CN107274503A - A kind of Work attendance method based on track and sensing data - Google Patents
A kind of Work attendance method based on track and sensing data Download PDFInfo
- Publication number
- CN107274503A CN107274503A CN201610207477.6A CN201610207477A CN107274503A CN 107274503 A CN107274503 A CN 107274503A CN 201610207477 A CN201610207477 A CN 201610207477A CN 107274503 A CN107274503 A CN 107274503A
- Authority
- CN
- China
- Prior art keywords
- work attendance
- similarity
- attendance person
- person
- work
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
Abstract
The present invention is a kind of Work attendance method based on track and sensing data, is related to educational management technical field, and the process of work attendance need not be carried out successively by realizing, and improved the efficiency of work attendance, solved largely the potential safety hazard existed during by work attendance person successively work attendance.In the present invention, reach work attendance time point Tk, according to determined by the track of work attendance person belonged to by work attendance person turn out for work, turn out for work it is undetermined, late it is undetermined in which kind of situation, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, differentiated using the sensing data of mobile terminal belonged to by work attendance person turn out for work, which kind of situation in allograph.Reach lagged time point Tc, differentiate the track by work attendance person for class undetermined of being late, according to by the track of work attendance person differentiate by work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, the late situation undetermined of differentiation can not be differentiated according to track, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal.
Description
Technical field
The present invention relates to educational management technical field, the automatic management technical field more particularly to educated.
Background technology
Routine attendance check result include turn out for work, be late, leaving early, it is absent from duty（The absence from duty of student, which is also referred to as, cuts classes）, ask for leave, allograph etc.
Situation.
Work attendance mode, use now have artificial roll-call, it is hand-written register, fingerprint mode, mode of checking card, recognition of face mode
Deng more commonly used by the way of manually calling the roll in education sector.
For fingerprint mode, mode of checking card, recognition of face mode, it is necessary to be equipped with specific fingerprint identification device, reading
Card machine, recognition of face instrument equipment.
Most of mode is required for being individually identified by the process of work attendance person carrying out work attendance, is required for carrying out successively, such as
Artificial call the roll will be carried out successively, such as fingerprint mode, the mode of checking card by fingerprint or will check card successively, such as recognition of face mode is first
Camera was faced by work attendance person before this, system acquisition is to by the clear image of face of work attendance person, and then camera will be by work attendance person
Clear image of face be sent to system and be identified, notify to be completed by work attendance person's work attendance after confirmation.When by work attendance person's quantity compared with
When many, it is often necessary to spend more time, or even the situation such as crowded occur, cause potential safety hazard occur.
Also the mode based on track carries out work attendance, and work attendance situation can be judged by the track data of work attendance time.
This mode is more effective for staff on official business.But for the industries such as education, factory exist it is very big the problem of, that is, exist
One people holds the situation that multiple location equipments carry out allograph work attendance.
The content of the invention
In order to improve the deficiency of above-mentioned existing scheme means, the invention provides a kind of based on track and sensing data
Work attendance method.
Method needs the mobile terminal software used server and carried by work attendance person, and mobile terminal is smart mobile phone, flat board electricity
Equipment that brain etc. can be carried with, can connecting internet, mobile terminal has the biography such as track record, acceleration, gyroscope
Sensor.
Track is a series of location point in the spaces gathered using track record device, each point at least include the date, when
Between, longitude, latitude, altitude information etc., wherein the time of a single point, date, longitude, latitude just constitute a location data.
Track record device such as cellphone GPS module etc..
A kind of Work attendance method based on track and sensing data, it is characterised in that comprise the following steps：
1）By work attendance person's typing by work attendance person's information, mobile terminal software is read by the IMSI codes and/or IMEI of work attendance person mobile terminal
Code, by the IMSI codes and/or IMEI code upload server of work attendance person, work attendance person's typing is set and examined by work attendance person's information, work attendance person
Diligent require information, sets readiness time point Ty, work attendance time point Tk, lagged time point Tc, end time point Tx, and work attendance place
Scope F；
2）Submit application of asking for leave to server by mobile terminal software before readiness time point Ty by work attendance person, work attendance person will ratify
It is sent to by the mobile terminal software of work attendance person, mobile terminal software is received after approval, that is, stops carrying out work attendance by work attendance person to this；
3）Reach after readiness time point Ty, running software initialization process in mobile terminal, makes its kimonos at the automatic calibration mobile terminal time
Being engaged in, device is synchronous, and startup attendance record subprocess is read by the location data and sensing data of work attendance person mobile terminal, if read
Failure, mobile terminal software is by this message upload server；If read normally, start to read exercise data and the movement of sensor
Hold the location data of locating module；
4）Reach after work attendance time point Tk, according to differentiated by the track of work attendance person belonged to by work attendance person turn out for work, turn out for work it is undetermined, late
Which kind of situation in undetermined, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, is sentenced using the sensing data of mobile terminal
Do not belonged to by work attendance person turn out for work, which kind of situation in allograph；
5）Reach after lagged time point Tc, differentiate the track by work attendance person for class undetermined of being late, sentence according to by the track of work attendance person
Not by work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, can not differentiate that the late of differentiation waits according to track
Condition, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal；
6）Inquiry belong to turn out for work, be late, allograph situation by work attendance person's location data in lagged time point Tc to end time point
It is fixed if had by the location data of work attendance person in work attendance outside point range F whether in work attendance in point range F during Tx
To leave early.
By work attendance person reach work attendance point range F time be Td, the examination period starting point Tdh1 to Td before Td
The period between examination period end point Tdh2 afterwards is examination period Tdh.
The parameter for characterizing similarity is a lot, all has dependency relation with similarity, what is had is proportionate, what is had is negatively correlated,
For example Euclidean distance is smaller, and similarity is bigger；Coefficient correlation is bigger, and similarity is bigger；The phase that dynamic time adjustment algorithm is calculated
Smaller like number of degrees value, similarity is bigger.
, be according to being just so characterize parameter and when the belonging to which kind of situation when being differentiated more afterwards of threshold value of similarity
Related or negative correlation is determined.Allograph is a kind of big behavior of similarity, and the parameter and similarity for characterizing similarity are negatively correlated
, numerical value is smaller, and similarity is bigger, and allograph possibility is bigger, and the numerical value for characterizing the parameter of similarity is less than threshold value, is judged as
Allograph suspicion；The parameter and similarity for characterizing similarity are positively related, and numerical value is bigger, and similarity is bigger, and allograph possibility is bigger,
The numerical value for characterizing the parameter of similarity is more than threshold value, is judged as allograph suspicion.
A kind of Work attendance method based on track and sensing data, it is characterised in that：
Reach after work attendance time point Tk, according to differentiated by the track of work attendance person belonged to by work attendance person turn out for work, undetermined, late treat of turning out for work
Which kind of situation in fixed, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, differentiated using the sensing data of mobile terminal
Belonged to by work attendance person turn out for work, which kind of situation in allograph, differentiate that process is as follows：
Whether the location data by work attendance person when inquiring about work attendance time point Tk first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this is set to by work attendance person turns out for work；
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to and turn out for work；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to and turns out for work；
By by work attendance person with examination period Tdh in reach other by the sign similarity of the track of the track of work attendance person
Parameter and threshold value are compared, if characterizing the parameter and similarity correlation of similarity, and characterize the ginseng of similarity
Number be more than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh
The similarity of interior sensing data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes phase
Seemingly the parameter of degree is less than threshold value, then carries out being examined by work attendance person by other of work attendance person with reaching in examination period Tdh
The similarity of sensing data in period Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, be then set to and turn out for work, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and the parameter of sign similarity is more than threshold value, then is set to and turns out for work,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms；
If in point range F, not being set to late class undetermined in work attendance by the location data of work attendance person during work attendance time point Tk.
A kind of Work attendance method based on track and sensing data, it is characterised in that：
Reach after lagged time point Tc, differentiate the track by work attendance person for class undetermined of being late, according to by the track differentiation of work attendance person
By work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, can not differentiate that the late of differentiation waits according to track
Condition, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal, differentiates that process is as follows：
Whether the location data by work attendance person when inquiring about lagged time point Tc first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
Inquiry within examination period Tdh by work attendance person other whether entrance work attendance place model is had by the location data of work attendance person
Enclose in F,
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this be set to by work attendance person it is late,
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to late；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to late；
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be more than
Threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh in biography
The similarity of sensor data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes similarity
Parameter be less than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination the period
The similarity of sensing data in Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, then be set to it is late, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and characterize the parameter of similarity and be more than threshold value, then be set to it is late,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms；
Checking-in result can be returned to by work attendance person, thought not being inconsistent with the fact by work attendance person, can be linked up and be corrected with work attendance person.
A kind of Work attendance method based on track and sensing data, it is characterised in that：
The scope for examining period Tdh is between -120 seconds 0.1 second.
A kind of Work attendance method based on track and sensing data, it is characterised in that：
Server use by the location data of work attendance person, track, sensing data, from server to soft by work attendance person mobile terminal
Part calls or is actively sent to server by work attendance person mobile terminal software.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described work attendance ground point range F, by
The accurate location information in work attendance place is determined plus location data error.Such as：The longitude and latitude in work attendance place is（N,W）, use
There is absolute error J in localization method, then work attendance point range F be with point（N,W）Centered on radius be J similar round region.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described location data is to use to defend
One or several kinds of knots in star positioning method, the method positioned based on network base station, the localization method based on WLAN
What the method for conjunction was obtained.
Positioning method has a variety of, such as american global positioning system（GPS）Positioning, Russian Glonass satellite navigation system
（GLONASS）Positioning, Chinese Beidou satellite navigation system（BDS）Positioning, European Union's galileo satellite navigation system（GALILEO）It is fixed
Position, the location technology based on Cell ID, LBS architectures, the location technology based on AFLT, based on AGPS（Wireless network is aided in
GPS positioning technology）Location technology, based on GPSOne location technologies, WLAN position etc..
During location technology is also evolving, any practical location technology may be used to the present invention.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described location data, includes day
Phase, time, the longitude residing for the time, the latitude residing for the time.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described location data, also comprising latitude
Spend hemisphere, longitude hemisphere, earth ellipsoid the face height of level surface, height above sea level, ground speed, the one of ground course relative to the earth
Plant or several.
A series of location datas sequentially in time constitute track, and the similarity of track differentiates there are many maturation methods
And new method, all it is available.
In Chinese invention patent such as leaf sword《A kind of space-time track similarity calculating method and system》Disclose one kind
Space-time track similarity calculating method and system, methods described include：Step 1, definition portrays user interest point apart from corner rate
Feature；Step 2, rule of thumb threshold value, recognizes user interest point；Its public interest is calculated according to the user interest point of track
Point；Step 3, the similarity and dissmilarity degree between segmentation are calculated, wherein described be segmented between two public points of interest
Segmentation；By defining split time, similar segments, similar route, the similarity and dissmilarity degree between track are calculated, so that
Obtain track similarity.
Also crown man is in its thesis for the doctorate《Spatiotemporal data structure based on GPS track and photo track》In review it is existing
Some known methods having.
The distance between track function, similarity factor etc. can be used by characterizing the parameter of similarity.
Distance function can use Euclidean distance, absolute distance, Minkowski distances, Chebyshev distances, variance to add
Weigh distance, mahalanobis distance etc., can be used in pedigree clusters beeline, longest distance, group average distance, centroidal distance,
Sum of squares of deviations distance etc., similarity factor can use coefficient correlation, included angle cosine etc..
The similarity of track can also be excavated with trajectory model, the algorithm with time-labeling sequential mining etc. is calculated.
The similarity of track is compared can be using the similarity algorithm based on distance function, based on the similar of pedigree clusters
Spend algorithm, the similarity algorithm based on similarity factor, based on trajectory model excavate similarity algorithm, based on band time-labeling sequence
Arrange excavate similarity algorithm, the similarity algorithm based on fit equation, the similarity algorithm based on collaborative filtering, based on path
The similarity algorithm of search, the similarity algorithm based on discrete Frechet distances, the similarity algorithm based on stencil matching, base
Similarity algorithm in map stencil matching, the similarity algorithm based on vector quantization, the similarity based on gauss hybrid models
Algorithm, the similarity algorithm based on HMM, the similarity algorithm based on artificial neural network, based on dynamic time
The hybrid algorithm of one kind and above-mentioned algorithm in regular similarity algorithm.These parameters, algorithm all can be used for this hair
Similarity identification in bright.
Because the error of some situations is larger, therefore the similarity algorithm based on map stencil matching can also be used, i.e.,
It will be compared by the location data of the track of work attendance person and the road on the track periphery, confirm immediate road, will be examined
The track of diligent person is characterized with the location data of immediate road, is then carried out again with the track of road location data formation
Similarity-rough set.
A kind of Work attendance method based on track and sensing data, it is characterised in that：The ratio of the similarity of described track
Centering, the requirement threshold value after actual measurement is by the location data of work attendance person, the precision of track according to the work attendance degree of accuracy.
Location data, the precision of track and device therefor, local environment relation are very big, based on location data, trajectory calculation
Distance function, similarity factor etc. result of calculation it is also very big with device therefor, local environment relation,
Location data, the precision of track measured with distinct device or in different environments has difference, uses same distance
The parameters such as function, similarity factor carry out similarity comparison when corresponding error range can also change therewith, in addition, work attendance person for
Recognize that the error requirements of accuracy are also different.So the determination of threshold value need to survey by work attendance person turn out for work after the situation of track according to
Determined according to the requirement of the degree of accuracy.I.e.：
Thr=f（δ, △）Wherein Thr is threshold value, and δ is precision；△ is the error requirements of work attendance accuracy.
The selection of the threshold value of similarity is relevant with measurement accuracy, the work attendance degree of accuracy, and the selection range of distance threshold is measurement
1-10 times of precision.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described sensing data, is to accelerate
Spend acceleration information, the acceleration information of y-axis, the acceleration information of z-axis of the x-axis of sensor, the angle of the x-axis of direction sensor
Degrees of data, the angle-data of y-axis, the angle-data of z-axis, the angular acceleration data of gyro sensor x-axis, the angle of y-axis accelerate
Degrees of data, the angular acceleration data of z-axis, the gravimetric data of gravity sensor, the acceleration number of degrees of linear acceleration sensors x-axis
According to, the acceleration information of the acceleration information of y-axis, z-axis, x*sin (θ/2) data based on x-axis of rotating vector sensor,
Y*sin (θ/2) data based on y-axis, one or more of combinations in z*sin (θ/2) data based on z-axis.
Sensing data is a series of data acquisition system of special parameters of sensor collection, and each set at least includes day
Phase, time, the numerical value of special parameter, special parameter can be the acceleration of acceleration transducer, the angle of direction sensor, top
Angular acceleration, the gravity of gravity sensor, the acceleration of linear acceleration sensors, the rotating vector sensor of spiral shell instrument sensor
Data in one or several kinds of combinations.
The data of each axle of every kind of sensor are exactly one kind in data recited above, rather than x-axis, y-axis, z
Three axles of axle just can be regarded as one kind in data altogether.
It is that the collective effect power of multiple directions synthesizes an initial vector during human motion.Different sensors are fixed on
Mobile terminal, experiences the different types of derivative vector that described initial vector is produced, such as acceleration, direction, angular acceleration or
Person's rotating vector etc..
Vector is used in spherical coordinate system（R, φ, θ）Represent, used in rectangular coordinate system（x、y、z）To represent, sensor number
According to usually using rectangular coordinate system representation, i.e.,（X, y, z）.
Although sensor is integrally fixed at mobile terminal, the relative position relation of mobile terminal and human body is change.Cause
Even if this same human motion, initial vector that mobile end sensor is experienced, the numerical value of derivative vector are different.
Initial vector, derivative vector are used in spherical coordinate system（R, φ, θ）Represent, no matter the relative position of mobile terminal and human body
How is relation, and r numerical value is consistent, and θ, φ can be by overlapping after rotation.
Due to the change of mobile terminal and the relative position relation of human body, cause initial vector, derivative vector in rectangular co-ordinate
The component of x, y, z axle is different in system, and the data of the x, y, z axle of the rectangular coordinate system of mobile end sensor output have larger difference,
Difficulty becomes big when causing the similarity to compare.
Therefore by rectangular coordinate system（x、y、z）Data change into spherical coordinate system（R, φ, θ）Carrying out similarity comparison is
More preferable method, it is of course also possible to directly with rectangular coordinate system（x、y、z）To carry out similarity comparison.
No matter the coordinate system used, sensing data is the complicated fluctuation by abscissa of the time, belongs to machinery and shakes
The fluctuation of dynamic section.
Therefore, the similarity comparison method of described sensing data can use the similarity comparison method of ripple, various
The similarity-rough set method of practical frequency spectrum can be used.
Mechanical oscillation, sound wave, electromagnetic wave etc. belong to ripple.
The comparison method of the similarity of fluctuation also has a lot, and mechanical oscillation, sound wave, electromagnetic wave etc. have many maturation sides
Method, also there is many new methods.
For example in electromagnetic wave field, Zhang Qilin etc. exists《Infrared spectrum spectrogram comparison method based on rice sedling》In
Based on rice sedling, using the peak intensity, peak position, peak area of infrared spectrum as similar meta design alignment algorithm, by testing
Into similar finite element there is certain influence in different characteristic value to calculating spectrogram similitude, then obtain similar using analytic hierarchy process (AHP)
The weight coefficient of different characteristic value, finally proposes a kind of improved alignment algorithm during degree is calculated.
The similarity of sensing data is compared can be using the similarity algorithm based on Euclidean distance, based on fit equation
Similarity algorithm, the similarity algorithm based on collaborative filtering, the similarity algorithm based on route searching, based on sampling frequency feature
The similarity algorithm of analysis, the similarity algorithm based on discrete Frechet distances, the similarity algorithm based on stencil matching, base
Similarity algorithm in vector quantization, the similarity algorithm based on gauss hybrid models, based on the similar of HMM
Spend algorithm, the similarity algorithm based on artificial neural network, one kind in the similarity algorithm based on dynamic time warping and
The hybrid algorithm of above-mentioned algorithm.
A kind of Work attendance method based on track and sensing data, it is characterised in that：Described sensing data it is similar
In the comparison of degree, the requirement threshold value after actual measurement is by the precision of the sensing data of work attendance person according to the work attendance degree of accuracy.
Based on dynamic time warping（DTW）Similarity algorithm have that calculating speed is fast, conclusion intuitively advantage.The present invention
Similarity comparison is just carried out by taking this algorithm as an example.
DTW is a kind of non-linear regular technology that Time alignment and distance measure calculations incorporated are got up, and it finds one
Warping function im=Ф(in), the time shaft n of test vector is non-linearly mapped on the time shaft m of reference template, and make this
Function is met：
(1)
D be in optimal time it is regular in the case of two vectors distance.Because DTW constantly calculates the distance of two vectors to seek
Look for optimal coupling path, thus obtain be two vector matchings when Cumulative Distance minimum corresponding to warping function, this is just protected
The maximum comparability existed between them is demonstrate,proved.The essence of DTW algorithms is exactly the thought with Dynamic Programming, utilizes local optimum
The processing of change carrys out the paths of Automatic-searching one, along this paths, and the cumulative distortion amount between two characteristic vectors is minimum, so that
Avoid the error that may be introduced because duration is different.
The schematic diagram of DTW algorithms such as Fig. 4, each horizontal strokes of frame number n=1 ~ N in a two-dimensional Cartesian coordinate system of sequence 1
Marked on axle, each frame m=1 ~ M of sequence 2 is marked on the longitudinal axis, drawing some by the rounded coordinate of these expression frame numbers indulges
Each crosspoint (t that horizontal line can be formed in a grid, gridi,rj) represent sequence 1 in a certain frame with it is a certain in sequence 2
Frame crosses.
DTW algorithms are carried out in two steps, and one is to calculate the distance between two each frames of sequence, that is, obtains frame matching distance square
Battle array, two be that an optimal path W (n) is found out in frame matching distance matrix.Searching for the process of this paths can be described as follows：
Search is set out from (1,1) point, for local path constraint such as Fig. 5, point (in,im) accessible previous lattice point be only possible to be
(in-1,im)、(in-1,im- l) and (in-1,im-2).So (in,im) necessarily select these three distance in reckling corresponding to
Point as its front and continued lattice point, at this moment the Cumulative Distance in this path is：
D(in,im)=D(T(in),R(im))+min{D(in-1,im),D(in-1,im-1),D(in-1,im-2)} (2)
So set out from (l, 1) point (make D (1,1)=0) search for, repeatedly recursion, until (N, M) can be obtained by optimal path, and
And D (N, M) is exactly the matching distance corresponding to best matching path.
Common DTW is more sensitive to end-point detection, and terminal point information is to be supplied to recognizer as the parameter of one group of independence
's.It requires two sequence start to starts, and terminal is to terminal, and the required precision to end-point detection is higher.When noise compares
When big, end-point detection is difficult to carry out, and this requires to give to consider during dynamic time warping.Loosen end points method for limiting not
End points alignment is strict with, overcomes what the starting and terminal point due to two sequences that endpoint algorithm is inaccurately caused can not align to ask
Topic.Generally, if beginning and end relax 2-3 frames in both direction in length and breadth just can be with, that is, starting point can (1,
1), (l, 2), (1,3), (2,1), (3, l), terminal is similar.Such as Fig. 6.
In the DTW algorithms for relaxing end points limitation, element in Cumulative Distance matrix (1, l), (l, 2), (l, 3), (2,
L), (3,1) do not calculate according to local decisions function and obtained, but directly insert the element of frame matching distance matrix, from
Dynamic select minimum therefrom one, as starting point, is also that a minimum is selected out of relaxation terminal allowed band for terminal
It is worth the matching distance as two sequences.
The present invention uses DTW algorithms, has carried out track, the similarity of sensing data is compared, has as a result met the expected requirements,
Embodied in a specific embodiment.
It is important to note that using DTW algorithms not limitation of the present invention, relevant technical field is not being departed from
In the case of the spirit and scope of the present invention, it can also make a variety of changes and modification, therefore all equivalent technical schemes
Belong to scope of the invention, protection scope of the present invention should be defined by the claims.
Compared with prior art, the beneficial effects of the invention are as follows：
1) the very big operating pressure and work difficulty that alleviate work attendance person；
2) accuracy of work attendance and the efficiency of work attendance are greatly improved；
3) avoid and be individually identified by the process of work attendance person, it is to avoid largely by work attendance person assemble may caused by potential safety hazard
Deng.
Brief description of the drawings
The core process schematic diagram of Fig. 1 present invention.
The main flow schematic diagram of Fig. 2 present invention.
Fig. 3 work attendances core differentiates schematic flow sheet.
Fig. 4 DTW algorithm principle figures.
Fig. 5 constrained paths.
The improved DTW algorithm principles figures of Fig. 6.
The small A AGPS tracks A1S small A of Fig. 7 walk with the same paths of small B simultaneously when.
The small B AGPS tracks B1S small A of Fig. 8 walk with the same paths of small B simultaneously when.
The small A AGPS tracks A1T small A of Fig. 9 walk with the same paths of small B simultaneously when.
The small B AGPS tracks B1T small A of Figure 10 walk with the same paths of small B simultaneously when.
The small A of Figure 11 and the same Origin And Destinations of small B but during different path sections small B AGPS tracks B2S.
The small A of Figure 12 and the same Origin And Destinations of small B but during different path sections small B AGPS tracks B2T.
θ A1, the θ B1 value volatility series of the acceleration transducer same paths of the small A of Figure 13 and small B are walked simultaneously when are compared and shown
It is intended to；Solid line is small A, and dotted line is small B；Small A curve has moved 0.5 unit in the Y-axis direction.
φ A1, the φ B1 value volatility series of the acceleration transducer same paths of the small A of Figure 14 and small B are walked simultaneously when are compared
Schematic diagram；Solid line is small A, and dotted line is small B；Small B curve has moved 1 unit in the Y-axis direction.
The θ A2 of the acceleration transducer of homogeneous, θ A3 values volatility series do not compare schematic diagram under the same paths of the small A of Figure 15；It is real
Line is for the first time, dotted line is second.
The φ A2 of the acceleration transducer of homogeneous, φ A3 values volatility series do not compare schematic diagram under the same paths of the small A of Figure 16；
Solid line is for the first time, dotted line is second；Secondary curve has moved down 1 unit in Y direction.
The small A of Figure 17 take the volatility series of the θ A4 of acceleration transducer during two mobile phones of M, N, θ B4 values to compare schematic diagram；
Solid line is M, and dotted line is N；M curve has moved 1 unit in the Y-axis direction.
The small A of Figure 18 take the volatility series of the φ A4 of acceleration transducer during two mobile phones of M, N, φ B4 values to compare signal
Figure；Solid line is M, and dotted line is N.
Embodiment
The present embodiment is that the present invention is further illustrated.Obtain track data using the AGPS methods positioned and use and add
Velocity sensor obtains sensing data, and smart mobile phone is in the mobile terminal used.
Downloaded by work attendance person and mobile terminal software be installed, mobile terminal software requirement fills corresponding personal information by work attendance person,
Including the information such as cell-phone number and mailbox, into system, for the first time using system, mobile terminal software inquiry is by work attendance person mobile terminal feature
Information, record user IMSI codes and/or IMEI code, mobile terminal software by by the IMSI codes and/or IMEI code of work attendance person and by
Work attendance person's information upload server, is sent to server.
Work attendance person enters system, and filling corresponding personal information includes the information such as cell-phone number and mailbox, and work attendance person, which enters, to be examined
Diligent information management module, manages work attendance require information, increases attendance information, sets work attendance time, including readiness time point Ty, examines
Diligent time point Tk, lagged time point Tc, end time point Tx, are set, and point range F, sets checked-in person person to believe with setting work attendance
Breath.
Be able to please be permit a leave in mobile terminal software entrance before the Ty moment module by work attendance person, proposition is asked for leave application, and by this
Message is sent to server, and work attendance person be able to please permit a leave module in mobile terminal software, check the request of asking for leave of checked-in person person, and right
Application request of asking for leave is examined, and approval results are uploaded onto the server.
Work attendance readiness time Ty is reached, not actuated mobile terminal software automatic start is detected, initialization module is to system
Initialized, calibrate the time by work attendance person and work attendance person, make it consistent with server, mobile terminal software detection is by work attendance person
IMSI codes and/or IMEI code, matched with this in server by the IMSI codes and/or IMEI code of work attendance person, the match is successful
Do not process, be directly entered next step, it fails to match does corresponding equipment changing record in server, and detection message is timely
Feedback notification this by work attendance person, into next step.
By work attendance person's information in detection service device, judge whether asked for leave by this work attendance of work attendance person, if being asked for leave by work attendance person,
Then system finishing, if not asked for leave by work attendance person, is opened by work attendance person mobile terminal software data record the process, the inspection of mobile terminal software
The AGPS positioning components of user equipment and the starting state of acceleration transducer are surveyed, if abnormal state, is set in server
Standby exception record, and message is fed back to by work attendance person in time, notify it please voluntarily to be contacted with work attendance person, otherwise can be considered as scarce
Duty, if state is normal, data record process reads the location data of the AGPS positioning components of mobile terminal and corresponding time in real time
Mobile end equipment is stored in the exercise data and corresponding read access time of acceleration transducer.
By work attendance person reach work attendance point range F time be Td, before Td arrival time section starting point Tdh1 to Td
The period between examination period end point Tdh2 afterwards is examination period Tdh.
Reach work attendance time point Tk, work attendance place of the server in work attendance require information, according to AGPS localization methods
Ten meter accuracy errors work attendance is calculated centered on objective point range F.
Reach after work attendance time point Tk, according to differentiated by the track of work attendance person belonged to by work attendance person turn out for work, turn out for work it is undetermined, slow
To which kind of situation in undetermined, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, the sensing data of mobile terminal is used
Differentiate belonged to by work attendance person turn out for work, which kind of situation in allograph, differentiate that process is as follows：
Whether the location data by work attendance person when inquiring about work attendance time point Tk first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this is set to by work attendance person turns out for work；
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to and turn out for work；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to and turns out for work；
By by work attendance person with examination period Tdh in reach other by the sign similarity of the track of the track of work attendance person
Parameter and threshold value are compared, if characterizing the parameter and similarity correlation of similarity, and characterize the ginseng of similarity
Number be more than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh
The similarity of interior sensing data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes phase
Seemingly the parameter of degree is less than threshold value, then carries out being examined by work attendance person by other of work attendance person with reaching in examination period Tdh
The similarity of sensing data in period Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, be then set to and turn out for work, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and the parameter of sign similarity is more than threshold value, then is set to and turns out for work,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms；
If in point range F, not being set to late class undetermined in work attendance by the location data of work attendance person during work attendance time point Tk.
Reach after lagged time point Tc, the track by work attendance person for class undetermined of being late is differentiated, according to by the track of work attendance person
Differentiate by work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, the late undetermined of differentiation can not be differentiated according to track
Situation, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal, process is differentiated such as
Under：
Whether the location data by work attendance person when inquiring about lagged time point Tc first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
Inquiry within examination period Tdh by work attendance person other whether entrance work attendance place model is had by the location data of work attendance person
Enclose in F,
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this be set to by work attendance person it is late,
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to late；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to late；
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be more than
Threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh in biography
The similarity of sensor data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes similarity
Parameter be less than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination the period
The similarity of sensing data in Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, then be set to it is late, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and characterize the parameter of similarity and be more than threshold value, then be set to it is late,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms.
Inquiry belong to turn out for work, be late, allograph situation by work attendance person's location data in lagged time point Tc to end time
Whether in work attendance in point range F during point Tx, if had by the location data of work attendance person in work attendance outside point range F,
It is set to and leaves early.
Server is stored the result of judgement, while server is read in attendance information by the work attendance knot of work attendance person
Really, it is illustrated in the mobile terminal software of correspondence work attendance person；Server process checking-in result, updates by the information of work attendance person, and will
Nearest checking-in result information after renewal is illustrated in by work attendance person mobile terminal software interface.
If being wanted to check my attendance record by work attendance person, in mobile terminal, software submits respective request message, and server is received
To the request message sended over by work attendance person mobile terminal software, this in server is read by the checking-in result data of work attendance person,
Progress collects the information such as number of times of turning out for work, number of times absent from duty, late number of times, number of times of leaving early, allograph number of times, number of times of asking for leave, after collecting
Message be sent to mobile terminal software, be presented on mobile terminal software interface.
Server use by the location data of work attendance person, track, exercise data, from server to by work attendance person mobile terminal
Software transfer is actively sent to server by work attendance person mobile terminal software.
Similarity algorithm based on dynamic time warping carries out the comparison of track similarity.DTW programs are worked out, program is first
First the time in AGPS tracks, longitude, latitude are normalized, then carried out not according to the algorithm of conventional dynamic time warping
With the similarity comparison of track.
The AGPS initial trace A1S of small A when small A and small B walks together are acquired, sees and AGPS rails is obtained after Fig. 7, processing
Mark A1T（TA1T, xA1T, yA1T）, see Fig. 9, small B AGPS initial trace B1S, see after Fig. 8, processing and obtain AGPS tracks B1T
（TB1T, xB1T, yB1T）, see Figure 10, acquire small A and the same Origin And Destinations of small B but the asynchronous small B of path sections
AGPS initial trace B2S, are shown in after Figure 11, processing and obtain AGPS tracks B2T（TB2T, xB2T, yB2T）, see Figure 12.
The similarity that the similarity that DTW programs calculate A1T and B1T is 1.10, A1T and B2T is 2.46, A1T and B2T
Similarity there is significant difference.
A1T and B2T paths difference are that A1 does not bypass building BU, B2 and bypassed building BU, and other parts are all identical, and this is built
Spacing is at 15 meters or so between building two road, and detail is shown in Fig. 7, Fig. 8, therefore it is in road that AGPS tracks, which can accurately be distinguished,
The different tracks advanced on road.
The similarity of AGPS tracks A1T and B1T small A walks with the same paths of small B simultaneously when are 1.10, show similarity
The difference of track can not be told in 1.10 or so interval.
It is pointed out that the similarity algorithm based on dynamic time warping is insensitive for the time, that is, will
Small A AGPS tracks A1T, small B AGPS tracks B1T, are handled, and the time series delay in wherein B1T obtains track in 3 seconds
Time series delay in B11T, B1T obtains track B12T in 5 seconds, according to about 6 kilometer per hours of walking speed, now phase respectively
When being separated by about 6 meters, 12 meters before and after two people, similarity is detected, similarity is also 1.10.
The situation of different arrival times, belongs to the translation of curve under same path, and translation can be solved based on Euclidean distance
Problem, using the time as a sequence, longitude and latitude is two other sequences, constitutes a three-dimensional array, compares different delayed time
Same path.Calculate A1 and B1, A1 and B11, A1 and B12 Euclidean distance, three Euclidean distances are respectively 2.76,3.55,
5.07, show that delay is more, Euclidean distance is bigger, and similitude is lower.
AGPS precision is 5-10 meters in the embodiment of the present invention, and threshold value selection range can be 1-10 times of precision, and scope is
5-100 meters, distance threshold is chosen for 10 meters in the similarity algorithm based on dynamic time warping, according to above practical measurement knot
Really, the similarity that A1T and B1T similarity is 1.10, A1T and B2T is that the road before and after 2.46, building is minimum road
Footpath difference, thus can 1.10 and 2.46 it is interval choose, can use the median method threshold value to be（1.10+2.46）/2=
1.78。
Similarity algorithm based on dynamic time warping enters the comparison of line sensor similarity.Based on conventional dynamic time rule
Whole algorithm has worked out program, gathers small A and small B acceleration transducer data, initial data is rectangular coordinate system（X, y,
z）, it is translated into spherical coordinate system（R, θ, φ）.
Small A and the same paths of small B are obtained while acceleration transducer when walking（θA1、φA1）、（θB1、φB1）
Value, is shown in Figure 13,14.Obtain under the same paths of small A not first time of the acceleration transducer of homogeneous（θA2、φA2）, second
Secondary（θA3、φA3）Value, is shown in Figure 15,16.Acceleration transducer when obtaining small A by M, N two mobile phones（θA4、φ
A4）、（θB4、φB4）Value, is shown in Figure 17,18.
The value of above-mentioned sequence is subjected to similarity-rough set, is normalized first before comparing.Similarity comparison result two-by-two
It is as follows：
Small A walks with the same paths of small B simultaneously when, θ A1, θ B1 similarity are 1.77；φ A1, φ B1 similarity are 3.22.
The θ A2 of homogeneous, θ rA3 similarity are not 1.54 under the small same paths of A；φ A2, φ A3 similarity are 1.80.Small A takes M, N
θ A4, θ B4 similarity are 0.83 during two mobile phones；φ A4, φ B4 similarity are 0.77.
Comprehensive similarity is calculated according to following formula：
（3）
Small A walks with the same paths of small B simultaneously when, Y=3.67；Under the small same paths of A not homogeneous when, Y=2.37；Small A takes M, N
During two mobile phones, Y=1.13.
Can 1.13 and 2.37 it is interval choose, can use the median method threshold value to be（1.13+2.37）/2=1.75.
Small A walks with the same paths of small B simultaneously when, comprehensive similarity is 3.67；The not synthesis of homogeneous under the small same paths of A
Similarity is 2.37；Comprehensive similarity is 1.13 when small A takes two mobile phones of M, N, and difference is obvious.Show that a people holds multiple
In the case that mobile phone carries out allograph work attendance, similarity is very high.
Distance threshold is chosen for 10 meters in the similarity algorithm based on dynamic time warping in the embodiment of the present invention, according to preceding
Face practical measurement result, A1T and B1T similarity are that 1.10, A1T and B2T similarity is the road before and after 2.46, building
Road is minimum path difference, therefore in 1.10 and 2.46 interval selections the median method threshold value can be used to be
（1.10+2.46）/2=1.78.
The comprehensive similarity of homogeneous is not 2.37 under the threshold value selection range of sensing data, the small same paths of A；Small A takes
Comprehensive similarity is that between 1.13, identified threshold range can use the median method threshold value to be during two mobile phones of M, N
（1.13+2.37）/2=1.75.
Embodiment of above is merely to illustrate the present invention, and not limitation of the present invention, relevant technical field, is not taking off
In the case of from the spirit and scope of the present invention, it can also make a variety of changes and modification, therefore all equivalent technical schemes
Scope of the invention is fallen within, protection scope of the present invention should be defined by the claims.
Claims (6)
1. a kind of Work attendance method based on track and sensing data, it is characterised in that comprise the following steps：
1）Submit application of asking for leave to server by mobile terminal software before readiness time point Ty by work attendance person, work attendance person will ratify
It is sent to by the mobile terminal software of work attendance person, mobile terminal software is received after approval, that is, stops carrying out work attendance by work attendance person to this；
2）Reach after readiness time point Ty, running software initialization process in mobile terminal, makes its kimonos at the automatic calibration mobile terminal time
Being engaged in, device is synchronous, and startup attendance record subprocess is read by the location data and sensing data of work attendance person mobile terminal, if read
Failure, mobile terminal software is by this message upload server；If read normally, start to read exercise data and the movement of sensor
Hold the location data of locating module；
3）Reach after work attendance time point Tk, according to differentiated by the track of work attendance person belonged to by work attendance person turn out for work, turn out for work it is undetermined, late
Which kind of situation in undetermined, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, is sentenced using the sensing data of mobile terminal
Do not belonged to by work attendance person turn out for work, which kind of situation in allograph；
4）Reach after lagged time point Tc, differentiate the track by work attendance person for class undetermined of being late, sentence according to by the track of work attendance person
Not by work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, can not differentiate that the late of differentiation waits according to track
Condition, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal；
5）Inquiry belong to turn out for work, be late, allograph situation by work attendance person's location data in lagged time point Tc to end time point
It is fixed if had by the location data of work attendance person in work attendance outside point range F whether in work attendance in point range F during Tx
To leave early.
2. a kind of Work attendance method based on track and sensing data according to claim 1, it is characterised in that：
Reach after work attendance time point Tk, according to differentiated by the track of work attendance person belonged to by work attendance person turn out for work, undetermined, late treat of turning out for work
Which kind of situation in fixed, the situation undetermined of turning out for work of differentiation can not be differentiated according to track, differentiated using the sensing data of mobile terminal
Belonged to by work attendance person turn out for work, which kind of situation in allograph, differentiate that process is as follows：
Whether the location data by work attendance person when inquiring about work attendance time point Tk first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this is set to by work attendance person turns out for work；
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to and turn out for work；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to and turns out for work；
By by work attendance person with examination period Tdh in reach other by the sign similarity of the track of the track of work attendance person
Parameter and threshold value are compared, if characterizing the parameter and similarity correlation of similarity, and characterize the ginseng of similarity
Number be more than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh
The similarity of interior sensing data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes phase
Seemingly the parameter of degree is less than threshold value, then carries out being examined by work attendance person by other of work attendance person with reaching in examination period Tdh
The similarity of sensing data in period Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, be then set to and turn out for work, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and the parameter of sign similarity is more than threshold value, then is set to and turns out for work,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms；
If in point range F, not being set to late class undetermined in work attendance by the location data of work attendance person during work attendance time point Tk.
3. a kind of Work attendance method based on track and sensing data according to claim 1, it is characterised in that：
Reach after lagged time point Tc, differentiate the track by work attendance person for class undetermined of being late, according to by the track differentiation of work attendance person
By work attendance person belong to it is late, late it is undetermined, absent from duty in which kind of situation, can not differentiate that the late of differentiation waits according to track
Condition, which kind of situation belonged to by work attendance person in late, allograph is differentiated using the sensing data of mobile terminal, differentiates that process is as follows：
Whether the location data by work attendance person when inquiring about lagged time point Tc first is in work attendance in point range F, if by work attendance
Person in point range F, proceeds following differentiation in work attendance：
Inquiry within examination period Tdh by work attendance person other whether entrance work attendance place model is had by the location data of work attendance person
Enclose in F,
If there is no other to enter work attendance place model by the location data of work attendance person within examination period Tdh by work attendance person
Enclose in F, this be set to by work attendance person it is late,
If having other within examination period Tdh by work attendance person by the location data of work attendance person with entering work attendance point range
In F, it will be entered by work attendance person with the track in other examination periods Tdh by work attendance person for being reached in examination period Tdh
The comparison of row similarity,
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be less than
Threshold value, then be set to late；If the parameter and the negatively correlated relation of similarity of similarity were characterized, and it would be big to characterize the parameter of similarity
In threshold value, then it is set to late；
By by work attendance person with reached in examination period Tdh other by the parameter of the sign similarity of the track of work attendance person and
Threshold value is compared, if the parameter and similarity correlation of similarity were characterized, and the parameter of sign similarity would be more than
Threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination period Tdh in biography
The similarity of sensor data is compared, if characterizing the parameter and the negatively correlated relation of similarity of similarity, and characterizes similarity
Parameter be less than threshold value, then carry out by work attendance person with examination period Tdh in reach other by work attendance person examination the period
The similarity of sensing data in Tdh is compared,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be less than threshold value, then be set to it is late, if characterized similar
The parameter of degree and the negatively correlated relation of similarity, and characterize the parameter of similarity and be more than threshold value, then be set to it is late,
If other reached by work attendance person and in examination period Tdh are by the sign similarity of the sensing data of work attendance person
Parameter and similarity correlation, and characterize the parameter of similarity and be more than threshold value, it is determined as allograph suspicion, if characterizing phase
Like the parameter and the negatively correlated relation of similarity of degree, and the parameter of similarity is characterized less than threshold value, it is determined as allograph suspicion, submit to
Work attendance person confirms.
4. a kind of Work attendance method based on track and sensing data according to claim 1,2,3, it is characterised in that：Institute
The work attendance stated ground point range F, is determined by the accurate location information in work attendance place plus location data error.
5. a kind of Work attendance method based on track and sensing data according to claim 1,2,3, it is characterised in that：Institute
The location data stated, includes the longitude residing for date, time, the time, the latitude residing for the time.
6. a kind of Work attendance method based on track and sensing data according to claim 1,2,3, it is characterised in that：Institute
The sensing data stated, is the acceleration number of degrees of the acceleration information, the acceleration information of y-axis, z-axis of the x-axis of acceleration transducer
According to the angle-data of the x-axis of direction sensor, the angle-data of y-axis, the angle-data of z-axis, the angle of gyro sensor x-axis
Acceleration information, the angular acceleration data of y-axis, the angular acceleration data of z-axis, the gravimetric data of gravity sensor are linear to accelerate
Spend sensor x-axis acceleration information, the acceleration information of y-axis, the acceleration information of z-axis, rotating vector sensor based on x
X*sin (θ/2) data of axle, y*sin (θ/2) data based on y-axis, one kind in z*sin (θ/2) data based on z-axis or
Several combinations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610207477.6A CN107274503A (en) | 2016-04-06 | 2016-04-06 | A kind of Work attendance method based on track and sensing data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610207477.6A CN107274503A (en) | 2016-04-06 | 2016-04-06 | A kind of Work attendance method based on track and sensing data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107274503A true CN107274503A (en) | 2017-10-20 |
Family
ID=60052250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610207477.6A Pending CN107274503A (en) | 2016-04-06 | 2016-04-06 | A kind of Work attendance method based on track and sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107274503A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544718A (en) * | 2018-11-20 | 2019-03-29 | 左凌云 | A kind of Work attendance method and its system |
| CN109784371A (en) * | 2018-12-14 | 2019-05-21 | 北京三快在线科技有限公司 | Net about vehicle monitoring and managing method, device and storage medium |
| CN110097327A (en) * | 2018-01-30 | 2019-08-06 | 阿里巴巴集团控股有限公司 | Work attendance method and device |
| CN110930107A (en) * | 2019-10-14 | 2020-03-27 | 平安科技（深圳）有限公司 | Attendance information processing method and device, computer equipment and storage medium |
| CN113450073A (en) * | 2021-06-30 | 2021-09-28 | 未鲲(上海)科技服务有限公司 | Attendance checking method, attendance checking device, electronic equipment and storage medium |
| CN113924464A (en) * | 2019-05-10 | 2022-01-11 | 萨克提斯有限责任公司 | Method for measuring a structure, and a procedure for defining an optimal method for measuring said structure |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488678A (en) * | 2013-08-05 | 2014-01-01 | 北京航空航天大学 | Friend recommendation system based on user sign-in similarity |
| JP3195768U (en) * | 2014-08-22 | 2015-01-29 | 勉志 劉 | Calling device |
| CN104573922A (en) * | 2014-12-18 | 2015-04-29 | 四川联友电讯技术有限公司 | Designated check-in system and method for enterprise mobile office platform |
| CN105023106A (en) * | 2015-07-28 | 2015-11-04 | 艾体威尔电子技术(北京)有限公司 | Attendance management method and system based on intelligent NFC tag |
| CN204833381U (en) * | 2015-08-14 | 2015-12-02 | 武汉中星北斗通信技术股份有限公司 | Attendance system with locate function |
| CN105224545A (en) * | 2014-06-03 | 2016-01-06 | 华为技术有限公司 | A kind of position recommend method and device |
-
2016
- 2016-04-06 CN CN201610207477.6A patent/CN107274503A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488678A (en) * | 2013-08-05 | 2014-01-01 | 北京航空航天大学 | Friend recommendation system based on user sign-in similarity |
| CN105224545A (en) * | 2014-06-03 | 2016-01-06 | 华为技术有限公司 | A kind of position recommend method and device |
| JP3195768U (en) * | 2014-08-22 | 2015-01-29 | 勉志 劉 | Calling device |
| CN104573922A (en) * | 2014-12-18 | 2015-04-29 | 四川联友电讯技术有限公司 | Designated check-in system and method for enterprise mobile office platform |
| CN105023106A (en) * | 2015-07-28 | 2015-11-04 | 艾体威尔电子技术(北京)有限公司 | Attendance management method and system based on intelligent NFC tag |
| CN204833381U (en) * | 2015-08-14 | 2015-12-02 | 武汉中星北斗通信技术股份有限公司 | Attendance system with locate function |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110097327A (en) * | 2018-01-30 | 2019-08-06 | 阿里巴巴集团控股有限公司 | Work attendance method and device |
| CN109544718A (en) * | 2018-11-20 | 2019-03-29 | 左凌云 | A kind of Work attendance method and its system |
| CN109784371A (en) * | 2018-12-14 | 2019-05-21 | 北京三快在线科技有限公司 | Net about vehicle monitoring and managing method, device and storage medium |
| CN113924464A (en) * | 2019-05-10 | 2022-01-11 | 萨克提斯有限责任公司 | Method for measuring a structure, and a procedure for defining an optimal method for measuring said structure |
| CN110930107A (en) * | 2019-10-14 | 2020-03-27 | 平安科技（深圳）有限公司 | Attendance information processing method and device, computer equipment and storage medium |
| CN113450073A (en) * | 2021-06-30 | 2021-09-28 | 未鲲(上海)科技服务有限公司 | Attendance checking method, attendance checking device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107274503A (en) | A kind of Work attendance method based on track and sensing data | |
| Renaudin et al. | Evaluating indoor positioning systems in a shopping mall: The lessons learned from the IPIN 2018 competition | |
| CN104838281B (en) | Figure is positioned and built based on virtually target | |
| CN104395696B (en) | Estimate the method for device location and implement the device of this method | |
| CN105190233B (en) | Position determines that processing unit, position determine that processing method, position determine processing routine, portable information processor, mobile message processing method, mobile message processing routine and storage medium | |
| EP3213031B1 (en) | Simultaneous localization and mapping by using earth's magnetic fields | |
| US10145707B2 (en) | Hierarchical context detection method to determine location of a mobile device on a person's body | |
| US11035915B2 (en) | Method and system for magnetic fingerprinting | |
| CN106017454A (en) | Pedestrian navigation device and method based on novel multi-sensor fusion technology | |
| CN105556547B (en) | The method and system with annotation capture and movable group modeling for data | |
| CN107110654A (en) | Location equipment and the method for positioning | |
| CN104380043B (en) | Information in the portable electric appts that user carries is determined | |
| CN107421535A (en) | A kind of indoor pedestrian's alignment system walked based on magnetic signature and acceleration information meter | |
| US20170272915A1 (en) | System and method for processing personal identification information based on position information | |
| CN109819406B (en) | Indoor positioning method based on crowdsourcing | |
| Liu et al. | Focusing matching localization method based on indoor magnetic map | |
| Gao et al. | XINS: The anatomy of an indoor positioning and navigation architecture | |
| CN108139458A (en) | For determining the method, apparatus and system in indoor orientation | |
| CN110285809A (en) | A kind of integrated integrated positioning device of indoor and outdoor | |
| CN108629295A (en) | Corner terrestrial reference identification model training method, the recognition methods of corner terrestrial reference and device | |
| Seco et al. | RFID-based centralized cooperative localization in indoor environments | |
| WO2018232607A1 (en) | System and method for positioning a gateway of an architecture | |
| Fu et al. | A survey on artificial intelligence for pedestrian navigation with wearable inertial sensors | |
| Xia et al. | Autonomous pedestrian altitude estimation inside a multi-story building assisted by motion recognition | |
| CN109520508A (en) | Localization method, device and positioning device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171020 |
|
| WD01 | Invention patent application deemed withdrawn after publication |