JPH0373914B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a character image processing device that can read information on a form including characters and images.

[Technical background of the invention]

The reading control method of a character image processing device that can distinguish and input character information and image information on a form involves specifying a character area where character information exists and an image area where image information exists using coordinate values, etc. The so-called FC control method, which is stored as format control data (hereinafter referred to as FC data),
Put special marks or red marks directly on the form,
There is a mark detection and identification method that detects and controls the mark etc. at the time of reading.

[Problems with background technology]

However, in the former method, as a control method for an input device, there is a problem in that it is complicated to create FC data for various formats and lacks speed. Furthermore, although the latter method is simple, there is a problem in that accuracy and aesthetics are impaired. Furthermore, recently, free-format reading methods have also been used to read forms typed in English, but one of the above two methods must be adopted to read forms that contain text and image information. The situation is such that it cannot be done.

[Purpose of the invention]

This invention was made in view of the above circumstances, and it is not necessary to create FC data, and there is no need to write marks on the form to distinguish between text areas and image areas. It is an object of the present invention to provide a character image processing device that can distinguish and read information and image information.

[Summary of the invention]

This invention determines whether a photoelectric conversion signal pattern output from a scanning unit that optically scans a form is text information or image information, and if it is text information, it is recognized and the recognition result is An identification unit that outputs a character code is provided, a photoelectric conversion signal pattern of an area determined as image information by this identification unit is stored in an image buffer, and the photoelectric conversion signal pattern stored in this image buffer is converted into a predetermined capacity unit. If there is an image, the image encoding circuit converts it into a special code (image code), and edits the character code and special code based on the instruction data from the input section and the position information output from the identification section. The data is displayed so that character information and image information on the form can be input separately.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram of an embodiment of the present invention. In the figure, 1 is a scanning unit that optically scans the surface of a form (not shown), converts the reflected light from the surface of the form into an electrical signal, and then converts it into a binary signal corresponding to black and white on the form. It has the ability to convert.
2 is an identification unit that stores the signal sent from the scanning unit 1 in a predetermined capacity unit, recognizes the stored pattern as a character pattern, and if it can be recognized, sets that part as a character area. Then, the relative position (or absolute coordinates) of the recognized character pattern on the form is stored. Reference numeral 3 denotes an image buffer, which stores patterns sent from the scanning section in which areas that are not determined to be character areas by the identification section 2, ie, image areas, are subjected to photoelectric conversion. Reference numeral 4 denotes an image encoding circuit which has a function of replacing the image pattern sent from the image buffer 3 with a special code (hereinafter referred to as image code) in a predetermined capacity unit depending on the presence or absence of an image. Reference numeral 5 denotes an input unit, specifically a keyboard, which allows input of various control data. 6
is an editing department that combines the character code of the recognition result in the character area sent from the identification section 2 and the image code (special code) sent from the image encoding circuit 4, matching them to the coordinates of the form. , and a function of correcting and editing a character code indicating unrecognizability (hereinafter referred to as a reject code) using data from the input unit 5. 7 is
The CRT controller has a function of converting the code data synthesized or edited by the editing section 6 into a display pattern and displaying it on the CRT 8. 9 is an FC control unit which stores the FC data of the previous read form sent from the editing unit 6, and when the next read form has the same format, the identification unit 2 stores the FC data of the previous read form sent from the editing unit 6. Data is now being transferred. That is, this
The FC control unit 9 can be operated selectively.

Next, the operation of the above embodiment will be explained. Scanning section 1
The binarized signal output from the character recognition unit 2 is stored in a certain capacity unit in the recognition unit 2, and the recognition unit 2 starts recognizing the stored pattern as a character pattern. (In this case, it is better to prepare a general-purpose format form than a completely free format form to improve accuracy.) At this time, the identification unit 2 can identify the parts that were unrecognizable. , portions that cannot be separated as character patterns are stored in the image buffer 3 as image information. Note that it is desirable that the storage capacity of the image buffer 3 is such that all the information on the form can be stored. In this way, the image information stored in the image buffer 3 under the control of the identifying section 2 is sequentially converted into an image code by the image encoding circuit 4 in units of a certain capacity when an image exists. The editing section 6 synthesizes and edits the character code of the recognition result sent from the identification section 2 and the image code sent from the image encoding circuit 4 based on the coordinate information obtained by the identification section 2. ,
Create data consisting of character code and image code for one form. This data is sent to the CRT controller 7 and converted to display data.
Displayed on CRT8. Specifically, when reading a form that includes text information and image information as shown in Figure 2A, the text information CD part is displayed as an identification section on the display screen of the CRT 8 as shown in Figure 2B. A display corresponding to the character code of the recognition result in step 2 is displayed. Unrecognized characters are displayed as a reject code RC. Further, in the image information ID portion of the form, an image code IC is displayed as shown in FIG. 2B. Furthermore, since the dirty portion DT on the form is also regarded as image information, an image code IC is displayed.

The operator looks at this display and checks the recognition result of the text information and the pattern of the displayed image code indicating the image, and if the information on the read form is correctly distinguished into text information and image information, After modifying the reject code by operating the input unit 5, the data is output to an external device. In addition, if the information on the read form is not correctly distinguished into text information and image information, the operator operates the input section 5 to input the correct data and make corrections, and then inputs the reject code. After making the corrections, the data is output to an external device. Note that when making such corrections, the editing section 6 replaces the code input from the input section 5 with a reject code when modifying reject characters, or displays a reject pattern. control. In addition, when there is an instruction to change the area of text information and image information, the editing department 6
performs controls such as displaying information on the area after the change on the CRT 8, reading information on the area to be changed from the image buffer 3, and displaying the information.

Furthermore, the coordinate position indicating the character area consisting of character information finally obtained by the editing unit 6 and the coordinate position of the image code in the image area are stored as FC data in case a form with a similar format is to be read. It is stored in the FC control unit 9.
In such a state, when reading a form with the same format as the FC data stored in the FC control unit 9, the identification unit will read the form based on the FC data stored in the FC control unit 9. 2 reads the form.

Note that the instructions from the input unit 5 not only display the reject character pattern on the CRT 8, but also display instructions to confirm that the image in the image area is a correct image and has been converted to a good image quality. By being able to display an image in which a region is displayed, it is possible to provide a character image processing device with even higher reliability.

Therefore, such a character image processing device is
There is no need to create FC data each time, characters,
Image input can be performed with relatively high accuracy.
Further, it is not necessary to write marks or the like to distinguish between image areas and text areas on a form, and it is possible to read image areas and text areas. Furthermore, it can be used to create FC data for forms in various formats.

〔Effect of the invention〕

As described above, according to the present invention, it is not necessary to create FC data, and there is no need to write marks on the form to distinguish between text areas and image areas. It is possible to provide a character image processing device that can read characters separately.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram of an embodiment of the present invention, FIG. 2A is a diagram showing an example of a form used in the same embodiment, and FIG. FIG. 3 is a diagram showing an example of a display displayed on the screen. 1...Scanning section, 2...Identification section, 3...Image buffer, 4...Image encoding circuit, 5...
Input section, 6...editing section, 7...CRT controller, 8...CRT, 9...FC control section.

Claims (1)

[Scope of Claims] 1. A scanning unit that photoelectrically converts characters and image information on a form; Character recognition means that performs character recognition processing in a predetermined unit on the photoelectric conversion signal output from this scanning unit; and This character recognition. Based on the result of character recognition processing by the means, it is determined whether or not it is character information, and if it is determined to be character information, the character information is converted to a predetermined character code, and if it is not determined to be character information. a discrimination means for outputting a photoelectric conversion signal of the predetermined unit; and an image code conversion section for converting the photoelectric conversion signal output from the discrimination means into a special code corresponding to the image information that can be distinguished from the text information. A character image processing device characterized by comprising the following.