JP2011028722A - Character search method using stroke order - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a character search method that improves accuracy of character search according to the level of knowledge about stroke order for a character such as Kanji (Chinese character), Kana character, or numeric character, and also improve search efficiency by additionally inputting a correct stroke order. <P>SOLUTION: In the character search method, in addition to a method for coding a search character by vertically and horizontally dividing a space between components of a search character and using codes showing whether the character can be divided or not, a method for coding the character by sequentially replacing the strokes of the character by codes starting with "5" of a numerical keypad or as a node, is stored in a classification storage means together with the corresponding character. The two types of codes are used in combination for input, thereby improving accuracy of searching and displaying the character from the storage means as compared with a conventional search method using a single code. Information as to whether the character can be divided or not and the vertical and horizontal direction of the stroke order are binarized, whereby achieving integrated processing. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a search method for accurately and quickly searching a wide range of characters including numbers, alphabets, kana characters and the like in addition to characters such as kanji and hangul.

As a search method related to characters, the one disclosed in Patent Document 1 is a method in which characters are coded and searched by paying attention to the presence or absence of gaps between elements constituting the characters.
According to the present invention, for example, the character division code described above is registered in place of “reading (pronunciation)” in the user dictionary or system dictionary of a front-end processor (kanji conversion software) widely used in personal computers and mobile phones. By doing so, it is possible to search and convert Chinese kanji characters that the Japanese does not know how to pronounce.

For example, Patent Document 1 shows a distinction between the Kanji search method and the engineering application that have been followed for about 300 years since the “Kangzung Character Dictionary” published in 1716, even when compared to the Kanji search method. There were two periodic features.

The first feature is that the conventional kanji search method encodes graphic information itself such as lines that are constituent elements of characters, whereas Patent Document 1 has a gap between lines. It is the point which coded the presence or absence of the space of whether or not. In other words, the novelty was recognized in that it encoded the presence or absence of spatial information that had been missed in the past, not the graphic information itself such as lines.

The second feature is that the conventional kanji search method classifies the character components into, for example, 100 or more components in the Chinese five-stroke type input method, and the components are memorized by humans and then combined into kanji. Since humans were not able to directly binarize because complicated encoding such as reconstruction or search was required, Patent Document 1 has a gap between lines constituting kanji. Therefore, novelty has been recognized in that it is possible to code directly by binarization because it is possible for humans to make an immediate determination.

Patent Document 1 having the above two features is related to engineering such as the traditional Chinese character search method such as radicals of the Chinese character dictionary that has been followed for the past 300 years, and the Chinese-style five-stroke input method applied on the extension line. A revolutionary novelty that is different from the traditional Kanji search method was recognized.

JP 2008-262248 A

The invention described in Patent Document 1 has three problems.
First, when Westerners who have no knowledge of characters search Chinese kanji characters, the search method using character division codes is effective, but Japanese people with kanji knowledge are proficient in character division codes. When converting Kanji using a front-end processor, etc., for example, “Jin” and “Double” of Kanji are classified into the same division code of “1101”, so we want to improve the search conversion accuracy of these characters. The hope came out.

Secondly, in the case of Japanese characters where alphabetic characters are used together with kanji, many alphabetic characters with few gaps are classified into the same division code “0000”, so the accuracy is higher than that of the kanji part. There is a desire to improve the search accuracy of numbers, alphabets, and kana characters that are inferior.

Third, although the search method for kanji using a 4-part character code is convenient, there is a possibility that the learner may misunderstand that there is no need to learn the stroke order of kanji, and the impact on education is concerned. It came out.

  The present invention has been made in view of the above points, and improves accuracy compared with the case where the above method is used alone depending on the degree of stroke order knowledge for characters such as kanji, kana and numbers, and correct stroke order. It provides a character search method that is motivated by the advantage that the additional character input improves the character search accuracy and the search efficiency is good, and can be expected to have an educational effect. In addition, since a general and necessary minimum input device such as a numeric keypad of a personal computer or a mobile phone is used, the convenience of being intuitive and easy to use can be provided.

In addition, the present invention can further simplify the “detailed stroke order code” composed of combinations of 10 numbers from 0 to 9, which are suitable for an input device such as a numeric keypad. If this is the case, the part can be replaced with, for example, the number 1 and if it is in the horizontal direction, for example, the number 0 can be replaced. Since it is possible to replace it with a “simple stroke order code” that can be binarized, it is possible to perform simple coding that is not possible with conventional stroke order coding. In the character search device having the same configuration as the above, the character division code of Patent Document 1 and the stroke order code of the present invention are combined and stored in the storage device, so that the accuracy of Patent Document 1 or higher can be improved.

  The gist of the present invention is that, in addition to the method of coding characters by dividing the gap between the constituent elements of the search character in the vertical direction and the horizontal direction and replacing the possibility of division with a code, the stroke order of the characters The method of coding characters by sequentially replacing the stroke order with a code starting from “5” of the numeric keypad of a personal computer or the like is stored in the classification storage means together with the corresponding characters, and the two types of codes are stored. This is a character search method that can improve the accuracy of searching and displaying characters from the storage means when compared with a search method using a single code.

Furthermore, since the detailed stroke order code consisting of 10 types of numbers from 0 to 9 can be binarized into a simple stroke order code consisting of 2 types of numbers 0 and 1, the same as in Patent Document 1 The accuracy of Patent Document 1 can be improved only by adding the stroke order code of the present invention while using the apparatus.

The simple stroke order binarization code has novelty different from the conventional stroke order code.
That is, the present invention strictly defines the ultimate character decomposable element in order to binarize all combinations of time-series strokes in the stroke order, either vertically or horizontally.

For example, the number 5 on the numeric keypad of a personal computer or calculator is regarded as the origin of the coordinate axes. The 8 keys and 2 keys in the vertical direction from the 5 keys that are the origin are regarded as the y axis of the coordinate axis, and the 4 keys and 6 keys that are in the left and right directions of the 5 keys that are the origin are regarded as the x axis of the coordinate axis.

According to the above definition of the present invention, the plane including the key 9 constituted by the plus x axis of the keys 5 and 6 and the plus y axis of 5 and 8 corresponds to the first quadrant. Hereinafter, the plane including the key 7 composed of the minus x-axis of the keys 5 and 4 and the plus y-axis of the keys 5 and 8 counterclockwise corresponds to the second quadrant. A plane including one key composed of the minus x-axis of keys 5 and 4 and the minus y-axis of keys 5 and 2 corresponds to the third quadrant. A plane including the key 3 constituted by the plus x-axis of the keys 5 and 6 and the minus y-axis of the keys 5 and 2 corresponds to the fourth quadrant.

On the premise of the above definition, the present invention provides a straight line and a curve starting from 5 in the second quadrant including the y axis in the positive and negative directions of the key 8 and the key 2 having the key 5 as the origin, and in the third quadrant. A straight line and a curve starting from 5 are defined as “longitudinal stroke order”, and for example, the number 1 is coded in a simplified manner. Also, a straight line and a curve starting from 5 in the first quadrant not including the y-axis in the plus and minus directions of the key 8 and the key 2 starting from the key 5, and a straight line starting from 5 in the fourth quadrant In addition, the curve is defined as “the stroke order in the horizontal direction”, and is coded in a simplified form, for example, with the numeral 0. Thereby, the binarized stroke order code can be strictly defined.

On the other hand, all the conventional stroke order codes follow the radicals, stroke order and stroke number specified in the “Kangzung Character Dictionary” published about 300 years ago. There wasn't. For example, the Japanese JIS code and the Chinese national standard GB code follow the traditional radical, stroke order, and stroke count since the traditional Kang Yin Dictionary, so the kanji "mouth" (Kuchigen, Kunigae) The number of strokes is defined as 3 strokes.

Since binarization is strictly defined in the present invention, the stroke order of the first stroke of the Chinese character “Mouth” is vertical one, the second stroke is horizontal 0, and the third stroke is Since the length is 1 and the fourth image is width 0, the binarized stroke order code result is 1010.

However, in the JIS code and GB code that follow the traditional stroke order, the stroke order of the first stroke is vertical, so even if it can be replaced with the number 1 as in the present invention, the second stroke is horizontal and vertical. Since the combined stroke order element “F” (break), which is an integrated unit, is the minimum structural unit, for example, binary numbers such as the numbers 2 and 3 are used to distinguish them from the horizontal number 0 in the last third stroke. It was necessary to introduce a code other than the code.

In the People's Republic of China, the code book of stroke order and stroke number of the Chinese character “GB13000.1 Character Collection Kanji Character Order (stroke order) norm” which was approved by the government was promulgated on October 1, 1999, and on January 1, 2000 We are carrying out. A total of 20902 kanji characters are recorded in a codebook that defines Chinese stroke order and stroke count, which is called GB code, which corresponds to Japanese JIS code.
Reference: “National Language and Character Work Committee Language Character Code GF3003-1999“ GB13000.1 Character Collection Kanji Character Order ”(Shanghai Education Publishing Co., Ltd.) May 2000

This kanji stroke order code for the purpose of information processing and the like defines the stroke order and the number of strokes by quinarization instead of binarization. The minimum unit of stroke order and stroke number is not a strict binarization rule such as vertical or horizontal as in the present invention, but the traditional ambiguous “one” (horizontal), “|” (vertical), “no” (payment) , “,” (Point), “F” (break), the smallest unit, and the numbers 1 (horizontal), 2 (vertical), 3 (payment), 4 (point), and 5 (fold) Is replaced with a code.

Further, in each of the five types of quinary stroke order codes, similar components are identified and belong to the five categories. For example, the code 3 includes “he” in addition to “,” (points) as similar components. Further, the code 5 includes not only “F” (fold) but also 14 types of similar components such as “B”, “L”, and “KU”.
As a result, it cannot be said that it is strictly quinary. For example, the above-mentioned “mouth” (Kuchigen, Kunigae) has a stroke order code of 251, 2 (vertical), then 5 (folded), 1 (horizontal), and 3 It is classified into the number of strokes. In this case, it is impossible to perform binarization as strictly and intuitively as in the present invention. Accordingly, it has been difficult for anyone to understand universally and intuitively, and it has been difficult to learn what combination of stroke order and stroke number is appropriate for 300 years.

Further, for example, although “dry” and “us” are different character shapes, both of them are the same in stroke order code 112 (horizontal and horizontal) and cannot be identified. The reason is that “splash” corresponding to a substantial fourth stroke following the so-called third stroke in 2 (vertical) is identified as a similar component and included in the same category. On the other hand, according to the present invention, “dry” can be encoded in different stroke order codes such as “001” (horizontal, horizontal, vertical) and “us” 0011 (horizontal, horizontal, vertical, and vertical). Can be strictly identified. In this way, binarization is realized by strict regulations.
Of course, the “simple stroke order code” of the present invention can also be applied to the traditional “stroke order and stroke count” concept of kanji.
The traditional Kanji stroke order and stroke count concept was “All strokes are treated as one stroke even if the direction changes in the middle of a continuous line within a stroke.”
Therefore, if only the vertical / horizontal direction information from the starting point in a single stroke writing section is coded, and the vertical / horizontal direction information changed up to the end point in the subsequent single stroke writing section is binarized, it is not coded. The “simple stroke order code” of the present invention can also be applied to the traditional stroke order and stroke count of Chinese characters.
For example, the traditional number of strokes for the Chinese character “Mouth” was three, not four. The reason for this is that, in the second stroke order, the line that goes from the starting point to the right lateral direction is a continuous line with a single stroke, and the line is bent in the vertical direction on the way without leaving the brush. In order to write, this one-stroke section was to be handled as one stroke instead of two strokes.
Focusing on this feature, the second line of the traditional stroke order of the Chinese character “Mouth” is coded to the number 0 as a horizontal line that goes from the starting point of the continuous line in a single stroke to the right, The line that is bent and written in the vertical direction in the middle of the subsequent one-stroke section is discarded as a vertical direction without encoding it into the number 1, and this one-stroke section is the number 0 in the horizontal direction from the starting point. This one-stroke section can be handled as one stroke if it is coded as a representative.
As a result, the kanji “mouth” can be binarized by applying it to the traditional stroke order and stroke number concept of 100 (vertical and horizontal) with a “simple stroke order code”.
In this code, the line drawn from the starting point of the first stroke of the first stroke is 1 (vertical direction), and the change in direction to the end point that follows in the first stroke of the first stroke is discarded. It shows that there is. In the case of the Chinese character “mouth”, there is no change in direction from the start point to the end point of the first stroke. Then, it is shown that the section of the new stroke of the second stroke that is not continuous with the end point of the first stroke is only coded as 0 in the horizontal direction from the starting point, and the vertical direction ( It shows that the coding up to the end point changed to 1) is not adopted. Similarly, the third stroke is the final stroke order of the first stroke of the Chinese character “mouth”, and at the same time, the horizontal direction from a new start point that is not continuous with the end point of the stroke of the first stroke section until the second stroke (0) Indicates that only the line is coded.
Similarly, for example, the Chinese character “U” can also be binarized to 001 (horizontal, horizontal, vertical) by this method, and these correspond to the traditional stroke order and the number of strokes.
In this case, the three digits of the stroke order code of the binary code “100” of the Chinese character “mouth” and the binary code “001” of the Chinese character “U” correspond to three strokes of the traditional Chinese character stroke number. .
The advantage of this method is that it can binarize the stroke order and number of strokes of Kanji characters that have been followed in traditional Kanji dictionaries and character codebooks, and has a great affinity for education and practical use.
However, in the case of characters that consist mainly of one-stroke writing notation, such as cursive alphabetic characters and Arabic characters, binarization that uses only the starting point information of such a one-stroke section is sufficient for code identification Since there are not a few characters that cannot be expected to have an effect, in the present invention, in describing the “simple stroke order code”, the description of the application of the traditional kanji “stroke order and stroke count” is omitted, The method of coding the “stroke order” of “all sections” from the starting point of writing, which is called character writing, to the final ending point where the character has been completely written, will be mainly described.
In general, most of the characters that are used worldwide do not have the concept of “number of strokes”. On the other hand, since there is a universal commonality that characters are composed of lines, characters around the world are always time-series. Because it is possible to code based on typical "stroke order" information, the character search method by coding using "stroke order" can be applied to East Asian cultural characters such as Kanji, Hangul and Kana without exception. That's why.

The present invention, for example, allows a Japanese who has knowledge of kanji to search for Chinese kanji whose pronunciation is not known, and simply input a character division code using a familiar Japanese front-end processor kanji conversion software. Useful for improving accuracy when converting to Kanji. If you have the stroke order knowledge of Japanese common kanji, you can analogize the stroke order of Chinese common kanji to some extent, so you can efficiently search and convert Chinese by entering the stroke order code in addition to the character division code.
In addition to the method of registering the character division code and stroke order code by the user dictionary registration method of the Japanese front-end processor Kanji conversion software in the storage device of a personal computer or mobile phone, for example, the above code published on the Internet and It is possible to search the target Chinese using the search function of the browsing software and the CGI text search function of the home page, which lists the corresponding list of Chinese, so you can use the icons and soft keys displayed on the home page. High-precision search is possible just by clicking.

It is a functional block diagram of the present invention. It is a search processing flowchart of the present invention. It is explanatory drawing which compared the conventional character division method and the character stroke order search method of this invention. It is a character database table | surface of this invention. This is a method for inputting a line with the number 5 key of the present invention as a starting point and a nodal point. It is the input encoding method of the character comprised from the vertical line of this invention. It is the input coding method of the character comprised from the line of the horizontal direction of this invention. It is the input encoding method of the character comprised from the continuous line of this invention. It is the input encoding method of the character containing the line which is not continuous of this invention. It is the input encoding method of the character comprised from the counterclockwise circle of this invention. It is the identification method of the character comprised from the character and line comprised from the circle of this invention. It is the character form which reproduced the Chinese character "mouth" of the present invention. It is the input coding method of the character comprised from the combination of the circle | round | yen and line | wire of this invention. It is the input coding method of the character comprised from the circular arc which does not close of this invention. It is the input coding method of the character comprised from the combination of the straight line and semicircle of this invention. It is a character shape reproducing the alphabet “D” of the present invention. It is the input encoding method of the character comprised from the several semicircle and line of this invention. It is a character shape reproducing the alphabet “S” of the present invention. This is an input coding method that represents various arcs with the three keys of the present invention. FIG. 3 is an input coding method for a closed circle and a non-closed arc according to the present invention. FIG. It is the character form which reproduced the stroke order which shifts from a closed circle to a line of the present invention. It is the character form which reproduced the stroke order which shifts from a non-closed arc of the present invention to a line. In the database of the present invention, it is an example consisting of “quadrant binary (binary) code” and “stroke order binary (binary) code” corresponding to “character”, from “1” to “s”. Show. # Is a binary identifier. In the database of the present invention, it is an example of “quadrant binary (binary) code” and “stroke order binary (binary) code” corresponding to “character”, from “t” to “ge”. Show. # Is a binary identifier. In the database of the present invention, it is an example of “quadrant binary (binary) code” and “stroke order binary (binary) code” corresponding to “character”, from “go” to “ya” Is shown. # Is a binary identifier. In the database of the present invention, it is an example of “quadrant binary (binary) code” and “stroke order binary (binary) code” corresponding to “character”, from “yu” to “po”. Show. # Is a binary identifier.

  In the embodiment of the present invention, in addition to the method of coding characters by dividing the gap between the constituent elements of the search character in the vertical direction and the horizontal direction and replacing the possibility of the division with a code, the stroke order of the characters is added to the personal computer. The character encoding method is stored in the classification storage means together with the corresponding characters by replacing the stroke order with the codes in order starting from “5” of the numeric keypad as a starting point or a nodal point, and the two kinds of codes are input. When used together, the accuracy of searching and displaying characters from the storage means can be improved as compared with a search method using a single code.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 of the present invention is a functional configuration block diagram illustrating a general embodiment in which Japanese kanji conversion software called a front-end processor is stored in advance in a storage means 30 mounted on a personal computer or a mobile phone. is there.
As the input means 10, a numeric key called a numeric keypad of a personal computer or a mobile phone, a space key or an enter key necessary for operating the kanji conversion software is used. In the storage means 30, the stroke order codes of the present invention are stored in the form of a table in the form of a table in the same record as the character corresponding to the character division code in the form of a user dictionary or system dictionary of Japanese conversion software. The code information input from the input means 10 and the code stored in the storage device 30 are collated in the arithmetic device 20.

  When the input divided code and the stroke order code match the stored database code, the corresponding character is displayed on the liquid crystal screen of the display means 40 or the like. Thus, by using a character code instead of Japanese kana reading, the user can convert and display characters such as Chinese that he / she wants to search as if he / she entered a phone number.

  FIG. 2 is a flowchart of search processing according to the present invention. The processing in the computing means 20 is shown. The divided code and the stroke order code input in S200 are collated with the database divided code and the stroke order code in S300, and the matched characters are displayed in S400. If the character search conversion is to be continued, the following steps are repeated from the input operation in S200.

FIG. 3 is an explanatory diagram comparing the conventional character division method and the character stroke order search method of the present invention.
For example, when the user determines whether or not there is a gap between lines in the order of up, down, left, and right, the characters “jin” and “double” in the kanji are both spaced in three directions, up and down, right and left. It can be seen that these kanji characters can be divided into three. In that case, the codes are sequentially replaced with “1204” or the like. “0” indicates that there is no gap. In addition, it may be replaced with “1101” and a 4-bit code in order. At this time, “0” indicates that there is no gap and “1” indicates that there is a gap and characters can be divided. However, since “Jin” and “Double” are classified into the same division code “1101”, the user Can not be identified as either Kanji simply by inputting the same code “1101”.

On the other hand, in the character stroke order code described in FIG. 5 and the subsequent figures, “Jin” can be replaced with “1100” and “Double” can be replaced with “11100101010”. The search conversion character can be specified as either one.

This stroke order code is replaced with a number as follows when, for example, the numeric keypad of the personal computer at the lower right of FIG. 3 is used as an input device. The user determines in which direction the line is drawn in the stroke order with the two keys starting from the number key 5 to complete the character. For example, “51” in “Nenben” is “51”, and the stroke order is written. This means a diagonally downward line starting at 5.

Next, “52” is input as the line drawn vertically in the vertical direction of “People”. When shifting from the diagonally left-down line of “51”, which constitutes a human bias, to a line that extends right below “52”, this is not a continuous line called a stroke, but this is a number. It is coded with the key “0”. As a result, the bias can be encoded as “51052”. Since the line does not continue when shifting to the left side of “Jin” and then to “N” on the right side, it can be encoded as “510520” in the next stroke order. The stroke on the line above “2” is “56” and the stroke order shifts to the right. Next, since the line is not continuous when moving to the line below “2”, it can be finally encoded as “51052056056”.

This is a method for coding the stroke order of characters composed of lines in the stroke order. In situations where detailed code information is not required, a simple stroke order code can be used.

The five types of direction lines “52”, “51”, “54”, “57”, and “58” with 5 as the starting point are defined as “vertical direction lines”, and for example, the numeric key “1” is substituted. A line of three types “59”, “56” and “53” starting from 5 is defined as a “horizontal line”, for example, “0” which is a numeric key is substituted. Whether the continuity of the line is possible or not is discarded from the information. If the stroke order of the character is only “whether vertical or horizontal”, “Jin” can be replaced with a simple stroke order code “1100”.

“Mouth” and “A” in FIG. 3 also indicate that a character that cannot be distinguished from “0000” in the divided code is replaced with “1010” and “100” in the stroke order code, and thus can be distinguished.
Thus, the character search accuracy can be improved by using the character division code and the stroke order code together.

FIG. 4 is a character database table of the present invention. A database as shown in FIG. 4 is stored in the storage means 30 of FIG. For example, if characters such as “B”, “C”, “D”, “M”, “M”, “M”, and “Y” are represented by a four-division code, they are all classified into “0000” and characters having no gap. In a database in which characters are associated with four-division codes, it is not possible to search for and specify a target character from among these four-division codes.
On the other hand, if the stroke order code of the present invention is made into a database together with the corresponding characters, one character can be searched and specified from the character examples shown in FIG.

23, FIG. 24, FIG. 25, and FIG. 26 show an improvement in accuracy when a stroke order binary code is used in addition to a character and a 4-split binary 4-digit code in the storage means 30 of FIG. FIG. 23, FIG. 24, FIG. 25, and FIG. 26 show examples of stroke order codes by binarization using MS Gothic fonts commonly used in personal computers. For example, 10 numbers (see FIG. 23) of Gothic font numbers “1” to “0” are all the same code of 0000 and 4 digits in the character division code, so the number cannot be specified from the code. . However, if binary stroke order codes are used in combination, all numbers except “1” and “7” in MS Gothic font can be uniquely identified, and one number can be identified from one stroke order code, improving search accuracy. I understand that

As shown in FIG. 4, if the detailed stroke order code is a more detailed version of the binarized simple stroke order code, the character search accuracy is improved over binarization. For example, if MS gothic fonts “1” and “7”, which could not be identified by the binary stroke order code, are coded by the detailed stroke order code, “1” of the MS gothic font can be replaced with 5952. Since “7” can be replaced with a code different from 5651, “1” and “7” of the MS Gothic font can be identified and specified. Therefore, the number search and conversion accuracy are further improved as compared with the binarized stroke order code. The method for coding the detailed stroke order code will be described below.

FIG. 5 shows a method of inputting a line having the number 5 key of the present invention as a starting point and a nodal point. The number “5” key is not only the starting point at the time of writing, but also the nodal point of a continuous line called one-stroke writing. That is, when a line is drawn on keys in 8 directions other than 5 to 0 and the process proceeds to the next stroke order as it is, 5 can be considered as a node (node) since it is a recursive starting point. FIG. 5 shows that the line is represented by two numeric keys with 5 as the starting point or nodal point.

FIG. 6 shows an input encoding method for characters composed of vertical lines according to the present invention. As described with reference to FIG. 3, lines in the direction of 180 degrees or more and 360 degrees or less can be classified as “vertical lines” with respect to 5 as a starting point. The vertical direction lines are defined as “vertical direction lines” in six types of directions of “52”, “51”, “54”, “57”, and “58”. By substituting the number “1” for the “vertical line” and substituting the number “0” (zero) for the three “horizontal lines” in FIG. "1" and "0" may be identified in a binary manner.
In this case, “52”, “51”, “54”, “57”, and “58” are represented as “1”, “1”, “1”, “1”, and “1”, respectively.

FIG. 7 shows an input encoding method for characters composed of horizontal lines according to the present invention. Regarding the horizontal line, three types of lines “59”, “56”, and “53” are defined as “horizontal lines”.
By substituting the number “1” for the “vertical line” described in FIG. 6 and substituting the number “0” (zero) for the three types of “horizontal lines”, The determination of “a horizontal line” may be binaryly identified by “1” and “0”. In this case, “59”, “56”, and “53” are all represented as “0”, “0”, and “0”.

  FIG. 8 shows an input encoding method for characters composed of continuous lines according to the present invention. For example, the character kana “to” with continuous lines is represented as “5953” or “00”. In this case, the numeric key 5 is a starting point and a nodal point. According to the present invention, a stroke order code can be input by an input device such as a general numeric keypad without using an input device composed of a huge key or button arrangement by a method starting from the numeric key “5”.

FIG. 9 shows an input encoding method for characters including discontinuous lines according to the present invention. For example, the Chinese character “10” with no continuous lines can be coded as “56052” or “01”. In the correct stroke order, continuity is lost when moving to the line of “|” in the vertical direction after “1” in the horizontal direction of “10”. In such a case, the detailed stroke order code is represented as “56052”. In the simple stroke order code, the continuity information is discarded, and “01” indicates that “the stroke order is to draw the vertical line next to the horizontal line”. In the case of this encoding, “0” has no meaning of discontinuity and simply indicates a horizontal line.

FIG. 10 shows an input encoding method for characters composed of counterclockwise circles according to the present invention. For example, a counterclockwise circle such as the numeral “0” or the alphabet “O” is represented as “54125”. The difference between a line and a curve is identified by coding a line such as a circle or an arc with three or four keys starting from 5, while a line is coded using two keys starting from 5. .
The closed circle is characterized in that the starting point “5” is also the end point as “54125” in FIG.

FIG. 11 shows a method of identifying a character composed of a character composed of a circle and a line according to the present invention.
Since the Chinese character “mouth” has a different code from “5205652056” and the alphabet “O” has “54125”, it is possible to identify the difference in characters even if the number keys used are common.

FIG. 12 is a character shape reproducing the Chinese character “mouth” of the present invention. ● indicates the location where the numeric key “0” is designated. For continuous circles, the numeric key “0” is not specified.

FIG. 13 shows an input coding method for characters composed of a combination of a circle and a line according to the present invention.
As an example of a combination of a clockwise circle and a straight line, the alphabet “Q” is expressed as “5477853”. Before transitioning to the line of the “53” direction, which is the next stroke order after drawing a circle, the number key “0” is inserted because the continuity is lost.

FIG. 14 shows an input encoding method for characters composed of non-closed arcs according to the present invention.
As an example of an arc that does not close, the alphabet “C” is represented by “5412”. When this is compared with the closed circle “54125” in FIG. 10, it is shown that there is a difference in whether “5” is specified at the end of the stroke order. Even if four common numeric keys are used, such a difference can be expressed.

FIG. 15 shows an input coding method for characters composed of a combination of a straight line and a semicircle according to the present invention.
The alphabet “D” is expressed as “5205632”, which also indicates that the continuity of the line is lost at the position where the numeric key “0” is inserted.

FIG. 16 is a character shape reproducing the alphabet “D” shown in FIG. 15.

FIG. 17 shows an input encoding method for characters composed of a plurality of semicircles and lines according to the present invention.
The alphabet “S” composed of two semicircles and a line can be expressed as “58745535214”. The number key “0” is not inserted because the continuity of the line has not been lost when moving to the next line “53” depicting a left-handed semicircle with “5874”.

FIG. 18 is a character shape reproducing the alphabet “S” of the present invention.

FIG. 19 shows an input encoding method that represents various arcs with the three numeric keys of the present invention. Various arcs are represented by three numeric keys including the starting point 5. The pseudonym “tsu” is represented by “563”.

FIG. 20 shows the difference in the input coding method of the closed circle and the non-closed arc of the present invention.
As for “541556”, a horizontal straight line is drawn in the right direction from “54125” to “56” from the node 5 after the circle is closed.
“541256” is a circle that does not close with “5412”, that is, a horizontal straight line is drawn to the right of “56” from the end point of the arc.

FIG. 21 is a character shape reproducing the stroke order of transition from a closed circle to a line according to the present invention.
The stroke order of “556” in “541556” indicates that “5” is a nodal point that serves as the end point of a circle and the starting point of a line.
On the other hand, FIG. 22 is a character shape that reproduces the stroke order of the present invention from a non-closed arc to a line. The stroke order of “256” in “541256” indicates that “5” is not the end point of the circle but simply the starting point of the horizontal line from the end point “2” of the arc.

10 Input means 20 Calculation means 30 Storage means 40 Display means

Claims (1)

In addition to the method of coding the character by dividing the gap between the constituent elements of the search character in the vertical direction and the horizontal direction and replacing the possibility of the division with the code, the stroke order of the character is set to “5” of the numeric keypad. A method of coding characters by sequentially replacing the stroke order with codes as knots is stored in the classification storage means together with the corresponding characters, and the two types of codes are used together at the time of input from the storage means. It is possible to improve the accuracy of searching and displaying characters compared to a search method using a single code, and further, it is possible to process uniformly by binarizing the possibility of division and the vertical and horizontal directions of the stroke order. Character search method.

Images