CN108596941A - Target body movements trajectory predictions method and system based on depth image - Google Patents
Target body movements trajectory predictions method and system based on depth image Download PDFInfo
- Publication number
- CN108596941A CN108596941A CN201810196470.8A CN201810196470A CN108596941A CN 108596941 A CN108596941 A CN 108596941A CN 201810196470 A CN201810196470 A CN 201810196470A CN 108596941 A CN108596941 A CN 108596941A
- Authority
- CN
- China
- Prior art keywords
- movement locus
- point
- objective body
- track
- objective
- 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
- 238000000034 method Methods 0.000 title claims abstract description 84
- 230000001133 acceleration Effects 0.000 claims description 5
- 230000005484 gravity Effects 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 210000003746 feather Anatomy 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to technical field of image processing, specifically provide a kind of target body movements trajectory predictions method and system based on depth image, it is intended to the technical issues of how solution easily obtains the objective body track with degree of precision.For this purpose, target body movements trajectory predictions method in the present invention, it can be based on objective body tracing point in objective body depth image, the a plurality of movement locus of piecewise prediction, and then it can be according to the space length of objective body tracing point and movement locus, the higher movement locus of accuracy rate is chosen, best target trajectory can be obtained according to these movement locus.Meanwhile the target body movements trajectory predictions system in the present invention is able to carry out and realizes the above method.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of target body movements track based on depth image is pre-
Survey method and system.
Background technology
Shuttlecock robot is a kind of intelligence control system based on the technologies such as computer and vision tracking, can be simulated
The mankind carry out badminton.Wherein, shuttlecock trajectory predictions are an important factor for influencing shuttlecock robot motion accuracy.
It is currently possible to using shuttlecock physical model or based on the computation model of neural network, feather sphere path curve is predicted.Wherein, it uses
Shuttlecock physical model predicts feather sphere path curve, easy to operate, be easily achieved, but precision of prediction is relatively low.Using based on nerve
The computation model of network predicts feather sphere path curve, although having higher precision of prediction, needs the training that the order of magnitude is larger
Sample can just obtain the higher prediction locus of precision.
Invention content
In order to solve the above problem in the prior art, in order to solve how easily to obtain the mesh with degree of precision
The technical issues of standard type track.For this purpose, the present invention provides a kind of target body movements trajectory predictions based on depth image
Method and system.
In a first aspect, the target body movements trajectory predictions method based on depth image in the present invention, including:
According to the objective body depth image obtained in advance, multiple continuous objective body tracing points are obtained;
Based on preset trajectory predictions method, and according to the 1st~3 continuous objective body tracing point, obtain first
Movement locus;
Based on the preset trajectory predictions method, and according to the 2nd~4 continuous objective body tracing point, obtain
Second movement locus;
Based on the preset trajectory predictions method, and according to the 3rd~5 continuous objective body tracing point, obtain
Third movement locus;
First movement locus, the second movement locus and third movement locus are modified, target body movements are obtained
Track.
Further, an optimal technical scheme provided by the invention is:
The preset trajectory predictions method specifically includes:
The spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, calculates target
The initial motion speed of body and initial motion direction;
Using the second target body tracing point in three continuous objective body tracing points as initial position;According to described
Initial position, initial motion speed and initial motion direction, and according to the following formula shown in method, calculate in current kinetic track
Objective body tracing point:
Wherein, (xi,yi,yi) be the current kinetic track in i-th of objective body tracing point spatial position coordinate;
(v0x,v0y,v0z) be the initial position spatial position coordinate;The viIt is objective body in i-th of objective body track
The movement velocity of point;The αiFor objective body i-th of objective body tracing point the direction of motion;The tsTo utilize camera shooting
Device obtains the sampling interval of objective body depth image.
Further, an optimal technical scheme provided by the invention is:
Movement velocity v of the objective body in i-th of objective body tracing pointiWith direction of motion αiIt is shown below:
Wherein, the vi-1For objective body (i-1)-th objective body tracing point movement velocity;The αi-1Exist for objective body
The direction of motion of (i-1)-th objective body tracing point;The k is coefficient of air resistance;The g is acceleration of gravity.
Further, an optimal technical scheme provided by the invention is:
" first movement locus, the second movement locus and third movement locus are modified, target body movements are obtained
The step of track ", specifically includes:
In the 4th movement locus point and the 5th movement locus point without departing from first movement locus,
And in the case that the 5th movement locus point is without departing from second movement locus, according to the following formula shown in method obtain
Target body movements track Tm:
Deviate first movement locus in the 4th movement locus point or the 5th movement locus point, and
In the case that the 5th movement locus point is without departing from second movement locus, according to the following formula shown in method obtain target
Body movement locus Tm:
Deviate first movement locus in the 4th movement locus point or the 5th movement locus point, and
In the case that the 5th movement locus point deviates second movement locus, the target body movements track Tm=T3;
Wherein, the T1、T2And T3Respectively the first movement locus, the second movement locus and third movement locus.
Further, an optimal technical scheme provided by the invention is:
" first movement locus, the second movement locus and third movement locus are modified, target body movements are obtained
Include before the step of track ":
Calculate the space length W of the 4th the movement locus point and first movement locusn+1_1, calculate the described 5th
A movement locus point space length W with first movement locus and the second movement locus respectivelyn+2_1And Wn+2_2;
According to the space length Wn+1_1With predetermined threshold value e, it is described to judge whether the 4th movement locus point deviates
First movement locus:If Wn+1_1> e, then the 4th movement locus point deviate first movement locus;
According to the space length Wn+2_1With the predetermined threshold value e, judge whether the 5th movement locus point deviates
First movement locus:If Wn+2_1> e, then the 5th movement locus point deviate first movement locus;
According to the space length Wn+2_2With the predetermined threshold value e, judge whether the 5th movement locus point deviates
Second movement locus:If Wn+2_2> e, then the 5th movement locus point deviate second movement locus.
The target body movements trajectory predictions system based on depth image in second aspect, the present invention, including:
Objective body tracing point acquisition module is configured to, according to the objective body depth image obtained in advance, obtain multiple continuous
Objective body tracing point;
First movement locus acquisition module is configured to preset trajectory predictions method, and according to described in the 1st~3
Objective body tracing point obtains the first movement locus;
Second movement locus acquisition module is configured to the preset trajectory predictions method, and according to the 2nd~4
The continuous objective body tracing point, obtains the second movement locus;
Third movement locus acquisition module is configured to the preset trajectory predictions method, and according to the 3rd~5
The continuous objective body tracing point, obtains third movement locus;
Target body movements track acquisition module is configured to the first fortune acquired in the first movement locus acquisition module
The second movement locus acquired in dynamic rail mark, the second movement locus acquisition module and the third movement locus acquisition module
Acquired third movement locus is modified, and obtains target body movements track.
Further, an optimal technical scheme provided by the invention is:
It the system also includes trajectory prediction module, is configured to execute following operation, so that the system can obtain
First movement locus, the second movement locus and third movement locus:
The spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, calculates target
The initial motion speed of body and initial motion direction;
Using the second target body tracing point in three continuous objective body tracing points as initial position;According to described
Initial position, initial motion speed and initial motion direction, and according to the following formula shown in method, calculate in current kinetic track
Objective body tracing point:
Wherein, (xi,yi,yi) be the current kinetic track in i-th of objective body tracing point spatial position coordinate;
(v0x,v0y,v0z) be the initial position spatial position coordinate;The viIt is objective body in i-th of objective body track
The movement velocity of point;The αiFor objective body i-th of objective body tracing point the direction of motion;The tsTo utilize camera shooting
System obtains the sampling interval of objective body depth image.
Further, an optimal technical scheme provided by the invention is:
The trajectory prediction module includes target body movements speed/direction calculating unit, is configured to execute following operation:
Movement velocity v of the method calculating objective body in i-th of objective body tracing point shown according to the following formulaiAnd the direction of motion
αi:
Wherein, the vi-1For objective body (i-1)-th objective body tracing point movement velocity;The αi-1Exist for objective body
The direction of motion of (i-1)-th objective body tracing point;The k is coefficient of air resistance;The g is acceleration of gravity.
Further, an optimal technical scheme provided by the invention is:
Target body movements track acquisition module includes the first track correct unit, the second track correct unit and third
Track correct unit;
The first track correct unit is configured in the 4th movement locus point and the 5th movement locus
Point is without departing from first movement locus, and the 5th movement locus point is without departing from the feelings of second movement locus
Under condition, according to the following formula shown in method obtain target body movements track Tm:
Wherein, the T1、T2And T3Respectively the first movement locus, the second movement locus and third movement locus;
The second track correct unit is configured in the 4th movement locus point or the 5th movement locus
The case where point deviates first movement locus, and the 5th movement locus point is without departing from second movement locus
Under, according to the following formula shown in method obtain target body movements track Tm:
The third track correct unit is configured in the 4th movement locus point or the 5th movement locus
In the case that point deviates first movement locus, and the 5th movement locus point deviates second movement locus,
The target body movements track Tm=T3。
Further, an optimal technical scheme provided by the invention is:
Target body movements track acquisition module further includes objective body deviation computing unit, the deviation judgement of first object body
Unit, the second objective body deviate judging unit and third objective body deviates judging unit;
The objective body deviation computing unit is configured to calculate the 4th movement locus point and the first movement rail
The space length W of markn+1_1, calculate the 5th movement locus point respectively with first movement locus and the second movement locus
Space length Wn+2_1And Wn+2_2;
The first object body deviates judging unit, is configured to according to the space length Wn+1_1With predetermined threshold value e, judge
Whether the 4th movement locus point deviates first movement locus:If Wn+1_1> e, then the 4th movement locus point
Deviate first movement locus;
Second objective body deviates judging unit, is configured to according to the space length Wn+2_1With the predetermined threshold value e,
Judge whether the 5th movement locus point deviates first movement locus:If Wn+2_1> e, then the 5th movement rail
Mark point deviates first movement locus;
The third objective body deviates judging unit, is configured to according to the space length Wn+2_2With the predetermined threshold value e,
Judge whether the 5th movement locus point deviates second movement locus:If Wn+2_2> e, then the 5th movement rail
Mark point deviates second movement locus.
Compared with the immediate prior art, above-mentioned technical proposal at least has the advantages that:
1, the target body movements trajectory predictions method based on depth image in the present invention, can be based on objective body depth map
The objective body tracing point as in, a plurality of movement locus of piecewise prediction, and then can be according to the sky of objective body tracing point and movement locus
Between distance, choose the higher movement locus of accuracy rate, best target trajectory can be obtained according to these movement locus.
2, the target body movements trajectory predictions method based on depth image in the present invention, according to photographic device precision set
Whether the threshold value of space length deviates a upper track so as to the tracing point in accurate judgement current track.
3, the target body movements trajectory predictions method based on depth image in the present invention, using three continuous objective bodies
One section of movement locus of track point prediction, and according to three sections of continuous movement locus, best target body movements track can be obtained
With objective body pick-up point.
Description of the drawings
Fig. 1 is a kind of key step of the target body movements trajectory predictions method based on depth image in the embodiment of the present invention
Schematic diagram;
Fig. 2 is the main step of target body movements trajectory predictions system of the another kind based on depth image in the embodiment of the present invention
Rapid schematic diagram;
Fig. 3 is a kind of primary structure of the target body movements trajectory predictions system based on depth image in the embodiment of the present invention
Schematic diagram.
Specific implementation mode
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this
A little embodiments are used only for explaining the technical principle of the present invention, it is not intended that limit the scope of the invention.
Refering to attached drawing 1, it is pre- that Fig. 1 illustrates a kind of target body movements track based on depth image in the present embodiment
The key step of survey method.As shown in Figure 1, the movement locus of objective body can be predicted in the present embodiment in accordance with the following steps:
Step S101:According to the objective body depth image obtained in advance, multiple continuous objective body tracing points are obtained.Specifically
Photographic device, such as depth camera may be used in the present embodiment for ground, obtain the depth image in objective body moving process, and
And conventional depth image processing method may be used and obtain objective body tracing point.For example, may be used in the present embodiment
Depth Basics-D2D programs disclosed in Kinect for Windows v2.0 SDK obtain objective body tracing point.
Step S102:Based on preset trajectory predictions method, and according to the 1st~3 objective body tracing point, obtain the first fortune
Dynamic rail mark.
Specifically, preset trajectory predictions method includes the following steps in the present embodiment:
First, the spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, is calculated
The initial motion speed of objective body and initial motion direction.
Secondly, using the second target body tracing point in three continuous objective body tracing points as initial position;Root
According to initial position, initial motion speed and initial motion direction, and method shown in (1) according to the following formula, calculate current kinetic
Objective body tracing point in track:
Each meaning of parameters is in formula (1):
(xi,yi,yi) be current kinetic track in i-th of objective body tracing point spatial position coordinate.(v0x,v0y,v0z)
For the spatial position coordinate of initial position.viFor objective body i-th of objective body tracing point movement velocity.αiFor objective body
In the direction of motion of i-th of objective body tracing point.tsTo obtain the sampling interval of objective body depth image using photographic device.
Movement velocity v of the objective body in i-th of objective body tracing point in the present embodimentiWith direction of motion αiSuch as following formula (2) institute
Show:
Each meaning of parameters is in formula (2):
vi-1For objective body (i-1)-th objective body tracing point movement velocity.αi-1It is objective body in (i-1)-th objective body
The direction of motion of tracing point.K is coefficient of air resistance.G is acceleration of gravity.
Further, it is based on above-mentioned preset trajectory predictions method, the method for target body movements trajectory predictions shown in Fig. 1 can be with
The first movement locus is obtained in accordance with the following steps:
Step S1021:Obtain the 1st~3 successive objective body tracing point P1、P2And P3Spatial position, i.e.,With
Step S1022:By objective body tracing point P2As initial position, and (3) and (4) can calculate according to the following formula
To initial motion speed (v of the objective body in the first movement locus0x,v0y,v0z) and initial motion direction α0:
Step S1023:According to initial position, initial motion speed and initial motion direction, and according to shown in formula (1)
Method, calculate other objective body tracing points in the first movement locus, and then obtain the first movement locus.
Step S103:Based on preset trajectory predictions method, and according to the 2nd~4 continuous objective body tracing point, obtain
Second movement locus.The acquisition methods of the second movement locus are identical as the acquisition methods of the first movement locus in the present embodiment, tool
Body may include steps of:
Step S1031:Obtain the 2nd~4 successive objective body tracing point P2、P3And P4Spatial position, i.e.,With
Step S1032:By objective body tracing point P3As initial position, and according to the side described in formula (3) and (4)
Initial motion speed (v of the objective body in the second movement locus is calculated in method0x,v0y,v0z) and initial motion direction α0。
Step S1033:According to initial position, initial motion speed and initial motion direction, and according to shown in formula (1)
Method, calculate other objective body tracing points in the second movement locus, and then obtain the second movement locus.
Step S104:Based on preset trajectory predictions method, and according to the 3rd~5 continuous objective body tracing point, obtain
Third movement locus.
Step S1041:Obtain the 3rd~5 successive objective body tracing point P3、P4And P5Spatial position, i.e.,With
Step S1032:By objective body tracing point P4As initial position, and according to the side described in formula (3) and (4)
Initial motion speed (v of the objective body in third movement locus is calculated in method0x,v0y,v0z) and initial motion direction α0。
Step S1033:According to initial position, initial motion speed and initial motion direction, and according to shown in formula (1)
Method, calculate other objective body tracing points in third movement locus, and then obtain third movement locus.
Step S105:First movement locus, the second movement locus and third movement locus are modified, objective body is obtained
Movement locus.Specifically, target body movements track can be obtained in the present embodiment as follows:
Step S1051:Judge objective body tracing point whether bias motion track:
First, the space length W of the 4th movement locus point and the first movement locus is calculatedn+1_1, calculate the 5th movement
The tracing point space length W with the first movement locus and the second movement locus respectivelyn+2_1And Wn+2_2。
Secondly, according to space length Wn+1_1With predetermined threshold value e, judge whether the 4th movement locus point deviates the first movement
Track;According to space length Wn+2_1With predetermined threshold value e, judge whether the 5th movement locus point deviates the first movement locus;Root
According to space length Wn+2_2With predetermined threshold value e, judge whether the 5th movement locus point deviates the second movement locus.
If Wn+1_1> e, then the 4th movement locus point deviate the first movement locus;
If Wn+2_1> e, then the 5th movement locus point deviate the first movement locus;
If Wn+2_2> e, then the 5th movement locus point deviate the second movement locus.
Step S1052:According to above-mentioned judging result, target body movements track T is obtainedm:
1, in the 4th movement locus point and the 5th movement locus point without departing from the first movement locus, and the 5th
In the case that movement locus o'clock is without departing from the second movement locus, method shown in (5) target body movements can be obtained according to the following formula
Track Tm:
Each meaning of parameters is in formula (5):T1、T2And T3Respectively the first movement locus, the second movement locus and third fortune
Dynamic rail mark.
2, deviate the first movement locus, and the 5th movement in the 4th movement locus point or the 5th movement locus point
In the case that tracing point is without departing from the second movement locus, method shown in (6) obtains target body movements track T according to the following formulam:
3, deviate the first movement locus, and the 5th movement in the 4th movement locus point or the 5th movement locus point
In the case that tracing point deviates the second movement locus, target body movements track Tm=T3。
Refering to attached drawing 2, Fig. 2 illustrates target body movements track of the another kind based on depth image in the present embodiment
The key step of prediction technique.As shown in Fig. 2, the movement locus of objective body can be predicted in the present embodiment in accordance with the following steps:
Step S201:Obtain objective body depth image.
Step S202:Obtain the spatial position of 1-3 continuous objective body tracing points.
Step S203:Calculate initial motion speed and initial velocity direction of the objective body in current kinetic track.Specifically,
It is initial in current kinetic track that objective body can be calculated according to method shown in formula (3) and (4) in the present embodiment
Movement velocity (v0x,v0y,v0z) and initial motion direction α0。
Step S204:According to initial position, initial motion speed and initial velocity direction, calculate in current kinetic track
Other objective body tracing points.Specifically, current kinetic rail can be calculated according to method shown in formula (1) in the present embodiment
Other objective body tracing points in mark.
Step S205:According to 1-3 continuous objective body tracing points and its other corresponding objective body tracing points, shape
At the first movement locus.
Step S206:The spatial position of the 4th objective body tracing point is obtained, and repeats step S203- step S204,
Obtain the second movement locus.
Step S207:The spatial position of the 5th objective body tracing point is obtained, and repeats step S203- step S204,
Obtain third movement locus.
Step S208:Judge whether the 4th objective body tracing point or the 5th objective body tracing point deviate the first movement rail
Mark.Specifically, if the 4th objective body tracing point or the 5th objective body tracing point deviate the first movement locus, step is gone to
S209.If the 4th objective body tracing point and the 5th objective body tracing point go to step without departing from the first movement locus
S213。
Step S209:Abandon the first movement locus.
Step S210:Judge whether the 5th objective body tracing point deviates the second movement locus.Specifically, if the 5th target
Body tracing point then goes to step S211 without departing from the second movement locus.If the 5th objective body tracing point deviates the second movement rail
Mark then abandons the second movement locus, and goes to step S212.
Step S211:According to second and third movement locus, target body movements track is obtained.Specifically, in the present embodiment
Target body movements track can be obtained according to method shown in formula (6).
Step S212:According to third movement locus, target body movements track is obtained.Specifically, objective body in the present embodiment
Movement locus is third movement locus.
Step S213:Judge whether the 5th objective body tracing point deviates the second movement locus.Specifically, if the 5th target
Body tracing point deviates the second movement locus, then goes to step S214.If the 5th objective body tracing point is without departing from the second movement rail
Mark then goes to step S215.
Step S214:Abandon the second movement locus.
Step S215:According to the first, second, and third movement locus, target body movements track is obtained.Specifically, this implementation
Target body movements track can be obtained according to method shown in formula (5) in example.
Although each step is described in the way of above-mentioned precedence in above-described embodiment, this field
Technical staff is appreciated that realize the effect of the present embodiment, is executed not necessarily in such order between different steps,
It (parallel) execution simultaneously or can be executed with reverse order, these simple variations all protection scope of the present invention it
It is interior.
Based on technical concept identical with embodiment of the method, the embodiment of the present invention also provides a kind of mesh based on depth image
Standard type movement locus forecasting system.To this, the target body movements trajectory predictions system based on depth image carries out below in conjunction with the accompanying drawings
It illustrates.
Refering to attached drawing 2, it is pre- that Fig. 2 illustrates a kind of target body movements track based on depth image in the present embodiment
The primary structure of examining system.As shown in Fig. 2, the target body movements trajectory predictions system based on depth image can be in the present embodiment
Including objective body tracing point acquisition module 11, the first movement locus acquisition module 12, the second movement locus acquisition module 13, third
Movement locus acquisition module 14 and target body movements track acquisition module 15.Specifically, objective body tracing point obtains in the present embodiment
Modulus block 11 is configurable to, according to the objective body depth image obtained in advance, obtain multiple continuous objective body tracing points.The
One movement locus acquisition module 12 is configurable to be based on preset trajectory predictions method, and according to the 1st~3 objective body track
Point obtains the first movement locus.Second movement locus acquisition module 13 is configurable to be based on preset trajectory predictions method, and
According to the 2nd~4 continuous objective body tracing point, the second movement locus is obtained.Third movement locus acquisition module 14 can match
It is set to and is based on preset trajectory predictions method, and according to the 3rd~5 continuous objective body tracing point, obtain third movement locus.
Target body movements track acquisition module 15 is configurable to the first movement rail acquired in the first movement locus acquisition module 12
Acquired in the second movement locus and third movement locus acquisition module 13 acquired in mark, the second movement locus acquisition module 13
Third movement locus is modified, and obtains target body movements track.
Further, in this embodiment the system of objective body trajectory predictions shown in Fig. 2 can also include trajectory prediction module,
It is configured to execute following operation, so that the system can obtain the first movement locus, the second movement locus and third movement locus:
First, the spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, is calculated
The initial motion speed of objective body and initial motion direction.
Secondly, using the second target body tracing point in three continuous objective body tracing points as initial position;Root
According to initial position, initial motion speed and initial motion direction, and according to method shown in formula (1), calculate current kinetic
Objective body tracing point in track.
Further, in this embodiment trajectory prediction module may include target body movements speed/direction calculating unit,
It is configured to according to method calculating objective body shown in formula (2) in the movement velocity v of i-th of objective body tracing pointiWith movement side
To αi。
Further, in this embodiment target body movements track acquisition module 15 may include the first track correct unit,
Second track correct unit and third track correct unit.
Specifically, the first track correct unit is configurable in the 4th movement locus point and the 5th movement locus point
Without departing from the first movement locus, and in the case that the 5th movement locus o'clock is without departing from the second movement locus, according to formula (5)
Shown in method obtain target body movements track Tm.Second track correct unit be configurable in the 4th movement locus point or
The case where 5th movement locus point deviates the first movement locus, and the 5th movement locus o'clock is without departing from the second movement locus
Under, obtain target body movements track T according to method shown in formula (6)m.Third track correct unit is configurable to the 4th
A movement locus point or the 5th movement locus point deviate the first movement locus, and the 5th movement locus point deviates the second fortune
In the case of dynamic rail mark, the target body movements track Tm=T3。
Further, in this embodiment target body movements track acquisition module 15, which can also include objective body deviation, calculates list
Member, first object body deviate judging unit, the second objective body deviates judging unit and third objective body deviates judging unit.
Specifically, objective body deviation computing unit is configurable to calculate the 4th movement locus point and the first movement locus
Space length Wn+1_1, calculate the 5th movement locus point space length with the first movement locus and the second movement locus respectively
Wn+2_1And Wn+2_2.First object body deviates judging unit and is configurable to according to space length Wn+1_1With predetermined threshold value e, judge
Whether the 4th movement locus point deviates the first movement locus:If Wn+1_1> e, then the 4th the first movement of movement locus point deviation
Track.Second objective body deviates judging unit and is configurable to according to space length Wn+2_1With predetermined threshold value e, the 5th fortune is judged
Whether dynamic tracing point deviates the first movement locus:If Wn+2_1> e, then the 5th movement locus point deviate the first movement locus.The
Three objective bodies deviate judging unit and are configurable to according to space length Wn+2_2With predetermined threshold value e, the 5th movement locus is judged
Whether point deviates the second movement locus:If Wn+2_2> e, then the 5th movement locus point deviate the second movement locus.
The above-mentioned target body movements trajectory predictions system embodiment based on depth image can be used for executing above-mentioned based on deep
Spend image target body movements trajectory predictions embodiment of the method, technical principle, it is solved the technical issues of and generation technology
Effect is similar, and person of ordinary skill in the field can be understood that, for convenience and simplicity of description, foregoing description
The specific work process and related explanation of target body movements trajectory predictions system based on depth image can refer to aforementioned be based on
Corresponding process in the target body movements trajectory predictions embodiment of the method for depth image, details are not described herein.
It will be understood by those skilled in the art that the above-mentioned target body movements trajectory predictions system based on depth image further includes
Some other known features, such as processor, memory etc., wherein memory include but not limited to random access memory, flash memory,
Read-only memory, programmable read only memory, volatile memory, nonvolatile memory, serial storage, parallel storage
Or register etc., processor includes but not limited to CPLD/FPGA, DSP, arm processor, MIPS processors etc., in order to unnecessary
Ground obscures embodiment of the disclosure, these well known structures are not shown in FIG. 2.
It should be understood that the quantity of the modules in Fig. 2 is only schematical.According to actual needs, each module can be with
With arbitrary quantity.
It will be understood by those skilled in the art that can adaptively be changed to the module in the system in embodiment
And they are arranged in the one or more systems different from the embodiment.It can be the module or unit group in embodiment
A module or unit are synthesized, and multiple submodule or subelement can be divided into addition.In addition to such feature and/
Or except at least some of process or unit exclude each other, it is (including adjoint to this specification that any combinations may be used
Claim, abstract and attached drawing) disclosed in all features and so disclosed any method or equipment all mistakes
Journey or unit are combined.Unless expressly stated otherwise, public in this specification (including adjoint claim, abstract and attached drawing)
The each feature opened can be replaced by providing the alternative features of identical, equivalent or similar purpose.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference mark between bracket should not be configured to limitations on claims.Word " comprising " does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in claims of the present invention, embodiment claimed
It is one of arbitrary mode to use in any combination.
So far, it has been combined preferred embodiment shown in the drawings and describes technical scheme of the present invention, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific implementation modes.Without departing from this
Under the premise of the principle of invention, those skilled in the art can make the relevant technologies feature equivalent change or replacement, these
Technical solution after change or replacement is fallen within protection scope of the present invention.
Claims (10)
1. a kind of target body movements trajectory predictions method based on depth image, which is characterized in that the method includes:
According to the objective body depth image obtained in advance, multiple continuous objective body tracing points are obtained;
Based on preset trajectory predictions method, and according to the 1st~3 continuous objective body tracing point, obtain the first movement
Track;
Based on the preset trajectory predictions method, and according to the 2nd~4 continuous objective body tracing point, obtain second
Movement locus;
Based on the preset trajectory predictions method, and according to the 3rd~5 continuous objective body tracing point, obtain third
Movement locus;
First movement locus, the second movement locus and third movement locus are modified, target body movements track is obtained.
2. the target body movements trajectory predictions method according to claim 1 based on depth image, which is characterized in that described
Preset trajectory predictions method specifically includes:
The spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, calculates objective body
Initial motion speed and initial motion direction;
Using the second target body tracing point in three continuous objective body tracing points as initial position;According to described initial
Location point, initial motion speed and initial motion direction, and according to the following formula shown in method, calculate the mesh in current kinetic track
Standard type tracing point:
Wherein, (xi,yi,yi) be the current kinetic track in i-th of objective body tracing point spatial position coordinate;(v0x,v0y,
v0z) be the initial position spatial position coordinate;The viFor objective body i-th of objective body tracing point movement
Speed;The αiFor objective body i-th of objective body tracing point the direction of motion;The tsTo be obtained using photographic device
The sampling interval of objective body depth image.
3. the target body movements trajectory predictions method according to claim 2 based on depth image, which is characterized in that described
Movement velocity v of the objective body in i-th of objective body tracing pointiWith direction of motion αiIt is shown below:
Wherein, the vi-1For objective body (i-1)-th objective body tracing point movement velocity;The αi-1It is objective body i-th-
The direction of motion of 1 objective body tracing point;The k is coefficient of air resistance;The g is acceleration of gravity.
4. according to target body movements trajectory predictions method of the claim 1-3 any one of them based on depth image, feature
It is, " first movement locus, the second movement locus and third movement locus is modified, target body movements rail is obtained
The step of mark ", specifically includes:
In the 4th movement locus point and the 5th movement locus point without departing from first movement locus, and
In the case that the 5th movement locus point is without departing from second movement locus, according to the following formula shown in method obtain target
Body movement locus Tm:
Deviate first movement locus in the 4th movement locus point or the 5th movement locus point, and described
In the case that 5th movement locus point is without departing from second movement locus, according to the following formula shown in method obtain objective body fortune
Dynamic rail mark Tm:
Deviate first movement locus in the 4th movement locus point or the 5th movement locus point, and described
In the case that 5th movement locus point deviates second movement locus, the target body movements track Tm=T3;
Wherein, the T1、T2And T3Respectively the first movement locus, the second movement locus and third movement locus.
5. the target body movements trajectory predictions method according to claim 4 based on depth image, which is characterized in that " right
First movement locus, the second movement locus and third movement locus are modified, and obtain target body movements track " the step of
Include before:
Calculate the space length W of the 4th the movement locus point and first movement locusn+1_1, calculate the 5th fortune
The dynamic tracing point space length W with first movement locus and the second movement locus respectivelyn+2_1And Wn+2_2;
According to the space length Wn+1_1With predetermined threshold value e, judge whether the 4th movement locus point deviates described first
Movement locus:If Wn+1_1> e, then the 4th movement locus point deviate first movement locus;
According to the space length Wn+2_1With the predetermined threshold value e, it is described to judge whether the 5th movement locus point deviates
First movement locus:If Wn+2_1> e, then the 5th movement locus point deviate first movement locus;
According to the space length Wn+2_2With the predetermined threshold value e, it is described to judge whether the 5th movement locus point deviates
Second movement locus:If Wn+2_2> e, then the 5th movement locus point deviate second movement locus.
6. a kind of target body movements trajectory predictions system based on depth image, which is characterized in that the system comprises:
Objective body tracing point acquisition module is configured to, according to the objective body depth image obtained in advance, obtain multiple continuous mesh
Standard type tracing point;
First movement locus acquisition module is configured to preset trajectory predictions method, and according to the 1st~3 target
Body tracing point obtains the first movement locus;
Second movement locus acquisition module is configured to the preset trajectory predictions method, and continuous according to the 2nd~4
The objective body tracing point, obtain the second movement locus;
Third movement locus acquisition module is configured to the preset trajectory predictions method, and continuous according to the 3rd~5
The objective body tracing point, obtain third movement locus;
Target body movements track acquisition module is configured to the first movement rail acquired in the first movement locus acquisition module
The second movement locus and the third movement locus acquisition module acquired in mark, the second movement locus acquisition module are obtained
The third movement locus taken is modified, and obtains target body movements track.
7. the target body movements trajectory predictions system according to claim 6 based on depth image, which is characterized in that described
System further includes trajectory prediction module, is configured to execute following operation, so that the system can obtain first movement
Track, the second movement locus and third movement locus:
The spatial position of three continuous objective body tracing points is obtained, and according to acquired spatial position, calculates objective body
Initial motion speed and initial motion direction;
Using the second target body tracing point in three continuous objective body tracing points as initial position;According to described initial
Location point, initial motion speed and initial motion direction, and according to the following formula shown in method, calculate the mesh in current kinetic track
Standard type tracing point:
Wherein, (xi,yi,yi) be the current kinetic track in i-th of objective body tracing point spatial position coordinate;(v0x,v0y,
v0z) be the initial position spatial position coordinate;The viFor objective body i-th of objective body tracing point movement
Speed;The αiFor objective body i-th of objective body tracing point the direction of motion;The tsTo be obtained using camera system
The sampling interval of objective body depth image.
8. the target body movements trajectory predictions system according to claim 7 based on depth image, which is characterized in that described
Trajectory prediction module includes target body movements speed/direction calculating unit, is configured to execute following operation:
Movement velocity v of the method calculating objective body in i-th of objective body tracing point shown according to the following formulaiWith direction of motion αi:
Wherein, the vi-1For objective body (i-1)-th objective body tracing point movement velocity;The αi-1It is objective body i-th-
The direction of motion of 1 objective body tracing point;The k is coefficient of air resistance;The g is acceleration of gravity.
9. according to target body movements trajectory predictions system of the claim 6-8 any one of them based on depth image, feature
It is, target body movements track acquisition module includes the first track correct unit, the second track correct unit and third rail
Mark amending unit;
The first track correct unit is configured to equal in the 4th movement locus point and the 5th movement locus point
Without departing from first movement locus, and the case where the 5th movement locus point is without departing from second movement locus
Under, according to the following formula shown in method obtain target body movements track Tm:
Wherein, the T1、T2And T3Respectively the first movement locus, the second movement locus and third movement locus;
The second track correct unit is configured to inclined in the 4th movement locus point or the 5th movement locus point
From first movement locus, and in the case that the 5th movement locus point is without departing from second movement locus, press
Target body movements track T is obtained according to method shown in following formulam:
The third track correct unit is configured to inclined in the 4th movement locus point or the 5th movement locus point
It is described from first movement locus, and in the case of the 5th movement locus point deviation second movement locus
Target body movements track Tm=T3。
10. the target body movements trajectory predictions system according to claim 9 based on depth image, which is characterized in that institute
It further includes objective body deviation computing unit, first object body deviation judging unit, second to state target body movements track acquisition module
Objective body deviates judging unit and third objective body deviates judging unit;
The objective body deviation computing unit is configured to calculate the 4th movement locus point and first movement locus
Space length Wn+1_1, calculate the 5th movement locus point sky with first movement locus and the second movement locus respectively
Between distance Wn+2_1And Wn+2_2;
The first object body deviates judging unit, is configured to according to the space length Wn+1_1With predetermined threshold value e, described in judgement
Whether the 4th movement locus point deviates first movement locus:If Wn+1_1> e, then the 4th movement locus point deviation
First movement locus;
Second objective body deviates judging unit, is configured to according to the space length Wn+2_1With the predetermined threshold value e, judge
Whether the 5th movement locus point deviates first movement locus:If Wn+2_1> e, then the 5th movement locus point
Deviate first movement locus;
The third objective body deviates judging unit, is configured to according to the space length Wn+2_2With the predetermined threshold value e, judge
Whether the 5th movement locus point deviates second movement locus:If Wn+2_2> e, then the 5th movement locus point
Deviate second movement locus.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810196470.8A CN108596941B (en) | 2018-03-09 | 2018-03-09 | Target body motion trajectory prediction method and system based on depth image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810196470.8A CN108596941B (en) | 2018-03-09 | 2018-03-09 | Target body motion trajectory prediction method and system based on depth image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108596941A true CN108596941A (en) | 2018-09-28 |
CN108596941B CN108596941B (en) | 2020-10-02 |
Family
ID=63625957
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810196470.8A Active CN108596941B (en) | 2018-03-09 | 2018-03-09 | Target body motion trajectory prediction method and system based on depth image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108596941B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112328086A (en) * | 2020-11-14 | 2021-02-05 | 上海卓腾展览展示有限公司 | Intelligent display method, system and device based on video identification and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012096907A1 (en) * | 2011-01-10 | 2012-07-19 | Hangzhou Conformal & Digital Technology Limited Corporation | Mesh animation |
CN102752855A (en) * | 2012-08-01 | 2012-10-24 | 重庆大学 | Indoor personnel positioning system and method based on path rule and prediction |
CN106454722A (en) * | 2016-09-09 | 2017-02-22 | 华南理工大学 | Dynamic differential positioning method based on map matching for mobile communication terminals |
CN107645701A (en) * | 2017-09-12 | 2018-01-30 | 咪咕互动娱乐有限公司 | A kind of method and device for generating movement locus |
-
2018
- 2018-03-09 CN CN201810196470.8A patent/CN108596941B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012096907A1 (en) * | 2011-01-10 | 2012-07-19 | Hangzhou Conformal & Digital Technology Limited Corporation | Mesh animation |
CN102752855A (en) * | 2012-08-01 | 2012-10-24 | 重庆大学 | Indoor personnel positioning system and method based on path rule and prediction |
CN106454722A (en) * | 2016-09-09 | 2017-02-22 | 华南理工大学 | Dynamic differential positioning method based on map matching for mobile communication terminals |
CN107645701A (en) * | 2017-09-12 | 2018-01-30 | 咪咕互动娱乐有限公司 | A kind of method and device for generating movement locus |
Non-Patent Citations (3)
Title |
---|
MICHAEL L. GLEICHER 等: "Re-Cinematography:Improving the Camera Dynamics of Casual Video", 《PROCEEDING MM "07 PROCEEDINGS OF THE 15TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA》 * |
TING, H. Y. 等: "Kinect-Based Badminton Movement Recognition and Analysis System", 《INTERNATIONAL JOURNAL OF COMPUTER SCIENCE IN SPORT》 * |
程照辉 等: "基于运动特征的视频检索", 《信号处理》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112328086A (en) * | 2020-11-14 | 2021-02-05 | 上海卓腾展览展示有限公司 | Intelligent display method, system and device based on video identification and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108596941B (en) | 2020-10-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
He et al. | The parameters selection of PSO algorithm influencing on performance of fault diagnosis | |
CN111176334B (en) | Multi-unmanned aerial vehicle cooperative target searching method | |
CN107677285B (en) | The path planning system and method for robot | |
CN107293115B (en) | Traffic flow prediction method for microscopic simulation | |
CN105259786B (en) | The inertial parameter discrimination method and device of target to be identified | |
CN102999664B (en) | Method for identifying position of indoor pollution source | |
CN110009037B (en) | Short-term engineering wind speed prediction method and system based on physical information coupling | |
CN104748735B (en) | Indoor orientation method and equipment based on intelligent terminal | |
US20130235047A1 (en) | Method for animating characters, with collision avoidance based on tracing information | |
AU2008242213A1 (en) | A method and a system for estimating the impact area of a military load launched from an aircraft | |
CN109663323A (en) | A kind of motion state monitoring method and wearable device based on wearable device | |
CN107677267A (en) | Indoor pedestrian navigation course feedback modifiers method based on MEMS IMU | |
CN109903367B (en) | Method, apparatus and computer readable storage medium for constructing map | |
CN108596941A (en) | Target body movements trajectory predictions method and system based on depth image | |
CN110456902A (en) | It is mobile to control the skeleton pattern in computer system to track user | |
CN102607591A (en) | Track data generation method for testing strap-down inertial navigation software | |
Mironov et al. | Applying neural networks for prediction of flying objects trajectory | |
Praczyk | Using evolutionary neural networks to predict spatial orientation of a ship | |
CN111414720B (en) | Flow field vortex detection method based on neural network | |
KR102480382B1 (en) | Wind load estimation system based on artificial intelligence | |
CN115542746A (en) | Energy control reentry guidance method and device for hypersonic aircraft | |
CN108054834A (en) | A kind of multistage energy coordinated control system | |
JP6831523B2 (en) | Information processing device | |
Lu et al. | LPNet: A reaction-based local planner for autonomous collision avoidance using imitation learning | |
CN103632008B (en) | A kind of vehicle movement analogue simulation method |
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 |