JPS638988A - Character reader - Google Patents

Abstract

PURPOSE:To easily know the format of a slip by detecting and outputting the position of respective character rows on the slip. CONSTITUTION:The image of a slip is inputted from a photoelectric converting part 1, the character image for one page is inputted to a page buffer, and then, a detecting cutting-out part 3 obtains the concentration value histogram of the character image for one page, detects, the presence of a character row and a character row detecting part 4 detects a distance (absolute distance) from the edge part of the slip of respective character rows. A character recognizing part 5 executes the recognizing processing of a character image. When the recognizing processing is completed, the recognizing result and the position information of respective character rows are outputted through an interface part 7 to a host system and successively stored into a magnetic disk device etc.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a character reading device particularly suitable for reading characters from pages such as books, magazines, and newspapers.

(Prior Art) Optical character reading devices have been widely used in various office processing fields, but in recent years they have also become popular in the field of journalism.

In this field, the objects to be read are often forms with various layouts of text and photographs, such as the pages of books, magazines, and newspapers, and are often not compatible with conventional text reading devices. .

When reading a form with a general character reading device,
For example, as shown in FIG. 3 (A>), it is necessary to specify in advance the position where a character line exists as format information.

In this case, the position of the character line is often indicated by absolute positional information such as position 11 from the edge of the form, position L2, and so on.

In recent years, character reading devices have been developed that can automatically detect character lines on a form and read them continuously. There are two types of forms with character lines (character lines exist at positions Ll and Ll from the end) and a form with character lines as shown in Figure 3 (C) (character lines exist at positions Ll and L3 from the end). If the same text is written in each character line of both forms, the reading results will be exactly the same.

In other words, in conventional automatic character line detection type character reading devices,
Only the results of reading the characters themselves are obtained; no information about their format is obtained. For this reason, it was impossible to later know where a particular character line was located on the form.

However, when reading books, magazines, newspapers, etc., there is a high possibility that format information will be needed later for reasons such as reprinting, so if you use a conventional character reading device, it is necessary to read information about the format on the paper. There was a problem in that it had to be detected and recorded separately by means of

(Problems to be Solved by the Invention) The present invention has been made in view of the above-mentioned circumstances, and is capable of automatically detecting not only characters but also the format of read forms. The present invention also aims to provide a character reading device that is particularly effective when reading newspapers and the like.

[Structure of the Invention] (Means for Solving the Problems) The character reading device of the present invention includes an image storage means for storing an image of a form, and detects the presence of each character line by obtaining a density value histogram from the image. a character line detection means for
A character line position detecting means for detecting the position of each character line on the form from the density value histogram, and a character recognition means for recognizing each character in the detected character line. The configuration is such that the information shown is output together with the information shown.

(Function) The character reading device of the present invention not only detects the presence of each character line from the density value histogram and recognizes each character, but also detects and outputs the position of each character line on a form. If the position information is recorded together with the recognition result in a magnetic disk device or the like, the format of the form can be easily known later.

(Example) Hereinafter, details of embodiments of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention.

In the figure, 1 is a photoelectric conversion unit that optically scans the form to be read and outputs an image signal, 2 is a page buffer that temporarily stores the image of one page, and 3 is the page buffer 2 in the X direction. and a detection/cutting unit that scans in the density value histogram to obtain density value histograms in the X and Y directions, detects the presence of each character line and each character, and performs cutting; A character line detection section detects the existing position, for example, in terms of how many centimeters from the edge of the form, a control section 6 controls each section, and an interface section 7 controls data input/output with the host system.

FIG. 2 is a diagram showing the functions of the detection cutout section 3 and the character line detection section 4.

The detection cutting unit 3 scans the image S stored in the page buffer 2 in the X direction and the Y direction, and
The total number of black dots per scanning line in the direction is calculated to obtain a density value histogram H.

Note that the density value histogram in the Y direction is omitted in FIG. The presence of a character line is then detected from this density value histogram H. At this time, the character line detection unit 4 counts the distance from the origin (Olo) of the location (peak position) where it is determined that a character line exists in the density value histogram H, and determines whether each character line is from the edge of the form. Detect how far away you are located.

That is, in the device of this embodiment, when an image of a form is input from the photoelectric conversion unit 1 and character images for one page are input to the page buffer, the detection cutting unit 3 converts the character image for one page. A density value histogram is obtained, the presence of a character line is detected, and a character line detecting section 4 detects the distance (absolute distance) of each character line from the edge of the form. The character recognition unit 5 then executes character image recognition processing. When this recognition process is completed, the recognition result and the position information of each character line are output to the host system via the interface section 7 and are sequentially stored in a magnetic disk device or the like.

In this embodiment, the information indicating the position of each character line once recorded is converted to a format suitable for the purpose, such as inches, centimeters, or millimeters, according to the specifications of the host system, and then output. .

In this way, the character reading device of this embodiment outputs data indicating the position of each character line along with the recognition result of each character, so it is particularly useful for reading pages with various layouts such as books, magazines, and newspapers. This is particularly effective when printing, for example, reprint databases, etc., which can be constructed relatively easily.

Note that in the above-described embodiment device, the position of the character line is detected as the distance from the edge of the form, that is, the absolute position, but it is also possible to detect the position as the distance between each character line, that is, as the relative position. .

Furthermore, in the above-described embodiment, the image of the form is stored in the page buffer 2, but if a means for counting the position of character lines is provided, the image of the form can be stored in the line buffer. You can.

[Effects of the Invention] As explained above, the character reading device of the present invention outputs not only the recognition result of each character but also the information indicating the position of each character line. This is extremely effective when performing reading, etc. where there are various layouts, since the format information of each page can be easily known later.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a device according to an embodiment of the present invention, FIG. 2 is a diagram showing the functions of a detection cutout section and a character line detection section provided in the same embodiment device, and FIGS. C) is
FIG. 2 is a diagram illustrating the layout of character lines in a form that is to be read by a conventional character reading device. 1...Photoelectric conversion unit 2...Page buffer 3...Detection cutting unit 4...Character line Detection unit 5...Character recognition unit 6...Control unit 7...Interface unit Applicant
Toshiba Corporation Representative Patent Attorney Sasa Suyama - Figure 1

Claims (2)

[Claims] (1) image storage means for storing an image of a form;
character line detection means for detecting the presence of each character line by obtaining a density value histogram from the image; character line position detection means for detecting the position of each character line on the form from the density value histogram; 1. A character reading device comprising: character recognition means for recognizing each character, and configured to output both a recognition result of each character and information indicating a character line position. (2) The character reading device according to claim 1, wherein the character line position detection means is configured to detect the absolute position of each character line on a form.