CN103080878A - Method and apparatus for segmenting strokes of overlapped handwriting into one or more groups - Google Patents
Method and apparatus for segmenting strokes of overlapped handwriting into one or more groups Download PDFInfo
- Publication number
- CN103080878A CN103080878A CN2010800687358A CN201080068735A CN103080878A CN 103080878 A CN103080878 A CN 103080878A CN 2010800687358 A CN2010800687358 A CN 2010800687358A CN 201080068735 A CN201080068735 A CN 201080068735A CN 103080878 A CN103080878 A CN 103080878A
- Authority
- CN
- China
- Prior art keywords
- stroke
- strokes
- series
- group
- features
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/333—Preprocessing; Feature extraction
- G06V30/347—Sampling; Contour coding; Stroke extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- 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
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/22—Character recognition characterised by the type of writing
- G06V30/226—Character recognition characterised by the type of writing of cursive writing
- G06V30/2268—Character recognition characterised by the type of writing of cursive writing using stroke segmentation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Character Discrimination (AREA)
- User Interface Of Digital Computer (AREA)
- Character Input (AREA)
Abstract
A method, apparatus and computer program product are provided in order to determine a plurality of features associated with the current stroke based upon geometrical properties of the series of strokes and to then segment a series of strokes that comprise overlapped handwriting into one or more groups of strokes based on the features associated with the strokes. This segmentation may be performed prior to any segmentation into one or more characters and, indeed, each group of strokes may be either a character or a portion of a character, but, in any event, no group of strokes includes strokes from more than one character.
Description
Technical field
Illustrative embodiments of the present invention relates to overlapping hand-written processing, and more specifically, relates to and will comprise that overlapping hand-written a series of strokes are divided into one or more groups stroke.
Background technology
Handwriting input is caught or received to various device by touch-screen or other input equipments.For example, the electronic equipment of a lot of computing machines, panel computer, mobile phone, PDA(Personal Digital Assistant) and other types comprises touch-screen, and it allows the user as the sense of touch input is provided with hand-written form.Handwriting input can be caught by electronic equipment, and processed to attempt to identify hand-written character in order to inform the future action of electronic equipment, comprise storage or the transmission of the expression of for example handwriting input.
In order to make character recognition easier, handwriting input is usually processed in order to handwriting input is divided into kinds of characters, such as letter, numeral or other characters.There are several different technology that are used for cutting apart handwriting input.A kind of technology is utilized the time-out between each character input.According to this technology, user interface such as touch-screen, only need to comprise that is write a zone.The user can write a character in this writes the zone.Then suspend, so as to receive this character and remove this write think next character in the zone receive ready this write the zone.Therefore, the time-out between each character input is used for hand-written cutting apart.
Another kind of technology employing has two or more and writes the zone, such as the user interface of two or more touch-screens.The user can write a character in these write in the zone one, then switch to other and write the zone in order to write next character.When the user just when these write a middle written character in the zone, write at other characters of writing in the zone before and can be received and process, and other write that the zone can then be eliminated in order to write the zone for the receive ready these other of next character.Therefore, the difference switching of writing between the zone is used for hand-written cutting apart.
Another technology adopts sizable single zone of writing.In writing the zone, the user can adopt and be comparable to the mode of character writing on a piece of paper several characters of continuous writing one by one.Space interval and the bottom of this technology utilization between institute's written character is intelligent in order to suitably cut apart handwriting input.
In another technology, user interface provides the single zone of writing, write in the zone at this, the user can adopt between the character without space interval the several characters of mode continuous writing that overlap each other.By utilizing bottom intelligence, overlapping character can be divided in order to a character and other characters are separated.Yet this program is more complicated than the technology of the identifying processing of the common employing of top use.Under maximal possibility estimation, this identifying processing can have error ground that a series of strokes are divided into character.This identifying processing has very large calculating strength, and for true-time operation mini-plant particularly, very challenging property.
Utilize overlapping continuously hand-written multiple technologies, the electronic equipment that allows to have relatively little user interface receives user's finger or the continuous handwriting input of other writing pencils, so that handwriting input can provide than the relatively fast mode of the input of next character.Yet, about cutting apart and user interactions, may challenge to overlapping continuously hand-written dependence.In this respect, overlappingly hand-writtenly can increase and effectively cut apart overlapping hand-writtenly in order to each character and other characters are separated relevant difficulty continuously, particularly between more general institute's write characters, have the hand-written continuously of space interval.In addition, the overlapping meeting of a plurality of characters causes difficulty to the user because the user may since the noisy and complicated background that causes of other overlapping characters can not check clearly that the user is current what is being write.
Summary of the invention
According to illustrative embodiments, a kind of method, device and computer program are provided, be divided into one or more groups stroke in order to will comprise overlapping hand-written a series of strokes.This is cut apart and can carry out before being divided into any cutting operation of one or more characters, and every group of part that stroke can be character or character in fact, in any case but do not have one group of stroke to comprise from the stroke more than a character.By a series of strokes are divided into one or more groups stroke, follow-uply should the series stroke be divided into one or more characters and become more effective.In addition, one or more groups stroke before can show in less obvious different mode, receives continuously simultaneously overlapping hand-written in order to allow the user more clearly to check up-to-date stroke.
In one embodiment, a kind of method is provided, and the method comprises: reception comprises overlapping hand-written a series of strokes, and for each stroke in these a plurality of strokes, by processor, based on the definite a plurality of features that are associated with current stroke of the geometrical property of this series stroke.The method of this embodiment also is divided into one or more groups stroke based on the feature that is associated with stroke with a series of strokes.As mentioned above, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
The method of an embodiment also can should be divided into one or more characters by the series stroke based on one or more groups stroke that this series stroke has been divided into, thereby increase the efficient that a series of strokes is divided into character.According to illustrative embodiments, the method can be by only determining a plurality of features of being associated with current stroke based on the geometrical property of this series stroke, thus definite a plurality of features that are associated with current stroke.In one embodiment, the method also can cause the demonstration at least some groups in described group, so that at least one group by the mode of another group shows obviously to be different from least.Therefore, overlapping hand-written demonstration image can be simplified to make things convenient for the user to check for example up-to-date stroke.
In one embodiment, the method a plurality of characteristic standards that also will be associated with current stroke.In this respect, can should a plurality of characteristic standards based on the overall dimensions of a series of strokes.When determining with this a plurality of feature that current stroke is associated, these a plurality of features can from by the geometric center of the starting point of the geometric center of the terminal point of current stroke, current stroke, next stroke, next stroke, comprise current stroke minimum rectangle, comprise the minimum rectangle of next stroke and comprise the group that the minimum rectangle of front unicursal consists of and select.
The method can be carried out with increment mode or batch mode.In increment mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving each continuous stroke.In batch mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving a plurality of strokes.
In another embodiment, provide a kind of device, comprised at least one processor and at least one storer that comprises computer program code.At least one storer and the computer program code of the device of this embodiment are configured to: comprise overlapping hand-written a series of strokes with at least one processor so that this device receives at least, and for each stroke in these a plurality of strokes, based on the definite a plurality of features that are associated with current stroke of the geometrical property of this series stroke.At least one storer of the device of this embodiment and computer program code also are configured to at least one processor so that this device is divided into one or more groups stroke based on the feature that is associated with stroke with a series of strokes.As mentioned above, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
At least one storer and the computer program code of the device of an embodiment also can be configured to: with at least one processor so that this device should be divided into one or more characters by the series stroke based on one or more groups stroke that this series stroke has been divided into, thereby increased the efficient that a series of strokes is divided into character.At least one storer and the computer program code of the device of an illustrative embodiments also are configured to: with at least one processor so that this device by only determining a plurality of features of being associated with current stroke, definite a plurality of features that are associated with current stroke based on the geometrical property of this series stroke.In one embodiment, at least one storer and the computer program code of this device also can be configured to: with at least one processor so that this device causes the demonstration at least some groups in described group, so that at least one group shown in the mode that obviously is different from another group at least.Therefore, overlapping hand-written demonstration image can be simplified to make things convenient for the user to check for example up-to-date stroke.
In one embodiment, at least one storer of this device and computer program code also can be configured to: with at least one processor so that a plurality of characteristic standards that this device will be associated with current stroke.In this respect, can be based on the overall dimensions of a series of strokes with a plurality of characteristic standards.When determining with a plurality of feature that current stroke is associated, a plurality of features can from by the geometric center of the starting point of the geometric center of the terminal point of current stroke, current stroke, next stroke, next stroke, comprise current stroke minimum rectangle, comprise the minimum rectangle of next stroke and comprise the group that the minimum rectangle of front unicursal consists of and select.
Can carry out analysis to stroke with increment mode or batch mode.In increment mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving each continuous stroke.In batch mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving a plurality of strokes.
In another embodiment, a kind of equipment is provided, it comprises the device that comprises overlapping hand-written a series of strokes for reception, and for each stroke in these a plurality of strokes, determines the device of a plurality of features of being associated with current stroke based on the geometrical property of this series stroke.The equipment of this embodiment also can comprise for the device that a series of strokes is divided into one or more groups stroke based on the feature that is associated with stroke.As mentioned above, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
The equipment of an embodiment also can comprise for one or more groups stroke that has been divided into based on a series of strokes should be divided into the device of one or more characters by the series stroke, thereby increase the efficient that a series of strokes is divided into character.In an illustrative embodiments, device that be used for to determine a plurality of features of being associated with current stroke comprises for only based on the device of the definite a plurality of features that are associated with current stroke of geometrical property of this series stroke.In one embodiment, this equipment also can comprise for the device that causes the demonstration of at least some groups of described group, so that at least one group by the mode of another group shows obviously to be different from least.Therefore, overlapping hand-written demonstration image can be simplified to make things convenient for the user to check for example up-to-date stroke.
In one embodiment, this equipment also comprises the device for a plurality of characteristic standards that will be associated with current stroke.In this respect, can be based on the overall dimensions of a series of strokes with a plurality of characteristic standards.When determining with a plurality of feature that current stroke is associated, a plurality of features can from by the geometric center of the starting point of the geometric center of the terminal point of current stroke, current stroke, next stroke, next stroke, comprise current stroke minimum rectangle, comprise the minimum rectangle of next stroke and comprise the group that the minimum rectangle of front unicursal consists of and select.
This equipment can be analyzed stroke with increment mode or batch mode.In increment mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving each continuous stroke.In batch mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving a plurality of strokes.
In another embodiment, a kind of computer program is provided, it comprises at least one computer-readable memory with the computer-executable code part that is stored in wherein.This computer-executable code partly comprises for reception and comprises the code instructions of overlapping hand-written a series of strokes and be used for each stroke for these a plurality of strokes, determines the code instructions of a plurality of features of being associated with current stroke based on the geometrical property of this series stroke.The computer-executable code part of this embodiment also comprises for the code instructions that a series of strokes is divided into one or more groups stroke based on the feature that is associated with stroke.As mentioned above, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
The computer-executable code of an embodiment partly also can comprise for one or more groups stroke that has been divided into based on a series of strokes should be divided into the code instructions of one or more characters by the series stroke, thereby increase the efficient that a series of strokes is divided into character.The computer-executable code of an illustrative embodiments partly also can comprise for by only determining a plurality of features of being associated with current stroke, the code instructions of definite a plurality of features that are associated with current stroke based on the geometrical property of this series stroke.In one embodiment, this computer-executable code partly also can comprise for the code instructions that causes the demonstration of at least some groups of described group, so that at least one group by the mode of another group shows obviously to be different from least.Therefore, overlapping hand-written demonstration image can be simplified to make things convenient for the user to check for example up-to-date stroke.
In one embodiment, this computer-executable code part also comprises the code instructions for a plurality of characteristic standards that will be associated with current stroke.In this respect, can be based on the overall dimensions of a series of strokes with a plurality of characteristic standards.When determining with a plurality of feature that current stroke is associated, these a plurality of features can from by the geometric center of the starting point of the geometric center of the terminal point of current stroke, current stroke, next stroke, next stroke, comprise current stroke minimum rectangle, comprise the minimum rectangle of next stroke and comprise the group that the minimum rectangle of front unicursal consists of and select.
This computer program can be analyzed stroke with increment mode or batch mode.In increment mode, the step of cutting apart of definite and a series of strokes of a plurality of features repeats after receiving each continuous stroke.In batch mode, a plurality of features determine and the step of cutting apart that should the series stroke repeats after receiving a plurality of strokes.
Description of drawings
Described some illustrative embodiments of the present disclosure with generic term, now with reference to accompanying drawing, accompanying drawing does not need to draw in proportion, wherein:
Fig. 1 is to the wherein overlapping hand-written explanation of four character overlaps;
Fig. 2 is the block diagram of equipment according to an illustrative embodiment of the invention;
Fig. 3 is the functional block diagram of the operation carried out according to an illustrative embodiment of the invention;
Fig. 4 is the process flow diagram that the operation of carrying out according to an illustrative embodiment of the invention is shown;
Fig. 5 is the overlapping hand-written explanation that according to an illustrative embodiment of the invention the wherein feature of three continuous strokes is identified;
Fig. 6 is can be processed in order to identify the explanation of the mode of four characters to the overlapping hand-written of Fig. 1 according to an illustrative embodiment of the invention;
Fig. 7 is the process flow diagram that the operation of increment mode execution according to an illustrative embodiment of the invention is shown; And
Fig. 8 is the process flow diagram that the operation of carrying out according to the batch mode of another illustrative embodiments of the present invention is shown.
Embodiment
Describe more fully hereinafter some embodiments of the present invention referring now to accompanying drawing, wherein, show more of the present invention but not whole embodiment.In fact, various embodiment of the present invention can a lot of multi-form embodiment and not should be understood to be subject to each embodiment of statement here; Or rather, provide these embodiments so that this instructions will satisfy applicable legal requiremnt.From start to finish, same Reference numeral refers to same element.As used herein, according to the embodiments of the present invention, term " data ", " content ", " information " and similar term can be used in reference to the data that generation can be transmitted, receives and/or store interchangeably.Therefore, the use of any this term should not be used for the spirit and scope of restriction the embodiments of the present invention.
In addition, as used herein, term " circuit " refers to: (a) only have the circuit of hardware to realize (as, with mimic channel and/or digital circuit); (b) combination of (respectively) computer program of the software that is stored on one or more computer-readable memories of circuit and comprising and/or firmware instructions, one plays operation so that device execution one or more functions described herein; And (c) circuit, for example, the part of microprocessor or microprocessor, it need to be used for software or the firmware of operation, exists even this software or firmware are physics.This definition of " circuit " is applicable to the whole uses here of this term, comprises in any claim.As another example, as used herein, term " circuit " also comprises and comprises one or more processors and/or its part and subsidiary software and/or the realization of firmware.As another example, as used herein, term " circuit " also comprises, for example, the base band integrated circuit or the application processor integrated circuit that are used for mobile phone, or the similar integrated circuit in server, cellular network device, other network equipments and/or other computing equipments.
Such as definition here, " computer-readable recording medium " that refer to the storage medium (for example, volatibility or non-volatile memory device) of permanent, physics can be different from " the computer-readable transmission medium " that refers to electromagnetic signal.
As shown in fig. 1, hand-written can the input via user interface such as touch-screen with overlap mode, in this overlap mode, character is overlapping continuously, and between the character without space or the time interval.In this respect, the example of Fig. 1 is illustrated in and lists respectively in the bottom of this figure and overlapping with the order of 100 four characters that indicate.In this respect, show that 102 illustrate the input of the first character, show that 104 illustrate the overlapping input of the first and second characters, show that 106 illustrate the overlapping input of first, second, and third character, and show that 108 illustrate the overlapping input of whole four characters.
Overlapping hand-written permission user takes full advantage of user interface, and this for example provides the user of handwriting input particularly useful at less and the low touch-screen of resolution for the finger that utilizes them.In addition, overlapping hand-written use provides relatively nature and the technology that is used for receiving fast handwriting input.With reference to the demonstration of crowding gradually of Fig. 1, will recognize, because other character and character overlap before, thereby depend on overlapping hand-written to provide handwriting input for having proposed challenge cutting apart of kinds of characters.In addition owing to showing when a lot of other overlapping characters, overlapping hand-written also can be so that the user checks the current character of writing of user or the last character difficult that becomes.Therefore, provide according to an illustrative embodiment of the invention a kind of for the overlapping hand-written technology of pre-segmentation.
Can be received via in the various input equipments any by the overlapping hand-written of pre-segmentation, such as user interface, such as touch-screen etc.Not only overlapping hand-written can the reception via various different input equipments, and these input equipments can be proved and can comprise the part of various dissimilar electronic equipments.For example, Fig. 2 illustrates the block diagram of the portable terminal 10 that can embody illustrative embodiments of the present invention.Yet, be to be understood that, the portable terminal 10 that illustrates and describe hereinafter only is the example of one type the equipment that can benefit from illustrative embodiments of the present invention, and, various types of portable terminals, such as portable digital-assistant (PDA), mobile phone, pager, mobile TV, game station, laptop computer, camera, video recorder, audio/video player, radio, positioning equipment such as GPS (GPS) equipment, or combination above-mentioned and sound other types and text communication system, can easily use illustrative embodiments of the present invention.
In some embodiments, processor 20 can comprise that expectation is used for realizing the audio frequency of portable terminal 10 and the circuit of logic function.For example, processor 20 comprises one or more digital signal processors and/or one or more microprocessor.This processor can further comprise one or more analog to digital converters, one or more digital to analog converter and/or other support circuit.The control of portable terminal 10 and signal processing function distribute between these devices according to their abilities separately.Therefore, processor 20 also can comprise modulation and transmit before the function of convolutional encoding and staggered message and data.Processor 20 can comprise internal voice coder in addition, and can comprise internal data modem.In addition, processor 20 can comprise that operation can be stored in the function of the one or more software programs in the storer.For example, processor 20 can operate the connectivity program, such as traditional Web browser.Then, the connectivity program can allow portable terminal 10 to transmit according to for example wireless application protocol (wap), HTTP (HTTP) and/or similar agreement and receive web content, such as location-based content and/or other web page contents.
As mentioned above, this user's input interface can comprise touch-screen display 28, and it can be presented as any known touch-screen display.Therefore, for example, touch-screen display 28 can constitute by any suitable technology and realize touch recognition, such as technology such as resistance, electric capacity, infrared, strainmeter, surface wave, optical imagery, decentralized signal technology, acoustic pulse recognitions.Touch-screen display 28 can constitute the indication that receives user's input and the expression that the user is inputted provides to processor 20.
No matter the type of input equipment and comprise this input equipment electronic equipment type how, overlapping handwriting input can be analyzed according to illustrative embodiments as shown in Figure 3.In this embodiment, overlapping hand-written can the reception via touch-screen 110 is such as the touch-screen display 28 of Fig. 2 etc.Then, each stroke of handwriting input can pass through feature extraction 112, wherein, and definite one or more features relevant with this stroke as described below.Then, for example by 114 pairs of current strokes of sorter part of the stroke group identical with front unicursal whether, or current stroke whether the part of another stroke group determine.Although various sorters 114 can be used to analyze the feature relevant with current stroke with the group under definite current stroke, but statistical sorter commonly used adopted, such as support vector machine or artificial neural network.
As shown in Figure 3, sorter 114 is cut apart the current stroke of rule 116 analyses by utilization.Can determine in every way to cut apart rule 116.Yet, in the embodiment that illustrates, providing hand-written data storehouse 118, hand-written data storehouse 118 comprises the handwritten stroke of a plurality of different series.Then, each stroke series in hand-written data storehouse 118 can be passed through feature extraction 120, wherein, analyzes each stroke series in this hand-written data storehouse so that the feature of identification and this stroke Series correlation and each stroke that comprises this series.In one embodiment, be extracted and related with the current stroke in hand-written data storehouse 118 these features are cut apart rule through training 122 in order to improve.In this respect, hand-written data storehouse 118 can comprise various words and/or a lot of samples of sentence of being write by different people.These words and/or sentence can manually be separated into character.Be separated into character based on this, each stroke can be classified as last stroke or the stroke except last stroke of character of character.Therefore, the parameter that the stroke learning that training 122 can be from hand-written data storehouse 118 in order to set up will be considered when being defined in last stroke of determining stroke yes or no character cut apart rule.As described below, can represent this stroke by proper vector based on the various parameters of stroke, then the value of this proper vector can be determined by sorter 114 based on cutting apart rule 116.Then, this sorter can represent relatively that the value of proper vector of this stroke and predefine threshold value are to determine last stroke of this stroke yes or no character.Therefore, training 122 can be used for determining to provide the suitably parameter set of the maximum success ratio of specified stroke aspect for the whole various strokes in the hand-written data storehouse 118.Based on training 122, cutting apart rule 116 is defined, it then defines and can be classified device 114 and utilize the parameter set of stroke being classified with the stroke of the different series of being stored based on hand-written data storehouse 118, for example according to the feature of each stroke Series correlation and/or comprise each stroke of each series.Based on value and the predefine threshold value of the proper vector of the feature of having extracted from current stroke, therefore sorter 114 can determine that current stroke is that a part or the current stroke of the stroke group identical from front unicursal starts from different stroke groups by relatively.
Because purpose is that stroke is divided into complete character, so if if the group of cutting apart includes only the part of character or the group cut apart comprises and then error may occur from the stroke more than a character.Because the method and apparatus of an illustrative embodiments can more easily adapt to and proofread and correct any error relevant with the group of cutting apart of a part that includes only character, because the group of cutting apart may make up following pre-segmentation, so in one embodiment, can regulate the predefine threshold value, for example by increasing this threshold value, will comprise from the possibility more than the stroke of a character in order to reduce the group of cutting apart.
In the illustrative embodiments of Fig. 3, the operation on the dotted line comprises the training stage 126.So, can carry out in advance these operations and need to when overlapping handwriting input is received by touch-screen 110, not repeat these operations.Yet the operation under the dotted line comprises the execute phase 124 and carries out in the reception of overlapping handwriting input or afterwards, as after the one or more strokes of touch-screen 110 inputs.
As further explanation, referring now to Fig. 4, the operation of being carried out by device according to an illustrative embodiment of the invention shown in it.This device can for example be used on the portable terminal 10.Yet alternately, this device may be embodied on various other equipment, comprises mobile and fixed equipment, for example, and any equipment listed above.This device can comprise be used to the device that receives a series of strokes, such as processor 20, and user's input interface (such as touch-screen display 28) etc.Operation 130 referring to Fig. 4.As mentioned above, a plurality of strokes that receive comprise overlapping hand-written, and wherein, a plurality of characters are write out one by one continuously.For each stroke, can determine a plurality of features.Referring to operation 132.In this respect, device can comprise for the device of determining a plurality of features relevant with each stroke, such as processor 20.Can determine various features according to an illustrative embodiment of the invention.Yet, in one embodiment, comprise the feature relevant with current stroke and the feature relevant with last and next stroke for the determined a plurality of features of current stroke.In this respect, Fig. 5 illustrates a plurality of overlapping hand-written character that comprises current stroke 200, front unicursal 210 and next stroke 220.
For example, for the determined feature of current stroke can comprise starting point, next stroke of geometric center, next stroke of the terminal point of current stroke, current stroke geometric center, comprise current stroke minimum rectangle, comprise the minimum rectangle of next stroke and comprise the minimum rectangle of front unicursal.Those features about definition specified point or position define those features with a pair of coordinate, usually such as x, y coordinate.Similarly, for those features with the definition of rectangle or other two-dimensional shapes, each shape such as each rectangle can be with four characterizing definitions, such as the coordinate of the upper and lower of the coordinate on the left side of this shape and right side and this shape.With reference to Fig. 5, according to illustrative embodiments, the feature of extracting from current stroke can comprise x, the y coordinate (endX with the terminal point of the 204 current strokes that indicate, endY), with the geometric center (currentGCX, currentGCY) of the 206 current strokes that indicate, starting point (startX with 222 next stroke that indicates, startY), with the geometric center (nextGCX, nextGCY) of 226 next stroke that indicates with comprise the minimum rectangle of current stroke, next stroke and front unicursal.With reference to Fig. 5, each rectangle can be by four characterizing definitions, that is, and and with left side, right side, coordinate that the upper and lower is relevant.In this respect, in Fig. 5, current with prefix current(respectively with left and right, the upper and lower coordinate of current stroke, minimum rectangle that next stroke is relevant with front unicursal), next(is next) and previous(previous) sign.For example, the feature relevant with the minimum rectangle of current stroke worked as front left with currentLeft(in Fig. 5), currentRight(works as front right), currentTop(is when front upper) and currentBottom(when front lower) sign.For directed and other points that indicate in Fig. 5 comprise the terminal point 224 of starting point with the 202 current strokes that indicate, next stroke and respectively with starting point and the terminal point of the 212 and 214 front unicursals that indicate, although these other point is not extracted as feature in this illustrative embodiments.
Although can define in every way the geometric center of stroke, according to an illustrative embodiments, the geometric center of stroke is defined as the equalization point of whole points in the stroke.For example, comprise terminal point (x
i, y
i) (i=0 ... n-1) stroke has geometric center (GCX, GCY), and it is defined as foloows:
GCX=sum(x
i)/n，GCY=sum(y
i)/n；i=0…n-1
A plurality of overlapping characters and comprise that the corresponding stroke of these characters may be of different sizes are because these characters may be write littlely or still want to pass on identical implication during than Datong District.So, this device can comprise for the characteristic standard that will extract from current stroke to consider the device of any impacts different on (for example by removing) identical stroke size, such as processor 20.Operation 134 referring to Fig. 4.In one embodiment, based on the overall dimensions of a series of strokes with these characteristic standards.In this respect, the minimum rectangle that comprises each handwritten stroke can be defined as shown in Figure 5.In this respect, comprising the minimum rectangle of whole strokes can be by defining with upper and lower, the left and right relevant coordinate of overlapping hand-written character.The minimum rectangle that comprises therein whole serial strokes is positioned in order to have a jiao of the initial point place of this coordinate system and have along in the illustrative embodiments on the limit that this coordinate axis is extended, and the minimum rectangle that comprises whole strokes can be by its width such as totalWidth(overall width) and its height such as totalHeight(overall height) represent in simplified form.
In one example, the feature that the above has described can be as follows like that by standardization, and wherein, prefix Current(is current), Next(is next) previous with Pre() refer to respectively and current stroke, feature that next stroke is relevant with front unicursal.
So, CurrentStrokeEndX and CurrentStrokeEndY are standardized x and the y coordinates of the terminal point of current stroke.NextStrokeStartX and NextStrokeStartY are standardized x and the y coordinates of the starting point of next stroke.CurrentGCX and CurrentGCY are standardized x and the y coordinates of the geometric center of current stroke.NextGCX and NextGCY are standardized x and the y coordinates of the geometric center of next stroke.CurrentLeft, CurrentRight, CurrentTop and CurrentBottom are respectively left and right, the upper and lower coordinates of current stroke.NextLeft, NextRight, NextTop and NextBottom are respectively left and right, the upper and lower coordinates of next stroke.At last, PreLeft, PreRight, PreTop and PreBottom are respectively left and right, the upper and lower coordinates of front unicursal.
In one embodiment, then these a plurality of features of extracting from current stroke can be combined, and for example are combined into proper vector by processor 20.Just analyzed current stroke is in the example of initial stroke therein, will not have front unicursal.So, the feature relevant with front unicursal can be set to the predefine value, for example-1.
In case determined a plurality of features for current stroke, and in one example, a plurality of strokes then can should be divided into one or more groups stroke by the series stroke based on the feature relevant with stroke by standardization.Operation 136 referring to Fig. 4.So, this device can comprise for the device that a plurality of serial strokes is divided into one or more groups stroke based on these features, such as processor 20.As mentioned above, these strokes can be slit into one or more groups stroke by sorter 114 and according to following in greater detail technical point.This device also can comprise one or more groups stroke that has been divided into for based on a plurality of serial strokes, a plurality of serial strokes is divided into the device of one or more characters, such as processor 20.Referring to operation 138.Can carry out in every way to cutting apart of character based on these groups, comprise by utilization and cut apart rule 116 etc., cut apart rule 116 definition and will be classified the parameter of device 114 utilizations in order to suitably estimate the proper vector of stroke.So, the processor 20 of an illustrative embodiments can comprise or carry out feature extraction 112, sorter 114 and cut apart rule 116, as above in conjunction with Figure 3.Than in the situation that stroke is not carried out the example that intermediate packets is cut apart a plurality of overlapping characters, by initially defining the stroke group and then based on these group definition characters, can carry out stroke cutting apart to character in the mode that counting yield is more arranged.
With reference to Fig. 6, a plurality of overlapping characters are shown at 230 places.By a plurality of serial strokes are divided into the stroke group, can shown in 232, define a plurality of groups.After this, can a plurality of serial strokes be divided into a plurality of characters based on the group shown in 234.Then, the character of generation can be identified, as by pattern-recognition etc., shown in 236, so that each series of characters can be cut apart and be identified as to a plurality of overlapping hand-written character very effectively.In the illustrative embodiments of Fig. 6, based on these groups stroke is divided into character, can identify a plurality of possible combination of these groups, for example whole possible combination of these groups is shown in 234 and 236.In this respect, can be identified by handwriting recognition engine by each character that the various combination of these groups represents, this handwriting recognition engine can keep another computing equipment embodiment of communicate by letter by processor 20 or with this processor.In one embodiment, this handwriting recognition engine also can be determined every kind of combination of these groups and the similarity between the current character collection.Then, can be analyzed by each character set that the combination of these groups represents, as by by processor 20 or the language model that embodies with another computing equipment that this processor keeps communicate by letter, so that whether definite each character set is meaningful.Then, this device (such as processor 20), handwriting recognition engine and/or language model can be to based on to the measurements of the similarity determined by handwriting recognition engine, to the whether significant measurement and in some embodiments each possible character set of the combination of the measurement of the predefine geometrical property of each group is specified mark of the current character collection determined by language model.In this respect, can for example analyze each by processor 20 and organize to determine whether this group satisfies one or more predefine geometrical properties, for example, geometrical property can with the Size dependence of group, be considered to too little rather than complete character less than the group of predefine threshold value.As another example of geometrical property, be positioned at along the limit of writing the zone, such as the group along the most left of touch-screen display 28 or right half, can not think complete character.The character set of for example determining to have highest score based on the one or more quilts in similarity, the meaningful and geometrical property such as processor 20 can be identified as representing the character set of these a plurality of overlapping characters, shown in 238.
In one embodiment, can apply one or more restrictions to a plurality of processes that may make up of identifying these groups, to increase these combinations of identification and with the efficient of these combinations of aftertreatment.For example, these possible combinations can limit by any potential combination or character limit are maximum four groups.
Except increasing the efficient that can cut apart a plurality of overlapping characters, a series of strokes are to the demonstration that also can make things convenient for handwriting input of cutting apart of one or more groups stroke.In this respect, this device also can comprise for so that at least some demonstrations of these groups so that at least one group of device that shows obviously to be different from the mode of another group at least, such as processor 20, display 28 etc.Operation 140 referring to Fig. 4.For example, only have the subset of these groups to be shown, so that at least one group is not shown.In this respect, can show up-to-date group of relatively small number purpose and remove before whole groups from display 28.So, the so mixed and disorderly and user of display 28 can more easily check the current stroke of inputting and back to back each stroke before.In other embodiments, can show these stroke groups in the mode that obvious difference is arranged between the different stroke groups.For example, the available different colours of these stroke groups and/or luminance level show, such as different colours or the luminance level that changes based on the received order of stroke.In an illustrative embodiments, up-to-date stroke group the darkest available color (and/or brightness) shows, the available slightly brighter color of back to back stroke group (and/or brightness) shows before, and the rest may be inferred until with the initial stroke group of the brightest color (and/or brightness) demonstration.Alternately, can represent by dissimilar line different stroke groups, for example represent up-to-date stroke group with solid line, back to back stroke group before representing with dot-and-dash line, the like.In each embodiment, the stroke group is shown so that different stroke groups are obviously different, and in some embodiments, newer stroke group is more obvious.
In one embodiment, definite and this stroke of the feature relevant with stroke can be carried out after each stroke of input to current group cut apart, for example as shown in Figure 7.About the subsequent descriptions of Fig. 7, current stroke and next stroke (as discussed above) are expressed as respectively stroke k-1 and stroke k.With reference to the operation 250,252 and 254 of Fig. 7, after the entering stroke k, determine whether initially stroke of this stroke, that is, determine whether k=0.The stroke of having inputted therein is in the example of initial stroke, but the actual zone of writing of initialization.Referring to operation 256.In this respect, the actual zone of writing can be initialized to the minimum rectangle of surrounding this initial stroke, and the overall width of the rectangle of available directed with respect to coordinate system as shown in Figure 5 and location and overall height or define with upper and lower, the left and right coordinate of the rectangle that is positioned at other directions and position.After this, counter k can increase progressively before the input of waiting for next stroke.Referring to operation 258.After next stroke of input, can recomputate actual write the zone in case recomputate actually write the minimum rectangle that region representation surrounds each stroke.Referring to operation 260.After this, can determine the feature of current stroke k-1.Referring to operation 262.Based on for the fixed feature of stroke k-1, sorter 114 can be based on the feature value of providing S of stroke k-1
PreIn this respect, sorter 114 can and be cut apart rule 116 value of providing S based on the proper vector of stroke k-1
Pre, cut apart the proper vector of stroke is estimated in rule 116 definition by this sorter utilization parameter.Operation 264 referring to Fig. 7.Then, the value S that is provided by sorter 114
PreCan with the predefine threshold value T that is used for classification
PreCompare.Referring to operation 266.If the value S that is provided by sorter 114
PreSurpass the predefine threshold value T that is used for classification
Pre, then stroke k-1 is considered to belong to different groups with stroke k.Referring to operation 268.On the contrary, if the value S that is provided by sorter 114
PreBe not more than the predefine threshold value T for classification
Pre, then stroke k-1 is considered to belong to identical group with stroke k.Referring to operation 270.The above describes and this process shown in Figure 7 can incrementally repeat in order to should suitably cut apart in groups by the series stroke for each stroke of input.
As after each stroke input, individually analyzing these strokes so that suitably with the replacement scheme of these stroke groupings, can be with a plurality of strokes of batch processing process analysis, for example as shown in Figure 8.In this respect, can determine the actual zone of writing, as surround a plurality of strokes 0,1 ..., the minimum rectangle of M-1.Referring to operation 280.Then, but count initialized device k such as k=1, and can determine whether whole strokes of this batch are considered, for example the number by this counter relatively and stroke M.Referring to operation 282 and 284.In the example that all stroke does not also have to be considered therein, determine the feature of stroke k-1 and then can for example pass through the value S that sorter 114 is determined stroke k-1 based on the feature of stroke k-1
PreReferring to operation 286 and 288.As before, can be worth S
PreWith threshold value T
PreComparison, wherein, stroke k-1 belongs to different group or identical groups with stroke k and depends on respectively value S
PreWhether greater than the threshold value T that is used for classification
PreReferring to operation 290,292 and 294.Can repeat this process for each stroke that increases progressively this indicated batch of operation 296 Counter k, until each stroke has been taken into account and suitably be divided into each group.
Fig. 4, Fig. 7 and Fig. 8 be according to an illustrative embodiment of the invention method and the process flow diagram of program product.Will be appreciated that, the combination of each piece in each piece of these process flow diagrams and these process flow diagrams, can implement in every way, for example hardware, firmware, processor, circuit and/or other equipment relevant with the execution of the software that comprises one or more computer program instructions.For example, above-described one or more flow process can be embodied by computer program instructions.In this respect, embody the computer program instructions of above-described each flow process, can carry out by the memory device, stores of portable terminal 10 and by the processor 20 in this portable terminal.Will be appreciated that, any this computer program instructions (for example can be loaded into computing machine or other programmable devices, hardware) on, producing machine, so that these instructions of carrying out at this computing machine or other programmable devices generate the device that is used for carrying out the specified function of (respectively) flow chart block.These computer program instructions also can be stored in bootable computing machine or the computer-readable memory of other programmable devices with the ad hoc fashion operation, so that these instructions that are stored in this computer-readable memory produce the goods that comprise the command device of specified function in execution (respectively) flow chart block.These computer program instructions also can be loaded into computing machine or other programmable devices so that sequence of operations carries out to produce the computing machine implementation at this computing machine or other programmable devices, so as in these instruction execution flow segments that this computing machine or other programmable devices are carried out specified function.
Therefore, the piece support of these process flow diagrams is used for the combination of carrying out the operation of specified function and being used for carrying out the programmed instruction of specified function for the combination of the device of carrying out specified function.It will also be understood that, one or more of these process flow diagrams, and the combination of each piece in these process flow diagrams can be by the special-purpose hardware based computer systems of carrying out specified function, or the combination of specialized hardware and computer instruction is implemented.
Under the useful instruction that presents in before description and relevant drawings, the technician of field that the present invention belongs to will expect of the present invention a lot of modifications and other embodiments of setting forth here.Therefore, it being understood that the present invention is not limited to disclosed embodiment, and revise with other embodiments and will comprise within the scope of the appended claims.In addition, although description before and relevant drawings have been described each illustrative embodiments in the context of some example combinations of each element and/or function, should be understood that, the various combination of each element and/or function also can be provided by alternate embodiments, and does not break away from the scope of claims.In this respect, for example, each element except top those that describe clearly and/or the various combination of function also be considered and can be in claims some in propose.Although adopt concrete term here, they only use and are not used in restriction in a generic and descriptive sense.
Claims (36)
1. method comprises:
Reception comprises overlapping hand-written a series of strokes;
For each stroke in a plurality of strokes, utilize processor to determine a plurality of features that are associated with current stroke based on the characteristic of described a series of strokes; And
Based on the described feature that is associated with described stroke described a series of strokes are divided into one or more groups stroke, wherein, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
2. the method for claim 1, further comprise: described one or more groups stroke that has been divided into based on described a series of strokes is divided into one or more characters with described a series of strokes.
3. method as claimed in claim 1 or 2, wherein, cutting apart described a series of stroke comprises described a series of strokes is divided into a plurality of groups, and wherein said method further comprises: cause the demonstration at least some groups in described group, so that at least one group by the mode of another group shows obviously to be different from least.
4. such as each the described method among the claim 1-3, further comprise: described a plurality of characteristic standards that will be associated with described current stroke.
5. method as claimed in claim 4 wherein, comprises described a plurality of characteristic standards: based on the overall dimensions of described a series of strokes with described a plurality of characteristic standards.
6. such as each the described method among the claim 1-5, wherein, determining that described a plurality of feature comprises determines a plurality of features select from the group that is made of the following: the starting point of the terminal point of described current stroke, the geometric center of described current stroke, next stroke, the geometric center of next stroke, comprise described current stroke minimum rectangle, comprise the minimum rectangle of described next stroke and comprise the minimum rectangle of previous adjacent stroke.
7. such as each the described method among the claim 1-6, wherein, determine described a plurality of feature and cut apart described a series of stroke after receiving each continuous stroke, to repeat.
8. such as each the described method among the claim 1-6, wherein, determine described a plurality of feature and cut apart described a series of stroke after receiving a plurality of strokes, to carry out.
9. such as each the described method among the claim 1-8, wherein, determine that the described a plurality of features that are associated with described current stroke comprise: only based on the definite described a plurality of features that are associated with described current stroke of the geometrical property of described a series of strokes.
10. device comprises at least one processor and comprises at least one storer of computer program code, described at least one storer and described computer program code be configured to described at least one processor so that described device at least:
Reception comprises overlapping hand-written a series of strokes;
For each stroke in a plurality of strokes, based on the definite a plurality of features that are associated with current stroke of the geometrical property of described a series of strokes; And
Based on the described feature that is associated with described stroke described a series of strokes are divided into one or more groups stroke, wherein, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
11. device as claimed in claim 10, wherein, described at least one storer and computer program code be further configured for: with described at least one processor so that described device based on described one or more groups stroke that described a series of strokes have been divided into, is divided into one or more characters with described a series of strokes.
12. such as claim 10 or 11 described devices, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described device is cut apart described a series of stroke by described a series of strokes are divided into a plurality of groups, and cause the demonstration at least some groups in described group, so that at least one group by the mode of another group shows obviously to be different from least.
13. such as each the described device among the claim 10-12, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described a plurality of characteristic standards that described device will be associated with described current stroke.
14. device as claimed in claim 13, wherein, described at least one storer and computer program code be further configured into described at least one processor so that described device by based on the overall dimensions of described a series of strokes with described a plurality of characteristic standards with described a plurality of characteristic standards.
15. such as each the described device among the claim 10-14, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described device is determined described a plurality of feature by a plurality of features of determining to select from the group that is made of the following: the terminal point of described current stroke, the geometric center of described current stroke, the starting point of next stroke, the geometric center of next stroke, the minimum rectangle that comprises described current stroke, the minimum rectangle and the minimum rectangle that comprises previous adjacent stroke that comprise described next stroke.
16. such as each the described device among the claim 10-15, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described device after receiving each continuous stroke, repeatedly determine described a plurality of features and cut apart described a series of stroke.
17. such as each the described device among the claim 10-15, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described device after receiving a plurality of strokes, determine described a plurality of features and cut apart described a series of stroke.
18. such as each the described device among the claim 10-17, wherein, described at least one storer and computer program code be further configured for described at least one processor so that described device by only determining a plurality of features of being associated with described current stroke based on the geometrical property of described a series of strokes, thereby definite described a plurality of features that are associated with described current stroke.
19. an equipment comprises:
Be used for to receive the device that comprises overlapping hand-written a series of strokes;
Be used for each stroke for a plurality of strokes, determine the device of a plurality of features of being associated with current stroke based on the geometrical property of described a series of strokes; And
Be used for based on the described feature that is associated with described stroke described a series of strokes being divided into the device of one or more groups stroke, wherein, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
20. equipment as claimed in claim 19 further comprises: the device that is used for based on described one or more groups stroke that described a series of strokes have been divided into described a series of strokes being divided into one or more characters.
21. such as claim 19 or 20 described equipment, wherein, described device be used to cutting apart described a series of strokes comprises for described a series of strokes are divided into a plurality of groups device, and wherein said equipment further comprises: be used for causing the device to the demonstration of at least some groups of described group, so that at least one group by the mode of another group shows obviously to be different from least.
22. each the described equipment as among the claim 19-21 further comprises: for the device of described a plurality of characteristic standards that will be associated with described current stroke.
23. equipment as claimed in claim 22 wherein, describedly comprises for based on the overall dimensions of the described a series of strokes device with described a plurality of characteristic standards for the device with described a plurality of characteristic standards.
24. such as each the described equipment among the claim 19-23, wherein, described device for determining described a plurality of features comprises the device for described a plurality of features of determining to select from the group that is made of the following: the starting point of the terminal point of described current stroke, the geometric center of described current stroke, next stroke, the geometric center of next stroke, comprise described current stroke minimum rectangle, comprise the minimum rectangle of described next stroke and comprise the minimum rectangle of previous adjacent stroke.
25. such as each the described equipment among the claim 19-24, wherein, described device and described device repeatedly operation after receiving each continuous stroke be used to cutting apart described a series of strokes for determining described a plurality of features.
26. such as each the described equipment among the claim 19-24, wherein, described device and described device operation after receiving a plurality of strokes be used to cutting apart described a series of strokes for determining described a plurality of features.
27. such as each the described equipment among the claim 19-24, wherein, the device of described a plurality of features for determining to be associated with described current stroke comprises for only based on the device of the definite described a plurality of features that are associated with described current stroke of geometrical property of described a series of strokes.
28. a computer program comprises at least one computer-readable memory that wherein stores the computer-executable code part, described computer-executable code partly comprises:
Be used for to receive the code instructions that comprises overlapping hand-written a series of strokes;
Be used for each stroke for a plurality of strokes, determine the code instructions of a plurality of features of being associated with current stroke based on the geometrical property of described a series of strokes; And
Be used for based on the described feature that is associated with described stroke described a series of strokes being divided into the code instructions of one or more groups stroke, wherein, every group of stroke is the part of character or character, but do not have one group of stroke to comprise from the stroke more than a character.
29. computer program as claimed in claim 28 further comprises: the code instructions that is used for based on one or more groups stroke that described a series of strokes have been divided into described a series of strokes being divided into one or more characters.
30. such as claim 28 or 29 described computer programs, wherein, comprise for described a series of strokes are divided into a plurality of groups code instructions for the described code instructions of cutting apart described a series of strokes, and wherein said computer-executable code part further comprises: be used for causing the demonstration at least some groups of described group, so that at least one group by the code instructions that shows obviously to be different from the mode of another group at least.
31. each the described computer program as among the claim 28-30 further comprises: for the code instructions of described a plurality of characteristic standards that will be associated with described current stroke.
32. computer program as claimed in claim 31 wherein, is used for described code instructions with described a plurality of characteristic standards and comprises for based on the overall dimensions of the described a series of strokes code instructions with described a plurality of characteristic standards.
33. such as each the described computer program among the claim 28-32, wherein, described code instructions that be used for to determine described a plurality of features comprises the code instructions for described a plurality of features of determining to select from the group that is made of the following: the starting point of the terminal point of described current stroke, the geometric center of described current stroke, next stroke, the geometric center of next stroke, comprise described current stroke minimum rectangle, comprise the minimum rectangle of described next stroke and comprise the minimum rectangle of previous adjacent stroke.
34. such as each the described computer program among the claim 28-33, wherein, be performed after receiving each continuous stroke for the described code instructions of determining described a plurality of features with for the described code instructions of cutting apart described a series of strokes.
35. such as each the described computer program among the claim 28-33, wherein, be performed after receiving a plurality of strokes for the described code instructions of determining described a plurality of features with for the described code instructions of cutting apart described a series of strokes.
36. such as each the described computer program among the claim 28-35, wherein, described code instructions that be used for to determine a plurality of features of being associated with described current stroke only comprises the code instructions based on the definite described a plurality of features that are associated with described current stroke of geometrical property of described a series of strokes.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2010/076285 WO2012024829A1 (en) | 2010-08-24 | 2010-08-24 | Method and apparatus for segmenting strokes of overlapped handwriting into one or more groups |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103080878A true CN103080878A (en) | 2013-05-01 |
| CN103080878B CN103080878B (en) | 2017-03-29 |
Family
ID=45722811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201080068735.8A Active CN103080878B (en) | 2010-08-24 | 2010-08-24 | One or more groups of method and apparatus are divided into for hand-written stroke will be overlapped |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP5581448B2 (en) |
| KR (1) | KR101486174B1 (en) |
| CN (1) | CN103080878B (en) |
| WO (1) | WO2012024829A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104657071A (en) * | 2013-11-20 | 2015-05-27 | 株式会社东芝 | Feature calculation device and method |
| WO2015161823A1 (en) * | 2014-04-25 | 2015-10-29 | 夏普株式会社 | Handwriting recognition method and device |
| CN108492349A (en) * | 2018-03-19 | 2018-09-04 | 广州视源电子科技股份有限公司 | Processing method, device and equipment for writing strokes and storage medium |
| CN108509955A (en) * | 2017-02-28 | 2018-09-07 | 柯尼卡美能达美国研究所有限公司 | Infer stroke information from image |
| CN110503101A (en) * | 2019-08-23 | 2019-11-26 | 北大方正集团有限公司 | Font evaluation method, device, equipment and computer readable storage medium |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103376998B (en) * | 2012-04-19 | 2016-06-15 | 中兴通讯股份有限公司 | Handwriting equipment Chinese character type-setting method and device |
| JP2013246732A (en) * | 2012-05-28 | 2013-12-09 | Toshiba Corp | Handwritten character retrieval apparatus, method and program |
| JP5717691B2 (en) * | 2012-05-28 | 2015-05-13 | 株式会社東芝 | Handwritten character search device, method and program |
| EP2712433B1 (en) * | 2012-05-31 | 2016-11-02 | MultiTouch Oy | User interface for drawing with electronic devices |
| CN105283882B (en) * | 2013-04-12 | 2019-12-27 | 诺基亚技术有限公司 | Apparatus for text input and associated method |
| US9465985B2 (en) | 2013-06-09 | 2016-10-11 | Apple Inc. | Managing real-time handwriting recognition |
| CN105247540B (en) * | 2013-06-09 | 2018-10-16 | 苹果公司 | Manage real-time handwriting recognition |
| CN103345365B (en) * | 2013-07-12 | 2016-04-13 | 北京蒙恬科技有限公司 | The display packing of continuous handwriting input and the hand input device of employing the method |
| KR102125212B1 (en) * | 2013-08-29 | 2020-07-08 | 삼성전자 주식회사 | Operating Method for Electronic Handwriting and Electronic Device supporting the same |
| US9224038B2 (en) | 2013-12-16 | 2015-12-29 | Google Inc. | Partial overlap and delayed stroke input recognition |
| US9881224B2 (en) | 2013-12-17 | 2018-01-30 | Microsoft Technology Licensing, Llc | User interface for overlapping handwritten text input |
| US9524440B2 (en) * | 2014-04-04 | 2016-12-20 | Myscript | System and method for superimposed handwriting recognition technology |
| DK179329B1 (en) | 2016-06-12 | 2018-05-07 | Apple Inc | Handwriting keyboard for monitors |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0720084B1 (en) * | 1994-12-29 | 2002-02-27 | Sharp Kabushiki Kaisha | Input and display apparatus for handwritten characters |
| CN1607540A (en) * | 2003-09-24 | 2005-04-20 | 微软公司 | System and method for detecting hand-drawn objects inputted by ink |
| CN1805459A (en) * | 2005-01-14 | 2006-07-19 | Lg电子株式会社 | Message display unit and message display method for mobile communication terminal |
| CN101299236A (en) * | 2008-06-25 | 2008-11-05 | 华南理工大学 | Method for recognizing Chinese hand-written phrase |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS60254384A (en) * | 1984-05-31 | 1985-12-16 | Fujitsu Ltd | Corresponding method of stroke |
| JPH05233599A (en) * | 1992-02-18 | 1993-09-10 | Seiko Epson Corp | Online character recognizing device |
| JPH06162266A (en) * | 1992-11-20 | 1994-06-10 | Seiko Epson Corp | Method for recognizing on-line handwritten character and device therefor |
| JPH0792817B2 (en) * | 1994-08-19 | 1995-10-09 | 沖電気工業株式会社 | Online character recognizer |
| JPH08161426A (en) * | 1994-12-09 | 1996-06-21 | Sharp Corp | Handwritten character stroke segmenting device |
| JPH09161011A (en) * | 1995-12-13 | 1997-06-20 | Matsushita Electric Ind Co Ltd | Handwritten character input device |
| US5970498A (en) * | 1996-12-06 | 1999-10-19 | International Business Machines Corporation | Object oriented framework mechanism for metering objects |
| JP3216800B2 (en) * | 1997-08-22 | 2001-10-09 | 日立ソフトウエアエンジニアリング株式会社 | Handwritten character recognition method |
| JP3024680B2 (en) * | 1998-01-13 | 2000-03-21 | 日本電気株式会社 | Handwritten pattern storage and retrieval device |
| JP3456931B2 (en) * | 1999-12-10 | 2003-10-14 | シャープ株式会社 | Handwritten character recognition device, computer readable recording medium storing a handwritten character recognition program, and method for correcting characters recognized by handwritten characters |
| JP3974359B2 (en) * | 2000-10-31 | 2007-09-12 | 株式会社東芝 | Online character recognition apparatus and method, computer-readable storage medium, and online character recognition program |
| JP2005141329A (en) * | 2003-11-04 | 2005-06-02 | Toshiba Corp | Device and method for recognizing handwritten character |
| JP2009289188A (en) | 2008-05-30 | 2009-12-10 | Nec Corp | Character input device, character input method and character input program |
-
2010
- 2010-08-24 JP JP2013525107A patent/JP5581448B2/en active Active
- 2010-08-24 CN CN201080068735.8A patent/CN103080878B/en active Active
- 2010-08-24 WO PCT/CN2010/076285 patent/WO2012024829A1/en active Application Filing
- 2010-08-24 KR KR1020137007240A patent/KR101486174B1/en active IP Right Grant
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0720084B1 (en) * | 1994-12-29 | 2002-02-27 | Sharp Kabushiki Kaisha | Input and display apparatus for handwritten characters |
| CN1607540A (en) * | 2003-09-24 | 2005-04-20 | 微软公司 | System and method for detecting hand-drawn objects inputted by ink |
| CN1805459A (en) * | 2005-01-14 | 2006-07-19 | Lg电子株式会社 | Message display unit and message display method for mobile communication terminal |
| CN101299236A (en) * | 2008-06-25 | 2008-11-05 | 华南理工大学 | Method for recognizing Chinese hand-written phrase |
Non-Patent Citations (1)
| Title |
|---|
| 高彦宇，杨扬: "无约束手写体汉字切分方法综述", 《计算机工程》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104657071A (en) * | 2013-11-20 | 2015-05-27 | 株式会社东芝 | Feature calculation device and method |
| WO2015161823A1 (en) * | 2014-04-25 | 2015-10-29 | 夏普株式会社 | Handwriting recognition method and device |
| CN105095924A (en) * | 2014-04-25 | 2015-11-25 | 夏普株式会社 | Handwriting recognition method and device |
| CN108509955A (en) * | 2017-02-28 | 2018-09-07 | 柯尼卡美能达美国研究所有限公司 | Infer stroke information from image |
| CN108509955B (en) * | 2017-02-28 | 2022-04-15 | 柯尼卡美能达美国研究所有限公司 | Method, system, and non-transitory computer readable medium for character recognition |
| CN108492349A (en) * | 2018-03-19 | 2018-09-04 | 广州视源电子科技股份有限公司 | Processing method, device and equipment for writing strokes and storage medium |
| CN110503101A (en) * | 2019-08-23 | 2019-11-26 | 北大方正集团有限公司 | Font evaluation method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2012024829A1 (en) | 2012-03-01 |
| KR20130058053A (en) | 2013-06-03 |
| CN103080878B (en) | 2017-03-29 |
| JP2013543158A (en) | 2013-11-28 |
| KR101486174B1 (en) | 2015-01-23 |
| JP5581448B2 (en) | 2014-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103080878A (en) | Method and apparatus for segmenting strokes of overlapped handwriting into one or more groups | |
| CN106681633B (en) | System and method for auxiliary information input control function of sliding operation of portable terminal equipment | |
| CN102449640B (en) | Recognizing handwritten words | |
| CN101308610B (en) | Calligraphy exercising method and apparatus based on mobile terminal | |
| US20090256808A1 (en) | Device and method for stroke based graphic input | |
| CN104205047A (en) | Apparatus and method for providing for remote user interaction | |
| CN112580643A (en) | License plate recognition method and device based on deep learning and storage medium | |
| CN110942004A (en) | Handwriting recognition method and device based on neural network model and electronic equipment | |
| CN109194689B (en) | Abnormal behavior recognition method, device, server and storage medium | |
| CN101697107A (en) | Method and device for inputting characters in touch mobile communication terminal | |
| CN103870799A (en) | Character direction judging method and device | |
| CN103854019A (en) | Method and device for extracting fields in image | |
| JP2014127188A (en) | Shaping device and method | |
| US9311523B1 (en) | Method and apparatus for supporting object recognition | |
| CN110413856A (en) | Classification annotation method, apparatus, readable storage medium storing program for executing and equipment | |
| CN110232381A (en) | License Plate Segmentation method, apparatus, computer equipment and computer readable storage medium | |
| CN113935349A (en) | Method and device for scanning two-dimensional code, electronic equipment and storage medium | |
| CN107885770B (en) | Target domain database construction method, target domain database sample identification method, terminal and storage medium | |
| KR102322606B1 (en) | Method for correcting typographical error and mobile terminal using the same | |
| CN111722717A (en) | Gesture recognition method and device and computer readable storage medium | |
| CN110555431B (en) | Image recognition method and device | |
| CN115700519A (en) | Text-to-image generation method and device, storage medium and terminal | |
| CN108052525B (en) | Method and device for acquiring audio information, storage medium and electronic equipment | |
| CN105094544B (en) | Method and device for acquiring characters | |
| CN104423867A (en) | Character input method and character input device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160115 Address after: Espoo, Finland Applicant after: Technology Co., Ltd. of Nokia Address before: Espoo, Finland Applicant before: Nokia Oyj |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |