JPS61187086A - Optical character reader - Google Patents

Abstract

PURPOSE:To read only characters to be read out of characters on a form without format information and with an inexpensively produced form by inputting optically various patterns on the form where a frame is printed in a non-dropout color and recognizing patterns existing in the detected frame and outputting them together with information. CONSTITUTION:A form 1 is irradiated with a lamp 2, and a pattern on the form 1 is detected by a CCD sensor 3. The detected pattern is stored in a memory after being converted to a digital signal by a photoelectric converting part 4. A frame detecting part 6 detects the frame from the pattern on the form 1 which is stored n the memory 5. A recognizing part 7 cuts out characters existing in the frame detected by the frame detecting part 6 and recognizes characters and outputs frame information and read character information to an output device 8. Characters 22 to be read are surrounded with a frame 20 printed preliminarily in the non-dropout color.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to an optical character reading device, and particularly to an optical character reading device suitable for reading characters without using format information.

[Background of the invention]

As disclosed in Japanese Patent Application Laid-Open No. 58-8385, an optical character reading device that saves the labor of conventional formatting information is disclosed in Japanese Patent Application Laid-open No. 58-8385, in which the area to be read on a form is surrounded with a writing instrument such as a pencil, and thereby the formatting is performed. 2. Description of the Related Art Devices are known that allow an optical character reader to recognize information and read characters. However, this optical character reading device requires format information, and it is troublesome that one person manually encloses the area to be read on the form. In addition, forms used in this optical character reading device must be printed using colors that the optical character reading device cannot detect (hereinafter referred to as dropout colors), which poses a problem in that the forms are expensive.

Furthermore, as disclosed in Japanese Patent Publication No. 59-21076, an optical character reading device has been proposed which reads characters by detecting character string lines without using any format information. However, this optical character reading device has a problem in that it reads all the characters on the form and cannot read only the characters that are desired to be read.

[Purpose of the invention]

The present invention was made in view of the problems of the prior art described above, and it is possible to read only the desired characters from among the characters on the form without using format information and using a form that can be produced at low cost. The purpose is to provide an optical character reading device.

[Summary of the invention]

The optical character reading device of the present invention uses a form with a frame printed in a color that the optical character reading device can sense, such as black (hereinafter referred to as non-dropout color).
Various patterns on a form with frames printed in non-dropout colors are optically input, the input patterns are memorized, frames printed in the above non-dropout colors are detected from the stored patterns, and the frames are printed in the non-dropout colors. It is characterized in that it is output as information, and furthermore, a pattern existing within the detected frame is recognized and output together with the frame information.

[Embodiments of the invention]

Hereinafter, the present invention will be explained in more detail using the accompanying drawings.

FIG. 1 is a block diagram showing an embodiment of the optical character reading device of the present invention. As shown, a form 1 is illuminated by a lamp 2, and a pattern on the form 1 is detected by a CCD sensor 3. The detected pattern is converted into a digital signal by the photoelectric conversion unit 4 and then stored in the memory 5.

The frame detection section 6 is.

A frame is detected from a pattern on a form 1 stored in a memory 5. The recognition unit 7 cuts out the characters existing inside the frame detected by the frame detection unit 6, performs character recognition, and outputs the frame information and the read character information to the output device 8. This output device 8 may be, for example, a floppy disk device or a line.

FIG. 2 is a diagram showing an example of the form 1 used in this embodiment. On the form 1, there are characters 22 to be read and characters 30 not to be read. As shown, the characters 22 to be read are surrounded by a frame 2o printed in advance in a non-dropout color.

Next, the operations of the frame detection section 6 and recognition section 7 shown in FIG. 1 will be explained using FIGS. 3 and 4.

FIG. 3 is a flowchart showing the operations of the frame detection section 6 and the recognition section 7, and FIG. 4 is an enlarged view of a part of the form 1 shown in FIG. 2. The frame 20 on the form 1 is detected as follows. That is, as shown in FIG. 3, in step 10, it is detected that there are pixels of a predetermined length or more in the horizontal direction of the form 1, and horizontal line detection is performed. Next, in step 11, the right end and left end of the plurality of detected horizontal lines are compared to associate the two horizontal lines forming the frame 20, and the space between the corresponding horizontal lines is decoded. The range to be decoded is as shown by the dotted line 21 in FIG. 4, extending the horizontal line by lll1ffl in the horizontal direction, and 0.
The area is shifted inward by 5m111. The reason for extending the horizontal line by 1 mm in the horizontal direction is to reliably detect the vertical line that exists at the end of the horizontal line.
The purpose of shifting uo is to prevent the two horizontal lines that have already been detected from being detected again. Vertical line detection is performed in step 12 by using the result of decoding the inside of point g (21).The range from the vertical line detected in step 12 to the vertical line and from the horizontal line detected in step 10 to the horizontal line That is, the inside of the frame 20 becomes the area where reading is performed.In step 13, characters 22 are cut out from inside the frame 20 by known means, and in step 14, the cut out characters 22 are recognized.

FIG. 5 is a diagram showing an example of the output after reading the form 1 shown in FIG. 2, in which read character information 41 and frame information 40 indicated by an at mark are mixed.

The character information 41 is obtained as a result of recognizing the characters 22 within the frame 20. Further, the frame information 40 is information that means a vertical line of the frame 20 shown in the second FM, and as a result, a portion where two pieces of frame information 40 are consecutive means that the line has changed. Therefore, a portion where two pieces of frame information 40 are consecutive may be changed to a so-called separate nine-line mark.

In the above embodiments, the shape of the frame 20 has been described as a square, but the present invention is not limited to this, and by changing the function of the frame detection unit 6, it can be formed into any shape, such as a circle or a dotted line. It can be done. Also, once frame information 4
If 0 is detected, even if characters or patterns that do not need to be read are surrounded by a frame when reading other forms, they can be skipped. Furthermore, according to this embodiment, when processing the frame information 40 and character information 41 shown in FIG.
It becomes possible to search for character information 41 using frame information 40 as key data.

Further, as shown in FIG. 6, by outputting the number of characters in one frame 20 as field character number information 50, character information 41 can be output continuously. Also,
In this case, instead of field character count information 50, frame 2
A similar effect can be obtained by outputting the relative address of the read result of 0.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, an inexpensive form that does not require format information and is printed in non-dropout color is used, and only the characters that are desired to be read from among the characters on the form can be freely read. It is possible to provide an optical character reading device that can read characters.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the optical character reading device of the present invention, FIG. 2 is a plan view showing an example of a form used in the optical character reading device of the present invention, and FIG. 3 is similar to FIG. FIG. 4 is a partially enlarged view of the form shown in FIG. 2, and FIGS. 5 and 6 are explanatory diagrams showing examples of output in the embodiment shown in FIG. DESCRIPTION OF SYMBOLS 1... Form, 2... Lamp, 3... CCD sensor, 4... Photoelectric conversion section, 5... Memory, 6... Frame detection section, 7... Recognition section, 8. ...Output device, 20...
Frame, 22... Characters to be read. 30... Characters that cannot be read, 40... Frame information, 41.
...Character information, 50...Field character number information.

Claims (1)

[Claims] a first means for optically inputting various patterns on a form with a frame printed in a non-dropout color; a second means for storing the inputted patterns; and printing from the stored pattern in the above-mentioned non-dropout color. Detect the frame that is
An optical character reading device comprising: a third means for outputting frame information; and a fourth means for recognizing a pattern existing within the detected frame and outputting it together with the frame information.