JP2705972B2 - Document processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an external character registration process required for reading characters or symbols that are not registered in dictionary data of a character reading device, characters or symbols that are easily misread, and the like. In accordance with an instruction from a computer connected to the character reading device.

[Conventional technology]

For a character to be read by the document reading device, it is necessary to extract a feature amount of the character or the like in advance and register the information as a dictionary. Therefore, for a character or the like that is not registered as dictionary data, it is necessary to extract a characteristic amount of the character or the like and register this as new dictionary data. At that time, in the case of the related art, it is only possible to register the variation of the character shape with respect to the registered characters and the like.

[Problems to be solved by the invention]

In other words, in order to newly register a character or the like in the conventional character reading device, it is necessary to rewrite the dictionary data of the character reading device by extracting the characteristics of the character serving as the dictionary data by another dedicated device. There's a problem. Therefore, a) it is difficult for the user to additionally register dictionary data.

b) It is difficult for users to create their own set of recognized characters (legal, technical, medical, etc.).

Inconvenience such as occurs.

[Means for solving the problem]

An image input unit for inputting a document image, a character reading unit for extracting a feature from the input document image, comparing the extracted character with dictionary data stored in advance, and reading a character; A computer unit for giving a predetermined instruction to the reading unit, when a predetermined character is designated from an image input to the image input unit and displayed on the display unit, and dictionary data is created and additionally registered. By giving an instruction to that effect to the character reading unit from the computer unit, the computer unit enlarges and displays a predetermined area including the character to be registered on the display screen, and when a registration frame of the character to be registered is designated. Displays the registration frame by changing the line type according to the size of the registration frame. The character reading unit extracts the characteristics of the character specified by the registration frame, creates and stores the dictionary data, and stores it in the computer unit. Stores creates the designated character corresponding font, enabling output.

[Action]

A computer is connected to the document reading device, and a dictionary pattern of external characters can be created according to an instruction from the computer.
The applicable range is to be expanded.

〔Example〕

FIG. 1 is a block diagram showing a system configuration example of the present invention.

That is, the computer 1 operates as a character reading device (OCR) 2
The computer 1 is connected to a CRT and a keyboard 3, a mouse 4 and the like used when operating the screen. Further, an image scanner 5 for inputting image information (image data) is connected to the character reading device 2.

FIG. 2 is a flowchart for explaining the operation of the computer, and FIG. 3 is a flowchart for explaining the operation of the character reading device.

2 and 3 will be described with reference to FIGS. 4 to 10.

Now, when a document including a character or a symbol to be registered as an external character is placed on the image scanner 5 and a predetermined instruction is given from the computer 1 to the character reader 2 (see FIG. 2), the information from the image scanner 5 becomes image information. Is sent to the character reading device 2 and the computer 1, and the computer 1 displays this image information (image data) on a CRT (not shown) (see above). An example is shown in FIG. Computer 1
According to the above operation, the character reading device 2 performs the operation as shown in FIG.

Next, a code to be used as data for each character or symbol is assigned to the character or symbol to be registered on the screen as shown in FIG. 5 (see above). In FIG. 5, "160
The code "1" is assigned. When the assignment of the codes is completed, the screen shown in FIG. 4 is displayed again. When the vicinity of the character or the symbol to be registered is designated by the mouse cursor, the image in the designated vicinity is enlarged and displayed as shown in FIG. . At this time, a range (enlarged display frame) 6 to be enlarged and displayed as shown in FIG. 7 can be displayed together with the mouse cursor. That is, in this way, it is possible to easily find which part of the enlarged display image the area to be subjected to feature extraction is located, and to display an enlarged image in which characters or symbols to be registered are at appropriate positions. Become like

When an instruction to enclose a character or a symbol to be registered in a quadrilateral registration frame 7 as shown in FIG. 8 is given on the enlarged display screen (see FIG. 8), the computer 1 displays the instruction. It is determined whether or not the range is within the specified value. If the range is within the specified value, the address is converted into an address for image data in the character reading device 2 based on the addresses of the four vertices of the enclosed quadrilateral registration frame 7. Then, the address data and the code specified in FIG. 5 are transferred to the character reading device (OCR) 2 (see above). As a result, the character reading device 2 extracts the characteristic amount necessary for character recognition as shown in the step of FIG. 3 based on these data and processes it into dictionary data. As described above, the dictionary data is stored in the dictionary data storage area (RAM) for additional registration, and the dictionary data is transferred to the computer 1 so that the dictionary data can be stored on a floppy disk or the like. The result of the determination in the step may be, for example, as shown in FIG. 9, depending on whether the set range (registration frame) is within a specified value. That is, FIGS. 7A and 7C respectively show the case where the registration frame is smaller than or equal to the specified size, and is indicated by a dotted line here. FIG. 11B shows a case where the registration frame is within the specified size, and is displayed by a solid line in this case. This makes it easy to determine whether the size of the registration frame is within the specified size.

After confirming that the external registration has been performed normally (see the same), the computer 1 creates a font from an image provided through the character reading device 2 and having a designated registration frame (see the same). That is, assuming that the information indicated by the quadrilateral 7 in FIG. 8 is indicated (shown in an enlarged manner) as shown in FIG. 10 (a), its vertical and horizontal sides are 24 sections each. In other words, the image is divided into 24 × 24 = 576 divisions, and it is determined whether there are many black images or white images as image information for each division. If there are many black images, black dot information is obtained. Instead of dot information, a font consisting of 24 × 24 dots is created as shown in FIG. 10 (b).

In this way, the recognition result for the newly registered characters and the like in the character reading device will be sent to the computer with the code shown in FIG. 5, and the font displayed on the CRT screen will be as shown in FIG. Will be used.

In addition, the character code corresponding to the registered character, the dictionary data for character recognition, and the character font information for displaying or printing the recognition result can be deleted or added as appropriate, and can be saved to a floppy disk. By replacing the information as needed, it becomes possible to utilize the device as a document reading device suitable for the purpose of use.

〔The invention's effect〕

According to the present invention, an external registration operation for inputting characters or symbols not present in the dictionary data of the character reading device via the scanner and registering the data in the dictionary data is performed by the computer connected to the character reading device and its CRT. Since the screen can be more easily displayed on the screen, the user of the character reading apparatus can easily and freely create his / her own dictionary, thereby expanding the range of use. If a character having a low recognition rate is registered, the recognition rate can be increased.

Furthermore, by displaying an enlarged display frame and changing the line type of the registration frame during registration, not only can the character to be registered be displayed at an appropriate position, but also it is easy to check whether the size of the registration frame is within the specified size. Can judge,
Registration work becomes more accurate and operability is improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the system configuration of the present invention, FIG. 2 is a flowchart for explaining the operation of the computer shown in FIG. 1, and FIG. 3 is an explanation of the operation of the character reading apparatus shown in FIG. FIG. 4 is an explanatory diagram for explaining an example of a display screen, FIG. 5 is an explanatory diagram for explaining an example of a character code allocation screen, and FIG. 6 is a predetermined diagram including a designated character. FIG. 7 is an explanatory diagram for explaining an enlarged display frame, FIG. 8 is an explanatory diagram for explaining a reading frame of a designated character, and FIG. 9 is an explanatory diagram for explaining an enlarged display frame. FIG. 10 is an explanatory diagram for explaining an example of displaying whether the size of the character registration frame is appropriate or not by changing the line type.
The figure is an explanatory diagram for explaining image data and font data. Description of symbols 1 ... computer, 2 ... character reading device (OCR), 3
... Keyboard, 4 mouse, 5 image scanner, 6 enlarged display frame, 7 registration frame.

 ──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Yukio Kuto 1 Fujimachi, Hino-shi, Tokyo Inside Fuji-Facom Control Corporation (56) References JP-A-63-70391 (JP, A)

Claims (1)

(57) [Claims] An image input unit for inputting a document image; a character reading unit for extracting a feature from the input document image and comparing the extracted character with dictionary data stored in advance to read characters; And a computer unit for giving a predetermined instruction to the character reading unit, and a dictionary is created by designating a predetermined character from an image input to the image input unit and displayed on the display unit. When additional registration is performed, the computer section gives an instruction to that effect to the character reading section, and the computer section enlarges and displays a predetermined area including the character to be registered on the display screen, and registers the character to be registered. When a frame is specified, the registration frame is displayed with a different line type according to its size,
The character reading unit extracts the characteristics of the character specified by the registration frame and creates and stores the dictionary data, while the computer unit creates and stores a font corresponding to the specified character so that it can be output. A character reading device characterized by the above-mentioned.