JPS58114171A - Optical character reader - Google Patents

Abstract

PURPOSE:To correct reject characters efficiently, by displaying characters before and after a reject character to be corrected on a display means. CONSTITUTION:When read character patterns corresponding to one-page components of a form are accumulated in a pattern storage device 3, they are transferred successively to a feature extracting circuit 11. A character recognizing circuit 12 accesses a feature storage device 13 and compares features of each character in a feature dictionary with extracted features to recognize characters of read character patterns and outputs a character code and a reject code to a system bus 10. After the character recognition for one-page components of the form, a display controlling circuit 7 develops characters to dot patterns and displays them on a display 17. In this case, if the reject code exists, a CPU5 sends a compression display command to the circuit 7. The circuit 7 compresses the character arrangement on all character lines of the form and displays it and clarify the part of the reject character. The operator designates a line to be corrected through a keyboard 14 to correct the reject character.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical character recognition device that optically reads character patterns on a form and recognizes characters, and in particular, the present invention relates to an optical character recognition device that recognizes characters by optically reading character patterns on a form, and particularly relates to character recognition results for read character patterns (reject characters) that fail in character recognition. Regarding improvements related to correction processing.

Optical character reading devices used in combination with word processors, etc. read documents and other forms page by page.
Normally, characters are recognized, and if characters cannot be recognized or are modified, one page of characters is corrected at once. If it is possible to understand the position and number of rejected character parts in the entire document, and the flow of sentences before and after the rejected character parts in the entire document, the efficiency of the correction work can be improved. However, with the configuration of a conventional optical character reading device, it is difficult to satisfy all of the above requirements without using a large-screen display device.

SUMMARY OF THE INVENTION An object of the present invention is to provide an optical character reading device that can satisfy the above requirements and can efficiently correct reject characters without using a large-screen display device.

However, in the optical character reading device according to the present invention,
The outline of the character arrangement of the entire form is reduced and displayed on a display device. At the time of this display, the reject character part is displayed separately from other character parts so that the operator can easily identify it. Further, when the operator specifies a specific reject portion through an input device such as a keyboard device, the character recognition result of a character line or one sentence including the rejected character portion is displayed on the display device. At this time, it is more preferable to display the read character pattern of the rejected character portion at the same time, as in an embodiment described later, in order to streamline the correction work. Other features will be made clear by the following description.

FIG. 1 shows a functional block diagram of an example of an optical character reading device according to the present invention.

In the figure, l is a scanner that optically scans a form to read an image on the form, and outputs an analog image signal. The preprocessing circuit 2 digitizes the analog image signal output from the scanner 1 and cuts out individual read character patterns. The cut out read character patterns are sequentially written and stored in the pattern storage device 3.

The central processing unit 5 not only centrally controls the operations of each part of the optical character reading device using a built-in program, but also controls the flow of data through the system bus 10 and performs logical operations and the like.

The feature extraction circuit 11 is a circuit that extracts features of the read character pattern and provides them to the character recognition circuit 12.

The character recognition circuit 12 performs character recognition by comparing the features extracted from the read character pattern with the features of each character registered in the feature dictionary in the feature storage device 13, and determines the code of the recognized character or the character that cannot be recognized. If not, output a reject code. A code storage device stores codes resulting from character recognition.

The input/output control circuit 6 is a part that controls input of information from the keyboard device 14. The display control circuit 7 is a circuit that controls the display operation of the display device 17. The display control circuit 7 is a circuit that controls the display operation of the display device 17, and develops a dot pattern by cooperating with a character generator (CG) 15 using code data transferred via the system bus 10. It functions to write to the screen storage device 16, etc. The output control circuit 9 controls data output operations by an output device such as a printer 18. The mechanism control circuit 8 is a circuit that controls various mechanisms (not shown) for conveying documents and the like.

Next, the operation of this embodiment will be explained with reference to the schematic flowchart shown in FIG.

When a specific key on the keyboard device 14 is pressed, a signal indicating the pressing of the key is sent from the input control circuit 6 to the central processing unit 5 via the system bus 10.

As a result, the central processing unit 5 starts executing the specific processing program that has failed in the internal storage device.

First, the central processing unit 5 instructs the mechanism control circuit 8 to carry in one form on a form hopper (not shown).

When a sensor (not shown) detects that the form carried in from the form hopper has been conveyed to the reading position by the scanner l, and the detection signal is sent from the mechanism control circuit 8,
The central processing unit 5 includes a scanner 1, a moe processing circuit 2, a pattern memory, and DMA (direct memory access) in a storage device 8.
Activate the mechanism. As a result, the character patterns read from the form are sequentially stored in the pattern storage device 8.

(Processing step 100 in FIG. 2) When the read character patterns for one page of a form are stored in the pattern storage device 5, the central processing unit 5 uses the DMA of the scanner 1, the preprocessing circuit 2, and the pattern storage device 3. After stopping the mechanism, the feature extraction circuit 11 and character recognition circuit 12 are activated. The central processing unit 5 then accesses the pattern storage device 5 via the system bus IO, reads out the read character patterns one after another, and transfers them to the feature extraction circuit 11. Feature extraction circuit 1
1 extracts the features of the given read character pattern and sends them to the character recognition circuit 12. The character recognition circuit 12 accesses the feature storage device 18 at both speeds, compares the features of each character in the feature dictionary with the extracted features, and performs character recognition of the read character pattern. If the character can be identified, the character code is output to the system bus 10, and if the character cannot be identified, the ejection code is output.

The character code or redirect code output to the system bus 10 is stored in the code storage device 4 under the control of the central processing unit 5. At this time, the central processing unit 5 increments an internal reject counter every time a reject code is issued. (Processing step 1 in Figure 2
01) When the character recognition for the read character pattern for one page of the form is completed, the central processing unit 5 stops the feature extraction circuit 11 and the character recognition circuit 12, and checks the value of the internal paste counter, Determine whether there are reject characters in the form. (Processing step 102 in FIG. 2) If the value of the reject counter is zero, the central processing unit 5 activates the display control circuit 7 and sends a termination message to the system bus 10. The display control circuit 7 uses the character generator 15 to develop the contents of the end message into a character dot pattern, writes it into the screen storage device 16, and displays it on the display device 17. (Processing step 108 in FIG.
) In this case, the processing for one document is now complete, and the optical character reading device is in a standby state until instructed by the keyboard device 14. Note that the read form is transported to a stacker (not shown) by a transport mechanism (not shown) under the control of the mechanism control circuit 8.

On the other hand, if the value of the reject counter is 1 or more (if there is a reject character), the central processing unit 5 sends a compressed display command to the display control circuit 7 via the system bus 10, and then outputs the compressed display command from the code storage device 4. The character recognition results (consisting of character codes or redirect codes) for one page of the form are sequentially read out and placed on the system bus. When the display control circuit 7 receives the compressed display command, it is subsequently transferred from the system bus 10 and displays the <'ya character recognition result.
It is displayed on the display device 17 in the format shown in FIG.

That is, the character arrangement on all character lines of the form is compressed and displayed as a solid line 50 or a broken line 51.

Each solid line 50 means that there are no reject characters in that line of characters. Each broken line 51 means that a reject character exists in that character line, and the rejected character portion is clearly indicated by a frame line 52. Also, line numbers 58 for each character line are displayed at the same time.

In order to perform such compressed display, the display control circuit 7
This is Cj, who converts the character recognition results for one page of the form transferred from the system bus 10 into a required dot pattern, edits them, and writes them into the screen storage device 16. (Processing step 103 in FIG. 2) When the compressed display described above is completed and a completion report is issued from the display control circuit 7, the central processing unit 5 waits for input of correction line designation information from the keyboard device 14. When the operator inputs the number of the character line to be corrected through the keyboard device 14, the central processing unit 5 attaches the correction line number and sends a correction line display command to the display control circuit 7. The character recognition results are sequentially read from the code storage device 4, placed on the system bus IO, and temporarily stored in a specific work area of the internal storage device. If there is a reject code in the character recognition result of the corrected line, the corresponding read character pattern is read from the pattern storage device 8 and placed on the system bus 10.

When the display control circuit 7 receives a correction line display command from the central processing unit 5, it develops the correction line number added thereto into a character dot pattern and writes it into a predetermined position in the screen storage device 16. Then, among the character recognition results transferred from the system bus 10, each character code is developed into a corresponding character dot pattern and sequentially stored in a predetermined position in the screen storage device 16. For Reject Code, leave blank. Then, when the read character pattern corresponding to the reject code is transferred, it is written in the corresponding position in the image storage device 16. On this occasion,
Add a dot pattern with a frame surrounding the rejected character part. In this way, as shown in FIG. displayed in an easily readable size. (Steps 104 and 105 in FIG. 2) The operator operates the keyboard device 14 to correct or align the cursor on the screen of the display device 16 to the position of the digit character, and inputs the code of the read character. The central processing unit 5 takes in the input character code and cursor position information through the input control circuit 6, and modifies the input character code to correspond to the cursor position in the work area of the internal storage device. , transfers the input character code to the display control circuit 7 together with character position information. At this time, the central processing unit 5 decrements the internal transfer counter by one. In the display control circuit 7, the central processing unit 5
According to the character position information attached to the character code transferred from Re-edit the contents of the screen storage device so that it is removed. (Step 1o6 in FIG. 2) In this way, all the paste characters in the correction line are corrected, and when the operator specifies another correction line number from the keyboard device 14, the central processing unit 5 Perform similar processing for rows. During this correction process, the central processing unit 5
checks the value of the internal transfer counter (step 107 in FIG. 2). When all the paste characters have been corrected and the value of the reject counter becomes zero,
As described above, the central processing unit 5 causes the display device 17 to display a termination message, completes the processing of the form, and enters a standby state.

Although one embodiment has been described in detail, the present invention is not limited thereto, and various modifications may be made without departing from the gist of the present invention.

For example, instead of displaying the rejected character part with a frame line,
It can also be clearly indicated by enclosing it in parentheses, displaying it in a specific color, blinking, etc. Also, characters with reject characters may be displayed in a different color from other character lines, or the same display may be performed, but the line number may be displayed in a different color or blinking so that it can be distinguished from other lines. In short,
It would be good if the reject character part could be made clear. Furthermore, it is easy to immediately erase the frame line 52 of the character line of the rejected character from the screen, which is even more immature. Furthermore, in the correction display, one sentence containing the eject character may be displayed instead of the correction line, or if the screen size allows, two sentences may be displayed.
It may be displayed over more than one line.

Furthermore, although the embodiment described above was configured using program control, it is of course possible to configure similar functions using only hardware.

As described in detail above, the present invention compresses the character arrangement of one page of a form and displays it on the display screen, clearly indicates the number of rejected character parts and the current position of the form, and also displays the number of rejected characters and the position of the form to be corrected. To display the characters before and after a character separately for easy reading. Therefore, the operator can efficiently correct the embossed characters. Furthermore, since the character arrangement of the entire form is displayed in a compressed manner, the size of the display screen does not need to be so large, and the device can be realized at low cost.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing an embodiment of the optical character reading device according to the present invention, Fig. 2 is a schematic flowchart showing the operation flow of the above embodiment, and Fig. 8 is an explanation of the display screen in the above embodiment. It is a diagram. DESCRIPTION OF SYMBOLS 1... Scanner, 2... Preprocessing circuit, 3... Pattern storage device, 4... Code storage device, 5... Central processing unit, 6... Input control circuit, 7... Display control circuit, 8... Mechanism control circuit, 9... Output control circuit, 1
0... System bus, 11... Feature extraction circuit, 12
...Character recognition circuit, 13...Feature storage device, 14.
...Keyboard device, 15...Character generator, 16...Screen storage device, 17...Display device. Agent Patent Attorney Makoto Suzuki Figure 1 Figure 2

Claims (1)

[Claims] L In an optical character reading device that optically reads a character pattern on a form and recognizes the characters, an outline of the character arrangement over the entire read form is detected by character portions that cannot be recognized (rejects). A means for specifying and displaying the character part (referred to as a character part) in a reduced size on a display means, and a limited range before and after the character part that is displayed on the display means or specified by the operator through the input means among the character part. means for displaying the result of character recognition for the read character pattern on the display means, and correcting the result of character recognition for the valley character pattern of the reject character portion in accordance with information input by the operator through the input means. An optical character reading device comprising: means. 2. Display the character recognition results for the character patterns before and after the paste character part. 2. The optical character reading device according to claim 1, wherein the means also displays a read character pattern of the rejected character portion.