CN108492870A - Picture clock test detection method based on digital pen and system - Google Patents
Picture clock test detection method based on digital pen and system Download PDFInfo
- Publication number
- CN108492870A CN108492870A CN201710100530.7A CN201710100530A CN108492870A CN 108492870 A CN108492870 A CN 108492870A CN 201710100530 A CN201710100530 A CN 201710100530A CN 108492870 A CN108492870 A CN 108492870A
- Authority
- CN
- China
- Prior art keywords
- stroke
- profile
- digital
- pointer
- data
- 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
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
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/36—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Character Discrimination (AREA)
Abstract
The present invention provides a kind of picture clock test detection method based on digital pen, and step includes：1) acquisition user draws the handwriting data of clock test；2) profile stroke, pointer stroke and the digital stroke for identifying handwriting data, extract stroke feature；3) according to stroke feature, profile stroke, pointer stroke and digital stroke are detected, realize the test of picture clock.The present invention also provides a kind of, and the picture clock based on digital pen tests detecting system, including：One digital pen, to graphing；One information collecting device, connect above-mentioned digital pen, including a liquid crystal numerical digit screen, a data acquisition applications and a data analysis application, the handwriting data of digital pen is inputted and recorded by liquid crystal display, handwriting data is acquired by data acquisition applications, and the stroke feature of handwriting data is identified by data analysis application and is given a mark.
Description
Technical field
The invention belongs to digital medical fields, and in particular to a kind of inspection of the picture clock test (CDT) based on digital pen track
Survey method and system.
Background technology
The innovation of science and technology plays an important role to the progress of tele-medicine and nursing field.Especially in terms of residential care,
There is technology that some chronic diseases can be monitored and be controlled, This effectively reduces (the references of the cost of medical monitoring
Document：Kayyali,Basel,Zeb Kimmel,and Steve van Kuiken."Spurring the market for
high-tech home health care."McKinsey Quarterly September(2011).).Domestic medicine is nursed
Technology can make people more initiatively be concerned about the health status (bibliography of itself：Horwitz,C.M.,et al."Is
home health technology adequate for proactive self-care."Methods InfMed 47.1
(2008):58-62.).For example people can be in and oneself detect the blood pressure situation of itself using electronic sphygmomanometer to prevent brain soldier
In generation.Suffer from person with diabetes and can also be in and does a simple finger pricking method experiment to detect the blood glucose water of itself
It is flat.
With increasing rapidly for aging populations, cognition dysfunction is such as dull-witted to become a kind of common disease
(bibliography：Esther Landhuis"Prevalence ofDementia,AD,in China Eclipses
Predictions"Alzforum,14Jun 2013,URL:http://www.alzforum.org/news/research-
News/prevalence-dementia-ad-china-eclipses-pr edictions), with it is some other physiologically have it is bright
The disease of aobvious feature is different, and the early stage detecting of cognitive disorder is relatively difficult.Old man and their household can be by dull-witted one
A little early signal such as decrease of memory are considered as age increased normal effects (bibliography：R.E.Powers,
J.W.Ashford and S.Peschin,Memory Matters,Alzheimer’s Foundation of America,
2008.).Patient is identification early stage dull-witted major obstacle to the deviation and scarcity of cognition of diseases.And health doctor is most of
All it is unwilling to implement dull-witted detection means (bibliography inside the office of oneself：P.Iracleous,J.X.Nie,
C.S.Tracy,R.Moineddin,Z.Ismail,K.I.Shulman and R.E.Upshur,Primary care
physicians’attitudes towards cognitive screening:Findings from a national
postal survey,International Journal ofGeriatric Psychiatry 25(1)(2010),23–
29.)。
Medically there are many means of detection cognitive function, wherein it is that a neuropsychologist is used for understanding them to draw
Patient essential information source (bibliography：SMITHAD.On the Use ofDrawing Tasks in
Neuropsychological Assessment[J].American PsychologicalAssociation,2009,23)。
Lai Zhake (Lexak.M.D.) thinks, draws task and is in core status in Neuropsychology.Cognitive function and hand function
There is this requirement of drawing in inspection, so the two combination can save the review time.Cognitive function and motor function in clinic
Evaluation (the especially evaluation of hand function) often influence each other, any function of single analysis is unfavorable for the visitor to clinical general symptom
See understanding.
It is a common, easy-operating cerebral nervous system detection instrument to draw clock test, what this test request was tested
People independently draws a clock on paper, and marks the specified time as requested.It is simple to draw clock test implementation, but completes it
Need the participation of many cognitive functions.The test simultaneously and cultural correlation are smaller, and the people of acceptance test need not necessary culture
It is horizontal, it is only necessary to simple deictic word can be understood, so that it may to complete test to draw the clock oneself praised.
However the current way of neurological examination is still to be operated in a manner of before 10 years, patient is required with pen one
Picture clock test is carried out on a given paper, is needed direct surveillance and is analyzed, labor intensive, energy judge that speed is slow, and
And judging that there is very strong subjectivity, different case results may have greater difference, error is larger.
The fast development of information technology promotes the progress of nervous system prevention and diagnostic techniques.Novel interaction technique side
Help others obtain the information that conventional method is unable to get, and convert fuzzy qualitative data to digitized quantitative number
According to.Digitlization input equipment provides a drafting input mode that can not be substituted, it can automatically be collected in two dimensional X-Y plane
Interior accurate handwriting data can effectively quantify cognition and hand movement function data, improve the comprehensive of disease information judgement
And objectivity.The present invention provides a kind of solution thus.
Invention content
The purpose of the present invention is to provide a kind of picture clock test detection method and system based on digital pen, are drawing clock test
When acquire digital pen handwriting data, can effectively quantify cognition with hand movement function data, improve disease information judge it is comprehensive
Conjunction property and objectivity.
In order to solve the above technical problems, technical scheme is as follows：
A kind of picture clock test detection method based on digital pen, step include：
1) acquisition user draws the handwriting data of clock test；
2) profile stroke, pointer stroke and the digital stroke for identifying handwriting data, extract stroke feature；
3) according to stroke feature, profile stroke, pointer stroke and digital stroke are detected, realize the test of picture clock.
Further, Stroke discrimination is carried out using RandomForest graders.
Further, data of identifying the handwriting before identification are pre-processed, including：
1) data of identifying the handwriting carry out resampling so that the distance between sampled point is equal to T；
2) data of identifying the handwriting are smoothed, and reduce the degree of stroke shake.
Further, the method for Stroke discrimination is：
1) the contour identification stroke in total stroke set S1 obtains profile stroke set O, finds the redundancy pen of profile stroke
Album closes RS1, and stroke set O and RS1 are got rid of in S1, obtains stroke set S2, i.e. S2=S1- (O+RS1)；
2) pointer stroke is identified in stroke set S2, obtains pointer stroke set H, finds the redundancy stroke of pointer stroke
Set RS2 gets rid of stroke set H and RS2 in S2, obtains stroke set S3, i.e. S3=S2- (H+RS2)；
3) digital stroke is identified in stroke set S3.
Further, mechanism is given a mark to profile stroke, pointer stroke and digital stroke using the picture clock of BJ Union Hospital
It is detected, which divides, and wherein profile stroke, pointer stroke, digital stroke respectively account for 1 point.
Further, the stroke feature of profile stroke includes：1) it whether there is profile stroke；2) whether profile stroke closes
It closes；3) stroke quantity N1；4) ratio R 1 of the height H and width W of profile stroke closure；5) profile stroke length L and profile pen
Draw the ratio R 2 of the inscribe circumference P of closure.
Further, the stroke feature of pointer stroke includes：1) whether hour hands stroke intersects with minute hand stroke；2) hour hands pen
Draw the ratio R 3 of length HL and minute hand stroke length ML；3) the quadrant Quadrant1 where hour hands stroke；4) where minute hand stroke
Quadrant Quadrant2.
Further, the stroke feature of digital stroke includes：Correct digital numerical N2.
Further, pointer stroke is split using ShortStraw methods, obtains hour hands stroke and minute hand stroke.
Further, digital stroke is polymerize according to the difference of affiliated number by K-means algorithms.
A kind of picture clock test detecting system based on digital pen, including：
One digital pen, to graphing；
One information collecting device connects above-mentioned digital pen, including a liquid crystal numerical digit screen, a data acquisition applications and a data
Analysis application, inputs by liquid crystal display and records the handwriting data of digital pen, and person's handwriting number is acquired by data acquisition applications
According to, and the stroke feature of handwriting data is identified by data analysis application and is given a mark.
Further, information collecting device is also showing that drafting task, task include：When drawing the profile of clock, drawing
Between number, draw pointer.
Further, drawn using Microsoft Tablet PC Platform SDK 1.7 as the Stroke discrimination of number
It holds up.
The beneficial effects of the invention are as follows：Compared with prior art, the present invention is obtained using digital pen and information collecting device
The Data Detection accuracy of picture clock test can be improved in accurate handwriting data.By data analysis application and recognizer and beat
Extension set system, it can be achieved that draw clock test handwriting data analysis and diagnosis automation, reduce tradition draw clock test it is required
Time and manpower.
Description of the drawings
Fig. 1 is the handwriting data acquisition method schematic diagram for drawing clock test of the present invention.
Fig. 2 is that the picture clock of the present invention tests detection method flow chart.
Fig. 3 is point set closure schematic diagram.
Fig. 4 is the person's handwriting figure that user draws clock test.
Fig. 5 is the effect diagram of ShortStraw algorithms.
Fig. 6 is the effect diagram of digital recognizer.
Specific implementation mode
To enable objects, features and advantages of the present invention to be clearer and more comprehensible, hereafter by specific embodiment, and combine attached
Figure, is described in detail.
The present embodiment provides a kind of picture clock test detection method and system based on digital pen, number is acquired when drawing clock test
The handwriting data of position pen, can effectively quantify cognition and hand movement function data, improve the comprehensive and objective of disease information judgement
The property seen.The handwriting data for acquiring digital pen carries out stroke feature extraction, it is established that be based on content understanding in conjunction with person's handwriting recognizer
CDT data models, establish basis for cognition-motor function evaluation method, as shown in Figure 1.
The present embodiment operating process is as shown in Fig. 2, specific as follows：
One, clock is drawn on liquid crystal numerical digit screen using digital pen, acquires the initial data of person's handwriting.
1. user opens data acquisition applications；
After 2. user's selection starts, the drafting task of user is shown on information collecting device：Specially " draw the wheel of clock
It is wide " " number that the time is indicated on profile subscript " " setting pointer makes the time showing of clock be 11：10”；
3. user carries out picture clock on liquid crystal numerical digit screen, user handwriting data during which can be recorded, which sits for X-Y
The bivector (X, Y) of system is marked, as shown in Figure 4.
In the present embodiment, digital pen uses Wacom KP-701E；Liquid crystal numerical digit screen uses Wacom
13HD, screen size are 13.0 cun, and resolution ratio is 1280 × 800, and operating system is Winodws 8, but be not limited to that
The equipment can use the other equipment that can acquire same information.
Two, initial data is pre-processed：
One) resampling is carried out to initial data, algorithm is as follows：
1. finding the closure rectangle of original sample point first, and calculate the catercorner length BbDiagonal of the rectangle.
Wherein, X is enabledmaxFor the maximum value of X in crude sampling point set Points, XminFor minimum value, YmaxFor crude sampling
The maximum value of Y, Y in point set PointsminFor minimum value, then the closure rectangle of original sample point is by point (Xmin,Ymin) and
(Xmax,Ymax) as the rectangle constituted to angular vertex, see attached drawing 3.
Wherein catercorner length BbDiagonal can be obtained by the following formula：
2. it is 0 to initialize sampled point distance D, by first original sample point Point0It is added to resampling point set
In RPoints.
3. calculating adjacent 2 points Point in crude sampling point set Points using above-mentioned formulai, Pointi-1Between
D is added in D (i.e. D=D+d) by Euclidean distance d.
4. if D is less than threshold value T, so that i is increased by 1, again rebound step 3, wherein T is the 1/40 of BbDiagonal.
5. otherwise just generating a new point Pointj, the PointjThe distance between upper resampling point needs approximate
Equal to T, by PointjIt is added in resampling point set RPoints, and is inserted into point PointiBefore, step is jumped back to again
Rapid 3.
Two) initial data is smoothed：
Three continuous point Point are taken out from crude sampling point set Pointsi-1, Pointi, Pointi+1, calculate
The average value of three point X and Y, respectively xj, yj.Generate a new point Pointj, its X, Y are not originally set as xj, yj, will
PointjIt is added to smooth sampled point set SPoints.
Three, person's handwriting identifies：
It identifies the handwriting and carries out Stroke discrimination, process is：The contour identification stroke in total stroke set S1, obtains profile stroke set
O is closed, the redundancy stroke set RS1 of the profile stroke about O is found before and after the O in each 3 strokes, stroke set is got rid of in S1
O and RS1 is closed, stroke set S2, i.e. S2=S1- (O+RS1) are obtained；Pointer stroke is identified in stroke set S2, obtains pointer
Stroke set H, H include hour hands set H1 and minute hand set H2, found in each 3 strokes before and after the H1 and H2 respectively about H1 and
The redundancy stroke of H2 is simultaneously included into pointer stroke redundant set RS2 together, and stroke set H and RS2 are got rid of in S2, obtains stroke
Set S3, i.e. S3=S2- (H+RS2)；Digital stroke is identified in stroke set S3.It is specific as follows：
One) profile Stroke discrimination：
By to each analysis for drawing clock file, finding general longest stroke in profile stroke.Specific algorithm is such as
Under：
1. finding longest stroke Stroke first from stroke set S1l, it is added into profile stroke set O.Judge O
Whether it is closed, if be not closed, continues to execute step 2, otherwise jump to step 6；
2. with StrokelOn the basis of respectively forwardly backward find N (N takes empirical value 3) a stroke, as candidate stroke set
C；
3. selection and Stroke from ClBoth ends apart from nearest stroke Strokei；
4. if StrokeiWith StrokelThe distance between be less than threshold value T, then by StrokeiO is added, is deleted from C
StrokeiAnd step 5 is continued to execute, otherwise jump to step 6；
5. judging whether O is closed, if be not closed, step 3 is executed, step 6 is otherwise continued to execute；
6. preserving Outline and exiting.
Two) pointer Stroke discrimination：
The main problem of pointer Stroke discrimination is that user may go out hour hands and minute hand using unicursal, this in order to solve
Problem needs first to be split stroke.(bibliography is split to stroke using ShortStraw methods：Wolin,
Aaron,Brian Eoff,and Tracy Hammond."ShortStraw:A Simple and Effective Corner
Finder for Polylines."SBM.2008.)。
ShortStraw methods are briefly described at this：Resampling is carried out to stroke Stroke first, method is as above
Shown in text, resampling point set RPoints is obtained.Point set is taken out from RPoints using the sliding window that length is WSuch as fruit dotWith pointThe distance between D and point set length L
Ratio R be less than threshold value T1, then it is assumed that point PointiFor a cut-point of stroke Stroke, Fig. 5 is to use ShortStraw
The design sketch of method, wherein (b) be segmentation after design sketch.
Wherein the value of T1 be in set RPoints between all consecutive points apart from 0.85 times of mean value.
The main algorithm of pointer Stroke discrimination is：The central point of profile stroke closure is Center, is found with Center's
Distance is less than the stroke of threshold value T2, constitutes the candidate stroke set C1 of pointer stroke, and all strokes in C1 are used
ShortStraw methods are split to obtain new stroke set C2, and longest stroke is found from C2 as minute hand stroke, is sought
The stroke of vice-minister is found as hour hands stroke, minute hand stroke and hour hands stroke composition pointer stroke set H.
Wherein the value of T2 is the 1/10 of the point set RPoints closure rectangle minimum length of sides.
Three) digital stroke identifies：
First main problem of digital stroke identification is to need stroke according to the difference of affiliated number polymerizeing,
Referring to Fig. 6, wherein (a) figure is initial data, it is (b) result after polymerizeing to stroke.
Polymerization is as follows：Using between stroke space length D and the temporal sequence S of stroke as two measure
Variable obtains the polymerization result of stroke using K-means algorithms.Wherein it is using D and S as the reason of gauge variable：1)
From number structure from, it is same number in stroke it is closely coupled, the stroke between different digital have larger space away from
From；2) from person writing's number traditionally from the point of view of, because of the stroke limited amount (generally 1 to 2) in a number,
People draw next stroke again after often having drawn a stroke, so in time sequencing, in the same number
Stroke also should be closely coupled.
We can obtain the set Clusters of stroke aggregation after polymerizeing to stroke, in next step needs pair
Each stroke aggregation in Clusters is identified, this belongs to Symbol recognition.Have very about the method for Symbol recognition at present
More (bibliography：Chhabra,Atul K."Graphic symbol recognition:An overview."Graphics
Recognition Algorithms andSystems.SpringerBerlin Heidelberg,1998.68-79.).At this
In system, from convenient angle is realized, identifications of the TabletPC Platform SDK 1.7 as individual digit is had chosen
Engine, but not to be limited.
Four) redundancy stroke differentiates：
Target stroke Stroke is judged by the following methodtWhether be stroke set U redundancy stroke：
First by stroke set U={ Stroke1,Stroke2,...StrokenIn stroke linked according to sequencing
Stroke is drawn at a backboneu, both by U in addition to the last one stroke StrokenOuter arbitrary stroke StrokeiTerminating point
With next stroke Strokei+1Starting point link.
For Strokeu, evenly distributedly take wherein 10 sampled points to be added in point set UPoints, calculate UPoints
Middle all the points and StroketDistance mean value DmeanIf DmeanLess than threshold value T3, then it is assumed that StroketIt is stroke set U
Redundancy stroke.
The wherein any point UPoints and StroketDistance definition be the point to StroketThe distance at middle any point
Minimum value.The value of threshold value T3 is StroketThe 1/20 of length.
The respective stroke feature that profile stroke, pointer stroke and digital stroke are extracted from above-mentioned four step (sees below step
Suddenly), according to these stroke features assess and give a mark as follows.
Four, it gives a mark to drawing situation
One) profile stroke is given a mark：
Judged using the feature being mentioned above, wherein judging whether that the method being closed is as follows：For profile stroke
Outline, it includes sampling point set be combined into Pointsoutline, from first point Point0L3 point, composition sampling are taken backward
Point set Pointsstart, from the last one sampled point Point|points|-1L3 point is taken forward, forms sampled point set
Pointsend, calculate two sampled point set Pointsstart, PointsendThe distance between D1, distance calculation formula it is as follows：
Euclid between wherein Distance (Point1, Point2) function representation point Point1 and Point2 away from
From.The D1 calculated is made comparisons with T4, if D1 is not more than T4, then it is assumed that profile stroke is closed.
The value of threshold value T4 is the 1/20 of BbDiagonal.
The ratio R 2 of profile stroke length L and the inscribe circumference P of profile stroke closure are calculated, method is as follows：Profile pen
The length L of Outline is drawn to be added successively for adjacent 2 points of Euclidean distance in wherein sampled point set.Inscribe circumference P
=π × min (width, height), wherein width, height are respectively the width and height of profile stroke Outline closures,
R2=L/P.
When meeting following condition (stroke feature)：1) there are profile strokes；2) profile stroke is closed；3)N1<T5；4) profile
The ratio R 1,1/T6 of the height H and width W of stroke closure<=R1<=T6；5) profile stroke length L and profile stroke closure
The ratio R 2,1/T7 of inscribe circumference P<=R2<=T7.Empirical value T5=5, T6=T7=1.5 are wherein taken according to experimental data.
1 score is made to profile at this time, otherwise makes 0 score.
Two) pointer stroke is given a mark：
Judged using feature mentioned above, wherein judging hour hands stroke HourHand and minute hand stroke
The method whether MinuteHand intersects is：Find out first sampled point Point of HourHand strokes0, the last one sampling
Point Point1；First sampled point Point of MinuteHand2, the last one sampled point Point3。
If D2<T4 then thinks two stroke intersections.
The computational methods of quadrant are where hour hands：It is obtained using formula (3)WithBetween with clockwise side
To quadrant where the rotation angle RotatedAngle1, HourHand of rotationMinute hand institute
It is identical as hour hands computational methods in quadrant computational methods, where quadrant be Quadrant2.
When meeting following condition (stroke feature)：1) hour hands stroke intersects with minute hand stroke；2) hour hands stroke length HL with
Minute hand stroke length ML ratio Rs 3 are less than 1；3) the quadrant Quadrant1, Quadrant1=4 where hour hands stroke；4) minute hand pen
Quadrant Quadrant2, Quadrant2=1 where drawing.When meeting conditions above, 1 score is made to pointer stroke, otherwise makes 0 score.
Three) digital stroke is given a mark：
Judged using feature mentioned above, the method for wherein correct judgment digital numerical is as follows：Digital stroke
Digit is correctly mainly by two conditional decisions：1) digital value of digital stroke Digit represented is i, 1<=i<=12, and i has
And only there are one；2) position of digital stroke Digit is correct.
Its conditional 2) detection method is：The central point Center1 of profile stroke closure is obtained, digital stroke Digit is closed
The central point Center2 of packet.Center2-Center1 is obtained into vectorIt chooses (0, -1) and is used as reference vectorIt chooses (0, -1) and is used as benchmark, because the coordinate system of liquid crystal Digitizing plate is using downwards as the positive direction of Y-axis.Using such as
Lower formula can obtainWithBetween the rotation angle that rotates in a clockwise direction：
After obtaining rotation angle RotatedAngle, when 1<=i<When=11, if
Then the positions digital stroke Digit are correct.As i=12, if 0<=RotatedAngle<=L8 or 2 × π-L8<=
RotatedAngle<=2 × π, then the positions digital stroke Digit are correct.Wherein
When meeting condition：The quantity N2 of correct digit>It when=T8, then gives number to make 1 score, otherwise digital makes 0 score to give.T8
Take empirical value 6.
The detection method and system of the present invention are described in detail above by embodiment, but the specific reality of the present invention
Existing form is not limited thereto, and those of ordinary skill in the art can be without departing from the spirit and principles of the present invention
Various obvious variations and modification are carried out to it.Protection scope of the present invention is subject to described in claims.
Claims (10)
1. a kind of picture clock based on digital pen tests detection method, step includes：
1) acquisition user draws the handwriting data of clock test；
2) profile stroke, pointer stroke and the digital stroke for identifying handwriting data, extract stroke feature；
3) according to stroke feature, profile stroke, pointer stroke and digital stroke are detected, realize the test of picture clock.
2. according to the method described in claim 1, it is characterized in that, carrying out Stroke discrimination using RandomForest graders.
3. according to the method described in claim 1, it is characterized in that, data of identifying the handwriting before identification are pre-processed, including：
1) data of identifying the handwriting carry out resampling so that the distance between sampled point is equal to T；
2) data of identifying the handwriting are smoothed, and reduce the degree of stroke shake.
4. according to the method described in claim 1, it is characterized in that, the method for Stroke discrimination is：
1) the contour identification stroke in total stroke set S1 obtains profile stroke set O, finds the redundancy stroke set of profile stroke
RS1 is closed, stroke set O and RS1 are got rid of in S1, obtains stroke set S2, i.e. S2=S1- (O+RS1)；
2) pointer stroke is identified in stroke set S2, obtains pointer stroke set H, finds the redundancy stroke set of pointer stroke
RS2 gets rid of stroke set H and RS2 in S2, obtains stroke set S3, i.e. S3=S2- (H+RS2)；
3) digital stroke is identified in stroke set S3.
5. according to the method described in claim 1, it is characterized in that, giving a mark mechanism to profile using the picture clock of BJ Union Hospital
Stroke, pointer stroke and digital stroke are detected, which divides, wherein profile stroke, pointer stroke, number
Stroke respectively accounts for 1 point.
6. according to the method described in claim 1, it is characterized in that,
The stroke feature of profile stroke includes：1) it whether there is profile stroke；2) whether profile stroke is closed；3) stroke quantity
N1；4) ratio R 1 of the height H and width W of profile stroke closure；5) inscribed circle of profile stroke length L and profile stroke closure
The ratio R 2 of perimeter P；
The stroke feature of pointer stroke includes：1) whether hour hands stroke intersects with minute hand stroke；2) hour hands stroke length HL with point
The ratio R 3 of needle stroke length ML；3) the quadrant Quadrant1 where hour hands stroke；4) quadrant where minute hand stroke
Quadrant2；
The stroke feature of digital stroke includes：Correct digital numerical N2.
7. according to the method described in claim 1, it is characterized in that, being divided pointer stroke using ShortStraw methods
It cuts, obtains hour hands stroke and minute hand stroke；Digital stroke is gathered according to the difference of affiliated number by K-means algorithms
It closes.
8. a kind of picture clock based on digital pen tests detecting system, including：
One digital pen, to graphing；
One information collecting device connects above-mentioned digital pen, including a liquid crystal numerical digit screen, a data acquisition applications and a data analysis
Using, the handwriting data of digital pen is inputted and recorded by liquid crystal display, and handwriting data is acquired by data acquisition applications, and
The stroke feature of handwriting data is identified by data analysis application and is given a mark.
9. system according to claim 8, which is characterized in that information collecting device is also showing drafting task, task
Including：The profile of clock is drawn, time figure is drawn, draws pointer.
10. system according to claim 8, which is characterized in that use Microsoft Tablet PC Platform
Stroke discrimination engines of the SDK 1.7 as number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710100530.7A CN108492870B (en) | 2017-02-23 | 2017-02-23 | Method and system for testing and detecting picture clock based on digital pen |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710100530.7A CN108492870B (en) | 2017-02-23 | 2017-02-23 | Method and system for testing and detecting picture clock based on digital pen |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108492870A true CN108492870A (en) | 2018-09-04 |
| CN108492870B CN108492870B (en) | 2022-01-18 |
Family
ID=63343885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710100530.7A Active CN108492870B (en) | 2017-02-23 | 2017-02-23 | Method and system for testing and detecting picture clock based on digital pen |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108492870B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409329A (en) * | 2018-11-14 | 2019-03-01 | 科大讯飞股份有限公司 | Method of discrimination, appraisal procedure, device, equipment and the storage medium of figure topic |
| CN110384484A (en) * | 2019-07-25 | 2019-10-29 | 科大讯飞股份有限公司 | A kind of pattern evaluation method, device, equipment and storage medium |
| CN110859599A (en) * | 2019-11-26 | 2020-03-06 | 复旦大学 | Cerebrovascular disease nerve injury patient cognitive function automatic screening system |
| CN111091910A (en) * | 2019-12-17 | 2020-05-01 | 中国科学院自动化研究所 | Intelligent evaluation system based on drawing clock test |
| CN113535042A (en) * | 2020-09-18 | 2021-10-22 | 厦门市和家健脑智能科技有限公司 | Method and device for generating image based on old people cognitive recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008115555A1 (en) * | 2007-03-21 | 2008-09-25 | Massachusetts Institute Of Technology | Method and apparatus for measuring representational motions in a medical context |
| CN105046730A (en) * | 2015-07-09 | 2015-11-11 | 北京盛世宣合信息科技有限公司 | Writing trace demonstration method and apparatus applied to writing brush |
| CN105807957A (en) * | 2016-03-01 | 2016-07-27 | 上海卓易科技股份有限公司 | Input method and intelligent pen |
| CN106446566A (en) * | 2016-09-29 | 2017-02-22 | 北京理工大学 | Elderly cognitive function classification method based on random forest |
-
2017
- 2017-02-23 CN CN201710100530.7A patent/CN108492870B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008115555A1 (en) * | 2007-03-21 | 2008-09-25 | Massachusetts Institute Of Technology | Method and apparatus for measuring representational motions in a medical context |
| CN105046730A (en) * | 2015-07-09 | 2015-11-11 | 北京盛世宣合信息科技有限公司 | Writing trace demonstration method and apparatus applied to writing brush |
| CN105807957A (en) * | 2016-03-01 | 2016-07-27 | 上海卓易科技股份有限公司 | Input method and intelligent pen |
| CN106446566A (en) * | 2016-09-29 | 2017-02-22 | 北京理工大学 | Elderly cognitive function classification method based on random forest |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409329A (en) * | 2018-11-14 | 2019-03-01 | 科大讯飞股份有限公司 | Method of discrimination, appraisal procedure, device, equipment and the storage medium of figure topic |
| CN110384484A (en) * | 2019-07-25 | 2019-10-29 | 科大讯飞股份有限公司 | A kind of pattern evaluation method, device, equipment and storage medium |
| CN110859599A (en) * | 2019-11-26 | 2020-03-06 | 复旦大学 | Cerebrovascular disease nerve injury patient cognitive function automatic screening system |
| CN110859599B (en) * | 2019-11-26 | 2022-05-20 | 复旦大学 | Automatic cognitive function screening system for cerebrovascular disease nerve injury patients |
| CN111091910A (en) * | 2019-12-17 | 2020-05-01 | 中国科学院自动化研究所 | Intelligent evaluation system based on drawing clock test |
| CN111091910B (en) * | 2019-12-17 | 2023-11-03 | 中国科学院自动化研究所 | Intelligent evaluation system based on painting clock test |
| CN113535042A (en) * | 2020-09-18 | 2021-10-22 | 厦门市和家健脑智能科技有限公司 | Method and device for generating image based on old people cognitive recognition |
| CN113535042B (en) * | 2020-09-18 | 2023-09-22 | 厦门市和家健脑智能科技有限公司 | Method and device for generating cognitive recognition image based on old people |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108492870B (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108492870A (en) | Picture clock test detection method based on digital pen and system | |
| CN104382570B (en) | Digital full-automatic health status detection device | |
| CN110853111B (en) | Medical image processing system, model training method and training device | |
| CN101129261A (en) | Device and method for acquiring and recognizing pulsation information and tongue inspection information | |
| Corradi et al. | Lung ultrasound and B-lines quantification inaccuracy: B sure to have the right solution | |
| WO2019047365A1 (en) | Medical cloud platform-based image big data analysis system and method | |
| CN100377691C (en) | Blood flow dynamic analyzer and its method, and image diagnostic apparatus | |
| Cheng et al. | Atrial fibrillation identification with PPG signals using a combination of time-frequency analysis and deep learning | |
| CN112309566A (en) | Remote automatic diagnosis system and method for intelligent image recognition and intelligent medical reasoning | |
| CN114358194A (en) | Gesture tracking based detection method for abnormal limb behaviors of autism spectrum disorder | |
| Tobias et al. | CNN-based deep learning model for chest X-ray health classification using tensorflow | |
| CN103735247B (en) | For the palm colour analysis method of health status identification | |
| WO2018176717A1 (en) | Medical cloud platform-based breast screening image analysis system and method | |
| CN111180026A (en) | Special diagnosis and treatment view system and method | |
| Radha | Analysis of COVID-19 and pneumonia detection in chest X-ray images using deep learning | |
| WO2008138219A1 (en) | Unifying and standardizing normal range reference value and actual measurement value of detection or laboratory report in clinical medicine | |
| CN107970027A (en) | A kind of radial artery detection and human body constitution identifying system and method | |
| CN108962379B (en) | Mobile phone auxiliary detection system for cranial nerve system diseases | |
| CN107045393A (en) | Line test multi-channel data collecting method and system based on digital pen | |
| Kumar et al. | CACHET-CADB: A contextualized ambulatory electrocardiography arrhythmia dataset | |
| CN112651976A (en) | Focal region brain network determination method and system based on low-resolution nuclear magnetic data | |
| WO2018176718A1 (en) | Intelligent identification system and method for mammary gland screening image | |
| Nguyen et al. | Quantitative assessment of cerebellar ataxia with kinematic sensing during rhythmic tapping | |
| CN113995383B (en) | Internet of things-based traditional Chinese medicine remote inspection informatization processing system and method | |
| KV et al. | Deep Learning Approach to Nailfold Capillaroscopy Based Diabetes Mellitus Detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |