CN103902100A - Speed characteristics based stroke partitioning method for intelligent freehand sketching - Google Patents
Speed characteristics based stroke partitioning method for intelligent freehand sketching Download PDFInfo
- Publication number
- CN103902100A CN103902100A CN201410128457.0A CN201410128457A CN103902100A CN 103902100 A CN103902100 A CN 103902100A CN 201410128457 A CN201410128457 A CN 201410128457A CN 103902100 A CN103902100 A CN 103902100A
- Authority
- CN
- China
- Prior art keywords
- curve
- point
- high speed
- correction
- stroke
- 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
Abstract
The invention provides a speed characteristics based stroke partitioning method for intelligent freehand sketching. The speed characteristics based stroke partitioning method includes extracting sampling points of input valid strokes, partitioning the valid strokes into normal-speed strokes and approximate-constant-speed strokes, judging speed partitioning points by three-line threshold values for the normal-speed strokes, smoothing and sharpening firstly for the approximate-constant-speed strokes, reducing the number of the sampling points and enhancing peak value characteristics of a speed curve, converting the approximate-constant-speed strokes into the normal-speed strokes, processing according to the normal-speed strokes, and finally converting. By the speed characteristics based stroke partitioning method, the problem that algorithm is sensitive to stroke noise can be effectively avoided, and compared with exiting Sezgin algorithm, the speed characteristics based stroke portioning method has better judgment accuracy.
Description
Technical field
The present invention relates to a kind of method of the Freehandhand-drawing figure stroke segmentation based on velocity characteristic, for intelligent Freehandhand-drawing input equipment, as the Freehandhand-drawing such as handwriting pad, smart mobile phone input equipment.
Background technology
Freehandhand-drawing figure be the mankind a kind of from however directly thinking and exchange way, the introducing of the Post-WIMP User Interface of new generation based on Freehandhand-drawing figure, has proposed new challenge to the mankind.More completely define based on Freehandhand-drawing figure human-computer interaction technology and be: user uses rough geometric configuration to the fuzzy a kind of graphical interaction input mode of approaching of its intention.The randomness of user's input makes to adopt Freehandhand-drawing drawing method can express fast creative thinking, but the ambiguity of Sketchy data causes Freehandhand-drawing figure and user's design idea to have certain deviation, and this has brought very large difficulty just to computer interpretation Freehandhand-drawing figure.The target of sketch recognition is to draw the minimum situation of constraint user to be issued to best identified efficiency and effect.At present, in the field by 2D sketch reconstruction 3D model, sketch explanation aspect generally all adopts the Freehandhand-drawing figure recognition methods based on basic line element.The method has solved the Uniqueness that figure is described, and comprises stroke segmentation, pattern classification, identification, matching, the step such as regular, and wherein stroke segmentation is the committed step that affects recognition effect.
Stroke segmentation technology has been passed through the existing a lot of achievement in research of long-term research, but mainly come from image processing field or develop thus, existing velocity characteristic extracts Sezgin algorithm and generally adopts the method based on mean value, using speed average 90% as threshold value, by the rate curve grouping lower than threshold value, in each group using velocity amplitude smallest point as velocity characteristic point.The method has solved the extraction problem of velocity characteristic to a certain extent, but still has many deficiencies: local noise may cause packet error, and threshold value is relatively fixing may cause algorithm to stroke noise-sensitive, and when plotting speed is more even, feature extraction was lost efficacy.
Summary of the invention
Technical scheme
Technical scheme of the present invention is:
A kind of described stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input, is characterized in that: the effective input stroke receiving is extracted to N sampled point, and sampled point comprises stroke starting point and stroke terminal, is not less than S the interval time of neighbouring sample point; Obtain respectively the instantaneous velocity of each sampled point, and calculate the mean value of the instantaneous velocity of all N sampled point
if the ratio of sampled point number and stroke length L is not more than ξ
maxand
be greater than v
max, be constant speed stroke by this effective input pen marking, otherwise be labeled as closely at the uniform velocity stroke;
For constant speed stroke, adopt following steps to determine speed cut-point:
Step 1: set up rate curve according to the instantaneous velocity of sampled point in coordinate system, described rate curve is connected each self-corresponding instantaneous velocity to obtain successively according to sampling order by each sampled point, and wherein the ordinate of coordinate system is instantaneous velocity value;
Step 2: instantaneous velocity is more than or equal to
sampled point be labeled as high speed point, calculate at a high speed point relatively
deviation
wherein H is for putting at a high speed number, v
ibe i the instantaneous velocity of point at a high speed; Instantaneous velocity is less than
sampled point be labeled as low speed point; Calculate low speed point relative
deviation
wherein L is low speed point number, v
jit is the instantaneous velocity of j low speed point;
Step 3: adopt v
lisoline is divided into several lower than v by rate curve
llow speed segment of curve and several higher than v
lhigh speed segment of curve; For all high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain high speed segment of curve, if this high speed segment of curve satisfies condition: this high speed segment of curve and v
hisoline intersect, and/or this high speed segment of curve with
the intersection point number of isoline is greater than K, and the sampled point of getting instantaneous velocity minimum in the front side low speed segment of curve adjacent with this high speed segment of curve is speed cut-point; If this high speed segment of curve does not meet described condition, both sides, the front and back low speed segment of curve this high speed segment of curve being adjacent merges the low speed segment of curve that composition is new, and participates in the judgement to next high speed segment of curve with this new low speed segment of curve;
For stroke closely at the uniform velocity, adopt following steps to determine speed cut-point:
Step 4: according to formula
Step 5: according to formula
Calculate the instantaneous velocity value v of stroke smothing filtering sharpening each correction sampled point after treatment
p', wherein
Step 6: according to the instantaneous velocity v of each correction sampled point
p' in coordinate system, set up erection rate curve, described erection rate curve is revised sampled point by each and according to sampling order, each self-corresponding instantaneous velocity is connected and is obtained successively; Calculate the mean value of the instantaneous velocity of all correction sampled points
Step 7: instantaneous velocity is more than or equal to
correction sampled point be labeled as and revise high speed point, calculate and revise at a high speed point relatively
deviation
wherein H' puts at a high speed number, v for revising
i' be i the instantaneous velocity of revising high speed point; Instantaneous velocity is less than
correction sampled point be labeled as and revise low speed point; Calculate and revise low speed point relatively
deviation
wherein L' is for revising low speed point number, v
j' be j the instantaneous velocity of revising low speed point;
Step 8: adopt v
l' isoline is that several are lower than v by erection rate curve segmentation
l' correction low speed segment of curve and several higher than v
l' correction high speed segment of curve; For all correction high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain correction high speed segment of curve, if this correction high speed segment of curve satisfies condition: this correction high speed segment of curve and v
h' isoline intersect, and/or this correction high speed segment of curve with
the intersection point number of isoline is greater than K, and the correction sampled point of getting instantaneous velocity minimum in the front side correction low speed segment of curve adjacent with this correction high speed segment of curve is erection rate cut-point; If this correction high speed segment of curve does not meet described condition, the both sides, front and back that this correction high speed segment of curve are adjacent are revised low speed segment of curve and are merged the correction low speed segment of curve that composition is new, and participate in next to revise with this new correction low speed segment of curve the judgement of high speed segment of curve;
Label p by erection rate cut-point obtains in original sample point, corresponding with erection rate cut-point, [(p-1) (T between label area
d-1)+1, (T
d-1) p+1]; Get label interval [(p-1) (T
d-1)+1, (T
d-1) p+1] sampled point of instantaneous velocity minimum is speed cut-point in corresponding original sample point.
A kind of described stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input, is characterized in that: N is not less than 5.
A kind of described stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input, is characterized in that: S is 0.015s.
A kind of described stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input, is characterized in that: ξ
maxvalue is not more than 0.2, v
maxbe not less than 200mm/s.
A kind of described stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input, is characterized in that: K is not less than 3 integer.
Beneficial effect
The present invention is directed to the deficiency of existing velocity characteristic extracting method, stroke is divided into constant speed and two classes closely at the uniform velocity, provide for constant speed stroke the method that adopts speed average and upper lower deviation thereof to cut apart as three line threshold values, effectively avoided the problem of algorithm to stroke noise-sensitive; For render speed more at the uniform velocity stroke be translated into constant speed stroke by filtering-sharpening processing and process.
Accompanying drawing explanation
Fig. 1: rate curve schematic diagram;
Fig. 2: constant speed pentagram velocity characteristic extracts; (a) Sezgin algorithm, (b) method of the present invention;
Fig. 3: closely at the uniform velocity pentagram velocity characteristic extracts; (a) Sezgin algorithm, (b) method of the present invention;
Fig. 4: constant speed pentagram velocity characteristic extracts waveform comparison; (a) Sezgin algorithm, (b) method of the present invention;
Fig. 5: closely at the uniform velocity pentagram velocity characteristic extracts waveform comparison; (a) Sezgin algorithm, (b) method of the present invention;
Embodiment
Below in conjunction with specific embodiment, the present invention is described:
Embodiment: the velocity characteristic of Freehandhand-drawing pentagram figure extracts
In the present embodiment, adopt respectively constant speed and Freehandhand-drawing pentagram figure at the uniform velocity, to verify method of the present invention, contrast the Sezgin algorithm of prior art simultaneously.
The stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input in the present embodiment, the effective input stroke receiving is extracted to N sampled point, and N is not less than 5, and sampled point comprises stroke starting point and stroke terminal, be not less than S the interval time of neighbouring sample point, and S gets 0.015s; The track that the input stroke here refers to button---movement---forms in the process of loose key.Obtain respectively the instantaneous velocity of each sampled point, and calculate the mean value of the instantaneous velocity of all N sampled point
wherein stroke starting point and stroke terminal are as the head and the tail sampled point of sampled point sequence, and instantaneous velocity is 0; If the ratio of sampled point number and stroke length L is not more than ξ
maxand
be greater than v
max, be constant speed stroke by this effective input pen marking, otherwise be labeled as closely at the uniform velocity stroke; ξ
maxvalue is not more than 0.2, v
maxbe not less than 200mm/s; ξ in the present embodiment
maxvalue 0.15, v
maxget 400mm/s.
For constant speed stroke, adopt following steps to determine speed cut-point:
Step 1: set up rate curve according to the instantaneous velocity of sampled point in coordinate system, described rate curve is connected each self-corresponding instantaneous velocity to obtain successively according to sampling order by each sampled point, wherein the ordinate of coordinate system is instantaneous velocity value, as shown in Figure 1;
Step 2: instantaneous velocity is more than or equal to
sampled point be labeled as high speed point, calculate at a high speed point relatively
deviation
wherein H is for putting at a high speed number, v
ibe i the instantaneous velocity of point at a high speed; Instantaneous velocity is less than
sampled point be labeled as low speed point; Calculate low speed point relative
deviation
wherein L is low speed point number, v
jit is the instantaneous velocity of j low speed point;
Step 3: adopt v
lisoline is divided into several lower than v by rate curve
llow speed segment of curve and several higher than v
lhigh speed segment of curve; Its feature of such partitioning scheme is that head and the tail two segment of curve are low speed segment curve; Low speed segment curve and high regime curved intersection occur; In each segment of curve, at least comprise the velocity amplitude of a sampled point, as shown in Figure 1, in research in the past, researcher it has been generally acknowledged that in each low speed segment of curve should a velocity characteristic point, but find in experiment, in the time that user draws degree of freedom increase, in low speed segment of curve, may not there is not velocity characteristic point.So just low speed segment of curve has been carried out with v
hwith
correction for benchmark: for all high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain high speed segment of curve, if this high speed segment of curve satisfies condition: this high speed segment of curve and v
hisoline intersect, and/or this high speed segment of curve with
the intersection point number of isoline is greater than K, and K is not less than 3 integer, and the sampled point of getting instantaneous velocity minimum in the front side low speed segment of curve adjacent with this high speed segment of curve is speed cut-point; If this high speed segment of curve does not meet described condition, both sides, the front and back low speed segment of curve this high speed segment of curve being adjacent merges the low speed segment of curve that composition is new, and participates in the judgement to next high speed segment of curve with this new low speed segment of curve.
The velocity characteristic that adopts this dividing method can well solve constant speed stroke extracts problem.As a comparison, the present embodiment has also adopted the Sezgin algorithm of prior art to cut apart for constant speed Freehandhand-drawing pentagram figure:
In the present embodiment, be 61 for the sampling number of constant speed Freehandhand-drawing pentagram figure, as shown in Figure 2, as shown in Figure 4, the key value of two kinds of methods is more as shown in table 1 for the rate curve of two kinds of methods for the speed cut-point of two kinds of methods:
Table 1: two kinds of key value comparisons that method is extracted
As seen from Figure 4, visible stroke of drawing for constant speed, the inventive method is on the basis of Sezgin algorithm, to have increased by three line threshold values to judge speed cut-point, thereby effectively avoid the mistake of the velocity characteristic causing due to noise to extract, in constant speed drafting situation, although the stroke segmentation time of the inventive method is a little more than Sezgin algorithm, but the accuracy of cutting apart for above-mentioned example the inventive method is 100%, and Segin algorithm cut apart accuracy not higher than 66.77%, therefore in the situation that accuracy obviously improves, have increased slightly and can receive the sliced time that constant speed is drawn stroke.
And for stroke closely at the uniform velocity, adopt following steps to determine speed cut-point:
Step 4: according to formula
Step 5: according to formula
Calculate the instantaneous velocity value v of stroke smothing filtering sharpening each correction sampled point after treatment
p', wherein
Step 6: according to the instantaneous velocity v of each correction sampled point
p' in coordinate system, set up erection rate curve, described erection rate curve is revised sampled point by each and according to sampling order, each self-corresponding instantaneous velocity is connected and is obtained successively; Calculate the mean value of the instantaneous velocity of all correction sampled points
Step 7: instantaneous velocity is more than or equal to
correction sampled point be labeled as and revise high speed point, calculate and revise at a high speed point relatively
deviation
wherein H' puts at a high speed number, v for revising
i' be i the instantaneous velocity of revising high speed point; Instantaneous velocity is less than
correction sampled point be labeled as and revise low speed point; Calculate and revise low speed point relatively
deviation
wherein L' is for revising low speed point number, v
j' be j the instantaneous velocity of revising low speed point;
Step 8: adopt v
l' isoline is that several are lower than v by erection rate curve segmentation
l' correction low speed segment of curve and several higher than v
l' correction high speed segment of curve; For all correction high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain correction high speed segment of curve, if this correction high speed segment of curve satisfies condition: this correction high speed segment of curve and v
h' isoline intersect, and/or this correction high speed segment of curve with
the intersection point number of isoline is greater than K, and the correction sampled point of getting instantaneous velocity minimum in the front side correction low speed segment of curve adjacent with this correction high speed segment of curve is erection rate cut-point; If this correction high speed segment of curve does not meet described condition, the both sides, front and back that this correction high speed segment of curve are adjacent are revised low speed segment of curve and are merged the correction low speed segment of curve that composition is new, and participate in next to revise with this new correction low speed segment of curve the judgement of high speed segment of curve;
Label p by erection rate cut-point obtains in original sample point, corresponding with erection rate cut-point, [(p-1) (T between label area
d-1)+1, (T
d-1) p+1]; Get label interval [(p-1) (T
d-1)+1, (T
d-1) p+1] sampled point of instantaneous velocity minimum is speed cut-point in corresponding original sample point.
In fact, be exactly by filtering sharpening processing for the nearly at the uniform velocity disposal route of stroke, minimizing sampled point number also strengthens rate curve peak value feature, thus nearly at the uniform velocity stroke be converted into constant speed stroke, process again the process of finally changing according to the mode of above-mentioned constant speed stroke.
In the present embodiment, be 342 for the nearly at the uniform velocity sampling number of Freehandhand-drawing pentagram figure, as shown in Figure 3, as shown in Figure 5, the key value of two kinds of methods is more as shown in table 2 for the rate curve of two kinds of methods for the speed cut-point of two kinds of methods:
Table 2: two kinds of key value comparisons that method is extracted
As seen from Figure 3, in closely at the uniform velocity drafting situation, Sezgin algorithm complete failure, the Sezgin algorithm accuracy providing in table 2 is not higher than 11.1%, and the extraction of the inventive method is more accurate.As seen from Figure 5, visible for nearly stroke of at the uniform velocity drawing, first the inventive method has increased filtering sharpening processing on Sezgin algorithm basis, thereby has effectively reduced sampled point number and strengthened rate curve peak value feature, and this has been also embodied in table 2 in sampled point number after treatment.
Claims (5)
1. for the stroke segmentation method based on velocity characteristic of intelligent Freehandhand-drawing input, it is characterized in that: the effective input stroke receiving is extracted to N sampled point, and sampled point comprises stroke starting point and stroke terminal, is not less than S the interval time of neighbouring sample point; Obtain respectively the instantaneous velocity of each sampled point, and calculate the mean value of the instantaneous velocity of all N sampled point
if the ratio of sampled point number and stroke length L is not more than ξ
maxand
be greater than v
max, be constant speed stroke by this effective input pen marking, otherwise be labeled as closely at the uniform velocity stroke;
For constant speed stroke, adopt following steps to determine speed cut-point:
Step 1: set up rate curve according to the instantaneous velocity of sampled point in coordinate system, described rate curve is connected each self-corresponding instantaneous velocity to obtain successively according to sampling order by each sampled point, and wherein the ordinate of coordinate system is instantaneous velocity value;
Step 2: instantaneous velocity is more than or equal to
sampled point be labeled as high speed point, calculate at a high speed point relatively
deviation
wherein H is for putting at a high speed number, v
ibe i the instantaneous velocity of point at a high speed; Instantaneous velocity is less than
sampled point be labeled as low speed point; Calculate low speed point relative
deviation
wherein L is low speed point number, v
jit is the instantaneous velocity of j low speed point;
Step 3: adopt v
lisoline is divided into several lower than v by rate curve
llow speed segment of curve and several higher than v
lhigh speed segment of curve; For all high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain high speed segment of curve, if this high speed segment of curve satisfies condition: this high speed segment of curve and v
hisoline intersect, and/or this high speed segment of curve with
the intersection point number of isoline is greater than K, and the sampled point of getting instantaneous velocity minimum in the front side low speed segment of curve adjacent with this high speed segment of curve is speed cut-point; If this high speed segment of curve does not meet described condition, both sides, the front and back low speed segment of curve this high speed segment of curve being adjacent merges the low speed segment of curve that composition is new, and participates in the judgement to next high speed segment of curve with this new low speed segment of curve;
For stroke closely at the uniform velocity, adopt following steps to determine speed cut-point:
Step 4: according to formula
Step 5: according to formula
Calculate the instantaneous velocity value v of stroke smothing filtering sharpening each correction sampled point after treatment
p', wherein
Step 6: according to the instantaneous velocity v of each correction sampled point
p' in coordinate system, set up erection rate curve, described erection rate curve is revised sampled point by each and according to sampling order, each self-corresponding instantaneous velocity is connected and is obtained successively; Calculate the mean value of the instantaneous velocity of all correction sampled points
Step 7: instantaneous velocity is more than or equal to
correction sampled point be labeled as and revise high speed point, calculate and revise at a high speed point relatively
deviation
wherein H' puts at a high speed number, v for revising
i' be i the instantaneous velocity of revising high speed point; Instantaneous velocity is less than
correction sampled point be labeled as and revise low speed point; Calculate and revise low speed point relatively
deviation
wherein L' is for revising low speed point number, v
j' be j the instantaneous velocity of revising low speed point;
Step 8: adopt v
l' isoline is that several are lower than v by erection rate curve segmentation
l' correction low speed segment of curve and several higher than v
l' correction high speed segment of curve; For all correction high speed segment of curve, adopt following methods to judge successively according to sampling order:
For a certain correction high speed segment of curve, if this correction high speed segment of curve satisfies condition: this correction high speed segment of curve and v
h' isoline intersect, and/or this correction high speed segment of curve with
the intersection point number of isoline is greater than K, and the correction sampled point of getting instantaneous velocity minimum in the front side correction low speed segment of curve adjacent with this correction high speed segment of curve is erection rate cut-point; If this correction high speed segment of curve does not meet described condition, the both sides, front and back that this correction high speed segment of curve are adjacent are revised low speed segment of curve and are merged the correction low speed segment of curve that composition is new, and participate in next to revise with this new correction low speed segment of curve the judgement of high speed segment of curve;
Label p by erection rate cut-point obtains in original sample point, corresponding with erection rate cut-point, [(p-1) (T between label area
d-1)+1, (T
d-1) p+1]; Get label interval [(p-1) (T
d-1)+1, (T
d-1) p+1] sampled point of instantaneous velocity minimum is speed cut-point in corresponding original sample point.
2. a kind of stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input according to claim 1, is characterized in that: N is not less than 5.
3. a kind of stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input according to claim 2, is characterized in that: S is 0.015s.
4. a kind of stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input according to claim 3, is characterized in that: ξ
maxvalue is not more than 0.2, v
maxbe not less than 200mm/s.
5. according to a kind of stroke segmentation method based on velocity characteristic for intelligent Freehandhand-drawing input described in claim 3 or 4, it is characterized in that: K is not less than 3 integer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410128457.0A CN103902100B (en) | 2014-04-01 | 2014-04-01 | Stroke segmentation method based on velocity characteristic for intelligence Freehandhand-drawing input |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410128457.0A CN103902100B (en) | 2014-04-01 | 2014-04-01 | Stroke segmentation method based on velocity characteristic for intelligence Freehandhand-drawing input |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103902100A true CN103902100A (en) | 2014-07-02 |
| CN103902100B CN103902100B (en) | 2016-08-17 |
Family
ID=50993460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410128457.0A Active CN103902100B (en) | 2014-04-01 | 2014-04-01 | Stroke segmentation method based on velocity characteristic for intelligence Freehandhand-drawing input |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103902100B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046287A (en) * | 2015-06-29 | 2015-11-11 | 西北工业大学 | Clustering and fitting method of on-line multi-stroke repeatedly drawn sketch |
| CN105957046A (en) * | 2016-05-04 | 2016-09-21 | 西北工业大学 | End point fusion method for intelligent sketching inputting planar three-dimensional projection drawings |
| CN110569800A (en) * | 2019-09-10 | 2019-12-13 | 武汉大学 | detection method of handwriting signal |
| CN111736745A (en) * | 2020-07-27 | 2020-10-02 | 深圳市千分一智能技术有限公司 | Stroke erasing method, device, equipment and readable storage medium |
| CN113487553A (en) * | 2021-06-30 | 2021-10-08 | 武汉楚精灵医疗科技有限公司 | Intestinal tract endoscope withdrawal speed smoothing method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515211A (en) * | 2008-02-22 | 2009-08-26 | 汉王科技股份有限公司 | Handwriting input method, device and user terminal |
| KR20100020389A (en) * | 2008-08-12 | 2010-02-22 | 주식회사 케이티테크 | Input method using size of touching spot and device thereof |
| CN101833788A (en) * | 2010-05-18 | 2010-09-15 | 南京大学 | Three-dimensional human modeling method by using cartographical sketching |
| CN102013010A (en) * | 2010-12-10 | 2011-04-13 | 哈尔滨工业大学深圳研究生院 | Interactive chapter-level handwriting recognition method and system |
| KR101329086B1 (en) * | 2011-01-20 | 2013-11-14 | 케이제이엘이디(주) | Including a multi-touch pad remote control apparatus using by stroke function and it's data processing method |
-
2014
- 2014-04-01 CN CN201410128457.0A patent/CN103902100B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515211A (en) * | 2008-02-22 | 2009-08-26 | 汉王科技股份有限公司 | Handwriting input method, device and user terminal |
| KR20100020389A (en) * | 2008-08-12 | 2010-02-22 | 주식회사 케이티테크 | Input method using size of touching spot and device thereof |
| CN101833788A (en) * | 2010-05-18 | 2010-09-15 | 南京大学 | Three-dimensional human modeling method by using cartographical sketching |
| CN102013010A (en) * | 2010-12-10 | 2011-04-13 | 哈尔滨工业大学深圳研究生院 | Interactive chapter-level handwriting recognition method and system |
| KR101329086B1 (en) * | 2011-01-20 | 2013-11-14 | 케이제이엘이디(주) | Including a multi-touch pad remote control apparatus using by stroke function and it's data processing method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046287A (en) * | 2015-06-29 | 2015-11-11 | 西北工业大学 | Clustering and fitting method of on-line multi-stroke repeatedly drawn sketch |
| CN105046287B (en) * | 2015-06-29 | 2018-05-18 | 西北工业大学 | A kind of online more strokes repeat the cluster and approximating method of skeletonizing |
| CN105957046A (en) * | 2016-05-04 | 2016-09-21 | 西北工业大学 | End point fusion method for intelligent sketching inputting planar three-dimensional projection drawings |
| CN110569800A (en) * | 2019-09-10 | 2019-12-13 | 武汉大学 | detection method of handwriting signal |
| CN110569800B (en) * | 2019-09-10 | 2022-04-12 | 武汉大学 | Detection method of handwriting signal |
| CN111736745A (en) * | 2020-07-27 | 2020-10-02 | 深圳市千分一智能技术有限公司 | Stroke erasing method, device, equipment and readable storage medium |
| CN113487553A (en) * | 2021-06-30 | 2021-10-08 | 武汉楚精灵医疗科技有限公司 | Intestinal tract endoscope withdrawal speed smoothing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103902100B (en) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102663378B (en) | Method for indentifying joined-up handwritten characters | |
| CN101398902B (en) | Natural hand-written Arabian letter on-line identification method | |
| CN101866417B (en) | Method for identifying handwritten Uigur characters | |
| CN103902100A (en) | Speed characteristics based stroke partitioning method for intelligent freehand sketching | |
| CN102013010B (en) | Interactive chapter-level handwriting recognition method and system | |
| CN103400109B (en) | A kind of cartographical sketching identified off-line and shaping methods | |
| CN104899601A (en) | Identification method of handwritten Uyghur words | |
| CN103927537B (en) | Extraction method of self-adapting intersection regions of strokes of Chinese characters | |
| CN104809481A (en) | Natural scene text detection method based on adaptive color clustering | |
| CN105260751B (en) | A kind of character recognition method and its system | |
| CN102750556A (en) | Off-line handwritten form Chinese character recognition method | |
| CN103927056B (en) | Stroke segmentation method based on composite character for intelligence Freehandhand-drawing input | |
| CN105046287B (en) | A kind of online more strokes repeat the cluster and approximating method of skeletonizing | |
| CN103295009B (en) | Based on the license plate character recognition method of Stroke decomposition | |
| CN107092917A (en) | A kind of Chinese-character stroke extraction method based on manifold learning | |
| CN104156730A (en) | Anti-noise Chinese character feature extraction method based on framework | |
| CN107609509A (en) | A kind of action identification method based on motion salient region detection | |
| CN101452531B (en) | Identification method for handwriting latin letter | |
| CN102314252B (en) | Character segmentation method and device for handwritten character string | |
| CN103500323A (en) | Template matching method based on self-adaptive gray-scale image filtering | |
| CN103176651B (en) | A kind of hand-written information Quick Acquisition method | |
| CN107220634A (en) | Based on the gesture identification method for improving D P algorithms and multi-template matching | |
| CN109271882A (en) | A kind of handwritten Chinese character extracting method for distinguishing color | |
| CN101604392B (en) | Chinese character stroke feature parallel extraction method used for on-line handwritten Chinese character recognition | |
| CN104346798B (en) | Objective contour Polygonal Approximation and its 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180104 Address after: 401220 No. 3 group 27, ferry boat street, longevity district, Chongqing Patentee after: Wu Xiaoze Address before: 710072 Xi'an friendship West Road, Shaanxi, No. 127 Patentee before: Northwestern Polytechnical University |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180817 Address after: 518000 Guangdong Shenzhen Longhua New District big wave street Longsheng community Tenglong road gold rush e-commerce incubation base exhibition hall E commercial block 706 Patentee after: Shenzhen step Technology Transfer Center Co., Ltd. Address before: 401220 No. 3 group 27, ferry boat street, longevity district, Chongqing Patentee before: Wu Xiaoze |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181206 Address after: 214026 Tiandi 27-9-1402, Xinghui, Shanghe Garden, Liangxi District, Wuxi City, Jiangsu Province Patentee after: Wuxi Tongchun New Energy Technology Co., Ltd. Address before: 518000 Guangdong Shenzhen Longhua New District big wave street Longsheng community Tenglong road gold rush e-commerce incubation base exhibition hall E commercial block 706 Patentee before: Shenzhen step Technology Transfer Center Co., Ltd. |