GB2359400A

GB2359400A - Character input

Info

Publication number: GB2359400A
Application number: GB0004072A
Authority: GB
Inventors: Shide Wang
Original assignee: Kenwood KK
Current assignee: Kenwood KK
Priority date: 2000-02-21
Filing date: 2000-02-21
Publication date: 2001-08-22
Anticipated expiration: 2020-02-21
Also published as: GB2359400B; CN1310371B; HK1036505A1; GB0004072D0; TWI249697B; CN1310371A

Abstract

In a method of e.g. Chinese character selection, a user of an input device enters the first two strokes of the Bu Shou part of a character, and optionally the last stroke. The user can then select the desired character without entering any further strokes, but if there are too many possible characters, the user then enters the first and second strokes, and optionally the last stroke, of the remaining part of the character until the desired character is selected for input. Preferably, at any stage in the character selection at which the number of possible characters which matches the user input up to that stage can conveniently be displayed on the input device, the possible characters are displayed and the user can select one of the displayed characters instead of entering further strokes.

Description

2359400 1 CHARACTER INPUT

TECHNICAL FIELD

The present invention relates to a method and apparatus for character input, particularly but not exclusively for input of Chinese characters, and particularly but not exclusively for input on a keyboard having a restricted number of keys, such as a telephone keypad.

BACKGROUND

As there are many more Chinese characters than keys on a standard typewriter or computer keyboard, an algorithm is needed to convert multiple key presses on such a keyboard to the required Chinese character.

Many input methods have been invented in past decades and some of them have been incorporated in computer software packages, but few have gained popularity among occasional users. Most methods are designed for coding efficiency and typing speed. Users must have intensive training to master the complicated and easily forgotten rules. These input methods are only suitable for professional typists rather than occasional users.

Up to now, the only widely accepted Chinese input method is by pronunciation, namely Pinyin in China and Bopornofo in Taiwan. However, most people speak their own dialects. Some of them can speak Mandarin but not necessarily with an accurate accent. Moreover, there are too many (as many as 100 or more) Chinese characters with the same or similar pronunciation. Therefore, Pinyin or Bopomofo input methods are slow and very frustrating.

Other input methods are based on strokes, which are the basic building blocks of writing Chinese characters. A stroke is the portion of a Chinese character written with one motion of brush to paper. Some input methods are based on the sequence of strokes in writing order. However, Chinese 2 characters are pictograms, so the writing order varies considerably among different people. Although there are basic rules, not everyone follows them correctly. Once a writing habit is developed in childhood, it is almost impossible to change later.

Most stroke-based input methods divide a Chinese character into a few parts or components. Strict rules define how to divide a character and assign these components to different keys. Users press the necessary keys to form a sequence of components to produce a valid character. It is too hard for occasional users to follow those artificial components and partition rules.

STATEMENT OF INVENTION

According to the present invention, there is provided a method of Chinese character selection, in which a user of an input device enters the first two strokes of the Bu Shou part of a character to select the Bu Shou part of the character, and optionally another stroke such as the last stroke. The user can then select the desired character without entering any further strokes, but if there are too many possible characters, the user then enters at least one stroke of the remaining part of the character, preferably the first and second strokes, and optionally the last stroke, until the desired character is selected for input.

Preferably, at any stage in the character selection in which the number of possible characters which matches the user input up to that stage is less than a threshold number, such as the number which can conveniently be displayed on the input device, the possible characters are displayed and the user can select one of the displayed characters instead of entering further strokes.

The present invention includes within its scope an input device arranged to carry out the method, software which enables the method to be carried out, and a database of Chinese characters indexed by Bu Shou part and 1 3 first and second strokes of the other part of the character, which facilitates the performance of the method.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a front view of a mobile telephone handset; Figure 2 is a diagram of the internal electronic components of the handset; Figure 3 shows a first step of inputting a Chinese character in an embodiment of the present invention; and Figure 4 shows a second step of inputting a Chinese character in an embodiment of the present invention.

MODES FOR CARRYING OUT THE INVENTION Figure 1 shows a mobile telephone handset H having a keypad K comprising numeric keys 0 to 9, star () and hash (#) keys, and function keys such as 'YES', 'NO', back/up forward/down (->). clear (CLR) and other function (f). A display D is able to display Arabic numerals, Roman letters and Chinese characters, and may be an LCD with sufficient resolution to display at least one line of numerals. letters and characters. A microphone M and speaker S are also present, to allow voice calls.

Figure 2 is a schematic diagram of the electronic components of the handset H. These components need not be discrete, and may be integrated. For example, the components may be integrated onto a microcontroller chip and an RF stage chip. A processor P is connected via a bus B to a volatile memory (V), a non-volatile memory (NV), an 1/0 interface (I/0) and an RF modem (RF). The 1/0 interface decodes input from the keypad K and microphone M, drives the display D and speaker S. The RF modem is connected to an antenna A so as to receive and transmit RF signals. The components are powered by a 4 battery (not shown) or a mains electricity connection (not shown) via a t ransformer.

The non-volatile memory (NV) stores software which is executed by the processor P in order to carry out the functions of the handset. Optionally, the non-volatile memory is reprogrammable to upgrade the software. The algorithm described below as an embodiment of the present invention may be installed as an upgrade to the software of an existing mobile telephone. The upgrade may be received as a wireless message, via the RF modem (RF). The non-volatile memory may include a removable component which can be transferred to other handsets H, such as a SIM card.

The handset H implements protocols which allow text messages to be sent and received. For example, the handset may be GSM-compatible and support the GSM SMS (short message service) protocols.

A Chinese character input method in an embodiment of the present invention will now be described. The method is based on "Radical Indexing", which is a common way of indexing Chinese dictionaries and is therefore familiar to most users of Chinese. Each character can be normally divided into two parts. The first part (Part 1) is called 'Bu Shou' which represents one of a number of different concepts, such as 'metal', 'head', 'dog', 'walk', 'father' and so on, in different categories (Nature, Body, Animals, Actions, Others).

The remaining part of the character (Part 2) has different possible purposes. It may represent how this character is pronounced or the meaning of the character. Everyone who learns the Chinese language must learn these basic rules from the beginning. Therefore, no extra training is necessary for those who have a few years' education in Chinese.

To check a Chinese dictionary using radical indexing, the Bu Shou is looked up first. Under the same Bu Shou. the required character is found by counting how many strokes the character has. However, stroke counting is not practical for character input.

In the method of this embodiment, all Chinese character strokes are classified into one of five basic strokes - horizontal, vertical. leftfalling, rightfalling and turning strokes, which is a fairly standard and well-known classification. The handset H stores a database of Bu Shous indexed according to their first, second and last strokes, each record including a Bu Shou code and optionally display data representing the appearance of that Bu Shou, if Bu Shous are not included in a Chinese font database also stored by the handset H. The handset also stores a database of Chinese characters in the Traditional or Simplified Character set, indexed by Bu Shou and by the first, second and last strokes of their Part 2, each record including a code indicating the appropriate Chinese character. More than one Chinese character may have the same Bu Shou and first and second strokes of Part 2, in which case there is a separate record for each character. The font database stores the display data for each Chinese character.

Some characters have more than one different possible partition and/or different possible orders of writing, and the database reflects this by storing multiple entries for those characters indexed according to each possible partition and/or order of writing. Therefore, users can keep their own writing habits and still use the input method. The multiple entries may reflect alternative correct possibilities, or common errors in writing, which occur for about 20% of Chinese characters. The rest may have only a single entry.

As the Bu Shou database is much smaller than the character database, all possible normal ways of entering each Bu Shou character are encoded in the database. Once the user finds the correct Bu Shou, the desired character can usually be found from the list of options without keying in further strokes, so that by allowing errors in keying in the Bu Shou, there is a greater likelihood of even inexperienced users finding the right character.

A symbol for each type of stroke is mapped onto one of a group of keys on the keypad K. In this embodiment. the strokes are assigned to 1 ty 6 numerical keys 2 to 6, as shown in Figure 1. The assignment of different keys in this mode is shown below in Table 1:

Table 1

Key Assignment 1 Input punctuation 2 3 4 6 7 Skip Bu Shou Input (see below) The user carries out the following steps to enter a character:

(1) The user keys in the first, second and optionally the last strokes of the Bu Shou. Software running on the handset H searches the Bu Shou database for records matching the input strokes. After the second stroke, the handset H displays all possible Bu Shous having the first and second strokes.

If there are too many possible Bu Shous to be displayed. the handset may indicate this. If the last stroke is entered, the software running on the handset selects, from among the records already selected by the first and second strokes, those records having the input last stroke, for which the Bu Shous are then displayed.

(2) The user selects the required Bu Shou by pressing appropriate keys. The software then selects from the character database those records having the selected Bu Shou code. If there are only a few characters with the selected Bu Shou, these are displayed and the user selects the required character.

7 (3) If there are too many characters from which to select, the user keys in the first, second and optionally the last strokes of Part 2 of the character to reduce the number of options. As each stroke is keyed in, the software selects a subset of the currently selected character records having the keyed stroke in the relevant position, and displays the characters of those records if there are not too many for display.

(4) The user selects the desired character from the displayed possible characters. This may be done before the first, second or last strokes are entered at stage (3), if the number of selected characters is few enough to be displayed.

Some characters cannot be divided easily. In this case, instead of carrying out steps (1) to (3) above, the user presses another predefined key (numeral '7' key in this case) to skip Bu Shou entry and then keys in the I st, 2nd and optionally the last strokes of the whole character. The handset then only selects, from those characters which cannot be divided easily, those characters having the input strokes. The characters which cannot be divided easily are indexed in the character database by first, second and last strokes. These characters may also be indexed by Bu Shou and first, second and last strokes of Part 2, according to the most normal division of these characters.

The above method allows characters to be entered using only rules which the user can easily follow. The partition rules are well-known, and no extra rules are required. Most users of Chinese can write the first two strokes in the correct order, but may get the order of the rest wTony. Keying in the last stroke is optional and only required for reducing the number of options.

This input method is suitable for everyone who has even a very basic knowledge of Chinese writing regardless of their education background and dialects. The probability of finding the correct Bu Shou and character is very high. Consequently. this reduces the frustration of Chinese input by occasional users. The input speed is not as fast as other more complicated

8 coding methods for professional typists, but is similar to or faster than Pinyin methods.

In one embodiment, numeric keys 2 to 6 keys are used to represent the five basic strokes. Effectively, a 5x5x5 coding space is available to cover the 247 possible Bu Shous. There are about 300 characters within the larger Bu Shou groups. Another 5x5x5 coding space is available to cover each group.

Therefore, the number of multiple options is much less than that of Pinyin. In fact, most of the multiple options are from 6 or fewer characters, which can fit on one screen of a mobile telephone handset. Hence, this method is particularly advantageous for use in mobile telephones, which have a restricted number of keys and display size. However, the method can also be applied to standard typewriter and computer keyboards. The assignment of strokes to different keys can be defined by the user.

The above method can be applied both to Simplified and Traditional Chinese character sets. For example, in Taiwan much emphasis is placed on learning the correct Bu Shou partition of Traditional characters, so the method can be easily used in Taiwan. Simplified Chinese teaching tends to emphasise Pinyin, but Bu Shou indexing is also used.

Once the desired character string has been entered by the user, it may be processed in any way appropriate for the input device. For example, on a mobile telephone the string may be transmitted as an SMS message, or entered in an address book or memo for later retrieval and display. On a computer. the string may be passed as an input to application software.

A more specific example will now be described with reference to Figures 3) and 4, which show the display D after steps (1) and (3) of the above method when the user keys in the last character for the word "arrive". The display has three lines which are used to indicate the progress of character entry, indicated as L 1 to L3. The first line L 1 shows the complete and partial characters entered so far, the second line L2 shows the strokes as theY are 9 entered by the user, and the third line L3 shows options for selection by the user. The stroke input mode is indicated by the letters 'SSB'.

(1) The user presses the '5' key twice and two vertical strokes are shown in the second line L2. In the third line L-3), the handset displays valid Bu Shou options beginning with two vertical strokes. The curved lines to the left of the characters indicate that they are not complete characters. The user can select between the Bu Shou options by pressing the or keys; the current selection is highlighted as shown in Fig.3.

(2) The user presses another key, such as 'YES', to confirm the selection and the selected Bu Shou is displayed on the first line L 1. All possible characters having the selected Bu Shou are displayed on the third line L3), as shown in Figure 4.

(3)) The user can make a selection now without keying in any more strokes because the desired character has already appeared. Otherwise, the user may key in the first, second and optionally the last strokes of the Part2 to reduce the number of options if there are too many. The display of possible characters is updated each time a stroke is input.

(4) The user selects the desired character, which is then processed as an input character.

In the above description, specific techniques for user input and display are described by way of example, but the present invention is not limited to such specific techniques. For example, input of strokes and selection of options may be driven by the user's voice or by drawing and pointing on a touch sensitive screen, since speech recognition of spoken or written Chinese characters is also difficult and may be simplified by the partition and partial stroke selection methods of the above invention.

As mentioned above, the invention is of particular utility in input devices having a restricted number of keys or a simple display, but may also be applied to devices having more complex keyboards and displays, such as general purpose computers.

It would be possible, though less desirable, to accept stroke input from the user in an order other than the 'first, second, last' order described above, for example 'first, second, third'. However, the use of the last stroke as a differentiator is particularly advantageous because it is easy for users to identify and often distinguishes between characters having the same first and second strokes.

Embodiments of the present invention may be applied to other character sets derived from Chinese characters, such as the Japanese Kanji character set.

11

Claims

A method of selecting as input a character from a predetermined character set, including:

a) receiving as distinct inputs from a user an indication of a first component of a desired character of said character set and an indication of a second component of a desired character of said character set; b) presenting to the user one or more characters selected from said character set, each displayed character having both said first and second components; and c) receiving a selection indication from the user indicating a subset of the characters presented at step b).

2. A method as claimed in claim 1, wherein the selection indication at step c) indicates a further component of the desired character.

1 3. A method as claimed in claim 1 or claim 2, wherein the selection indication at step c) indicates the desired character.

4. A method as claimed in any preceding claim. further including:

d) presenting to the user one or more characters selected from a second character set according to the selection indication by the user at step c). and e) receiving a further selection indication from the user indicating a subset of the characters indicated at step d).

5. A method as claimed in claim 4, wherein step e) includes: receiving as distinct inputs from the user an indication of a first component of a desired character from the second character set and an 12 indication of a second component of the desired character of the second character set.

6. A method as claimed in claim 5, wherein step e) further includes: receiving an indication of a further component of the desired character from the second character set.

7. A method as claimed in claim 5 or claim 6, including:

f) presenting to the user the subset of the characters displayed at step d), wherein the subset comprises one or more characters having the or each component indicated by the user.

8. A method as claimed in claim 7, including receiving a yet further selection indication from the user, indicating one of the characters presented at step f).

9. A method as claimed in claim 4, wherein the further selection indication at step e) indicates one of the characters presented at step d).

10. A method as claimed in any one of claims 1 to 3, wherein the predetermined character set is a Chinese character set and the first and second components are the first and second strokes of the desired character of the Chinese character set.

11. A method as claimed in claim 10 when dependent on claim 2, wherein the further component is the last stroke of the desired character of the Chinese character set.

13 12. A method as claimed in any one of claims 4 to 9, wherein the predetermined character set is a set of Bu Shou parts of a Chinese character set and the second character set is said Chinese character set.

1 '). A method as claimed in claim 12, wherein the first and second components of the predetermined character set are the first and second strokes of a desired Bu Shou part.

14. A method as claimed in claim 13 when dependent on claim 2, wherein the further component of the predetermined character set is the last stroke of the desired Bu Shou part.

15. A method as claimed in any one of claims 12 to 14 when dependent on claim 5, wherein the first and second components of the desired character from the second character set are the first and second strokes of a second part of the desired character separate from the Bu Shou part.

16. A method as claimed in any one of claims 12 to 15 when dependent on claim 6, wherein the further component of the desired character from the second character set is the last stroke of the desired character separate from the Bu Shou part.

17. A method of input for Chinese characters, including: a) receiving as input from a user separate indications of the first, second and last strokes of the desired Chinese character, or a part thereof; and b) selecting as input a Chinese character having the indicated first, second and last strokes.

14 A method of input as claimed in claim 17, wherein the part indicated at step a) is a Bu Shou part of the character.

19. A method as claimed in claim 18, wherein step b) includes:

c) receiving as input from a user an indication of the first and/or second strokes of an other part of the Chinese character other than the Bu Shou part, and d) selecting as input a Chinese character having the indicated first, second and last strokes of the Bu Shou part and the indicated first andlor second strokes of the other part of the Chinese character.

20. A method as claimed in claim 19, wherein step c) further includes receiving as input from the user an indication of the last stroke of the other part of the Chinese character; wherein the Chinese character selected as input includes the indicated last stroke of the other part.

21. A method of input for Chinese characters, including:

a) receiving as input from a user separate indications of the first and second strokes of a Bu Shou part of the desired character; b) receiving as input from the user separate indications of the first and second strokes of the remaining part of the desired character; and c) selecting as input a character having the strokes indicated at steps a) and b).

22). A method as claimed in claim 21, wherein step a) further includes receiving from the user an indication of the last stroke of the Bu Shou part of the desired character, and the character selected at step c) includes the indicated last stroke of the Bu Shou part.

23. A method as claimed in claim 21 or 22, wherein step b) includes receiving from the user an indication of the last stroke of the remaining part, and the character selected at step c) includes the indicated last stroke of the remaining part.

24. A method as claimed in any preceding claim, including, when the or each selection indication by the user indicates one specific character, processing that character as input.

25. An electronic input device arranged to perform the method of any preceding claim.

26. A portable data terminal arranged to perform the method of any one of claims 1 to 24.

27. Software arranged to perform the method of any one of claims 1 to 24 when executed by a suitably arranged electronic input device.

28. A digitally represented database of Chinese characters indexed by at least Bu Shou part, and first and second strokes of the other part.

29. A method substantially as herein described with reference to Figures 3) and 4 of the accompanying drawings.