified by the operation unit by the operator, and the display unit displays the display range calculated by the display range calculation unit. A document reading device displaying image data. 2. The document reading device according to claim 1, wherein the display range changing unit is interlocked with a change of a designated character to be confirmed / corrected on a screen on which the recognition data is displayed as characters. And a size of image display of the image data is changed. 3. The document reading device according to claim 1, wherein the display range changing unit is interlocked with a change of a designated character to be confirmed / corrected on a screen on the character display side of the recognition data. And a document reading device for changing the image display range by enlarging or reducing the image of the image data.