JPS60201486A - Handwritten document reading method - Google Patents

Abstract

PURPOSE:To read a handwritten Japanese document with high speed and high precision by detecting a character kind designation in advance, detecting the character lines of a character without designation so as to select the character kind and using a dictionary proper to the character kind so as to identify the character. CONSTITUTION:A character type designation detecting sections 5 divides a pattern register 2 into four areas corresponding to a character kind designation describing frame 22, number of black points in each area is counted, compared with a threshold value, the presence or absence of the designation of the character kind is detected and the presence or absence of designation of the character kind is transmitted to an identification section 6. A character line amount detection section 4 detects the character lines from the character pattern in the pattern register 2 at the same time, the complicated degree D of the character is detected by normalizing the result with the size of the character, and when the complicated degree D of the character is detected, the D is fed to the character kind designation detection section 5. When the character kind designation is not detected on the character kind designation area at the character kind designation detection section 5, the complicated degree D is used, the character kind is decided and the result is fed to the identification section. In the identification section 6, the feature fed from the feature extraction section 3 and the dictionary are collated by using the dictionary memory of the character kind having the designation of character kind corresponding to each character in advance.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to a method for reading handwritten documents at high speed and with high accuracy.

(Background technology) As a method for reading handwritten documents, the presence or absence of a specification in a writing frame that specifies the character type of the written character is detected, and the characters in the column where the specification is detected are read only in a dictionary of the specified character type. A method has been proposed that refers to However, this method has the problem that if no character type is specified, all dictionaries must be referenced, which slows down the processing speed.

(Objective and Summary of the Invention) An object of the present invention is to improve the above-mentioned drawbacks of the conventional technology and provide a method for reading handwritten documents at high speed and with high accuracy.
Its feature is that it detects the character dose of a character and uses it as the complexity of the character, and if the character type is not specified, it selects the dictionary of the character style that contains the character based on the complexity and performs identification. It is in.

(Embodiment of the Invention) FIG. 1 is a block diagram showing an embodiment of the handwritten Japanese document reading method of the present invention. In the figure, 1 is a photoelectric conversion unit;
2 is a button register, 3 is a feature extraction unit, 4 is a character dose detection unit, 5 is a character type detection unit, 6 is an identification unit, 7 is a character name output, 8 is a hiragana dictionary memory, 9 is a katakana dictionary memory,
10 is an alphanumeric symbol dictionary memory, and 11 is a kanji dictionary memory.

Furthermore, Figure 2 is a diagram showing an example of the form used in this example.
21 is a form, 22 is a character type specification entry frame, 23 is a kanji specification column, 24 is a hiragana specification column, 25 is a katakana specification column, 26 is an alphanumeric symbol specification column, 27 is a character entry frame,
Reference numeral 28 is a character type specification entry line, and 29 is a character entry line, in which entries are made, for example, as in the form entry example shown in FIG.

The operation of the present invention will be explained below using this example of the form.

First, the presence or absence of a character type designation corresponding to each character in the character entry line 29 is detected for the character type designation entry line 28 of the form 21 in FIG. The operation is such that the photoelectric conversion unit 1 performs photoelectric conversion on the character type specification entry line 28 and converts it into a binary quantized electric signal.
Cut out an area for one character and store it in the turn register 2.

The character type designation detection unit 5 divides the turn register 2 into four areas corresponding to the character type designation entry frame 22, and counts the number of black dots in each area (character line parts are black dots). Then, it compares it with the threshold value to detect whether each character type has been specified, and sends the presence or absence of the character type, ie, kanji, hiragana, katakana, symbol, etc., to the identification unit 6. By the above operation, the character entry line is 2
The character type designation corresponding to each character of 9 is detected. Next, the second
The character entry line 29 in the figure is read. In this operation, the photoelectric converter 1 performs photoelectric conversion on the character entry line 29, converts it into a binary quantized electric signal, cuts out an area for one character, and stores it in the turn register 2. The feature extraction section 3 extracts various features from the character patterns in the turn register 2 and sends the features to the identification section 6.

At the same time, the character dose detection unit 4 detects the character dose from the characters and patterns in the pattern register 2 and normalizes it by the size of the character, depending on the complexity of the character. The complexity is expressed by the following equation.

However, K is a constant for converting D into an integer, A is the total number of black dots within the character frame, PB is the size of the circumscribed frame of the character in the height direction,
Similarly, PR indicates the size in the width direction. WL is the line width of the character and is determined by the following formula.

However, Q represents the number of points in which all four points are black points when observing all points within the character frame using a 2×2 window.

When the degree of complexity of a character is detected, the degree of complexity is sent to the character type designation detection section 5. If the character type designation detecting unit 5 cannot detect a character type designation in the character type designation area, it uses the degree of complexity to determine the following conditions, determines the character type, and sends it to the identification unit.

D (a) Refer to all dictionaries.

D≧a Assuming that the character type is a kanji, look up a kanji dictionary.

However, in this example, a=10 and K=5.

(5) The identification unit 6 compares the features sent from the feature extraction unit 3 with the dictionary, and finally outputs a one-character color name to the character bubble cuff.

There are four types of dictionary memories used in the identification unit 6: a hiragana dictionary memory 8, a katakana dictionary memory 9, an alphanumeric symbol dictionary memory 10, and a kanji dictionary memory 11. The feature is compared with the dictionary using the dictionary memory of the character types in which the character types corresponding to each character have been designated in advance.

(Effects of the Invention) As described in detail below, the present invention detects the character type designation in advance, and for characters without the above designation, the character type is selected by detecting the character dose of the character. Since characters are identified using a dictionary appropriate for the species, reading can be performed at high speed and with high accuracy, and therefore, handwritten Japanese documents can be read at high speed and with high accuracy.

[Brief explanation of drawings]

Figure 1 is a block diagram showing an example of the handwritten document reading method (6) according to the present invention, the flow side, Figure 2 is a diagram showing an example of a form used in an embodiment of the present invention, and Figure 3 is the entry of the form. It is a figure which shows an example. 1... Photoelectric conversion unit, 2... Pattern register, 3...
・Feature extraction unit, 4... Character dose detection unit, 5... Character type specification detection unit, 6... Identification unit, 7... Character name output, 8
...Hiragana dictionary memory, 9...Katakana dictionary memory, lO...Alphabet/numeric symbol dictionary memory, 11...Kanji dictionary memory, 21...Form, 22...Character type specification entry frame, 23 ...Hiragana specification field, 24...Katakana specification field, 25...Alphabet, numeric symbol specification field, 26...Kanji specification field, 27...Character entry box, 28...Character type specification entry box, 29...Character entry line. Patent Applicant Oki Electric Industry Co., Ltd. Patent Application Agent Megumi Yamamoto - (7) Figure 1 Procedural Amendment (Voluntary) August 14, 1980 - Japan Patent Office Commissioner Manabu Shiga 1, Indication of Case 1982 Patent Application No. 57420 2 Title of the invention Handwritten document reading method 3 Relationship with the case of the person making the amendment Patent applicant name (029) Oki Electric Industry Co., Ltd. 5 Patent claim for the specification subject to amendment scope, detailed description of the invention, and brief description of the drawings, drawing 6, and contents of the amendment (11) Claims are amended as shown in the attached sheet. (2) Page 2 of the specification, line 20 to "One embodiment of the present invention..." in the first line of page 3 is amended to read "an embodiment of the handwritten Japanese document reading method according to the present invention." (3) Page 3, line 8, Line 11, line 16, line 19~
Line 20, page 4, line 3, lines 3-4, page 5, line 12
Correct the "Character type designation" in line 1, line 13, and lines 5, 9, and 12 of page 7 to "1 character type designation." (4) Page 4, lines 2, 3, 14, and 15,
Lines 16, 18, 19 and 1 of the 4th line of page 7
Correct the pattern with a bang. (5) Correct "text type" in line 6 of page 4 to "character type". (6) Correct "the character type to the identification unit 6" on the seventh line of the fourth page to "the character type". (7) Store "Send." in the 8th to 9th lines of the 4th page in the character type designation memory in the 1 character type detection unit 5. ” he corrected. (8) On page 4, line 10, "specify character type" is corrected to "specify character type". (9) On the 4th page, line 20, the letter 1 is replaced by the letter 1.
and correct it. (10) "Specify the character type in the character type specification area" on the 14th line of page 5 is changed to "Sequentially refer to the character type specification memory and send the character type specification to the identification unit 6.
2. Specify the character type as shown in Figure 2. (11) Change “To identification section” to “Identification section 6” on page 5, line 16.
"To," he corrected. (12) Figure 3 of the drawings will be amended as shown in the attached sheet. Above (3) Claims (1) A handwritten Japanese document has a character frame in which characters are written, and a character type designation area provided near the character frame to designate a character type, and a character type designation area for specifying a character type. In a handwritten document reading method that recognizes handwritten characters written in a character frame using a dictionary of character types, when the character type is not specified, the character dose of the handwritten character is considered as the complexity of the character, and the complexity is A handwritten document reading method comprising selecting a dictionary of corresponding character types and recognizing handwritten characters using the selected dictionary. (2) The handwritten document reading method according to claim 1, wherein the dictionary is created for each of hiragana, katakana, alphanumeric characters, and kanji.

Claims (2)

[Claims] (1) A character frame in which to write the characters of a handwritten Japanese document,
In a handwritten document reading method that has a character type specification area provided near the character frame and specifies the character type, and recognizes handwritten characters written in the character frame using a dictionary of character types specified in the character type specification area, the character type is specified. is not specified, the character dose of handwritten characters is taken as the character complexity, a dictionary of character types corresponding to the complexity is selected, and the handwritten characters are recognized using the selected dictionary. A handwritten document reading method. (2) The handwritten document reading method according to claim 1, wherein the dictionaries are respectively created for hiragana, katakana, alphanumeric characters, and kanji.