CN104102835B - Throw training score detection method based on real-time video images - Google Patents
Throw training score detection method based on real-time video images Download PDFInfo
- Publication number
- CN104102835B CN104102835B CN201410332067.5A CN201410332067A CN104102835B CN 104102835 B CN104102835 B CN 104102835B CN 201410332067 A CN201410332067 A CN 201410332067A CN 104102835 B CN104102835 B CN 104102835B
- Authority
- CN
- China
- Prior art keywords
- target object
- raindrop
- throwing
- training
- real
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims description 15
- 238000002372 labelling Methods 0.000 claims description 13
- 230000009471 action Effects 0.000 claims description 10
- 238000006073 displacement reaction Methods 0.000 claims description 10
- 239000012530 fluid Substances 0.000 claims description 8
- FWQHNLCNFPYBCA-UHFFFAOYSA-N fluoran Chemical compound C12=CC=CC=C2OC2=CC=CC=C2C11OC(=O)C2=CC=CC=C21 FWQHNLCNFPYBCA-UHFFFAOYSA-N 0.000 claims description 7
- 230000000007 visual effect Effects 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 5
- 230000003247 decreasing effect Effects 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 241000669244 Unaspis euonymi Species 0.000 claims description 3
- 238000013500 data storage Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 230000035939 shock Effects 0.000 claims 1
- 238000004364 calculation method Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000006467 substitution reaction Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a throw training score detection method based on real-time video images. The method includes the steps of S1, setting a throw training field, marking a target object, and setting weather information for throw training; S2, calculating a scale factor by scaled calibration; S3, throwing the target object, tracing and shooting the target object, establishing a two-dimensional coordinate system in each frame image, capturing spatial coordinates of the target object in the frame image, in the two-dimensional coordinate system, and storing the captured spatial coordinates of the image in a fixed-length linked list in real time; S4, reading a plurality of nodes at the tail end of the linked list, calculating image spatial velocities of the target object, with horizontal and vertical speeds of the target object in the image being expressed as pixels per second; S5, calculating actual horizontal speed and vertical speeds of the target object at a moment before throwing, and calculating a horizontal throw distance and a vertical throw distance of the target object so as to detect throw training scores.
Description
Technical field
The invention belongs to training achievement horizontal detection field is thrown, more particularly to a kind of throwing based on real time video image
Training achievement detection method.
Background technology
As the economic military strength of China constantly strengthens in recent years, the military power of construction of modern becomes adaptation country and sends out
Exhibition and adapt to future war one kind in the urgent need to.And construction of modern army requires that the science and technology improved in army building contains
Amount, including features such as mechanization, automatization, informationizations.Arrive greatly advanced weaponry to update, the little daily instruction to army
Practice manoeuvre, will constantly incorporate new science and technology and method.
In the daily throwing training of army：Grenade in army of China is thrown training and mainly also rests on live shell at present
Field training.On the one hand, the very open place of training need is thrown, the normalization of training and exercises is which limits, is needed
Substantial amounts of manpower and materials are once trained to organize；On the other hand, due to practical training the characteristics of, training the data obtained is not
Can be typically canonicalized, informationalized record and perfect utilization, lose the value of contrast and reference, the effect of training will be beaten greatly
Discount；Sixty-four dollar question is, if carrying out live-fire training, inevitably to cause danger, and for example grenade is slipped out of the hand not
Circle can be successfully flown away from, so as to injure the personal safety for participating in trainer, in fact, such accident has been shown in not repeatly
It is fresh.
It can be seen that traditional throws the demand that training achievement level detection method cannot meet modern military forces use training.Therefore
When the daily throwing of army is trained, we can monitor the kinestate for throwing target object with Real-time Collection, using less field
Ground can just complete extensive person-time of throwing training.After analysis obtains scientific and normal data in addition, trainee can also be corresponded to
Member's input database is stored, and to compare analysis in the future, enables trainee to improve throwing skill to improve results.
The content of the invention
According to problems of the prior art, the invention discloses a kind of throwing based on real time video image is trained to
Achievement detection method, specifically includes following steps：
S1：Arrange and throw training court, process is marked to target object, weather conditions letter during training is thrown in collection
Breath；When such as throwing the live possibly rainy day or having landscape condition；
S2：Scale calculates scale factor：Record is taken the photograph in the case that between target object and photographic head, distance is certain
As in head, physical length is the image space length of one meter of object, i.e. scale factor using each how many meters of expression of pixel correspondence；
S3：Target object is thrown, the moving image letter before selling in throwing using photographic head Real-time Collection target object
Breath, is tracked shooting, the video image that real-time processing is collected to target object：Two-dimensional coordinate is set up in every two field picture
System, spatial coordinated information of the target object in two-dimensional coordinate system in the every two field picture of capture, by the image space coordinate for capturing
Information is stored in the chained list with regular length in real time, realizes the real-time update to capturing information in chained list；
S4：After determining that target object is sold, multiple nodes of end in reading chained list, are located at, shoot adjacent according to photographic head
The principle that time difference between two field picture is fixed, using the spatial coordinated information and correspondence moment letter of target object in multiple nodes
Breath, calculates target object space velocity in the picture, i.e., in image, the speed of target object horizontal direction and vertical direction is adopted
Represented for a how many pixel with per second；
S5：Image space is carried out with the ratiometric conversion of real space so as to count according to the scale and scale factor in S2
Practical level speed and vertical speed of the target object in previous moment of selling are calculated, so as to calculate the throwing of target object horizontal direction
The parabolic distance of thing distance and vertical direction, realizes throwing the detection of training achievement.
Further, in S1：When being marked process to target object：Mark on the target object and be easy to photographic head
The label information is simultaneously stored by the labelling of collection, when the moving image letter using photographic head Real-time Collection target object in throwing
During breath, photographic head can be identified to the above-mentioned labelling for having done accordingly.
Further, in S3：Throwing action is detected when throwing target object, by object in each two field picture for extracting
The spatial coordinated information of body is stored in chained list, when target object cannot be recognized by photographic head, from this two field picture corresponding moment
Start timing, when timing time length exceedes threshold value, then target object leaves the visual field of video camera completely.
Further, the node is the data structure that stores sequentially in time in chained list, in S4：Read data storage
In chained list positioned at multiple nodes of end when choose last two node, by target object sell the moment state using last
In two nodes, the coordinate parameters of target object are calculated.
Further, when weather conditions information when training is thrown in setting：When weather conditions are that fine day is calm then according to even
Fast equation of motion is calculated：Water is calculated with time data according to the target object location coordinate recorded in two last nodes
The speed of gentle vertical direction, then the displacement of target object horizontal direction and the displacement of vertical direction are calculated accordingly.
Further, when weather conditions information when training is thrown in setting：When weather conditions are to rain：It is fixed according to momentum
The collision of raindrop and target object is considered as momentum by reason, is raindrop and target object extremely short collision time action time, for
One drop raindrop, this momentum are expressed asWherein, Δ v is the end speed before raindrop impingemest target object, and Δ t is
Raindrop are decreased to zero time experienced to end speed by colliding, and m is the Approximation Quality of raindrop,Suffered by target object
To the active force of momentum；By the end speed Δ v of raindrop radius and raindrop, the size of the force of rain determines that the end speed of raindrop is by raindrop
Radius is derived from：
Wherein：A, B, C are constant, r1For raindrop radius (raindrop are approximately spheroid in calculating), n is for object
For body during whole projectile motion, the approximate number value of each moment raindrop impingemest target object is raindrop number density,
Can be obtained according to actual Weather information；Its theoretical value is calculated using equation below：N (D)=N0Exp (- λ D), wherein parameter lambda=
4.1I-0.21(mm-1)；N0=8000 (individual/m3)；In formula, N (D) is raindrop number density, and D is raindrop diameter, and I is raininess, then target
Object be subject to rain directed force F be：According to：Solve
The density (being similar to the density of water) of F, wherein ρ for raindrop, vtFor raindrop end speed, n and vtBy being obtained according to actual force of rain parameter
, therefore target object is affected by raindrop by the active force of gravity and raindrop, target object in vertical direction in the horizontal direction
Ignore, finally calculate in the case where considering target object by raindrop active force and throw training achievement.
Further, when weather conditions information when training is thrown in setting：When weather conditions have wind, in the horizontal direction
On：When marginal value of the target object movement velocity less than setting：If target object is acted on by viscosity resistance, viscosity resistance
Solution formula is：F=6 π η vr, wherein：Viscosities of the η for fluid, v are movement velocity of the target object relative to wind, and r is
Roll in target object motion the radius of the approximate circumsphere of attitude, the cylinder radius R of the radius of the circumsphere by target object
Calculated according to equation below with length h of target object：
The then fluoran stream surface area of target object
When marginal value of the target object movement velocity more than setting, if target object pressure difference resistance f,Wherein：C be form drag coefficient, ρ1For fluid density (atmospheric density), areas of the S for fluoran stream surface, v is mesh
Movement velocity of the mark object relative to wind, Re is Reynolds number,Wherein：ρ1For fluid density, v is object body phase
For the movement velocity of wind, η is viscosity, and d is characterized the length i.e. diameter of circumsphere；
In the vertical direction, only need to consider viscosity resistance, the calculating phase of its calculating process and viscosity resistance in horizontal direction
Together, finally calculate in the case where the active force of target object wind-engaging is considered and throw training achievement.
As a result of above-mentioned technical proposal, the present invention is simultaneously divided by appliance computer video capture technology, real time record
The analysis target object (can be grenade model etc.) thrown of trainee will be disengaged from shortly and depart from position after arm,
The information such as time, speed, even if realizing in more narrow space, it is also possible to simulate various natural environment to calculate throwing
The critical datas such as distance, height of throw and angle of trajectory.After obtaining scientific and normal data, can also be by data correspondence trainee
Member's input database is stored, and to compare analysis in the future, enables trainee to improve throwing skill to improve results, from
And realize the lifting of army's entirety fighting capacity.
The present invention's also has the advantages that：
1st, tradition is thrown training to be combined with computer video acquisition technique and computer virtual reality technology, by training knot
Fruit digitized, real time implementation show, and can utilize the achievement of each training of Database Systemsization ground storage.
2nd, detected to throwing training achievement using this method, can will throw the place reduced space to of training
In smaller area, to reach the purpose of save space.
3rd, training data can be digitized during actually detected, be stored with standardizing, and consider environment
The impact (the such as factor such as rainy day, air drag) of factor, makes testing result more accurate.
4th, when data analysiss being carried out to throwing level and being calculated, the mode of adoption rate calibrated and calculated scale factor is by mesh
The mode that the image space coordinate and real space coordinate, image space speed and real space speed of mark object combines is to throwing
The achievement level for throwing training is analyzed calculating.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present application or technical scheme of the prior art, below will be to embodiment or existing
Accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments described in application, for those of ordinary skill in the art, on the premise of not paying creative work,
Can be with according to these other accompanying drawings of accompanying drawings acquisition.
Fig. 1 is the simulation drawing of the throwing scene that the present invention is used；
The step of Fig. 2 is to throw training achievement level detection method in present invention schematic diagram；
Fig. 3 is the simplified program flow diagram of throwing training achievement level detection method in the present invention；
Fig. 4 is the detailed procedure flow chart of throwing training achievement level detection method in the present invention；
Fig. 5 be throw in the present invention in data acquisition process output data flow diagram；
Fig. 6 is the schematic diagram of target object change in location in every two field picture in the present invention；
Fig. 7 is schematic diagram of the target object in two-dimensional coordinate system in every two field picture in the present invention；
Fig. 8 is schematic diagram of the target object in two-dimensional coordinate system in every two field picture in the present invention；
Fig. 9 is simulation drawing when scale calculates scale factor in the present invention.
Specific embodiment
To make technical scheme and advantage clearer, with reference to the accompanying drawing in the embodiment of the present invention, to this
Technical scheme in inventive embodiments carries out clearly complete description：
A kind of throwing training achievement detection method based on real time video image, is specifically included following as shown in Figure 1 and Figure 2
Step：
S1：Arrange and throw training court, process is marked to target object, weather conditions letter during training is thrown in collection
Breath.
When being marked process to target object：The labelling general for being easy to photographic head collection is marked on the target object
The labelling is stored in a computer, and so the target object can be accurately identified by photographic head.According to the positional information of photographic head
Labelling is thrown personnel and makes the position of throwing action and the direction of throwing.Wherein correspond to training environment weather conditions (as wind,
Rain etc.), select different parameters inputs, it is ensured that analysis result it is accurate.
S2：Scale calculates scale factor：Record is taken the photograph in the case that between target object and photographic head, distance is certain
As in head, actual range is the image space length of one meter of object, i.e. scale factor using each how many meters of expression of pixel correspondence.
In order to make it easy to understand, as shown in figure 9, for the scale in the present invention calculates scale factor and illustrates being：
In the case that in throwing training court, setting completed, operator can be with the hard meter ruler that hand-held length is 1 meter, and stationary stations are in throwing
Throwing point position (i.e. at 2 meters of photographic head dead ahead), intercepts the video image of photographic head collection, is measured using Photoshop softwares
The image space length (i.e. 160 pixel units) of hard meter ruler in image, therefore the scale factor is each pixel correspondence
0.00625 meter.The scale factor is applied in S5 and is calculated.
S3：As shown in Fig. 6, Fig. 7 and Fig. 8：Target object is thrown, using photographic head Real-time Collection target object in throwing
Moving-picture information before selling, is tracked shooting, the video image that real-time processing is collected to target object：In every frame figure
Two-dimensional coordinate system is set up as in, spatial coordinated information of the target object in two-dimensional coordinate system in the every two field picture of capture will capture
To image space coordinate information be stored in the chained list with regular length in real time, realize reality to capturing information in chained list
Shi Gengxin.It is spatial coordinated information storage by target object in the picture in the step.
S4：After determining that target object is sold, multiple nodes of end are located in reading chained list, using target in multiple nodes
The spatial coordinated information and correspondence time information of object is calculated according to the principle that the time difference between the adjacent two field picture of photographic head is fixed
The image space speed of target object, i.e., in image, the speed of target object horizontal direction and vertical direction adopts per second for how many
Individual pixel is represented.
Time difference between adjacent two field picture is fixed, and is because that the frame per second of photographic head is fixed, i.e. each second can gather
Video frame images number be fixed, so the time difference is 1/ photographic head frame per second, therefore consecutive frame figure can be calculated
As the image space speed of X-direction and Y-direction within this time, that is, move how many pixel/seconds.
S5：Image space is carried out with the ratiometric conversion of real space so as to count according to the scale and scale factor in S2
Practical level speed and vertical speed of the target object in previous moment of selling are calculated, so as to calculate the throwing of target object horizontal direction
The parabolic distance of thing distance and vertical direction, realizes throwing the detection of training achievement.
And pass through early stage scale, it may be determined that the ratiometric conversion of image space and real space, i.e., 1 pixel unit
Correspondence is how much rice, therefore the real space speed that also can be just calculated within this time in X-direction and Y-direction, that is, move
Dynamic how many meter per seconds.
Further, in S1：When being marked process to target object：Mark on the target object and be easy to photographic head
The label information is simultaneously stored by the labelling of collection, when the moving image letter using photographic head Real-time Collection target object in throwing
During breath, photographic head can be identified to the above-mentioned labelling for having done accordingly.
Further, in S3:Throwing action is detected when throwing target object, by object in each two field picture for extracting
The spatial coordinated information of body is stored in chained list, when target object cannot be recognized by photographic head, from this two field picture corresponding moment
Start timing, when timing time exceedes threshold value, then target object leaves the visual field of video camera completely.In actually detected we
One time range of setting is threshold value, when target object exceedes the time range of threshold value, cannot be recognized by photographic head, then I
Think that target object is gone out by throwing.
Further, the node is the data structure that stores sequentially in time in chained list, in S4：Read data storage
In chained list positioned at multiple nodes of end when choose last two node, by target object sell the moment state using last
In two nodes, the coordinate parameters of target object are calculated.
When reading the multiple key nodes in data link table：Last two node or last two or more can be chosen
Node because in last several nodes target object kinestate closer to target object sell when kinestate.
Embodiment：
As shown in Fig. 3, Fig. 4 and Fig. 5：Step 1：Software condition and the hardware loop wanted needed for detection process is set first
Border, runs and initializes the graphical interface program of correspondence test system, setting weather and ambient parameter.Hardware device is as follows：One
Platform computer, the video camera of the high frame per second of a frame have carried out the target ammunition model (such as grenade model) of labelling, accept net
Pocket.Software arrangements are as follows：The virtual throwing training system of the present invention, records the data base of trainee's information.On computers
The virtual throwing training software of the present invention is installed, circle mark line is set in suitable region, target object is thrown to specify
Position.In circle, video camera is erected in one end of mark line direction perpendicular to the ground, and which should highly be close to the body of an adult
Height, and make video camera photographic head to direction it is parallel with circle mark line.Video camera is connected to into computer, is made it
Enough normal works.The string bag is accepted in the correct position placement apart from circle mark line certain distance.So far, the ring of system operation
Set up and finish in border.
Step 2：Run and initialize graphical interface program, i.e., virtual throwing training software of the invention on computers.
Program starts normal operation.The photographic head configuring condition of Automatic Program detection afterwards.If configuration is wrong or does not install photographic head, can
Eject prompting frame to require to check photographic head configuring condition.If photographic head is working properly, can be in the top-right form of program interface
Environmental aspect in Real Time Observation reality；If having detected that target object, bottom-right form can real-time real-world objects object
Coordinate information.It is the parameter setting area of program on program interface upper left side, according to function and can needs to arrange corresponding parameter.
Situation during virtual environment is shown in the form of the larger scope in program interface lower left, can set corresponding parameter
To change virtual environment situation, labelling (being likely to not be target object, its differentiation foundation is colouring information) is being had detected that
In the case of, visual human therein can make corresponding attitudes vibration according to the action of trainee in reality and (be only limitted to simple
Attitude).
Step 3：Setting weather and ambient parameter.These parameters can be set in graphic interface, after the completion of setting, interface
In virtual environment can according to the parameter of setting produce it is corresponding change, such as rain simulation environment etc..While corresponding parameter
Can be introduced in program, for the formula of corrected Calculation result.Default contexts are that fine day windless condition is (public using uniform motion
Formula is calculated).
Step 4：Identification target object.Before training starts, first have to definitely identify target object.Normal conditions
Under, photographic head can automatically identify the labelling with certain color characteristic, can identify in the same manner and do such labelling or band
There is the target object of such color characteristic.(such as the interior of dark) photographic head cannot automatic identification bid in certain environments
Clock, target object can be chosen manually in top-right form.Photographic head captured in real-time image, image is stored as accordingly
The data of type；Image is filtered, the target object in image is marked.Scale calculates scale factor：Record is in object
In the case that between body and photographic head, distance is certain, in photographic head, actual range is the image space length of one meter of object, that is, compare
The example factor is using each how many meters of expression of pixel correspondence.
Step 5：Start to throw training while following the trail of target object and recording its position data.If target object is occurred in
The target object for recognizing in picture frame, is shown () in the visual field of video camera, and be subjected to displacement (i.e. horizontal or vertical direction
Speed is not zero), then carry out the Real-time Collection of image：Two-dimensional coordinate system is set up in every two field picture, mesh in calculating per two field picture
Image space coordinate information in the two-dimensional coordinate system of mark object.Above-mentioned image space coordinate information data are stored in chained list,
What is be stored in simultaneously also has the data such as corresponding temporal information.
Step 6：Detection throwing action, while the positional information of target object in each two field picture that recorded is extracted
To be stored in chained list.When target object disappears from the visual field of video camera (be no longer able to identify target object), timing
Device variable counter starts timing from this two field picture corresponding moment immediately, until the value of time counter variable counter reaches
Default time threshold, now it can be assumed that target object has left the visual field of video camera completely, that is to say, that target object
Trainee is had been detached from, is successfully thrown away.Now throwing action has occurred and that and smoothly terminates.
Step:7：Choose key node and be analyzed calculating.The data structure for being stored in chained list sequentially in time is just
It is node.And the data required for finally calculating are exactly several nodes of last record.Indeed, it is possible to directly choose last
Two nodes.Because most latter two nodal test to data message closer to target object sell when velocity information.It is first false
If the weather conditions for setting in step 3 are calm as fine day, can be calculated according to basic uniform motion formula.
1) (can now be recognized come the speed of calculated level and vertical direction with time data using the position recorded in node
It is thrown out for target object), can be obtained according to the relation of time, displacement and speed.
The velocity magnitude of horizontal direction can be expressed as
The velocity magnitude of vertical direction can be expressed as
2) according to equation of uniformly accelerated motion, first by vertical direction calculate target object motion (doing projectile motion) when
Between, then the move distance that horizontal direction is drawn by movement time.The move distance of horizontal direction is exactly required throwing result.
By equation of uniformly accelerated motion
The maximum height that target object vertical direction reaches can be obtained is
The movement time of target object is
The maximum distance (throwing achievement) that target object horizontal direction reaches is
S=vxt
Can obtain after arrangement
If 3) weather environment for setting in step 3 adds rain, raindrop are when falling to in-flight and quality is long-range
Collision process should be regarded as in the impact of raindrop object, physically using momentum theorem, the collision of raindrop and target object is seen
Momentum is done, action time is raindrop and the extremely short collision time of target object, and for a drop raindrop, this momentum can be expressed as
Wherein, Δ v is the end speed before raindrop impingemest target object, and Δ t is decreased to zero by the end speed of raindrop and experiences
Time, m for raindrop Approximation Quality,The active force of momentum suffered by target object.For the different forces of rain, m and Δ v
Value it is all different, so under the different force of rainAlso it is different.Herein for convenience of calculation, derive through experience, it is right
The radius (raindrop are approximated to spheroid to facilitate calculating) and the end speed Δ v of raindrop of raindrop are defined in the different forces of rain.Its
In, the raindrop size that the force of rain is determined is by the parameter determination for setting, and the end speed of raindrop can be derived from by the radius of raindrop,
Empirical equation is accordingly
Wherein A, B, C are constant, r1For raindrop radius, now raindrop are similar to spheroid.
For target object, during whole projectile motion, clashing into its raindrop number at each moment can be with
It is set as that experience is derived by approximation n.Its theoretical value can be calculated by empirical equation, be suitable for Marshall-handkerchief here
Er Mo (Marshall-Palmer) is distributed, and formula is
N (D)=N0exp(-λD)
Wherein parameter lambda=4.1I-0.21(mm-1)；N0=8000 (individual/m3)；In formula, N (D) is raindrop number density；Raindrop diameter
The unit of D is (mm)；I is raininess, unit (mm/h).
Then the suffered active force from raindrop of target object can be approximated to be
So for each variable above, we can be solved with physical equation, are drawn and are solved the concrete of F
Formula：
ρ therein is the density of raindrop, can be approximately the density of water；N is raindrop number density；vtFor raindrop end speed.Afterwards
Both actual values are determined by the force of rain parameter of setting.By now it should be appreciated that the stress for target object in vertical direction removes weight
Again many active forces from raindrop outside power.The impact of horizontal direction raindrop is negligible relative to target object.
The formula derived in step 7 can be modified to
4) weather environment for setting in step 3 adds wind, then should also be according to the intensity of wind point situation discussion.
In the horizontal direction, (200m/s is typically taken) when target object movement velocity is less than marginal value, be considered as viscous
Resistance, the size of viscosity resistance are directly proportional to velocity magnitude, can obtain the size of this resistance by Stokes' law.
F=6 π η vr
Viscosities of the wherein η for fluid, v are movement velocity of the target object relative to wind, during r is target object motion
The radius of the approximate circumsphere of rolling attitude.In fact, the motion of target object is typically tumbling state, so it is corresponding to build which
Circumsphere carrys out approximate calculation fluoran stream surface.The radius of circumsphere can be calculated by Pythagorean theorem, and computing formula is as follows
Cylinder radiuses (be regarded as known quantity) of the R therein for target object (the usually column such as grenade model), h is
The length (being regarded as known quantity) of target object.Thus obtain the fluoran stream surface area (spheroid crosses the area of section in the center of circle) of target object
For
When target object movement velocity is more than marginal value, pressure drag is considered as, drag size can be also obtained.Pressure reduction hinders
The size of power and speed of moving body square are directly proportional, and its computing formula is as follows
Wherein C is form drag coefficient (its value is constant), ρ1For fluid density (i.e. atmospheric density), S is the area of fluoran stream surface,
V is movement velocity of the target object relative to wind, and Re is Reynolds number (Reynolds number), is that one kind can be used to characterize stream
The dimensionless number of body mobility status, its computing formula is
Wherein ρ1For fluid density, v is movement velocity of the target object relative to wind, and η is viscosity, and d is characterized length
Degree (herein refers to the diameter of target object correspondence circumsphere).
In the vertical direction, only need to consider viscosity resistance.The calculating phase of its calculating process and viscosity resistance in horizontal direction
Together.
It should be noted that when target object speed is less than critical velocity, using viscosity resistance computing formula, but if
The direction of wind is identical with the target object direction of motion, then the speed of target object may gradually increase even more than marginal value,
Now should conversion Calculation formula, using pressure drag；On the other hand, similarly, when the movement velocity of target object
During more than marginal value, using pressure drag computing formula, but if the direction of wind with the target object direction of motion conversely, then target
The speed of object can be gradually reduced even less than marginal value, now should conversion Calculation formula, using viscosity resistance computing formula；
Further, since the direction of wind may be with the direction of motion of object not in the same plane, we only consider wind in object here
The component in vertical plane that the direction of motion is located, the impact in other directions are negligible.
Next with the resistance calculation formulae for having obtained above correcting the result computing formula in step 7.
A. for vertical direction, first consider the stage for rising.Resistance formula can be regarded as
F=kv,
Obtain acceleration
Equal sign two ends are quadratured, and obtain
As t=0, there is v=vy, substitution obtains
Work as vtWhen=0, can be in the hope of the time
Distance (lifting height) is tried to achieve again
Thus, obtained throw maximum height be
H=h+hPeople
B. the stage of decline is considered further that, resistance formula is similarly
F=kv
Obtain acceleration
Quadrature and obtain in equal sign two ends
As t=0, there is v=0, substitution is obtained
Deformation obtains the expression formula of speed
Can the condition according to known to height be listed below equation solution whereabouts time
Obtain after abbreviation
As the equation is transcendental equation, it is impossible to which obtaining analytic solutions, i.e. result can not be represented with the formula containing known quantity,
Need by computer to solve.We can first assume that the result tried to achieve is T2.Certainly, it is recognized in ensuing calculating
To be a known quantity.
Thus obtained projectile motion total duration is
T=T1+T2
The cast during cast, i.e. resistance f=kv that c. calculating resistance is directly proportional to speed below.
Acceleration can be expressed as
Quadrature and obtain in equal sign two ends
As t=0, there is v=vx, substitution obtains
Deformation obtains the expression formula of speed
The displacement of horizontal direction is obtained to rate integrating
Indefinite integral is asked to obtain
As t=0, there is S1=0, substitution is obtained
Then the displacement of horizontal direction has been tried to achieve, that is, the distance thrown is
D. square cast being directly proportional of resistance and speed is calculated again, i.e.,
F=kv2
Acceleration can be expressed as
Quadrature and obtain in equal sign two ends
As t=0, there is v=vx, substitution obtains
The expression formula of speed and time is obtained after deformation
Time in speed formula is integrated, the displacement of horizontal direction is obtained
Indefinite integral is asked to obtain
As t=0, there is S2=0, substitution is obtained
Then the displacement of horizontal direction has been tried to achieve, that is, the distance thrown is
This completes the calculating process for throwing result.
Step 8：Output result of calculation, is displayed in graphical interface program.After the completion of the calculating process of step 7, immediately
The result for obtaining is fed back in the graphical interfaces of program.While showing throwing result, can also be in the virtual environment of lower left
The flash demo of this throwing is carried out in form, i.e., is virtually thrown according to throwing result, reach corresponding Jing after projectile motion
Position, its movement locus and position of dropping are that equal proportion scales relation, basic simulation of real scenes with the data in reality.
Step 9：Preserve calculation result data.The throwing result of step 7 can correspond to accordingly numbering trainee and store automatically
In data base, in graphical interfaces, corresponding trainee's numbering can be selected to be checked, can be with the void in lower left
The flash demo of this throwing, convenient contrast and analysis are carried out in the form of near-ring border.
Through above-mentioned steps, the throwing training of a trainee is just completed.Next trainee can be carried out afterwards
The throwing training of member.
The above, the only present invention preferably specific embodiment, but protection scope of the present invention is not limited thereto,
Any those familiar with the art the invention discloses technical scope in, technology according to the present invention scheme and its
Inventive concept equivalent or change in addition, should all be included within the scope of the present invention.
Claims (6)
1. a kind of throwing training achievement detection method based on real time video image, it is characterised in that：Comprise the following steps：
S1：Arrange and throw training court, process is marked to target object, weather conditions information during training is thrown in collection；
S2：Scale calculates scale factor：Record is in the case that between target object and photographic head, distance is certain, photographic head
Middle physical length is the image space length of one meter of object, i.e. scale factor using each how many meters of expression of pixel correspondence；
S3：Target object is thrown, the moving-picture information before selling in throwing using photographic head Real-time Collection target object is right
Target object is tracked shooting, the video image that real-time processing is collected：Two-dimensional coordinate system is set up in every two field picture, is captured
Spatial coordinated information of the target object in two-dimensional coordinate system in per two field picture, will be the image space coordinate information for capturing real-time
Be stored in the chained list with regular length, realize real-time update to capturing information in chained list；
S4：After determining that target object is sold, multiple nodes of end are located in reading chained list, consecutive frame figure is shot according to photographic head
The principle that time difference as between is fixed, using the spatial coordinated information and correspondence time information of target object in multiple nodes, meter
Target object space velocity in the picture is calculated, i.e., the speed of target object horizontal direction and vertical direction is using per second in image
Represent for a how many pixel；
S5：Image space is carried out with the ratiometric conversion of real space so as to calculate mesh according to the scale and scale factor in S2
Mark object previous moment of selling practical level speed and vertical speed, so as to calculate the parabolic of target object horizontal direction away from
With a distance from the parabolic with vertical direction, realize throwing the detection of training achievement；
In S3：Throwing action is detected when throwing target object, by the space coordinatess of target object in each two field picture for extracting
Information is stored in chained list, when target object cannot be recognized by photographic head, is started timing from this two field picture corresponding moment, is worked as meter
When time span exceed threshold value when, then target object leaves the visual field of video camera completely.
2. a kind of throwing training achievement detection method based on real time video image according to claim 1, its feature is also
It is：In S1：When being marked process to target object：The labelling for being easy to photographic head to gather is marked on the target object simultaneously
The label information is stored, when the moving-picture information in throwing using photographic head Real-time Collection target object, photographic head
The above-mentioned labelling for having done is identified.
3. a kind of throwing training achievement detection method based on real time video image according to claim 1, its feature is also
It is：The node is the data structure that stores sequentially in time in chained list, in S4：End is located in reading data storage chained list
Last two node is chosen during multiple nodes at end, mesh in state employing most latter two node at moment that target object is sold
The coordinate parameters of mark object are calculated.
4. a kind of throwing training achievement detection method based on real time video image according to claim 1 or 3, its feature
Also reside in：When weather conditions information when training is thrown in setting：When weather conditions are that fine day is calm then according to uniform motion formula
Calculated：According to the target object location coordinate recorded in two last nodes and time data calculated level and vertical side
To speed, then accordingly calculate target object horizontal direction displacement and vertical direction displacement.
5. a kind of throwing training achievement detection method based on real time video image according to claim 1 or 3, its feature
Also reside in：When weather conditions information when training is thrown in setting：When weather conditions are to rain：According to momentum theorem by raindrop and
The collision of target object is considered as momentum, and action time is raindrop and the extremely short collision time of target object, for a drop raindrop, this
One momentum is expressed asWherein, Δ v starts to speed to be decreased to zero speed difference for raindrop impingemest target object,
Δ t was decreased to for zero time experienced to speed by raindrop by collision, and m is the Approximation Quality of raindrop,For target object
The active force of suffered momentum；The size of the force of rain is by end speed v before raindrop radius and raindrop impingemest target objecttDetermine, rain
End speed before drop shock target object is derived from by the radius of raindrop：
Wherein：See raindrop as spheroid, A, B, C are constant, r1For raindrop radius, then target object is subject to the total contribution of rain
Power F is：Wherein n be for target object during whole projectile motion each moment raindrop impingemest target
The approximate number value of object is raindrop number density, is obtained according to actual Weather information；Its theoretical value is calculated using equation below：n
=N0Exp (- λ D), wherein parameter lambda=4.1I-0.21(mm-1)；N0=8000 (individual/m3), D is raindrop diameter, and I is raininess, root
According to：Solve F, density of the wherein ρ for raindrop water, vtFor rain
Drop clashes into the end speed before target object, and a starts to speed to be decreased to zero acceleration, n and v for raindrop impingemest target objectt
By actual force of rain gain of parameter, therefore target object receives the active force of gravity and raindrop in vertical direction, and target object is in water
Square affected to ignore to by raindrop, finally calculate in the case where considering target object by raindrop active force and throw training
Achievement.
6. a kind of throwing training achievement detection method based on real time video image according to claim 4, its feature is also
It is：When weather conditions information when training is thrown in setting：When weather conditions have wind：
In the horizontal direction：When marginal value of the target object movement velocity less than setting：Target object is made by viscosity resistance
With the solution formula of viscosity resistance is：F=6 π η vr, wherein：Viscosities of the η for fluid, v are target object relative to wind
Movement velocity, r are the radius of the approximate circumsphere of rolling attitude in target object motion, and the radius of the circumsphere is by target object
Cylinder radius R and target object length h according to equation below calculate：
The then fluoran stream surface area of target object
When marginal value of the target object movement velocity more than setting：If pressure drag suffered by target object is f,Wherein：C be form drag coefficient, ρ1For fluid density, areas of the S for fluoran stream surface, v are that target object is relative
In the movement velocity of wind, Re is Reynolds number,Wherein：ρ1For atmospheric density, v is fortune of the target object relative to wind
Dynamic speed, η is viscosity, and d is characterized the length i.e. diameter of target object correspondence circumsphere；
In the vertical direction, only need to consider viscosity resistance, and its calculating process is identical with the calculating of viscosity resistance in horizontal direction, most
Calculate in the case where the active force of target object wind-engaging is considered afterwards and throw training achievement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410332067.5A CN104102835B (en) | 2014-07-11 | 2014-07-11 | Throw training score detection method based on real-time video images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410332067.5A CN104102835B (en) | 2014-07-11 | 2014-07-11 | Throw training score detection method based on real-time video images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104102835A CN104102835A (en) | 2014-10-15 |
| CN104102835B true CN104102835B (en) | 2017-04-12 |
Family
ID=51670980
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410332067.5A Active CN104102835B (en) | 2014-07-11 | 2014-07-11 | Throw training score detection method based on real-time video images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104102835B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104486595A (en) * | 2014-12-25 | 2015-04-01 | 安徽寰智信息科技股份有限公司 | Video acquisition and processing method for bowling teaching and training |
| CN107256082B (en) * | 2017-05-11 | 2020-04-14 | 大连理工大学 | Throwing object trajectory measuring and calculating system based on network integration and binocular vision technology |
| CN107158701B (en) * | 2017-05-16 | 2018-08-31 | 广州四三九九信息科技有限公司 | The reminding method and device of outgoing dynamics |
| CN114777682A (en) * | 2017-10-06 | 2022-07-22 | 先进扫描仪公司 | Generating one or more luminance edges to form a three-dimensional model of an object |
| CN110766756B (en) * | 2019-10-21 | 2022-09-30 | 大连理工大学 | Multi-direction projection-based drop point positioning method |
| CN112819301A (en) * | 2021-01-21 | 2021-05-18 | 莆田市军源特种装备科技有限公司 | Intelligent bullet shooting examination system |
| CN113379801B (en) * | 2021-06-15 | 2024-04-30 | 江苏科技大学 | High-altitude parabolic monitoring and positioning method based on machine vision |
| CN115114466B (en) * | 2022-08-30 | 2022-12-13 | 成都实时技术股份有限公司 | Method, system, medium and electronic device for searching target practice information image |
| CN116597340B (en) * | 2023-04-12 | 2023-10-10 | 深圳市明源云科技有限公司 | High altitude parabolic position prediction method, electronic device and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540833A (en) * | 2009-04-13 | 2009-09-23 | 浙江大学 | Anti-interference real-time tracking method for profile of object |
| CN102214301A (en) * | 2011-07-26 | 2011-10-12 | 西南交通大学 | Multi-target tracking method for associated cooperation of adaptive motion |
| CN202404111U (en) * | 2011-11-18 | 2012-08-29 | 天津理工大学 | DSP-based video velocity-measuring system |
| CN103322984A (en) * | 2013-05-13 | 2013-09-25 | 成都理工大学 | Distance measuring and speed measuring methods and devices based on video images |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120231840A1 (en) * | 2011-03-08 | 2012-09-13 | Bank Of America Corporation | Providing information regarding sports movements |
-
2014
- 2014-07-11 CN CN201410332067.5A patent/CN104102835B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540833A (en) * | 2009-04-13 | 2009-09-23 | 浙江大学 | Anti-interference real-time tracking method for profile of object |
| CN102214301A (en) * | 2011-07-26 | 2011-10-12 | 西南交通大学 | Multi-target tracking method for associated cooperation of adaptive motion |
| CN202404111U (en) * | 2011-11-18 | 2012-08-29 | 天津理工大学 | DSP-based video velocity-measuring system |
| CN103322984A (en) * | 2013-05-13 | 2013-09-25 | 成都理工大学 | Distance measuring and speed measuring methods and devices based on video images |
Non-Patent Citations (4)
| Title |
|---|
| Adaptive pattern recognition in real-time video-based soccer analysis;Marc Schlipsing 等;《Journal of Real-Time Image Processing》;20140225;第1-17页 * |
| 基于DSP的视频测速系统研究;黄兴;《中国优秀硕士学位论文全文数据库 信息科技辑》;20121015(第10期);第I138-2389页 * |
| 基于图像特征的铅球投掷数据同步方法;任泓宇 等;《计算机工程与设计》;20140531;第35卷(第5期);第1699-1704页 * |
| 高速投掷物体跟踪与姿态测量方法的研究;王继方;《中国优秀硕士学位论文全文数据库 信息科技辑》;20090515(第5期);正文第15页第3段,第27页第2-3段,第32页第1-6段,第41页第2段,图3-1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104102835A (en) | 2014-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104102835B (en) | Throw training score detection method based on real-time video images | |
| CN108635858B (en) | Interface display method, device, electronic device and computer readable storage medium | |
| US11676312B2 (en) | Object trajectory simulation | |
| US9311727B2 (en) | Device, system, method and computer-readable storage medium for analyzing tennis swing motion | |
| CN108635857A (en) | Interface display method, device, electronic device and computer readable storage medium | |
| CN104504694A (en) | Method for acquiring three-dimensional information of moving ball | |
| AU2015258753B2 (en) | Monitoring surfing | |
| KR20130032857A (en) | Method and system for photographing object in movement with plural camera, and based on taken images therefor, obtaining actual movement trace of same object | |
| US11850498B2 (en) | Kinematic analysis of user form | |
| US11747128B2 (en) | System and method for testing a surface | |
| CN201269764Y (en) | Multi-parameter real-time detection system for toxophily training | |
| CN105797319A (en) | Badminton motion data processing method and device | |
| CN106611153B (en) | Intelligent ball training action recognition system and method | |
| KR20090045640A (en) | System and method for a correcting pose of golfer using an artificial intelligence caddy | |
| CN107256082A (en) | A kind of ammunition ballistic trajectory calculating system based on network integration and binocular vision technology | |
| CN113532193A (en) | Intelligent combat confrontation training system and method for team tactics | |
| CN209295789U (en) | A kind of laser simulated firing automatic scoring round target system | |
| CN105879349A (en) | Method and system for displaying golf ball falling position on putting green on display screen | |
| CN107126688A (en) | A kind of basketball shooting training device | |
| CN110472185B (en) | Method for calculating trajectory of window with inclination angle between round projectile impact and ground | |
| CN110472180B (en) | Method for calculating trajectory of round projectile striking window perpendicular to ground | |
| JP7339053B2 (en) | Evaluation methods, evaluation systems, and evaluation programs | |
| CN109045662A (en) | The shuttlecock service judge auxiliary system of view-based access control model | |
| CN115845349A (en) | General training method for ball game items for moving target detection based on deep learning technology and auxiliary referee system | |
| CN112245893A (en) | Method and device for measuring rotation speed of table tennis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |