WO2018149250A1 - 带屏幕提示界面的汉字骨架码输入法及系统 - Google Patents
带屏幕提示界面的汉字骨架码输入法及系统 Download PDFInfo
- Publication number
- WO2018149250A1 WO2018149250A1 PCT/CN2018/000072 CN2018000072W WO2018149250A1 WO 2018149250 A1 WO2018149250 A1 WO 2018149250A1 CN 2018000072 W CN2018000072 W CN 2018000072W WO 2018149250 A1 WO2018149250 A1 WO 2018149250A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- code
- word
- skeleton
- key
- input
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
Definitions
- the invention relates to a "Chinese character skeleton code input method with screen prompt interface", based on the “character interface interaction and information processing system” and device which are extended on display.
- the shape code input method is rarely used because of the complicated rules for splitting, different rules, and difficulty in learning, including the five-stroke type:
- the four-corner number check method invented by Wang Yunwu can also be compiled into a Chinese character code input method.
- the code is obtained from the four corners of Chinese characters, and the results are obtained in turn, but there are about 10 kinds of symbol types. It is still difficult to remember and use them, and the amount of thinking about splitting and taking code is large, and the splitting boundary often occurs. Ambiguity.
- the present invention provides a Chinese character skeleton code input method with a very low thinking speed and high speed efficiency, with a screen prompt guiding interface, and a Chinese character character interface based on the Chinese character character interface.
- Information interaction processing system the same applies to Korean;
- a Chinese character skeleton code input method with a screen prompt interface characterized in that the input is completed by the following steps:
- each skeleton can be represented by a representative character of the invention, forming a skeleton type. Tables and their respective in-class character sets;
- the Chinese character skeleton code input method with a screen prompt interface according to claim 1.
- the utility model is characterized in that the internal components of the Chinese characters are classified and respectively coded to form a "skeleton code" of the Chinese characters, and the whole of the skeleton code is a whole word code, and a whole word code represents a class of the same skeleton, the whole word
- the code consists of its internal code, and the code of its internal code:
- Hybrids such as "multiply” to "ten” branches
- Left and right classes [left and right], [left middle right], [first left and right, then up and down - such as "antagonism”], [first left and right, then up and down, then left and right - such as " ⁇ "...;
- Up and down classes [up and down], [upper middle and lower], [first up, down, left and right - such as “goods are blue”], [first up, down, left and right, then up and down - such as "winning”...:
- the basic structure is mainly divided into: single body, heterozygous (such as "ten” type “multi-branch"), surrounded, partially enclosed, up and down, upper middle and lower ("three” type), left and right, left middle right (" Chuan” type);
- the symbols in the middle of the four sides of the layout of the 3 rows x 3 columns of the Jiugong structure are called "middle" "Dimensions"
- the four blocks of the 2 rows x 2 columns can be equivalent to the blocks at the four corners of the 3 rows x 3 columns:
- the words “no middle yuan” and “including middle yuan” refer to whether there are symbols in the middle points of the four sides of the structure of the nine palaces, and there is no word in the four rows and two columns in the four rows and the upper/lower sides.
- the point with the middle element is 2 lines x3 column total 6 areas of the word
- the left / right midpoint has the middle element is 3 lines x 2 column total 6 areas of the word up / down
- left / right midpoint has a middle element is 3
- the word x3 column has a total of 9 areas (this sentence "/" means "or”).
- any Chinese character forms a two-dimensional network. All strokes can be mapped on the corresponding line segments of the corresponding "grid model".
- the model is formed by a plurality of mutually parallel horizontal lines, vertical lines, and oblique lines, that is, a plurality of "field” characters spread in a two-dimensional direction, and a diagonal line;
- Each intersection in the model has multiple lines intersecting.
- generally four oblique diagonal lines can be added.
- a total of eight branches of the "meter" shape are actually mapped.
- the number of strokes intersecting at one point is generally no more than eight, but many of the slanting pens in Chinese characters are written in 30°, 60° and other non-standard angles, so it can be added or set at a non-45° angle. More lines, for example, add one every 22.5 degrees. Then there are 16 branches at each intersection;
- subnet may have a subnet composed of partial symbol combinations, and the component area exceeds the national standard component. Listed):
- the Chinese character "mother network” is dismantled in units of intersections. On the macro level, it can be split into two basic types: “closed enclosing class” subnet and “open cross class” subnet. Some Chinese character parts also contain the above two components, and an “opening and closing hybrid class” subnet with a combined structure may be selected;
- closed symbol The symbol for the closed enclosing subnet is called “closed symbol”, which mainly refers to the “mouth” without pen, and can also add a “block” with a pen on the corner to form a new supplementary scheme:
- Mouth A closed stroke group without a pen at the corner, also known as a “narrow block”.
- the fully enclosed mouth contains 4 corners (ie “mouth”), and the mouth surrounded by three contains 2 corners (such as “Oka, Fierce”) ), the mouth surrounded by two contains 1 corner (such as " ⁇ , ⁇ ”):
- Block There is a closed stroke group at the corner, also known as "generalized block”.
- the fully enclosed block contains 0 corners (such as “common” upper, “and”), 1 (rare), or 2 Corners (such as “and”), or 3 corners (such as “mother”); three enclosed blocks contain 1 (such as “force”) to 2 corners (such as "use”), and the two enclosed blocks contain 1 corner (such as "spoon, wide”);
- the triangle block the full enclosure contains 2 corners such as “again”, 2 surrounds one corner (such as ⁇ , "light” upper right), can be regarded as incomplete triangle; "The interior contains two incomplete two-enclosed “ ⁇ ” with a pen ",”, which is special for the "two pieces", and the "female” contains irregular quadrilateral blocks.
- open cross-class subnet The symbol for the open cross-class subnet is called “open symbol”, which mainly refers to “crossing”, “branch” and “multi-branch”. “Intersection” means that multiple strokes intersect at one intersection, and “twig” refers to multiple A linear combination of complete or incomplete "crossing” in one dimension, and a linear combination of multiple complete or incomplete "crossings" in a two-dimensional direction is "multi-branch”:
- Crossing If an intersection of branches in 1-8 directions (referred to as “crossing” or “meter”) is regarded as a complete “crossing", then some Chinese characters are incomplete “crossing", such as (5) Branch), ten (4 branches), Ding (3 branches), etc., "-" can be regarded as 2 branches special "crossing”:
- Block, intersection, and branch code symbols in addition to integrity and quantity, can also be subdivided according to their orientation. For example, the corners of the incomplete block "force, seven, and wei” are respectively oriented. Fang, this is the three enclosing area, but the method of "prioritizing the block” as follows can be attributed to the "two enclosing block” class:
- Symbol combination azimuth refers to the relative azimuth when the symbols are combined, and can be further classified according to the three types of “combination of the types of components”: the symbol of “only closed class symbols” Combination of symbols, a combination of symbols containing only open class symbols, and a combination of symbols that contain both open and closed symbols:
- the splitting rule is the same as the five-stroke input method, “scatter” refers to the natural disconnection between components, or large or small gaps, and seamlessly the intersection or connection between components or stroke groups, But the shape categories have distinctly different boundaries.
- the “key position” may refer to “screen virtual key position” and may present the “skeletal code” overall code representative word, and represent all or part of the same skeleton weight code word to be selected as needed to help
- the user prompts input can also be mapped to the physical key position in the same way, and can synchronously present the "skeletal code” overall code representative word and all or part of the representative of the skeleton weight code word on the screen as needed;
- the key layout can be adjusted according to needs, according to a certain regular layout, such as relative orientation, especially according to the principle that "the orientation of the code in the word and the mapping of the key mapping are basically the same", such as "central radiation structure” can be taken according to
- the code symbol especially the position of the "block” in the entire Chinese character, is placed in the corresponding key area of the "two-dimensional keyboard".
- the closed area of "shop, temple” is in the lower right corner of the word, and the word is also used.
- Mapped in the corresponding key area in the lower right corner of the keyboard (forming a holographic fractal structure);
- the internal component orientation of the whole word is visually and visually presented at one time, and the simple and ambiguous humanized prompt is free from active learning by the user, and is more comfortable, and the input is free to make the operation more efficient. (Because the component symbols eventually form a whole-word coded "symbol", they are finally mapped to different locations of the keyboard according to different results, thereby effectively combining with the computer hardware, which embodies a beneficial technical effect).
- the features 2 and 4 are cross-classified according to claim 1, and the corresponding assignment and coding are respectively performed according to the characteristics of the size categories of the various symbol categories and the orientation structures, including:
- Azimuth basic code can also be used as a simple 10-key keypad "four corners to find blocks" or "four-corner angle" shape code input method, Chinese character encoding code length is indefinite length, a total of 10 parts code, available computer keyboard
- the 9 numeric keys of the 3 rows and 3 columns above indicate that one key corresponds to one component, and the key name corresponding to the component is the code of the component, and one component (narrow or broad) is selected from the group component component constituting the Chinese character.
- “block” can be based on the principle that the orientation of the code-taking and keystroke is basically the same: according to the position of the selected code "block” in the entire Chinese character, it is mapped to the corresponding 3 rows and 3 columns.
- the corresponding position of the "Nine Palaces" two-dimensional structure keyboard, the mapping relationship between the components and the corresponding keys of the computer keyboard are respectively -
- the second line the 4-key mapping is located in the block of the "left middle ( ⁇ ) part" in the whole word, the 5 key is the “medium ( ) part”, and the 6 key is the "right middle ( ⁇ ) part",
- the third line: 7-key mapping is located in the whole world "lower left
- the section contains the block, the 8 key is the “lower ( ⁇ ) part", and the 9th key is the "lower right” unit",
- Each Chinese character (except for Chinese characters without blocks) must be split out of the corresponding coding component - "block", which can take the four corner "corner block” first, and then the middle area "edge block” of each side.
- the sequence of numeric keys corresponding to the "block” component that is sequentially removed is the code of the word, and the code of the word can be obtained by inputting the code of the word on the computer keyboard (generally there are a small number of duplicate codes);
- a "block” does not contain the word, and the key is set, or incorporated into a certain key, as the alternative is set to the fourth line "0" key;
- “Upper left corner – right lower corner with diagonal block” and “upper right corner – lower left corner with diagonal block” can also be set with another key, such as the fourth line, or merge into a certain key, such as 5 or 9 key
- Cong 14769 (or a brief take 4769), waist 14766 (or a brief take 4766 (),
- the characters of the upper and lower structures do not take the 4, 5, and 6 keys because they lack components located in the middle of the horizontal direction.
- the left half of the " ⁇ " is integrated, and the upper half is divided into upper and lower parts, and is separated (ie, disconnected, with obvious or minute gaps).
- the left half of the "eye” is integrated, and the upper half is divided into upper and lower parts, and connected (the “day” is seamlessly connected with the lower stroke group).
- closed class symbols can be connected to open class symbols, such as ⁇ ⁇ , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,
- the branch or the opening and closing combination type of two surrounding symbols such as the corpse, the internal (via the gap) contains the content: such as "cancer", in the Jiugong way to take the code, the code position is limited, the priority to take up the area Compared with the large symbol, the two ports in the middle horizontal direction are ignored, and the block code is 69, and the branch is 713;
- the keyboard of the Jiugong is simple, but the capacity is limited. Some of the more complicated simplified Chinese characters need to be ignored. It is impossible to accurately and completely cover the codes.
- the "azimuth basic code” only takes the closed symbol “block”, which can roughly reduce the re-code range of the Chinese character encoding result, which ignores the code acquisition of the open symbol, and can also take the code together, and can further detail according to the details.
- Feature subdivision coding, expansion information such as the block in the upper right corner of the word "Cheng, product, ⁇ ", etc., the left and right parts of the components are open symbols, the subcategories are different, and some are intersecting, some are branches
- the "character component in-code information database” is abbreviated as “partial code information library” (the nine-square block is equivalent to an input code, and it is equivalent to an extended outer code, or an abstract glyph font code):
- each interior of the nine-square structure (called the mother's palace) can be re-inlaid, or a small nine-square grid (called the uterus) can be separated, and the symbols in the original mother grid are grouped according to the intersection point. After dismantling, it is mapped into each uterus of different orientations, and each of the uteri can contain a "meter-shaped" line segment.
- the mother's palace a small nine-square grid
- Method 2 Encoding by intersection point and boundary: splitting the mesh (both child and parent mesh) into multiple intersection points and coding separately;
- Method three coding in the square and the boundary: splitting the grid (both child and parent grid) into multiple regions, respectively coded;
- Method 4 coded separately by category, divided into closed symbol "border” and open symbol “m”, respectively assigned and encoded;
- Intermittent information mainly refers to the structure of more than one maternal palace, such as the "ear” of the above “photograph” accounted for two or three palaces, the same as the internal division of the uterus, For example, if the "factory” in the "Liao” accounts for 74,231 squares in parallel, this is also an important feature, and the key positions and corresponding codes can be classified and divided accordingly.
- the 123 districts are integrated into one, the 456 districts are integrated, and the 7th and 9th districts are separated.
- the distance between the gaps or the gaps between the components is not obvious or connected, and it needs to be separated by category. It is the same as Chinese characters, and it is also preferred.
- the code mapping is performed, and then the symbol size categories in the newly divided regions are subdivided;
- splitting rules can be formulated, only the disconnection (the upper and lower parts of the national standard “ ⁇ ” are disconnected, but can also be dismantled) and the large categories, without the sub-classes, basically match the connection between the parts. ,
- the upper right corner of the Jiugong of “Cheng” is a “mouth”, and the orientation of the coded area is determined as “upper right corner”.
- the content of the coded information is “the type is mouth, the quantity is one”, and the lower right corner is an open “ Branch, containing 3 intersections, the number of branches is 3, 4, 3,
- the mother cell contains an open symbol-horizontal " ⁇ ", and the uterus compartment code is 46.
- the mother cell contains an open symbol - "dry”, and the uterus location code is 28,
- the mother cell contains an open symbol - " ⁇ ", and the uterus compartment code is 46.
- the mother cell contains a closed symbol - " ⁇ ", and the uterus compartment code is 05 (0 is the block, 5 is its position).
- the uterine compartment is coded as 08 (0 for block, 5 for its position),
- each Chinese character can have an accurate and unique coding value
- the processing module converts the corresponding "character code information library" into a corresponding character inner code and machine code that needs to be hexadecimal, thereby displaying the Chinese character result;
- the code encoding order of the "partial code information library” is read and reversed: the actual use of the internal code processing for the internal recognition operation of the computing device such as a computer, the data of the word is taken, and the component symbols are taken.
- the code can be carried out in a plurality of steps according to a certain regularity, for example, the first angle is behind, the first is outside, the other is from left to right, from top to bottom, or the first four corners, the other is followed by clockwise and progressive; It is also possible to formulate a corresponding special order for several overall structures formed by symbol disconnection, and the left and right Chinese characters are first left upper, lower left lower, then upper right, then lower right, and the surrounding Chinese characters are first and last;
- xxxxxxxx can represent a computer-readable Chinese character internal code, and the corresponding word set can be adjusted:
- the member accessing it takes the form of "structure variable. member”; for the structure pointer, accessing the member of the structure variable pointed to by it, takes the form of "structure pointer -> member”.
- the method of the present invention splits and takes code, and does not cause boundary ambiguity.
- the two words of the word "Fen” are placed on the two corners of the field, and the computer ocr is not easy to be separately recognized, the method can be avoided, and the skeleton can be avoided.
- the representative word can be accurate to the actual details of the Chinese character such as the pen, the non-existent strokes and pens, and the intersection points and inflection points can be highlighted by a circle, etc., and the standard information can be provided to form a new type that can be scanned and accurately identified. "Dimensional code", see [ Figure 8]
- the user selects any area in the Jiugong maternal grid, and after the processing module receives the selection, reads the skeleton of the area component, and re-codes the information, and causes the display module to output the display as needed ⁇ the user selects the specific skeleton of the area or the specific A certain component, and query ⁇ computer query enumerates one or more types of skeleton whole words that may contain it, and prompts for selection, such as “pound” word. After the user selects the upper right corner area, the upper right corner can be queried with “ ⁇ ”.
- the weight code of several skeletons if necessary, can also specifically select the components in the upper right corner, or Feng...
- the feature information of the parts is divided into categories according to the non-glyph attribute information, that is, the parts themselves are some kind of radicals, the side, the homonym, the near meaning, etc. Construct a kind of information and relationship information between components in different skeletons and unified skeletons, and find and process the processing module;
- Typical dimensions such as: meaning (such as “an industry”, “trademark name”), input background scene, common level, attribute, part of speech (such as “verb” or “noun”%), scope (such as “homophone” ");
- the module of the homophone prompt interface can also be connected to external pinyin input methods such as Sogou, Google, qq, Microsoft, etc., to obtain the input result thereof, and display the same method and other operations and functions according to the method of the present invention;
- the software accepts the words that the user input will not be split or cannot be found, and is highlighted in the interface to indicate its location;
- the rule base can be set, which is different from the font library and the thesaurus, and can be processed by the processing module according to the logic rules, including multiple words. Combine processes, structures, results, and transfer to display module display; similar functions are intelligent semantic algorithm modules;
- the Chinese character skeleton code input method with a screen prompt interface according to any one of claims 1 to 4, wherein the Chinese character scheme is completely applicable to Korean characters and Korean characters because only Korean vowel consonants are used.
- the vowel consonants in Korean are classified as follows. According to their size, they can be mapped to different keyboard keys and key names:
- the representative words of the same skeleton can be stacked and stacked according to their skeletons.
- different formats such as line segments of different thicknesses
- It can be rendered with dynamic slides and inter-frame gradient animations, so that users can be prompted in a small display area - after the user clicks or selects a certain skeleton, expands or enlarges the "like," under the skeleton category.
- the layout and presentation are arranged side by side in a certain order, and the user directly selects or clicks and clicks the key.
- the first part and the tail part of the word can be taken in the order of left first, second right, first up and then down, and each line is aligned according to the first part, and each column is aligned by the tail part; wherein, the same can be followed.
- Cross-classification, alignment rows or columns of attribute features such as radicals and adjacent sounds can be arranged in the same way in the corresponding two-dimensional array and database structure to facilitate retrieval operations;
- the processing unit When displayed, the processing unit also marks and highlights the format in which the display content and the object are distinguished:
- the “format, style” refers to different shapes, patterns, and scaling, thickness, coloring, line type, grayscale, shadow, etc., for the coded symbols, especially as the most conspicuous "block” in the visual,
- the key name, "corner point”, “intersection point”, “end point”, “eight directions of the five strokes of the horizontal and vertical folds” are marked differently to form a contrast;
- a connection can be displayed to indicate.
- a Chinese character skeleton code input system with a screen prompt interface characterized in that
- the receiving module can acquire keyboard operations from the user, character input of other input methods, and even operation instructions of other programs or actions and information of the connected object;
- the processing module reads, retrieves, and calls the corresponding input result according to the key operation obtained from the accepting module, and sends the corresponding input result to the display module display, and can also compare the information of the storage module, and perform operations to determine the associated and special display words.
- the key position (or coordinates) of the word, as well as the highlighted format, the connection between the words, and the like, generate the result of the interactive interface change, and output to the interface for display:
- a storage module configured to access the code information database, the word library, the corpus, and the recode result coding table, and the word association database, the intelligent algorithm, the logic rule, and the semantic grammar rule;
- a display module configured to display the input method prompt interface of the present invention, and provide a text input prompt of the character structured layout to the user to perform the input by using the input device;
- connection module can connect the characters on each key position in the interface to the calculation object, and perform bidirectional transmission and acquisition.
- connection object of the connection module is:
- the various "words" in the interface represent the corresponding computer instructions or commands that are related to the meaning of the word. They are collectively referred to as the computing object. Clicking on a word can execute or run the computing object of the character link. Similarly, in the interface Each character can be directed to receive operations or information from related others, external programs, and the objects that can be interacted or connected are:
- URL links programming commands, data objects, database models, mesh data structures, modules, containers, strings, structures, unions, arrays, function variables, controls, source code, compilers, underlying machine code instructions, mathematical models , computing flow, docking into the corresponding computer underlying compilation, machine code, instructions, commands, code, statements, corresponding objects, dynamic link library, application interface API, services, modules, related links, addresses, procedures, specific content,
- the world-wide external standards can be established or performed according to the interface object database in the system.
- the word “room” is linked to the website of SouFun.com, and the information is exchanged with the website, or the execution of specific instructions is connected to each other; click the word “song” to execute the command or program for playing the song:
- An input device with a screen prompt interface characterized in that the input method according to any one of the above 1 to 7 can be implemented by software or firmware in a general-purpose computer, an embedded computer, a single-chip microcomputer, or other device, mainly include
- the keyboard operation receiving unit receives a key operation signal of the physical keyboard, or receives a key operation signal for the virtual keyboard, such as a mouse click or a touch signal detected by the touch screen, the keyboard may refer to a physical keyboard, or may be the same
- the utility model includes a virtual keyboard system with the same structural layout and basic correspondence, including an electronic device screen keyboard, a touch screen screen keyboard, a teaching whiteboard keyboard, a projection virtual keyboard, a mouse, a finger, a touch screen, a light pen, a sensing device, and a brain-computer interface. , viewpoint tracking device, etc.;
- the processing unit is configured to query and retrieve the coding table according to the input signal, generate corresponding prompt information, and output the result for display and display by the display module;
- a storage unit configured to access the various types of component coding information libraries, word banks, corpora and re-code result coding tables, and word association databases, intelligent algorithms, logic rules, and semantic grammar rules
- the display unit is configured to display the input method prompt interface of the present invention, and provide a text input prompt of the character structured layout to the user for performing the input by using the input device.
- a networked input terminal with a screen prompt interface comprising: the device for implementing intelligent association in an input method according to any one of the above 1-8;
- the invention comprises a memory, a processor and a computer program stored on the memory and operable on the processor, the processor executing the computer program to implement the steps of any one of claims 1 to 9.
- a computer readable storage medium having stored thereon a computer program, wherein the program is executed by a processor to implement the steps in the vocabulary recommendation method according to any one of claims 1 to 5.
- the first part is divided into “rows”, then the tail parts are divided into “columns”, that is, the same words on the upper or left part are aligned on the same line, and then the same words are aligned on the lower or right part.
- rows the same words on the upper or left part are aligned on the same line, and then the same words are aligned on the lower or right part.
- the representation of the border is of the same kind, and the location of other classes is occupied because the table is not large enough.
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Document Processing Apparatus (AREA)
- Input From Keyboards Or The Like (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
一种带屏幕提示界面的汉字骨架码输入法,主要步骤包括:1)根据汉字的字形特征分为相应几种骨架类别,并分别将骨架相同的汉字聚类,并可按一定的规则排列,每种骨架可用一个代表字代表,形成骨架种类表和各自的类内字符集;2)由计算机显示各种骨架的代表字,及各自的全部或部分同骨架字,并于不同的"虚拟键位"区域上显示提示界面;3)使用者选择需要的骨架类别的代表字或类内具体某个字进行输入。提供了一种思考量极低,速度效率却很高的,带屏幕提示引导界面的汉字骨架码输入法。
Description
本发明涉及一种“带屏幕提示界面的汉字骨架码输入法”,基于它面延展出的“字符界面交互及信息处理系统”及装置.
目前已有的各种拼音输入法,均存在大量同音字词重码、需要再次选择的弊病;
而形码输入法因为拆字规则复杂、规则不一、难以学会,包括五笔字型,都已经很少有人使用:
王云五发明的四角号码检字法,其实也可编制成一种汉字编码输入法程序。从汉字的四角取码,依次输入就可得到结果,但其码元种类有约10种,记住和使用它们仍有难度,而且拆分取码的思考量较大、还常出现拆分界限的歧义。
发明内容:
本发明为了克服上述不足,提供了一种思考量极低、速度效率却很高的、带屏幕提示引导界面的汉字骨架码输入法,及在此基础上延展成的一种基于汉字字符界面的信息交互处理系统,同理适用于韩文;
技术方案:
实现上述目的所采用的计算机汉字输入法技术方案:
(一)输入法部分
1.一种带屏幕提示界面的汉字骨架码输入法,其特征在于,由以下步骤完成输入:
1)根据汉字的字形特征分为相应几种骨架类别、并分别将骨架相同的汉字聚类、并可按一定的规则排列,每种骨架可用一个本发明自造的代表字代表,形成骨架种类表和各自的类内字符集;
2)由计算机显示各种骨架的代表字、及各自的全部或部分同骨架字,并于不同的“虚拟键位”区域上显示提示界面;
3)使用者选择需要的骨架类别的代表字或类内具体某个字进行输入。
2.根据权利要求1所述的带屏幕提示界面的汉字骨架码输入法。其特征在于,对汉字内部成分进行分类、并分别取码,组成所述各汉字的“骨架码”,骨架码的整体即整字编码,一个整字编码代表同骨架的一类字,整字编码由其内部编码组成,其内部编码的取码:
1)可只按总体结构特征来分类——
方式一,按总体组合次序:
独体类:如“米”;
杂合类:如“乘”为“十”型枝;
左右类:[左右]、[左中右],[先左右、后上下-如“拮”]、[先左右、后上下、再左右-如“櫭檞”…;
上下类:[上下]、[上中下],[先上下、后左右-如“品皆蓝”]、[先上下、后左右、再上下-如“藉赢”…:
纵横类:[上下左右-如“解”],[上(中)下左(中)右-如“繲、器”…;
方式二,按总体组合次数:
a.基本结构,主要分成:独体,杂合(如含“十”型“多枝”),包围、部分包围,上下、上中下(“三”型),左右、左中右(“川”型);
b.混合结构,先分左(中)右、后分上(中)下,先分上(中)下、后分左(中>右(≈“品”型),及各自分别含包围、半包围(≈“L”型);
多次混合,如先分左右、后分上下、再分左右(≈含“品”型),先分上下、后分左右、再分上下(≈含“品”型),及各自分别含包围、半包围(≈含“L”型):
方式三,按总体部件数量,大致分2行x2列、3行x3列的部件结构、及相应键位布局,在九宫结构的3行x3列的布局的四边中点的码元称为“中元”,2行x2列的四个区块尤其可相当于3行x3列中的四角处的区块:
本发明附图中所说“无中元、“含中元”类,指在九宫结构的四边中点是否有码元,都没有的是2 行x2列共4区的字、上/下边中点有中元的是2行x3列共6区的字、左/右边中点有中元的是3行x2列共6区的字上/下和左/右边中点有中元的是3行x3列共9区的字(本句“/”表示“或”)。
2)可只按码元种类特征来分类——
汉字笔划只有横坚撇捺折五种,除了“一”字之外,任何一个汉字均构成二维的网状,所含所有笔划均可映射在相应结构的“网格模型”的相应线段上,该模型由多条相互平行的横线、竖线、斜线交织而成,即二维方向铺开的多个“田”字格、外加对角线组成;
模型中的每个交点均有多条线相交,除了横线x轴、竖线y轴之外,一般可加上4条斜对角线,共8条呈“米”形的分支,实际映射汉字时,相交于一点的笔划数量也一般不超过8条,但是汉字中的不少斜笔写成30°、60°等非标准角度更加美观,所以还可加上或按非45°夹角设置更多条线,如,以每22.5度增加一条。则每个交点共有16条分支;
若以一个整字称为母网、则取出部分区域的笔画组(即汉字部件、或其再组合)称为“子网(可能有部分码元组合组成的子网,成分面积超过国家标准部件所列)”:
以交点为单位拆解汉字“母网”,宏观上可只拆分为两类基本结构:“封闭包围类”子网、“开放交叉类”子网。有些汉字部件兼含上述两种成分,可选再设组合结构的“开闭混合类”子网;
大类:封闭码元{[口][块]},开放码元{[交][枝][多枝]},
——开放和封闭的区分标准是:多个交点之间是否有笔划首尾相连、组成了的“口”形封闭笔画组区域(或称封闭的部件),有则为封闭结构、无则为开放结构;
a.对封闭包围类子网所取码元称“封闭码元”,主要指无须笔的“口”,亦可增加角上有须笔的“块”构成新的增补方案:
口:拐角处无须笔的封闭型笔画组,又称“狭义的块”,全包围的口含有4个拐角(即“口”),三包围的口含有2个拐角(如“冈、凶”),二包围的口含有1个拐角(如“匕、习”):
块:拐角处有须笔的封闭型笔画组,又称“广义的块”,全包围的块含有0个拐角(如“共”上部、“并”),1个(少见)、或2个拐角(如“且”)、或3个拐角(如“母”);三包围的块含有1(如“力”)到2个拐角(如“用”),二包围的块含有1个拐角(如“勺、广”);
——另外,须有时候是斜笔(不多见,如“井”的左下角的“丿”、“身”右下角的“丿”);
三角△:在现代汉字中很少使用这个部件,在甲骨文中倒是常出现,“了”上部为不完整二包围的“△”,“乙”内部含有两个不完整二包围的“△”:
及可选的,三角块:全包围含2个拐角如“又”,2包围含有1个拐角1个(如厶、“轻”的右上部)、可视为不完整的三角块;“幺”内部含两个不完整二包围的“△”且带有一个须笔“、”,归“两块”类特殊的,“女”是含有不规则四边形块。
b,对开放交叉类子网所取码元称“开放码元”,主要指“交”、“枝”“多枝”,“交”指多笔划相交于一个交点,“枝”指多个完整或不完整的“交”在一维方向的线性组合,多个完整或不完整的“交”在二维方向的线性组合则为“多枝”:
交:若以1-8个方向的分支的一个交点(简称“交”或“米”)视为完整的一个“交”,则有些汉字部件是分支不完整的“交”,如才(5分支)、十(4分支)、丁(3分支)等,“-”可视为2分支特殊“交”:
枝和多枝:多个完整或不完整的“交”在一维方向的线性组合,如“丰、干上”等,称为开放的一个“枝”,多个完整或不完整的“十”在二维方向的超线性组合,如“止”等,且所含3各交点间呈
形结构,是一个开放类的“2枝”,同理,“寸乍生失”同是
骨架的“2枝”,“电虫”的下半部则为
骨架的“2枝",(说明书附图中各“止”区的内容,清楚的展现了“2枝”不同旋转朝向的情况)而“不家衣”是“介≈n”型骨架的左半部【见说明书附图2】:
*它们均可按以下细节特征再细分小类:完整度、朝向、绝对位置、内含下级码元个数(“枝”中所含的交点数目、“交”的每交点上的分支数、“块”的须笔数、块内分块的数量及所含内容):
块的不同完整度,如“卫”=“里”的部分,只需“”补齐所含另外两个边,另外“专”同理:
——它们做汉字中的成分时,并不一定在整个字的外边框上,如“指”的右上部1/4区域中,角在该区域中的左下方、而非在整字区域的右上方,同理“萨”右下角的“厂”的拐角在右下区域中的左上角;
*单个汉字内部的不同成分,如笔划、笔画组(或称部件)之间、不同子网之间,均可分为相离、相连、相交的方位关系。
3)可只按码元组合特征来分类——
a.码元组合方位——指划分码元间组合时的相对方位,并可再按其中“成分的种类的组合”的三类情况交叉分类:“只含封闭类码元”类的码元组合、“只含开放类码元”类的码元组合、及“兼含有开放、封闭类码元”的码元组合:
——四角处部件相离散的“∷”型字很少,如“能”,可并入◇类,另有“筏筷符簸簿符”可并入品类中“上合下分”的“∴”细类;
——“十”型“多枝”,简体汉字很少、故此表中为节省空间,暂时并入“含L”类;
c.组合次序,
d.交连程度,是离断/相连/相交,
4)同理,可对上述特征进一步交叉分类,如上述特征1、3交叉分类,主要分成:
上述“按总体结构特征来分类”一般为汉字结构为部件之间的有缝隙、离散的相互方位情况,适用“能散不连”的拆分界限规则:
上述“码元种类特征、码元组合特征来分类”,则取码时有可能拆开一个内部成分无缝相交连的部件,如“里”中“田”与“土”相连于一点,适用“能连不交”的拆分界限规则;
——即,拆分规则与五笔字型输入法同理,“散”指部件之间的自然离断、或大或小的缝隙,无缝的则是部件或笔画组之间相交或连接、但是形状类别有明显不同而划分的界限。
*“骨架码”内部编码的拆分和映射、输入:将整个汉字按内部离断缝隙(含各取码区域“宫”间、各取码区域“宫”内的离断)、码元的类别不同(主要指对相交连的部件的再划分)为界限,“能散不连、能连不交”的进行拆分后,骨架码编码相同的汉字聚类映射于同一“键位”(形成了相应细类的“汉字子集”),由用户选取、进行输入:
所述“键位”,可以指“屏幕虚拟键位”上、并可呈现该“骨架码”整体编码代表字,及据需要呈现其代表的、全部或部分同骨架重码字供选择,来帮助用户提示输入;还可以同理映射于实体键位上,并可据需要在屏幕上同步呈现该“骨架码”整体编码代表字及其代表的全部或部分同骨架重码字供选择;
所述键位布局可以据需要调整,按一定规律布局,如相对方位,尤其可按“在字中取码和键位映射的方位基本一致原则”来布局,如“中心放射结构”可按照取码码元尤其是“块”在整个汉字中所处位置,放置在“二维键盘”的相应键位区域,比如“店、庙”的封闭区域在字的右下角,则也将该类字映射在键盘中右下角相应键位区域里((形成一种全息分形的结构);
——“中心反向放射结构”同理,是将相反方位的各键对调;
同理,还可以按码元种类来分区设置键位布局,或根据个性喜好任意设置键位布局;
——详细按码元种类分类设键、并按取码和映射的方案例,又见【附图3】
——不管以什么规则、界限来拆分,不管被映射在不同键区,汉字中的笔划、笔画组即部件,不外乎可取上述两大类码元;
将整字的内部成分方位一次性直观可视化呈现,其简单模糊的人性化提示又免于用户主动学习,更令人舒适,其免于输入则令操作使用的速度效率更高。(由于部件码元最终组成整字编码“码元”,最终按不同结果映射于键盘的不同区位,从而有效的和计算机硬件这一工具结合,体现了有益的技术效果)。
3.根据权利要求1和2所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,所述各“骨架”内部取码、拆分和映射后,可进行编码,并录入“字内部件编码数据库”
优选方案,按述权利要求1所述特征2和4交叉分类,按照各种码元类别、方位结构的大小类别的特征,分别进行相应的赋值和编码,其中包括:
1)方位基本码,也可作为一种简单的10键小键盘“四角找块”或“四角找角”形码输入法,汉字编码码长为不定长,共10个部件代码,可用计算机键盘上的3行x3列的9个数字键表示,一键对应一种部件,部件对应的键名即为该部件的代码,从构成汉字的组字部件中选取一种部件(狭义或广义的)“块”作为对汉字进行编码的部件,可按“取码和击键的方位基本一致”原则:根据选取的码“块”在整个汉字中所处位置,映射到相应3行x3列的“九宫”二维结构键盘的相应位置,所述该部件与计算机键盘相应键位的映射关系分别为——
第二行:4键映射位于整字中“左中(←)部”含有的块,5键则为“中( )部”,6键则为“右中(→)部”,
——即将单字内部划分成九宫结构的9部分,每部分中编码部件与相应九宫方位的9个键位一一对应映射;
每个汉字(不含有块的汉字除外)都要拆分出相应的编码部件——“块”,可先取四角“角块”、其次再取各边的中间区域“边块”,取完为止,依次拆出的“块”部件对应的数字键序列即为该字的编码,在计算机键盘上输入该字的编码就可得到该字的所属的细类类型(一般都有少量重码);
——“一个“块”也不含”的字,另外单设键位、或并入某个键位,如可选方案设于第四行“0”键;
——“左上角-右下角对角含有块”、“右上角-左下角对角含有块”也可另外单设键位、如第四行,或并入某个键位,如5或9键
*取码(优先度)规则,尤其适用于码位所限、不能全部取码时,:
优先对占整字面积比例大的部件取码,依次是贯穿整字、或字的一半、1/3、1/4、1/6、1/8、1/16……的部件,如“品”内上部的被拉长的“口”;
一个取码区域(即宫格)中同时有开放、封闭码元时,优先取封闭码元,忽略多出的开放码元;如“镑”中宝盖头下方的T被忽略【见附图6】
例:形如下面的键盘
聪14769(或简略取法4769),腰14766(或简略取法4766(),
廖5,固5,
铜369(或6),侗369(或6),桶39。
——可选“打乱、改变输入编码的先后次序”来输入容错码,如“桶”71=17。
另外,说明书附图中也列出了几个拆分取码实例【见附图11、18、19、30】
左右结构的字,因比左中右结构缺少位于纵向中部的部件,故不取2、5、8键,
上下结构的字,因比上中下结构缺少位于横向中部的部件,故不取4、5、6键,;
如“睁”的左半部一体、右半部内又分上下部分,且相离(即离断、有明显或微小的缝隙)
如“眼”的左半部一体、右半部内又分上下部分,且相连(其“日”与下部笔画组无缝相连接)。
*进一步鉴别重码,在同一骨架码键区内部,可以进一步按权利要求2所述“内部微观结构特征”,再细分区域或子类别、进一步提示,如按“相对方位”:
——闭合类码元的角、边处都可连接开放类码元,如巳≠它≠马,尸≠甲,又如日中的两个块是纵向组合。再如【附图6】的右中部的“方”:
——对于各种方向的部件间的“品”字形相对结构,如:摄=356[右上角“耳”的实际覆盖区域按“键码方位一致”其实已经含2、3两键,可设容错编码;同理,螌=9745;
——对于块中块:完整或不完整的块,内部如包含内容,则又可同理按是否含块、块的方位是在角上还是边上来进行细分编码、提示,参见【附图3】,
——对于拐笔、拐枝或开闭组合型二包围码元如尸,内部(隔着缝隙)包含内容的:如“癌”,在九宫方式取码,码位所限,优先取占面积比大的码元,就忽略了中间横向并列的两个口,只取块码则为69,所含的枝为713;
2)键位数量范围扩展方案:
九宫键盘键位简单、但容量有限,较为复杂的简体汉字尚有部分部件需要忽略,不能一一精确、全部覆盖取码,而繁体汉字如
成分更多、甚至含“块”达到十个左右,所以所述九宫结构可以同理扩展,3x3=九宫4x4=十六宫5x5=二十五宫……则更能一一精确对应到每个部件;
3)码元种类层次扩展编码方案
上节“方位基本码”中只取封闭码元“块”,能够粗略的缩小汉字编码结果的重码范围,其忽略了开放码元的取码,可以也一并取码,并可进一步按细节特征细分编码、扩充信息,如块在右上角的字有“程、积、悍”等,其左部、右下部的部件都是开放码元,其细类却不同,有的是交、有的是枝,编制“字符部件机内编码信息数据库”简称“部件编码信息库”(九宫寻块相当于一种输入码,而它相当于一种扩展了的外码、或抽象的字形字体编码):
a.可将九宫结构内部的每个宫格(称母宫格)内部再镶嵌、或分出一个小的九宫格(称子宫格),将原本已映射母宫格中的码元按交点为单位拆解后、映射于相应不同方位的各子宫格中,其中每个子宫格内部可再包含“米形”线段,拆分取码例见【附图5、6】:
方式一.以笔段为单位编码:对每个“米”内部可依次按其各个方向上是否有分支线段来赋值为1或0, 可按顺时针从0°线段开始,如该线段上映射有笔划则赋值“1”、然后45°线段上映射有笔划则赋值“1”、……最后315°线段上映射有笔划则赋值“1”,无笔画则赋值为0,如“才”=10101011;同理“米”形四周的正方形子宫格边框,其四条边可分别赋值,也可将每边从中拆断、形成八段后分别赋值【附图7】;
方式二.以交点为单位、界限来编码:将网格(子、母网格均可)拆分成多个交点,分别编码;
方式三.以宫格为单位、界限来编码:将网格(子、母网格均可)拆分成多个区域,分别编码;
方式四.按类别分别来编码,分成闭合码元“边框”、开放码元“米”后分别赋值、编码;
——例如【附图6】的右中部的“方”的“B”的上下部分全不忽略取码时、同属闭合类;
b.选取权利要求二所述多项特征,设置相应多种变量、赋值来有规律、精确的表示,包括数字、字母、字符组成的一维或二维、多维编码序列,组成“部件信息库”:
连断信息:主要指一个以上母宫格之间连成一体式的结构,如上文“摄”的“耳”占了2、3两个宫格,子宫格内部细分其结构时同理,如三包围部件若“廖”中的“厂”占了74123宫格并联成一体,这也是重要的特征,可据此分类并分设键位及相应的编码,
——在韩文中同理,如
的123区连成一体、456区连成一体、7和9区是离断的,离断或部件间缝隙距离不明显或相连、需按类别分开的情况、同汉字同理,而且亦是优先按九宫区块之间的连断分类取码映射、其次再按所取各新划分的区域内的码元大小类别细分;
——可制定拆分规则,只拆离断(国标部件“钅”中上下部是离断的、但也可以拆开)和大类、不拆细类,就基本吻合了部件间连断情况,
如“程”的九宫的右上角区域是1个“口”,取码区域方位确定为“右上角”,其编码信息内容为“种类为口、数量1个”、右下角为一个开放的“枝”、含有3个交点、分支数为3、4、3,
如“镑”字——其在子宫格九宫内部细分的“九宫交点码”与母宫格方案同理
在1区母宫格内含有开放码元-横向的“亻”,子宫格区位编码为46,
在7区母宫格内含有开放码元-“干”,子宫格区位编码为28,
在3区母宫格内含有开放码元-“立”,子宫格区位编码为46,
在6区母宫格内含有闭合码元-“宀”,子宫格区位编码为05(0代表块、5为其位置),
——各交点位置坐标被具体取码,已经大体反映和区分了整个部件的结构方位和朝向
可见,每个汉字都能有精确、唯一的编码数值;
c.然后,由处理模块根据读取的“部件编码信息库”转换成需要进制的相应字符内码、机器码,从而显示汉字结果;
*“部件编码信息库”的取码编码次序和被读取是一个相反的过程:实际用作为计算机等计算装置内部识别运算的机内编码处理、录入该字的数据时,部件码元的取码可按一定规律、分多步依次进行,比如一律先角后边、先外后内,又如从左到右、从上到下,或先四角、后其他,再如顺时针、逐行;还可对按码元离断形成的几种总体结构制定相应的特殊次序,对左右型汉字先左上、后左下、再右上、再右下,包围型汉字先外围、后内容;
上述整个过程,具体形式很多、不可能一一列举,例如,
1.定义结构体类型:
2.定义结构体变量、。
如word xxxxxxxx={″xxxx″,xxx,xxx}
前面的xxxxxxxx可以代表计算机可读的汉字内码,可调出相应字集:
3.访问结构体:
访问结构体成员要用直接成员运算符″.″或间接成员运算符″->″。
word xxxxxxxx={″xxxx″,xxx,xxx};
cout<<a.name<<a.xxxx<<a.zonebit;
对于结构体变量,访问其中的成员采取″结构体变量.成员″的形式;而对于结构体指针,访问它所指向的结构体变量中的成员,则采取″结构体指针->成员″形式。
3)本发明所述方法拆分取码,不会产生界限歧义,如“奋”字的撇捺两笔搭在田的两个角上,计算机ocr不易分开识别,本方法可以避免,而且骨架代表字可以精确到汉字实际细节如须笔的情况、不显示不存在的笔划及须笔,并可用圆形等突出标注交点、拐点,提供标准信息,则形成可扫描并准确识别的“新型二维码”,例见【附图8】
4)上述“部件编码信息库”方法可实现“按部件检索”功能
与总体方案同理,用户选择九宫母宫格中的任意区域,处理模块接收选择后、读取该区部件的骨架、重码信息,据需要致使显示模块输出显示→用户选择具体该区域骨架或具体某种部件、并查询→计算机查询列举可能包含它的一类或多类骨架整字、并提示供选择,如“镑”字,用户选中右上角区域后,可查询右上角含“立”的几种骨架的重码字,若需要还可具体选择右上角的部件立、还是丰……。
——补充一点,部件特征信息除了字形之外、还有按非字形的属性信息划分类别、分别编码,即部件本身是某种偏旁部首声旁、形旁、同音、近义等,据此构建不同骨架间、统一骨架内的部件间种类、关系信息库,共处理模块查找、处理;
5)字词库的设置和调用
a.因为汉字字库数目几千到几万、词库数目更大,所以在全虚拟键盘展示时,为了更简洁、易找到需要的字,可据需要调整从字库调出、以供展示的字集内容和范围,可以进行分级、分类或分子集输出;
典型的维度如:意义(如“某行业”、有关“商品商标名称的”),输入背景场景、常用程度,属性、词性(如“动词”或“名词”…),范围(如“同音字”);
——尤其字词库的范围可调小至仅显示“同音字”时,重码字的结果一般只有几十个左右,见【附图1】所列,同音字子集据示例可见,与其他用“拼音+形码补码”减少同音字的输入法相比,它们片面、割裂的只取字首、字尾的码,本发明提示的是整字的骨架,更加一目了然、自然、直观;
——该同音字提示框界面的模块亦可对接于搜狗、谷歌、qq、微软等外部拼音输入法,获取其输入结果,并以本发明的方法,同理展示、及后续其他运算和功能;
b.本软件除了接受点击选择、也接受用户输入不会拆分或找不到的字,在界面中突出显示,提示其位置;
c.输入或选定某单字后,可以检索词库,同步显示该字所组成的词,同理可显示和相关同组甚至句子:
——在输入>1个字组成词时,通过“检索词库、组词去重(相当于公知技术“以词定字”)”也能过滤大部分重码,如只输入“1枝”骨架,重码会有很多,如丰天、羊美等,而输入“1枝”+“口”两个骨架,检索词库,只得出“美国、美团”等少量结果。
——对于界面中汉字之间的关联关系、即谁应与谁组合,可设置规则库,它不同于字库、词库,可由处理模块据其中逻辑规则、进行运算处理,包括多字词间的组合流程、结构、结果,并传送到显示模块显示;类似作用的还有智能语义算法模块等;
e.另外,对于拆分容易歧义或拆错的字,可做为容错码在字库中收录;
4.根据权利要求1到4任一项所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,上述汉字方案完全同理适用于朝鲜文字、韩语——因为韩文中只有元音辅音两种字母,其码元种类特征、总体结构特征、码元组合特征基本等同汉字,故码元进行组合的取码也完全等同,进而进行映射、输入及编码、的方法也可完全照搬(故略):韩文中各元音辅音字母分类如下、按其大小类可映射于键盘不同键位、键名即代码:
5.根据权利要求1和2所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,所述步骤2中“虚拟键位”区域上显示提示界面,
用户输入和机器输出、甚至还有交互过程,都合为一体、合于同屏,甚至同一位置(其他有些专利只有是屏幕选择后输出结果至光标处,本发明还能呈现在统一规定的固定位置、便于查找),并具有统一的总体布局和动态结构,且所见即所得,;
a.有两种显示方式:
方式一,动态可将同骨架的代表字,按其骨架合并叠放在一起,对于各字所含须笔的数量、方向的差异,用不同的格式(如不同粗细的线段)区别呈现,还可用动态的幻灯片、帧间渐变动画来呈现,这样就可以在很小的显示区域给用户以提示——用户点击或选定某类骨架后、再展开或放大该骨架类别下的“同类、同族”重码字;
例如,【附图】中点划虚线所围住的区域内“洁猪堵墙秸、活潘括播插、信清猜靖、倍滴摘培”共18字,它们所有的笔划都可以合并在一个骨架(“骨架码”整字编码)的范围内【见附图】,其他所有字同理,都已严谨计算后——归入各类骨架,无一带有主观任意性;
方式二,按一定次序并列分开布局、展现,共用户直接选择或点击、击键,
【附图】中一般可按先左后右、先上后下的顺序取字的首部件、尾部件,并每行按首部件对齐、每列按尾部件对齐;其中,还可按照相同的偏旁部首、声旁形旁等属性特征交叉分类、对齐行或列,在相应的二维数组、数据库结构中可以同理排列布局,便于检索运算;
b.可多盘同屏、同时呈现;指同理设置一个以上的屏幕键盘能并行使用、并互相打通关联,增加信息呈现量;
c.显示时,由处理单元还对显示内容、对象加以区别的格式标注、突出显示:
所述“格式、样式”,指采用不同形状、图案,及缩放、粗细、着色、线型、灰度、阴影等修饰方法,对取码码元,尤其如视觉上最显眼的“块”、键名、“拐角点”、“交点”、“端点”“横竖撇捺折五种笔划的八个方向”进行区别标注,形成对比;
对于界面提示中的不同内容(如有关联的几个字词)之间的关系(如相关、联想、推导、步骤、流程等),可显示连线来表示。
(三)输入系统部分
6.一种带屏幕提示界面的汉字骨架码输入系统,其特征在于,
接收模块,可获取来自用户的键盘操作、其他输入法的字符输入、甚至其他程序的操作指令或连接对象的动作及信息;
处理模块,根据从接受模块获取的键位操作,读取、检索、调出相应输入结果并发送至显示模块显示,还可以比对存储模块的信息,并进行运算,确定相关联、特显示字词的键位(或坐标),以及突出显示的格式、呈现字间组合关系的连线等,生成交互界面变化的结果,输出至界面供显示:
存储模块,用以存取所述各类部件编码信息库、字词库、语料库及重码结果编码表,以及字词关联关系数据库,智能算法、逻辑规则、语义文法规则;
显示模块,用以显示本发明的输入法提示界面,将字符结构化布局的文字输入提示,提供给使用者藉由上述输入装置进行上述输入;
可选的,连接模块,可将界面中各键位上的字符,连接计算对象,进行是双向发送、获取。
7、根据权利要求6所述带屏幕提示界面的汉字骨架码输入系统,其特征在于:所述连接模块的连接对象指:
每个汉字(甚至细分到细分到内部笔画、部件为单位),因具有不同的形状、编码,可代表东方悠久文明、及人类所理解的各种意义,其间多次连接组合的流程路径,又具有千变万化、却不离其宗的几何结构(与图论类似),较的模拟体现了人类思维活动、吻合了人类智能的结构:
界面中的各种“字”,代表着与该字意义相关相近的相应计算机指令或命令,统称计算对象,点击选择某字、即可执行或运行该字符链接的计算对象,同理反之界面中每个字符,可定向接收来自相关他人、外部程序的操作或信息,可交互或连接的对象主要有:
网址链接、编程命令、数据对象、数据库模型、网状数据结构、模块,容器、字符串、结构体、联合体、数组、函数变量、控件、源代码、编译器、底层机器代码指令、数学模型、运算流程,对接转换成相应的计算机底层编译、机器码、指令、命令、代码、语句、相应对象,动态链接库、应用程序接口API,服务、模块、相关链接、地址、程序,具体内容、对应规则的确定和运算,可建立世界通用的外部标准、或按系统内的界面对象数据库进行。
如“房”字链接到搜房网的网址,并与该网站互通信息、或互相连通特定指令的执行;点击“歌”字执行播放歌曲的命令或程序等:
从而以统一界面结构,通用于人机界面交互(输入查找、输出及信息呈现、操作等)、人人界面(交流教学协作活动,成为用户的万能软件客户端)、机机界面(各类计算设备的后台运算的编码、编译、编程开发、存储、传输等各层面,来实现各种编程运算方法的通用新工具,成为编程开发者所见即所得的可视化编程界面)。
8.一种带屏幕提示界面的输入装置,其特征在于,可在通用计算机、嵌入式计算机、单片机、或其它装置中通过软件或固件实现上述1-7任一项所述的输入法,主要包括
键盘操作接收单元,接收实体键盘的按键操作信号、或接收针对该虚拟键盘的按键操作信号,如鼠标点击、或触摸屏检测到的触碰信号——所述键盘,可以指实体键盘,也可以同理包括采用同样结构布局、键位基本对应的虚拟键盘系统,包括电子设备屏幕键盘、触摸屏屏幕键盘、甚至教学白板键盘、投影虚拟键盘,可用鼠标、手指、触摸屏、光笔、感应装置、脑机接口、视点追踪装置等;
处理单元,用以根据上述输入信号,查询检索上述编码表,生成对应的提示信息,并将结果供显示模块输出显示等;
存储单元,用以存取所述各类部件编码信息库、字词库、语料库及重码结果编码表,以及字词关联关系数据库,智能算法、逻辑规则、语义文法规则
显示单元,用以显示本发明的输入法提示界面,将字符结构化布局的文字输入提示,提供给使用者藉 由上述输入装置进行上述输入。
9.一种带屏幕提示界面的联网输入终端,其特征在于,包括上述1-8任一项所述的在输入法中实现智能联想的装置;
包括:存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如权利要求1至9任一项所述步骤。
10.一种计算机可读存储介质,其上存储有计算机程序,其特征在于,该程序被处理器执行时实现如权利要求1至5中任一项所述的词汇推荐方法中的步骤。
说明书附图1:字符界面编码提示方案;
说明书附图2-8:其说明见正文;
说明书附图9-35:取码编码分类布局总表,采用通用规范汉字表3500字
表格中除特殊注明外,一般先按首部件分“行”,后按尾部件分“列”,即将上或左部相同的字对齐于同一行、再将下或右部相同的字对齐于同一列;
按含中元的种类和个数,分别用下部标有:
不含块:细虚线=1开中元-连-1开中元
含1块:细虚线=2开中元-连-1闭中元,粗虚线=2开中元-夹-1闭中元
含2块:细实线=2闭中元-夹-1开中元,粗实线=2闭中元-夹-2开中元
粗实线=2闭中元-连-2开中元,双实线=2或3闭中元连续相邻
栏区内大多是含一开一闭中元、或只有闭中元的一般不标线
拐枝的排列顺序:下左角-下右角(较少)-上左角(较少)-上右角--各朝向的n、几
加边框的表示属同一细类,因表格容量不够而占用其他类的位置。
四角寻块法检字规则:
(一).取码规则之补充细则;
1.细表,表中【】中为需要忽略取码的情况
至全部忽略,形成备选的取码方案,虽技术效果下降、但可以减少思考。
Claims (10)
- 一种带屏幕提示界面的汉字骨架码输入法,其特征在于,由以下步骤完成输入:1)根据汉字的字形特征分为相应几种骨架类别、并分别将骨架相同的汉字聚类、并可按一定的规则排列,每种骨架可用一个代表字代表,形成骨架种类表和各自的类内字符集;2)由计算机显示各种骨架的代表字、及各自的全部或部分同骨架字,并于不同的“虚拟键位”区域上显示提示界面;3)使用者选择需要的骨架类别的代表字或类内具体某个字进行输入。
- 根据权利要求1所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,对汉字内部成分进行分类、并分别取码,组成所述各汉字的“骨架码”,骨架码的整体即整字编码,一个整字编码代表同骨架的一类字,整字编码由其内部编码组成,其内部编码的取码;1)可只按总体结构特征来分类——方式一,按总体组合次序;方式二,按总体组合次数;a.基本结构,主要分成:独体,杂合(如含“十”型“多枝”),包围、部分包围,上下、上中下(“三”型),左右、左中右(“川”型);b.混合结构,先分左(中)右、后分上(中)下,先分上(中)下、后分左(中)右(≈“品”型),及各自分别含包围、半包围(≈“L”型);多次混合,如先分左右、后分上下、再分左右(≈含“品”型),先分上下、后分左右、再分上下(≈含“品”型),及各自分别含包围、半包围(≈含“L”型);方式三,按总体部件数量,大致分2行x2列、3行x3列的部件结构、及相应键位布局,在九宫结构的3行x3列的布局的四边中点的码元称为“中元”,2行x2列的四个区块尤其可相当于3行x3列中的四角处的区块;2)可只按码元种类特征来分类——汉字笔划只有横竖撇捺折五种,除了“-”字之外,任何一个汉字均构成二维的网状,所含所有笔划均可映射在相应结构的“网格模型”的相应线段上,该模型由多条相互平行的横线、竖线、斜线交织而成,即二维方向铺开的多个“田”字格、外加对角线组成;模型中的每个交点均有多条线相交,除了横线x轴、竖线y轴之外,一般可加上4条斜对角线,共8条呈“米”形的分支,还可加上或按非45°夹角设置更多条线,如,以每22.5度增加一条,则每个交点共有16条分支;以交点为单位拆解汉字“母网”,宏观上可只拆分为两类基本结构大类:封闭码元{[口][块]},开放码元{[交][枝][多枝]},——开放和封闭的区分标准是:多个交点之间是否有笔划首尾相连、组成了的“口”形封闭笔画组区域(或称封闭的部件),有则为封闭结构、无则为开放结构;a.“封闭码元”,主要指无须笔的“口”,亦可增加角上有须笔的“块”构成新的增补方案;——及可选的,三角块“开放码元”,主要指“交”、“枝”“多枝”,“交”指多笔划相交于一个交点,“枝”指多个完整或不完整的“交”在一维方向的线性组合,多个完整或不完整的“交”在二维方向的线性组合则为“多枝”;*它们均可按以下细节特征再细分小类:完整度、朝向、绝对位置、内含下级码元个数(“枝”中所含的交点数目、“交”的每交点上的分支数、“块”的须笔数、块内分块的数量及所含内容);*单个汉字内部的不同成分,如笔划、笔画组(或称部件)之间、不同子网之间,均可分为相离、相连、相交的方位关系;3)可只按码元组合特征来分类——a.码元组合方位c.组合次序,d.交连程度,是离断/相连/相交,4)同理,可对上述特征进一步交叉分类,如上述特征1、3交叉分类,主要分成:*“骨架码”内部编码的拆分和映射、输入:将整个汉字按内部离断缝隙、码元的类别不同为界限,“能散不连、能连不交”的进行拆分后,骨架码编码相同的汉字聚类映射于同一“键位”,由用户选取、进行输入;所述“键位”,可以指“屏幕虚拟键位”上、并可呈现该“骨架码”整体编码代表字,及据需要呈现其代表的、全部或部分同骨架重码字供选择,来帮助用户提示输入;还可以同理映射于实体键位上,并可据需要在屏幕上同步呈现该“骨架码”整体编码代表字及其代表的全部或部分同骨架重码字供选择;所述键位布局可以据需要调整,按一定规律布局,如相对方位,尤其可按“在字中取码和键位映射的方位基本一致原则”来布局,如“中心放射结构”可按照取码码元尤其是“块”在整个汉字中所处位置,放置在“二维键盘”的相应键位区域;——“中心反向放射结构”同理,是将相反方位的各键对调;同理,还可以按码元种类来分区设置键位布局,或根据个性喜好任意设置键位布局。
- 根据权利要求1和2所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,所述各“骨架”内部取码、拆分和映射后,可进行编码,并录入“字内部件编码数据库”优选方案,按述权利要求1所述特征2和4交叉分类,按照各种码元类别、方位结构的大小类别的特征,分别进行相应的赋值和编码,其中包括:1)方位基本码,也可作为一种简单的10键小键盘“四角找块”或“四角找角”形码输入法,汉字编码码长为不定长,共10个部件代码,可用计算机键盘上的3行x3列的9个数字键表示,一键对应一种部件,部件对应的键名即为该部件的代码,从构成汉字的组字部件中选取一种部件(狭义或广义的)“块”作为对汉字进行编码的部件,可按“取码和击键的方位基本一致”原则:根据选取的码“块”在整个汉字中所处位置,映射到相应3行x3列的“九宫”二维结构键盘的相应位置,所述该部件与计算机键盘相应键位的映射关系分别为——第二行:4键映射位于整字中“左中(←)部”含有的块,5键则为“中( )部”,6键则为“右中(→)部”,可先取四角“角块”、其次再取各边的中间区域“边块”,取完为止,依次拆出的“块”部件对应的数字键序列即为该字的编码,在计算机键盘上输入该字的编码就可得到该字;——“一个“块”也不含”的字,另外单设键位;*取码(优先度)规则,尤其适用于码位所限、不能全部取码时,;优先对占整字面积比例大的部件取码,依次是贯穿整字、或字的一半、1/3、1/4、1/6、1/8、1/16……的部件,如“品”内上部的被拉长的“口”;一个取码区域(即宫格)中同时有开放、封闭码元时,优先取封闭码元,忽略多出的开放码元;*进一步鉴别重码,在同一骨架码键区内部,可以进一步按权利要求2所述“内部微观结构特征”,再细分区域或子类别、进一步提示,如按“相对方位”;——闭合类码元的角、边处都可连接开放类码元;——“品”字形相对结构;2)九宫结构可以同理扩展,3x3=九宫 4x4=十六宫 5x5=二十五宫……则更能一一精确对应到每个部件;3)码元种类层次扩展编码方案开放码元的取码,可以也一并取码,并可进一步按细节特征细分编码、扩充信息,编制“字符部件机内编码信息数据库”简称“部件编码信息库”;a.可将九宫结构内部的每个宫格(称母宫格)内部再镶嵌、或分出一个小的九宫格(称子宫格),将原本已映射母宫格中的码元按交点为单位拆解后、映射于相应不同方位的各子宫格中,其中每个子宫格内部可再包含“米形”线段,;方式一.以笔段为单位编码:对每个“米”内部可依次按其各个方向上是否有分支线段来赋值为1或0;同理“米”形四周的正方形子宫格边框,其四条边可分别赋值,也可将每边从中拆断、形成八段后分别赋值;方式二.以交点为单位、界限来编码:将网格(子、母网格均可)拆分成多个交点,分别编码;方式三.以宫格为单位、界限来编码:将网格(子、母网格均可)拆分成多个区域,分别编码;方式四.按类别分别来编码,分成闭合码元“边框”、开放码元“米”后分别赋值、编码;b.选取权利要求二所述多项特征,设置相应多种变量、赋值来有规律、精确的表示,包括数字、字母、字符组成的编码序列,组成“部件信息库”;连断信息:主要指一个以上母宫格之间连成一体式的结构——在韩文中同理,离断或部件间缝隙距离不明显或相连、需按类别分开的情况、同汉字同理,而且亦是优先按九宫区块之间的连断分类取码映射、其次再按所取各新划分的区域内的码元大小类别细分;c.然后,由处理模块根据读取的“部件编码信息库”转换成需要进制的相应字符内码、机器码,从而显示汉字结果;*“部件编码信息库”的部件码元的取码可按一定规律、分多步依次进行3)本发明所述方法拆分取码,骨架代表字可以精确到汉字实际细节如须笔的情况、不显示不存在的笔划及须笔,并可用圆形等突出标注交点、拐点,提供标准信息,则形成可扫描并准确识别的“新型二维码”4)上述“部件编码信息库”方法可实现“按部件检索”功能与总体方案同理,用户选择九宫母宫格中的任意区域,处理模块接收选择后、读取该区部件的骨架、重码信息,据需要致使显示模块输出显示→用户选择具体该区域骨架或具体某种部件、并查询→计算机查询列举可能包含它的一类或多类骨架整字、并提示供选择——除了字形之外、还有按非字形的属性信息划分类别、分别编码,即部件本身是某种偏旁部首声旁、形旁、同音、近义等,据此构建不同骨架间、统一骨架内的部件间种类、关系信息库,共处理模块查找、处理;5)字词库的设置和调用a.可据需要调整从字库调出、以供展示的字集内容和范围,可以进行分级、分类或分子集输出——尤其字词库的范围可调小至仅显示“同音字”——该同音字提示框界面的模块亦可对接于搜狗、谷歌、qq、微软等外部拼音输入法,获取其输入结 果,并以本发明的方法,同理展示、及后续其他运算和功能;b.本软件除了接受点击选择、也接受用户输入不会拆分或找不到的字,在界面中突出显示,提示其位置;c.输入或选定某单字后,可以检索词库,同步显示该字所组成的词,同理可显示相关词组甚至句子;——在输入>1个字组成词时,通过“检索词库、组词去重”也能过滤大部分重码,——对于界面中汉字之间的关联关系、即谁应与谁组合,可设置规则库,它不同于字库、词库,可由处理模块据其中逻辑规则、进行运算处理,包括多字词间的组合流程、结构、结果,并传送到显示模块显示;类似作用的还有智能语义算法模块等;e.另外,对于拆分容易歧义或拆错的字,可做为容错码在字库中收录;
- 根据权利要求1和2所述的带屏幕提示界面的汉字骨架码输入法,其特征在于,所述步骤2中“虚拟键位”区域上显示提示界面,用户输入和机器输出、甚至还有交互过程,都合为一体、合于同屏,甚至同一位置,并具有统一的总体布局和动态结构,且所见即所得,;a.有两种显示方式:方式一,动态可将同骨架的代表字,按其骨架含并叠放在一起,对于各字所含须笔的数量、方向的差异,用不同的格式(如不同粗细的线段)区别呈现,还可用动态的幻灯片、帧间渐变动画来呈现,这样就可以在很小的显示区域给用户以提示——用户点击或选定某类骨架后、再展开或放大该骨架类别下的“同类、同族”重码字;方式二,按一定次序并列分开布局、展现,共用户直接选择或点击、击键,可按先左后右、先上后下的顺序取字的首部件、尾部件,并每行按首部件对齐、每列按尾部件对齐;其中,还可按照相同的偏旁部首、声旁形旁等属性特征交义分类、对齐行或列,在相应的二维数组、数据库结构中可以同理排列布局,便于检索运算;b.可多盘同屏、同时呈现:指同理设置一个以上的屏幕键盘能并行使用、并互相打通关联,增加信息呈现量;c.显示时,由处理单元还对显示内容、对象加以区别的格式标注、突出显示:所述“格式、样式”,指采用不同形状、图案,及缩放、粗细、着色、线型、灰度、阴影等修饰方法,对取码码元,尤其如视觉上最显眼的“块”、键名、“拐角点”、“交点”、“端点”“横竖撇捺折五种笔划的八个方向”进行区别标注,形成对比:对于界面提示中的不同内容(如有关联的几个字词)之间的关系(如相关、联想、推导、步骤、流程等),可显示连线来表示。
- 一种带屏幕提示界面的汉字骨架码输入系统,其特征在于,接收模块,可获取来自用户的键盘操作、其他输入法的字符输入、甚至其他程序的操作指令或连接对象的动作及信息;处理模块,根据从接受模块获取的键位操作,读取、检索、调出相应输入结果并发送至显示模块显示,还可以比对存储模块的信息,并进行运算,确定相关联、待显示字词的键位(或坐标),以及突出显示的格式、呈现字间组合关系的连线等,生成交互界面变化的结果,输出至界面供显示;存储模块,用以存取所述各类部件编码信息库、字词库、语料库及重码结果编码表,以及字词关联关系数据库,智能算法、逻辑规则、语义文法规则;显示模块,用以显示本发明的输入法提示界面,将字符结构化布局的文字输入提示,提供给使用者藉由上述输入装置进行上述输入;可选的,连接模块,可将界面中各键位上的字符,连接计算对象,进行是双向发送、获取。
- 根据权利要求6所述带屏幕提示界面的汉字骨架码输入系统,其特征在于:所述连接模块的连接对象指:界面中的各种“字”,代表着与该字意义相关相近的相应计算机指令或命令,统称计算对象,点击选择某字、即可执行或运行该字符链接的计算对象,同理反之界面中每个字符,可定向接收来自相关他人、外部程序的操作或信息;可交互或连接的对象主要有:网址链接、编程命令、数据对象、数据库模型、网状数据结构、模块,容器、字符串、结构体、联合体、数组、函数变量、控件、源代码、编译器、底层机器代码指令、数学模型、运算流程,对接转换成相应的计算机底层编译、机器码、指令、命令、代码、语句、相应对象,动态链接库、应用程序接口API,服务、模块、相关链接、地址、程序,具体内容、对应规则的确定和运算,可建立世界通用的外部标准、或按系统内的界面对象数据库进行;从而以统一界面结构,通用于人机界面交互(输入查找、输出及信息呈现、操作等)、人人界面(交流教学协作活动,成为用户的万能软件客户端)、机机界面(各类计算设备的后台运算的编码、编译、编程开发、存储、传输等各层面,来实现各种编程运算方法的通用新工具,成为编程开发者所见即所得的可视化编程界面)。
- 一种带屏幕提示界面的输入装置,其特征在于,可在通用计算机、嵌入式计算机、单片机、或其它装置中通过软件或固件实现上述1-7任一项所述的输入法,主要包括:键盘操作接收单元,接收实体键盘的按键操作信号、或接收针对该虚拟键盘的按键操作信号,如鼠标点击、或触摸屏检测到的触碰信号——所述键盘,可以指实体键盘,也可以同理包括采用同样结构布局、键位基本对应的虚拟键盘系统,包括电子设备屏幕键盘、触摸屏屏幕键盘、甚至教学白板键盘、投影虚拟键盘,可用鼠标、手指、触摸屏、光笔、感应装置、脑机接口、视点追踪装置等;处理单元,用以根据上述输入信号,查询检索上述编码表,生成对应的提示信息,并将结果供显示模块输出显示等;存储单元,用以存取所述各类部件编码信息库、字词库、语料库及重码结果编码表,以及字词关联关系数据库,智能算法、逻辑规则、语义文法规则显示单元,用以显示本发明的输入法提示界面,将字符结构化布局的文字输入提示,提供给使用者藉由上述输入装置进行上述输入。
- 一种带屏幕提示界面的联网输入终端,其特征在于,包括上述1-8任一项所述的在输入法中实现智能联想的装置包括:存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如权利要求1至9任一项所述步骤。
- 一种计算机可读存储介质,其上存储有计算机程序,其特征在于,该程序被处理器执行时实现如权利要求1至5中任一项所述的词汇推荐方法中的步骤。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201880010326.9A CN110494825A (zh) | 2017-02-15 | 2018-02-12 | 带屏幕提示界面的汉字骨架码输入法及系统 |
JP2019567772A JP2020530601A (ja) | 2017-02-15 | 2018-02-12 | 画面提示画面付き漢字骨格コード入力法とシステム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710101325 | 2017-02-15 | ||
CN201710101325.2 | 2017-02-15 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018149250A1 true WO2018149250A1 (zh) | 2018-08-23 |
Family
ID=63169677
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2018/000072 WO2018149250A1 (zh) | 2017-02-15 | 2018-02-12 | 带屏幕提示界面的汉字骨架码输入法及系统 |
Country Status (3)
Country | Link |
---|---|
JP (1) | JP2020530601A (zh) |
CN (1) | CN110494825A (zh) |
WO (1) | WO2018149250A1 (zh) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111078907A (zh) * | 2018-10-18 | 2020-04-28 | 中华图象字教育股份有限公司 | 汉字树处理方法及其装置 |
CN111275049A (zh) * | 2020-01-19 | 2020-06-12 | 佛山市国方识别科技有限公司 | 一种文字图像骨架特征描述符获取的方法及装置 |
CN111611503A (zh) * | 2020-05-27 | 2020-09-01 | 百度在线网络技术(北京)有限公司 | 页面的处理方法、装置、电子设备及存储介质 |
CN112597876A (zh) * | 2020-12-20 | 2021-04-02 | 湖北工业大学 | 基于特征融合的书法汉字评判方法 |
CN113589955A (zh) * | 2020-04-30 | 2021-11-02 | 北京搜狗科技发展有限公司 | 一种数据处理方法、装置和电子设备 |
CN117608402A (zh) * | 2023-11-23 | 2024-02-27 | 首都医科大学附属北京天坛医院 | 一种基于汉字书写想象的隐蔽汉语言处理系统和方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1032247A (zh) * | 1987-11-16 | 1989-04-05 | 刘树根 | 骨架词法智能编码专业汉字输入、记录和键盘系统 |
CN1694046A (zh) * | 2004-05-05 | 2005-11-09 | 宗刚 | 一种计算机编码汉字键盘输入方法及信息码 |
CN101699373A (zh) * | 2009-10-28 | 2010-04-28 | 华南理工大学 | 一种基于结构、基元和拼音的汉字输入方法 |
CN102778951A (zh) * | 2012-06-15 | 2012-11-14 | 惠州华阳通用电子有限公司 | 使用虚拟按键的输入设备及输入方法 |
-
2018
- 2018-02-12 WO PCT/CN2018/000072 patent/WO2018149250A1/zh active Application Filing
- 2018-02-12 JP JP2019567772A patent/JP2020530601A/ja active Pending
- 2018-02-12 CN CN201880010326.9A patent/CN110494825A/zh active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1032247A (zh) * | 1987-11-16 | 1989-04-05 | 刘树根 | 骨架词法智能编码专业汉字输入、记录和键盘系统 |
CN1694046A (zh) * | 2004-05-05 | 2005-11-09 | 宗刚 | 一种计算机编码汉字键盘输入方法及信息码 |
CN101699373A (zh) * | 2009-10-28 | 2010-04-28 | 华南理工大学 | 一种基于结构、基元和拼音的汉字输入方法 |
CN102778951A (zh) * | 2012-06-15 | 2012-11-14 | 惠州华阳通用电子有限公司 | 使用虚拟按键的输入设备及输入方法 |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111078907A (zh) * | 2018-10-18 | 2020-04-28 | 中华图象字教育股份有限公司 | 汉字树处理方法及其装置 |
CN111275049A (zh) * | 2020-01-19 | 2020-06-12 | 佛山市国方识别科技有限公司 | 一种文字图像骨架特征描述符获取的方法及装置 |
CN111275049B (zh) * | 2020-01-19 | 2023-07-21 | 佛山市国方识别科技有限公司 | 一种文字图像骨架特征描述符获取的方法及装置 |
CN113589955A (zh) * | 2020-04-30 | 2021-11-02 | 北京搜狗科技发展有限公司 | 一种数据处理方法、装置和电子设备 |
CN111611503A (zh) * | 2020-05-27 | 2020-09-01 | 百度在线网络技术(北京)有限公司 | 页面的处理方法、装置、电子设备及存储介质 |
CN112597876A (zh) * | 2020-12-20 | 2021-04-02 | 湖北工业大学 | 基于特征融合的书法汉字评判方法 |
CN112597876B (zh) * | 2020-12-20 | 2024-03-26 | 湖北工业大学 | 基于特征融合的书法汉字评判方法 |
CN117608402A (zh) * | 2023-11-23 | 2024-02-27 | 首都医科大学附属北京天坛医院 | 一种基于汉字书写想象的隐蔽汉语言处理系统和方法 |
CN117608402B (zh) * | 2023-11-23 | 2024-05-14 | 首都医科大学附属北京天坛医院 | 一种基于汉字书写想象的隐蔽汉语言处理系统和方法 |
Also Published As
Publication number | Publication date |
---|---|
JP2020530601A (ja) | 2020-10-22 |
CN110494825A (zh) | 2019-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2018149250A1 (zh) | 带屏幕提示界面的汉字骨架码输入法及系统 | |
US9911211B1 (en) | Lens-based user-interface for visualizations of graphs | |
CN111259653B (zh) | 基于实体关系消歧的知识图谱问答方法、系统以及终端 | |
US9990417B2 (en) | Boolean-query composer | |
US10817527B1 (en) | Systems and methods of using natural language processing for visual analysis of a data set | |
CN101694668B (zh) | 网页结构相似性确定方法及装置 | |
KR102521765B1 (ko) | 인과 관계의 판별 방법, 장치, 전자 기기 및 저장 매체 | |
BR112016013584B1 (pt) | Sistema de computação e método executado por processador | |
Goguen et al. | 7 Information visualisation and semiotic morphisms | |
US10747506B2 (en) | Customizing operator nodes for graphical representations of data processing pipelines | |
WO2020000764A1 (zh) | 一种面向印地语的多语言混合输入方法及装置 | |
US11842154B2 (en) | Visually correlating individual terms in natural language input to respective structured phrases representing the natural language input | |
TW201820172A (zh) | 對話模式分析系統、方法及非暫態電腦可讀取記錄媒體 | |
WO2023173631A1 (zh) | 编程方法和装置、设备、存储介质及计算机程序产品 | |
Atzberger et al. | Software Forest: A Visualization of Semantic Similarities in Source Code using a Tree Metaphor. | |
Chowdhury et al. | MIVA: Multimodal interactions for facilitating visual analysis with multiple coordinated views | |
Langenhan et al. | Sketch-based methods for researching building layouts through the semantic fingerprint of architecture | |
US12067358B1 (en) | Using a natural language interface to explore entity relationships for selected data sources | |
Liu et al. | Spatial specification and reasoning using grammars: from theory to application | |
Richardson et al. | Computational processes for map generalization | |
Wang et al. | An application oriented and shape feature based multi-touch gesture description and recognition method | |
Sánchez-Zamora et al. | Visualizing tags as a network of relatedness | |
CN117634605B (zh) | 一种基于知识图谱的岭南文化可视化展示方法及系统 | |
Yang et al. | Scene layout in text-to-scene conversion | |
CN114385873B (zh) | 基于长文本语义一致的家具场景重建与推荐方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18754656 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2019567772 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18754656 Country of ref document: EP Kind code of ref document: A1 |