WO2012152039A1

WO2012152039A1 - Method and device for determining candidate character in handwriting input

Info

Publication number: WO2012152039A1
Application number: PCT/CN2011/084849
Authority: WO
Inventors: 江桂凤
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-09-29
Filing date: 2011-12-28
Publication date: 2012-11-15
Also published as: CN102360265A; CN102360265B

Abstract

Disclosed are a method and a device for determining a candidate character in handwriting input, which are applied in the technical field of input to electronic apparatuses. The method comprises: determining, according to a radical, of a pre-input Chinese character, input by a user, radical information of the pre-input Chinese character; matching radical handwriting in the radical information with a stored personal handwriting sample, and if the matching is successful, acquiring a radical index number corresponding to the personal handwriting sample, the personal handwriting sample being determined according to handwriting of a radical input by the user and the personal handwriting sample being corresponding to a radical index number; and acquiring, according to the radical index number, a sequence of candidate characters comprising the radical, and determining a candidate character of the pre-input Chinese character from the candidate character sequence. In the method and the device of the present invention, by storing personal radical handwriting samples generated according to the user's personal handwriting habit, a character library is easily implemented and requires small storage space.

Description

Method and device for determining candidate word in handwriting input

TECHNICAL FIELD The present invention relates to the field of electronic device input technologies, and in particular, to a method and apparatus for determining a candidate word in handwriting input. Background technique

At present, data entry into an electronic terminal is mainly through two methods of keyboard input and handwriting input. Handwriting input is more in line with Chinese writing habits than keyboard input, so various fast handwriting input methods have emerged.

Chinese characters are a kind of block characters with a special shape structure. The order of Chinese characters can be divided into whole words, parts and strokes. Accordingly, the Chinese character handwriting input method can be roughly classified into three types: handwriting input method based on whole word recognition, handwriting input method based on part recognition, and handwriting input method based on stroke recognition. Among them, the handwriting input method based on whole word recognition is: the complete Chinese characters input by the user are matched by the identification system for sample or template, and the matching samples with the largest similarity and the smallest difference are used as the recognition result, and the input strokes are many and the recognition is complicated. The characteristics, recognition efficiency and accuracy are relatively low. The Chinese character handwriting input method based on the stroke sequence only considers the stroke itself and the sequence between strokes and strokes, regardless of the spatial structure information of the Chinese characters. Although the normative requirements for the writing of the user are not high, the recognition efficiency is relatively low. Handwriting input method based on component recognition By identifying the component input by the user and obtaining the candidate word containing the component, the number of strokes input by the user can be reduced, and the recognition efficiency and accuracy are higher than the other two handwriting input methods.

In the prior art, a fast Chinese character handwriting input method is provided, and the specific implementation manner thereof is: encoding a Chinese character according to a radical and a location to form a code font. Each field of the character field in the writing area is divided into 16 radicals. The user inputs a radical, and the encoded string consisting of the radical encoding and the radical encoding is obtained. The encoded string is encoded as a search key. The search in the font is satisfied All Chinese characters of the condition. This fast Chinese character handwriting input method can improve the handwriting input speed of Chinese characters to a certain extent. However, it applies Chinese characters according to the radicals and location information for encoding. There will be a complicated implementation of the coded font library, and the storage space is relatively large, and the user must be correct. In the small square area, writing radicals can find problems such as pre-entering Chinese characters.

For example, the encoding of the Chinese character "part" consists of three radical encodings and the location encoding of the positions of the three radicals. It is assumed that the user writes "standing" in the upper left corner of the writing area, according to the radical encoding containing the "standing" and the upper left. The code string of the corner location code can be queried from the code font library to obtain the candidate word sequence whose position is "right" in the upper left corner, and then find the "part". If the user writes "立" in the left half of the writing area, according to the radical encoding containing the "leaf" and the encoding code of the left half location encoding, the candidate word sequence of "part" cannot be queried from the encoding font. However, the above-mentioned Chinese character handwriting input method has strict requirements on the writing position. If the writing position of the radical is not correct, the Chinese character cannot be obtained; and in the portable terminal device, since the writing area is small, the user writing standardization is difficult to ensure. Therefore, there is a problem that the above method application has difficulty in input recognition. Summary of the invention

The invention provides a method and a device for determining a word to be selected in a handwriting input, which solves the problem that the handwriting input terminal in the prior art has difficulty in ensuring handwriting input recognition because the writing area is small and the user writing standardization is difficult to ensure.

The present invention provides a method for determining a word to be selected in a handwriting input. The method is applied to a smart device including a handwriting input device, including:

Determining the radical information of the pre-input Chinese character according to the radical input of the pre-entered Chinese character input by the user; matching the radical handwriting in the radical information with the stored personal handwriting sample, and if the matching is successful, obtaining the individual according to the a radical index number corresponding to the handwriting sample; wherein, the personal handwriting sample is determined by a radical handwriting input by a user;

Obtaining a sequence of candidate words including the radical according to the radical index number, and determining a candidate word of the pre-entered Chinese character from the sequence of candidate fonts. In the above solution, after determining the candidate word of the pre-entered Chinese character from the candidate word sequence, the method further includes:

Obtaining an input coordinate of the radical from the radical information, and determining a glyph structure of the pre-entered Chinese character according to the input coordinate;

According to the glyph structure, Chinese characters in the candidate word that are identical to the glyph structure are preferentially arranged.

In the above solution, after determining the candidate word of the pre-entered Chinese character from the candidate word sequence, the method further includes:

If the candidate word sequence includes N candidate words, and each candidate word of the N candidate words corresponds to a usage frequency value, according to the usage frequency value of the N candidate words The N candidate words are sorted, wherein the usage frequency value of the M-1th candidate word in the ranking is greater than or equal to the usage frequency value of the Mth candidate word in the ranking, and the M is greater than 2 or less than or An integer equal to N, resulting in an optimized sequence of candidate words.

In the above solution, after determining the to-be-selected word of the pre-entered Chinese character from the candidate word sequence, if receiving the first Chinese character stroke of the pre-input Chinese character except the radical, the method further includes: The first stroke of the remaining Chinese characters is prioritized, and the Chinese characters having the same strokes as the first stroke of the remaining Chinese characters are preferentially arranged to obtain an optimized sequence of the selected words.

In the above solution, the personal handwriting sample is determined by the radical handwriting input by the user, and includes: after receiving the radical input by the user, determining whether the radical is included in the stored personal handwriting sample, if included, displaying Outputting a sequence of to-be-selected words including the identified radicals; if not, determining the input radicals from the candidate radical sequence according to the input information of the user, and querying whether the input radicals have a radical sample, if If the input radical has a radical sample, the personal handwriting sample is used to update the radical handwriting sample; if the input radical does not have the radical handwriting sample, the personal handwriting sample is saved as the radical handwriting sample.

According to the above method, the present invention also provides a device for determining a word to be selected in handwriting input, the device The device includes: an input module, a radical determination module, and a candidate word determination module; wherein, the input module is configured to determine a radical information of the pre-input Chinese character according to the radical input of the pre-input Chinese character input by the user, and send the radical information to Radical determination module;

a radical determining module, configured to match a radical handwriting in the radical information sent by the input module with the stored personal handwriting sample, and if the matching is successful, obtaining a radical index number corresponding to the personal handwriting sample Sending a radical index number to the candidate word determining module; wherein, the personal handwriting sample is determined by a radical handwriting input by the user;

a candidate word determining module, configured to obtain, by the radical index number sent by the radical determining module, a sequence of to-be-selected words including the radical, and determine, from the candidate word sequence, the pre-entered Chinese character to be selected Select words.

In the above solution, the device further includes:

a glyph structure sorting module, configured to receive a candidate word sequence sent by the candidate word determining module and obtain the input coordinates of the radical in the radical information, and determine a glyph structure of the pre-input Chinese character according to the input coordinate According to the glyph structure, the Chinese characters in the candidate word sequence that are identical to the glyph structure are preferentially arranged;

Correspondingly, the candidate word determining module is further configured to send the candidate word sequence and the radical information to the font structure sorting module.

In the above solution, the apparatus further includes: a frequency ordering module, configured to receive the sorted candidate word sequence sent by the font structure sorting module, if the candidate word sequence includes N candidate words, and N candidates Each candidate word in the word corresponds to a usage frequency value, and the N candidate words are sorted according to the usage frequency value of the N candidate words, wherein the M-1 candidate words in the ranking The usage frequency value is greater than or equal to the usage frequency value of the Mth candidate word in the ranking, the M is an integer greater than 2 less than or equal to N, and the optimized sequenced candidate word sequence is sent to the candidate word to be determined. Correspondingly, the font structure sorting module is further configured to send the sorted candidate word sequence to the usage frequency sorting module; The candidate word determining module is further configured to receive a candidate word sequence after being sorted by using a frequency sorting module.

In the above solution, the device further includes: a remaining stroke sorting module, configured to receive a candidate word sequence sent by the candidate word determining module, and a remaining Chinese character first stroke of the pre-input Chinese character except the radical, according to the The first stroke of the remaining Chinese characters is prioritized, and the Chinese characters with the same strokes of the remaining Chinese characters are preferentially arranged, and the sequence of the selected candidate words is sent to the candidate word determination module;

Correspondingly, the candidate word determining module is further configured to send the candidate word sequence to the remaining stroke sorting module, and receive the candidate word sequence after the optimal sorting by the remaining stroke sorting module.

In the above solution, the device further includes: a personal handwriting sample determining module, configured to check whether the radical is included in the existing personal handwriting sample according to the radical handwriting sent by the radical determining module, and if included, the The first determining module sends a radical index number corresponding to the personal handwriting sample, and if not, whether the radical input sample exists in the input radical according to the radical handwriting, and if the input radical has a radical sample, Updating to the personal handwriting sample determined by the radical handwriting, if there is no radical handwriting sample in the input section, the radical handwriting in the radical information is saved as a personal handwriting sample and the radical number corresponding to the personal handwriting sample is added. ;

Correspondingly, the radical determining module is specifically configured to send a radical handwriting to the personal handwriting sample determination module, receive a radical index number corresponding to the personal handwriting sample sent by the personal handwriting sample determination module, and send the radical index number to The word selection module to be selected;

The candidate word determining module is configured to obtain, according to the radical index number sent by the radical determining module, a sequence of to-be selected words including the radical.

One or both of the above technical solutions have at least the following technical effects: Firstly, the method and device provided by the present invention store a sample of a personal handwriting generated according to a user's personal handwriting habit, and the font library is simple to implement and has a small storage space, and does not need to be The Chinese characters in the font are encoded, and only the font is indexed by the radicals for easy retrieval; in addition, since each user's handwriting is stable and clear The explicit personal characteristics, using the user's radical handwriting as a matching sample, can reduce the number of matches, improve the matching efficiency, and can improve the recognition efficiency and accuracy of the Chinese handwriting input method, reduce the word storage space; further because the user writes the radical The handwriting and the Chinese character radicals are established-corresponding relationship, which makes the user's writing more random and user-adaptive. DRAWINGS

1 is a flowchart of a method for determining a word to be selected in handwriting input according to an embodiment of the present invention; FIG. 2 is a flow chart of a method for implementing handwriting input by applying the method provided by the embodiment of the present invention;

3 is a schematic diagram of a user inputting a head according to an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of an apparatus for determining a word to be selected in handwriting input according to an embodiment of the present invention. detailed description

An embodiment of the present invention provides a method for determining a word to be selected in a handwriting input, the method comprising: determining, according to a radical of a pre-input Chinese character input by a user, a radical information of a pre-input Chinese character; The first handwriting is matched with the stored personal handwriting sample, and if the matching is successful, obtaining the radical index number corresponding to the personal handwriting sample; wherein the personal handwriting sample is determined by the radical handwriting input by the user; The first index number acquires a candidate word sequence including the radical, and the candidate word of the pre-entered Chinese character is determined from the candidate word sequence.

As shown in FIG. 1 , an embodiment of the present invention provides a method for determining a word to be selected in handwriting input, where the method is applied to a smart device including a handwriting input device, and the specific embodiments of the present invention are described below with reference to the accompanying drawings. Detailed description:

Step 101: Determine the radical information of the pre-entered Chinese character according to the radical of the pre-entered Chinese character input by the user.

Specifically, in the embodiment of the present invention, when the user inputs the Chinese character, according to the prompt of the system, the radical input of the pre-input Chinese character is first input, and when the user inputs the radical, the user needs to input the Chinese character according to the pre-input. The shape structure writes the radicals at the corresponding positions of the input device; the radicals of the input Chinese characters are saved as the radical information. For example, if you need to enter the word "part", you first enter "; and in the glyph structure of the "part", in the right half, so when you input the word "part", write 1" in the right half of the input device. .

Here, the radical information includes: a radical handwriting and input coordinates.

Step 102: Matching the radical handwriting in the radical information with the stored personal handwriting sample to obtain a radical index number corresponding to the personal handwriting sample.

Here, the personal handwriting sample is determined by the radical handwriting input by the user, and the personal handwriting sample can store different writing methods of the same radical, thereby improving the recognition of the input handwriting by the system. The personal handwriting sample can be the same for the user. The writing habits change and are updated instantly;

The radical index number is a radical index number of the middle of the font searched by the radical. After receiving the radical input by the user, determining the radical handwriting of the radical, and matching the existing personal handwriting sample with the radical handwriting, and returning the personal handwriting sample if matching the same personal handwriting sample Corresponding radical index number; if there is no match to the same personal handwriting sample, then according to the radical handwriting in the radical information, it is queried from the candidate radical sequence whether there is a matching radical sample, if any, Some radical handwriting samples are updated to the personal handwriting samples determined by the radical handwriting; if not, the radical handwriting in the radical information is saved as a personal handwriting sample and the radical index number corresponding to the personal handwriting sample is saved.

The radical index number corresponding to the personal handwriting sample is the number of the middle of the font searched by the radical, and the personal handwriting sample and the middle of the font are established by the radical number-corresponding relationship.

Step 103: Obtain a candidate word sequence including the radical according to the radical index number, and determine a candidate word of the pre-entered Chinese character from the candidate font sequence.

Here, the sequence of the selected words may be: Sorting the words to be selected according to the sorting condition Sequence of composition

The sorting conditions include the following:

(1) acquiring the input coordinates from the radical information, and determining a glyph structure of the pre-input Chinese characters according to the input coordinates; according to the glyph structure, the candidate characters are in the glyph structure Chinese characters are prioritized.

(2) According to the user's usage, each user has some common words, and the candidate words include N, and each of the candidate words has a usage frequency value. In this embodiment, the N candidate words are also sorted according to the usage frequency values of the N candidate words, where the usage frequency of the M-1 candidate words in the sorting is greater than or equal to the sorting. The frequency of use of the Mth candidate word, the M being an integer greater than 2 less than or equal to N.

(3) if receiving a user inputting the first stroke of the remaining Chinese characters other than the radicals of the pre-input Chinese characters, according to the first stroke of the remaining strokes, excluding the radicals from the first stroke and the first stroke Chinese characters with the same strokes are prioritized.

In a specific application, one or two sorting conditions may be selected to sort the selected words, or the above three sorting conditions may be freely combined and the selected words are sorted to form a candidate word sequence.

When the to-be-selected words are formed into a candidate word sequence according to the sorting condition, the sorting method may be selected according to the following judgment method:

If the coordinates of the user's writing radicals can well reflect the glyph structure of the pre-input Chinese characters, the Chinese character glyph structure is used to sort the selected character sequences. If the Chinese characters containing the radical are frequently used, the user frequency records are used to sort the sequences to be selected. If the user inputs the first stroke and then inputs the first stroke of the remaining Chinese characters, the remaining Chinese characters are used to sort the selected words.

As shown in FIG. 2, when a pre-entered Chinese character is a "part", a method for determining a candidate word in a handwriting input provided by an embodiment of the present invention includes:

Step 201: Receive a radical input by the user in the right half of the input device and record the radical Information.

The radical information is: a radical handwriting (as shown in Figure 3) and input coordinates.

Step 202: Match the radical handwriting in the radical information with the stored personal handwriting sample, and determine a radical index number corresponding to the personal handwriting sample.

Step 203: Obtain a sequence of candidate words containing "卩" according to the radical index number.

The sequence of to-be-selected words obtained in this embodiment may include Xl{Team Deng's suburbs accompanying ...).

Since the glyph structure of the "part" is a relatively obvious left and right structure, inputting "p" from the right half of the input device can further determine the structure of the pre-input Chinese character, and further includes:

Step 204: Obtain the input coordinate from the radical information, and determine a glyph structure of the pre-input Chinese character according to the input coordinate; according to the glyph structure, the font structure in the candidate word sequence is the glyph structure The Chinese characters are prioritized.

After further optimization, the order of the selected word sequences can be 1{Deng Suburbs' team accompaniment...}.

If the Chinese characters containing the radical are used frequently, then further include:

Step 205: Prioritize Chinese characters with high frequency of use in the candidate word sequence according to Chinese character usage frequency.

If it is further required to locate the pre-input Chinese characters more accurately, the method may further input, by the user, the first strokes of the remaining Chinese characters except the radicals of the pre-input Chinese characters, if the user inputs the remaining first strokes after inputting "P" ", ", then further includes:

Step 206: Align the Chinese characters of the to-be selected words with the same first strokes except the radicals according to the first stroke of the remaining strokes.

After further optimization, the sequence of the selected words can be arranged as 1{suburban with Dengdu team...}.

As shown in FIG. 4, according to the method shown in FIG. 1, the embodiment of the present invention further provides a handwriting input. The device for determining a word to be selected, the device comprising: an input module 401, a radical determination module 402, and a candidate word determination module 403;

The input module 401 is configured to determine a radical information of the pre-input Chinese character according to the radical input of the pre-input Chinese character input by the user, and send the radical information to the radical determining module 402;

The radical determining module 402 is configured to match the radical handwriting in the radical information sent by the input module 401 with the stored personal handwriting sample, and if the matching is successful, acquire the radical index corresponding to the personal handwriting sample. And sending the radical index number to the candidate word determining module 403; wherein the personal handwriting sample is determined by the radical handwriting input by the user, and the personal handwriting sample corresponds to the radical index number;

The candidate word determining module 403 is configured to obtain the candidate word sequence including the radical according to the radical index number sent by the radical determining module 402, and determine the pre-entered Chinese character from the candidate word sequence. The word to be selected.

Because the user inputs the radicals according to the glyph structure of the pre-input Chinese characters, the apparatus provided by the embodiment of the present invention further includes: a glyph structure sorting module 404, configured to receive the to-be-selected word determining module 403 Determining a font structure of the pre-input Chinese character according to the input coordinates; and, according to the font structure, prioritizing Chinese characters in the candidate word sequence that are identical to the glyph structure Correspondingly, the candidate word determining module 403 is further configured to send the to-be-selected word sequence and the radical information to the font structure sorting module 404.

In addition, if the user inputs the remaining word strokes after inputting the radicals, the remaining word strokes may be used to sort the selected word sequences. The apparatus provided by the embodiment of the present invention further includes: a remaining stroke ordering module 406, configured to receive a sequence of to-be-selected words sent by the candidate word determining module 403, and a first stroke of the remaining Chinese characters of the pre-input Chinese characters except the radicals And the candidate word determining module 403 is further configured to send the sorted candidate word sequence and the pre-input Chinese characters except the radicals to the remaining stroke sorting module 406 according to the remaining strokes. . The candidate word determining module 403 is further configured to receive the candidate word sequence that is optimized and sorted by the remaining stroke sorting module 406. Correspondingly, the remaining stroke sorting module 406 is further configured to optimize the sorted candidate words. The sequence is sent to the candidate word determination module 403.

The device provided by the embodiment of the present invention further includes:

The frequency ordering module 405 is configured to receive the sorted candidate word sequence sent by the font structure sorting module 404, if the candidate word sequence includes N candidate words, and each of the N candidate words Corresponding to the frequency of use, the N candidate words are sorted according to the frequency of use of the N candidate words, wherein the frequency of use of the M-1 candidate words in the sorting is greater than or And is equal to the use frequency value of the Mth candidate word in the sorting, wherein the M is an integer greater than 2 and less than or equal to N, and the sequence of the selected word to be selected is fed back to the candidate word determining module 403; The font structure is arranged in the candidate word determining module 403, and is further configured to receive the selected word sequence after being sorted by using the frequency sorting module 405.

Since the handwriting of each user has a stable and obvious personal feature, the user's radical handwriting is used as a matching sample, which can reduce the number of matching and improve the matching efficiency. Therefore, the apparatus provided by the embodiment of the present invention further includes:

The device provided by the embodiment of the present invention further includes: a personal handwriting sample determining module 407, configured to query, according to the radical handwriting sent by the radical determining module 402, whether the radical is included in the existing personal handwriting sample, if included And sending a radical index number corresponding to the personal handwriting sample to the radical determining module 402. If not, querying, according to the radical handwriting, whether the input radical has a radical sample, if the input radical already exists. The first handwriting sample is updated to the personal handwriting sample determined by the radical handwriting. If the input radical does not have the radical handwriting sample, the radical handwriting in the radical information is saved as a personal handwriting sample and the corresponding department is added. Correspondingly, the radical determining module 402 is specifically configured to send the personal handwriting sample determining module 407 The first handwriting is sent, and the radical index number corresponding to the personal handwriting sample sent by the personal handwriting sample determining module 407 is received.

The above one or more technical solutions in the embodiments of the present application have at least the following technical effects:

The method and apparatus provided by the embodiments of the present invention first store a sample of a personal handwriting generated according to a user's personal handwriting habit, and the font library is simple to implement and has a small storage space, and does not need to encode the Chinese characters in the font library, and only needs to press the font library according to the department. The first index is convenient for retrieval; in addition, since each user's handwriting has stable and obvious personal characteristics, using the user's radical handwriting as a matching sample can reduce the number of matches, improve matching efficiency, and improve the handwriting input method of Chinese characters. Identify the efficiency and accuracy, reduce the word storage space; further, because the user writes the handwriting and the Chinese character radicals to establish a corresponding relationship, the user is more casual and has the characteristics of user self-adaptation.

In addition, the method and apparatus provided by the embodiments of the present invention further apply the Chinese character glyph structure, the user frequency record, and the remaining word first stroke to optimize the sorting of the to-be-selected word sequences including the same radical, so that the sorted candidate words are more accurate. And can reflect the user's input habits. Other embodiments are obtained by the skilled person in accordance with the technical solution of the present invention, and are also within the scope of the technical innovation of the present invention.

It is apparent that those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Thus, it is intended that the present invention cover the modifications and modifications of the invention

Claims

Claim

A method for determining a word to be selected in a handwriting input, the method being applied to a smart device including a handwriting input device, comprising:

Determining the radical information of the pre-input Chinese character according to the radical input of the pre-entered Chinese character input by the user; matching the radical handwriting in the radical information with the stored personal handwriting sample, and if the matching is successful, acquiring the personal handwriting a radical index number corresponding to the sample; wherein, the personal handwriting sample is determined by a radical handwriting input by a user, and the personal handwriting sample corresponds to a radical index number;

And obtaining, by the radical index number, a candidate word sequence including the radical, and determining, from the candidate font sequence, the candidate word of the pre-entered Chinese character.

2. The method according to claim 1, wherein after determining the candidate word of the pre-entered Chinese character from the candidate word sequence, the method further comprises:

3. The method according to claim 1, wherein after determining the candidate word of the pre-entered Chinese character from the candidate word sequence, the method further comprises:

The method according to claim 1, wherein after determining the candidate word of the pre-input Chinese character from the candidate word sequence, if the pre-input Chinese character is received, except for the radical The first stroke of the remaining Chinese characters, the method also includes:

And according to the first stroke of the remaining Chinese characters, the Chinese characters in the candidate words that are identical to the first strokes of the remaining Chinese characters are preferentially arranged to obtain an optimized sequence of the selected words.

The method according to any one of claims 1 to 4, wherein the personal handwriting sample is determined by a radical handwriting input by a user, and includes:

After receiving the radical input by the user, determining whether the radical is included in the stored personal handwriting sample, if included, displaying the output candidate sequence containing the identified radical; if not, according to the user input The information determines an input radical from the candidate radical sequence, and queries whether the radical input sample exists in the input radical. If the radical input sample already exists in the input radical, the personal handwriting sample is used to update the radical writing Sample; if there is no radical sample in the input section, the personal handwriting sample is saved as a radical handwriting sample.

6. A device for determining a word to be selected in handwriting input, wherein the device comprises: an input module, a radical determination module, and a candidate word determination module; wherein

The input module is configured to determine a radical information of the pre-input Chinese character according to the radical input of the pre-input Chinese character input by the user, and send the radical information to the radical determination module;

a radical determining module, configured to match the radical handwriting in the radical information sent by the input module with the stored personal handwriting sample, and if the matching is successful, acquire the radical index number corresponding to the personal handwriting sample, Sending a radical index number to the candidate word determining module; wherein, the personal handwriting sample is determined by a radical handwriting input by a user, and the personal note sample corresponds to a radical index number;

7. The device of claim 6, wherein the device further comprises:

a font structure sorting module, configured to receive a candidate word sequence and a sentiment sent by the candidate word determining module Obtaining an input coordinate of the radical in the radical information, and determining a font structure of the pre-input Chinese character according to the input coordinate; according to the font structure, the same sequence of the font is selected in the candidate font sequence Chinese characters are prioritized;

The apparatus according to claim 6, wherein the apparatus further comprises: a frequency ordering module, configured to receive the sorted candidate word sequence sent by the font structure sorting module, if the candidate word sequence includes N candidate words, and each candidate word of the N candidate words corresponds to a usage frequency value, and the N candidate words are sorted according to the usage frequency value of the N candidate words, wherein The usage frequency value of the M-1th candidate word in the sorting is greater than or equal to the usage frequency value of the Mth candidate word in the sorting, and the M is an integer greater than 2 less than or equal to N, and the optimized sorting is performed. The candidate word sequence is sent to the candidate word determining module;

Correspondingly, the glyph structure sorting module is further configured to send the sorted candidate word sequence to the usage frequency sorting module;

The candidate word determining module is further configured to receive a sequence of candidate words after being sorted by using a frequency sorting module.

The apparatus of claim 6, further comprising: a remaining stroke ordering module, configured to receive a sequence of to-be-selected words sent by the candidate word determining module, and the pre-input Chinese characters except the part The first stroke of the remaining Chinese characters in the first foreign language is sent to the candidate word determining module according to the sequence of the selected words after the first optimization of the remaining Chinese characters;

The device according to any one of claims 6 to 9, wherein the device further comprises: a personal handwriting sample determining module, configured to: according to the radical handwriting sent by the radical determining module, to check whether the radical is included in the existing personal handwriting sample, and if included, send the personal handwriting sample corresponding to the radical determining module a radical index number, if not, according to the radical handwriting query whether the input radical has a radical sample of the radical, if the input radical has a radical sample, the individual is determined by the radical handwriting a handwriting sample, if the input portion does not have a radical handwriting sample, the radical handwriting in the radical information is saved as a personal handwriting sample and a radical number corresponding to the personal handwriting sample is added;