CN114998439A - Movement distance measuring system based on machine vision and application - Google Patents
Movement distance measuring system based on machine vision and application Download PDFInfo
- Publication number
- CN114998439A CN114998439A CN202210928996.7A CN202210928996A CN114998439A CN 114998439 A CN114998439 A CN 114998439A CN 202210928996 A CN202210928996 A CN 202210928996A CN 114998439 A CN114998439 A CN 114998439A
- Authority
- CN
- China
- Prior art keywords
- distance
- coordinates
- markers
- line
- target object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000033001 locomotion Effects 0.000 title claims abstract description 32
- 238000004364 calculation method Methods 0.000 claims abstract description 31
- 239000003550 marker Substances 0.000 claims description 19
- 238000001514 detection method Methods 0.000 claims description 18
- 238000003825 pressing Methods 0.000 claims description 16
- 238000002360 preparation method Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 8
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 16
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 238000012360 testing method Methods 0.000 abstract description 4
- 238000004422 calculation algorithm Methods 0.000 description 9
- 239000007787 solid Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- RZSCFTDHFNHMOR-UHFFFAOYSA-N n-(2,4-difluorophenyl)-2-[3-(trifluoromethyl)phenoxy]pyridine-3-carboxamide;1,1-dimethyl-3-(4-propan-2-ylphenyl)urea Chemical compound CC(C)C1=CC=C(NC(=O)N(C)C)C=C1.FC1=CC(F)=CC=C1NC(=O)C1=CC=CN=C1OC1=CC=CC(C(F)(F)F)=C1 RZSCFTDHFNHMOR-UHFFFAOYSA-N 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/14—Measuring arrangements characterised by the use of optical techniques for measuring distance or clearance between spaced objects or spaced apertures
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a movement distance measuring system based on machine vision and application thereof, wherein the system comprises: the device comprises a calibration device, video acquisition equipment and a calculation server, and is used for measuring the movement distance in the designated area. The invention can realize the intelligent measurement of the movement distance by utilizing artificial intelligence and machine vision, has high test efficiency and simple use, and reduces artificial measurement errors, thereby ensuring the accuracy and fairness of the measurement result.
Description
Technical Field
The invention belongs to the field of artificial intelligence, and particularly relates to a movement distance measuring system based on machine vision.
Background
At present, sports subjects are brought into the horizontal examination of the junior middle school and high school academic levels. In various sports, the movement distance is measured. In various sports events, there are important requirements for measuring the movement distance, such as measuring the long jump distance by setting the long jump, calculating the throwing distance for throwing a shot and a solid ball, and the like.
The existing moving distance measuring schemes comprise manual measurement, pressure sensor schemes and infrared correlation sensor schemes, and have the defects while the moving distance measuring requirements are met. The manual measurement mode cannot guarantee the accuracy and fairness of the measurement result, and certain manpower needs to be consumed. In the scheme of the pressure sensor, after a tester or a target object falls to the ground, the pressure sensor is triggered to generate a level signal, and the achievement is calculated. Because the pressure sensors need to be arranged on the ground in a surface mode, the number of the required sensors is large, and the equipment is heavy and inconvenient to move. Pressure sensor is also unable accurate detection sportsman's line ball is violating regulations. The infrared correlation sensor realizes distance measurement through continuous correlation, an infrared light path is arranged above the infrared correlation sensor, and when normal infrared pulse irradiates, the infrared receiving end outputs a low level signal, and when a tester or a target object falls to the ground and then is shielded, the infrared receiving end in a shielding range outputs a high level, so that the score is calculated. The infrared correlation sensors need to be arranged above the jump blanket and are easy to touch in the using process, so that the angles of the correlation sensors are changed, certain errors can be brought to the measurement precision of scores, and the method is difficult to calibrate and cannot be used in a normalized mode. In addition, the pressure sensor scheme and the infrared correlation sensor scheme are applied to different projects, different sensor carrier devices are required, the devices and specifications cannot be unified, and the maintenance and the management are inconvenient; the two sensor schemes cannot record the motion video of the athlete or a target object and accurately capture the illegal action of the athlete, and if the two sensor schemes are applied to large-scale examination scenes such as a middle test, additional video acquisition equipment needs to be erected, and the recorded video is used for later-stage review and filing.
Disclosure of Invention
The invention aims to solve the defects of the prior art, and provides a movement distance measuring system based on machine vision and application thereof, so that accurate measurement of movement distance can be realized by utilizing machine vision and artificial intelligence, thereby reducing errors, improving efficiency, meeting the requirements of distance measurement in various sports projects, and ensuring the accuracy and fairness of measurement results.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention relates to a movement distance measuring system based on machine vision, which is characterized by comprising the following components: the device comprises a calibration device, video acquisition equipment and a calculation server, and is used for measuring the movement distance of a target object in a target area; the computing server includes: the device comprises a calibration module, a landing frame calculation module and a distance calculation module;
the target area is divided into a preparation area and a ranging area by a dividing line, and two sides of the ranging area are symmetrically distributed with a distance measuring areamIndividual marker, 2 in totalmA marker; the distance between two adjacent markers on any side is equal, a pair of markers are formed by the markers on the symmetrical positions on two sides, and a virtual connecting line formed by the 1 st pair of markers closest to the dividing line is used as a starting line; the distance between the dividing line and the starting line is set asL；
The video acquisition equipment acquires the image of the target area, and takes one vertex of the image as an origin point, and two edges of the image connected with the origin point are respectivelyxShaft andyaxes, thereby establishing a planar coordinate system; adjusting the video capture device so that the markers on both sides follow the planar coordinate systemxArranged in the axial direction;
the calibration module detects the markers in the image of the target area to obtain the central coordinates { (C { (R) of the markers at one side of the ranging areax i,0 ,y i,0 )|i=1,2,…,mCentral coordinates of the marker on the other side { (S) } and the marker on the other side { (S)x i,1 ,y i,1 )|i=1,2,…,m}; wherein (A), (B), (C) and Cx i,0 ,y i,0 ) Indicates one side ofiThe center coordinates of each marker, ((ii))x i,1 ,y i,1 ) Indicates the other side is the firstiThe center coordinates of each marker;
assuming the target object toxThe forward direction of the axis is taken as the moving direction, and the video acquisition equipment acquires each frame of moving image in the moving process of the target object and sends the moving image to the floor frame calculation module;
the landing frame calculation module detects each frame of moving image to obtain a boundary frame containing a target object; comparing whether the displacement of the boundary box containing the target object in the two adjacent frames of moving images is less than a threshold valueδIf the number of the frames is smaller than the preset threshold value, judging that the previous frame of the two corresponding adjacent frames is a landing frame image, and detecting a boundary frame containing the target object in the corresponding landing frame image; otherwise, continuously comparing two adjacent frames;
the distance calculation module calculates the coordinate of any point on the edge closest to the starting line in the boundary frame containing the target object in the image in the landing frame (x’ end ,y’ end ) And then (a)x’ end ,y’ end ) Making a strip with said ranging areamLines having intersections with virtual lines of markers, obtainedmThe coordinates of the intersection point, which is expressed as { (x j , y j ),j=1,2,…,mIn which (are)x j , y j ) Represents the line and the firstjCoordinates of intersection points of virtual connecting lines of the markers are determined;
the distance calculation module pairmAn intersection coordinate { (x j , y j ),j=1,2,…,mGo through the traversal, find out so thatx’ end >x j Maximum subscript value of established intersection point coordinatesj max (ii) a Thereby calculating the maximum intersection coordinates by the equation (1) ((x jmax ,y jmax ) Distance from the dividing lined 0 ：
d 0 =∆× (j max -1)+L (1)
Calculating the compensation distance using equation (2)d 1 ：
d 1 =Δ×(x’ end -x jmax-1 )/(x jmax -x jmax-1 ) (2)
Calculating the distance from the target object to the dividing line by using the formula (3)d：
d= d 0 +d 1(3)
The invention also discloses a moving distance measuring system based on machine vision, which is characterized in that the system also comprises: a wire pressing detection module; after sending a preparation signal, the video acquisition equipment acquires video data of athletes in the preparation area and sends the video data to the line pressing detection module;
the line pressing detection module detects the boundary frames of the two feet of the athlete in the video data and the width of the boundary framesw f And heighth f And center coordinates (x f , y f ) And the coordinates of the toe are calculated by the formula (4) < CHEM >x foot , y foot )：
x foot =x f +0.5w f ；y foot =y f (4)
The line pressing detection module detects that the coordinates of two ends of the dividing line are respectively (x start,0 , y start,0 ),(x start,1 , y start,1 ) Passing through the tiptoe coordinate (x foot , y foot ) To be parallel toxThe coordinates of the intersection points of the straight lines in the axial direction and the dividing lines are expressed as (x t , y t ) Calculating the coordinates of the intersection point by using the formula (5) ((x t , y t )：
x t =x start,1 +((y start,1 - y t )/(y start,1 - y start,0 ))×(x start,1 -x start,0 )；y t =y foot (5)
If it is notx foot >x t If not, the line is not pressed.
The electronic device comprises a memory and a processor, and is characterized in that the memory is used for storing programs for supporting the processor to execute the system, and the processor is configured to execute the programs stored in the memory.
The invention relates to a computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, performs the steps of the system.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention is based on the artificial intelligence machine vision technology, and the movement distance is automatically measured by the calculation server after the video acquisition equipment acquires the movement process of the athlete or the target object, so that the labor input is saved, and the accuracy and the fairness of the measurement result are ensured.
2. The line pressing detection module included in the movement distance measurement system based on machine vision detects whether line pressing exists on the two feet by judging whether the toe position coordinates of the two feet of the athlete in the preparatory movement image of the athlete exceed the dividing line, and effectively solves the problem that the existing pressure sensor scheme cannot accurately detect line pressing violation of the athlete.
3. The invention can adopt a unified machine vision scheme to measure the distance between the feet of a player and the ground of a solid ball and a shot ball by properly adjusting a calibration device for a plurality of different sports events, such as setting up long jump, throwing the solid ball, shooting the shot ball and other sports events, for example, expanding the area of a spray drawing, thereby solving the problems that the prior sensor scheme customizes different sensor equipment aiming at different sports events, can not unify the equipment and standardize, and also reducing the maintenance complexity.
4. According to the movement distance measuring system based on the machine vision, only the calibration device is arranged near the test area, the calibration device can be a regular pattern sprayed on the ground of a playground, the video acquisition equipment and the calculation server can be placed at a far distance outside the test area, and athletes cannot touch the calibration device in the using process, and cannot easily touch the calibration device like an infrared correlation sensor to change the angle of the correlation sensor, so that certain errors can be brought to the measurement precision of the performance. Moreover, the calibration module contained in the calculation server can automatically identify the regular pattern on the calibration device, automatically calibrate, avoid the problem of difficult calibration of the existing scheme and facilitate normalized use.
5. The invention relates to a movement distance measuring system based on machine vision, which comprises three parts: the calibration device can be a mat paved on the ground and drawn with figures, a sticker pasted on the ground and drawn with figures or a series of figures directly drawn on the ground, the video acquisition device is generally a camera, the size of the calculation server is similar to that of a computer host, and the calibration device, the video acquisition device and the calculation server can be stored in a trunk, so that the problems of heavy equipment and unchanged movement of the existing pressure sensor scheme are solved.
The video acquisition equipment can record the videos of the preparation actions of the athletes and the motion states of the athletes or the target objects in real time, automatically stores the videos, can be used for rechecking and filing at the later stage, solves the problem that the motion videos of the athletes or the target objects cannot be recorded by the existing sensor scheme, and does not need to additionally erect a video recorder at an examination site to record the site videos.
Drawings
FIG. 1 is a diagram of a target area of the present invention, and the moving direction of the target object is consistent with the x-axis direction of a planar coordinate system.
Fig. 2 is a diagram of the target area of the present invention, and the moving direction of the target object is consistent with the x-axis direction of the plane coordinate system, when L =0, the dividing line coincides with the starting line of the 1 st pair of markers.
FIG. 3 is a schematic view of the arrangement of markers in a shot item having a target area in the form of a sector.
FIG. 4 is a diagram of the target area of the present invention, wherein the moving direction of the target object is opposite to the x-axis direction of the planar coordinate system.
Fig. 5 is a specific flowchart of the moving distance measurement according to the present invention.
Detailed Description
In this embodiment, a system for measuring a movement distance based on machine vision includes: the device comprises a calibration device, video acquisition equipment and a calculation server, and is used for measuring the movement distance of a target object in a target area; the computing server includes: the system comprises a calibration module, a landing frame calculation module and a distance calculation module;
referring to fig. 1, a target area is divided into a preparation area and a ranging area by a dividing line, and both sides of the ranging area are symmetrically distributedmIndividual markers, in total 2mAnd (4) a marker. The markers can be circular, rectangular, triangular, etc. with distinct shapes from the surrounding environment, can be various shapes of painted paint, stickers, etc., and can be painted on different carriers, such as playgrounds or cushions, etc. The distance between two adjacent markers on any side is equal, a pair of markers are formed by the markers on the symmetrical positions on two sides, and a virtual connecting line formed by the 1 st pair of markers closest to the dividing line is used as a starting line; setting the distance between the dividing line and the initial line as L; for different sports, the distance between the markers and the distance L between the dividing line and the starting line are different, for example, the distance for standing long is usually 0.5-10 cm, L is 0-100 cm; the project of throwing the lead ball and the solid ball is normally between 20cm and 200cm, and L is between 0cm and 500 cm. Fig. 2 is a schematic view of the target region of L = 0.
The distance between the markers on two sides, the marker on one side and the corresponding marker on the other side are in different motion items, and the distances are not necessarily equal. For the shot item ranging, since the shot ranging is a sector area, the markers are arranged in a group of 2 or 4 markers, and the distance between the two sides is gradually increased, as shown in fig. 3.
As shown in fig. 5, the video capture device obtains an image of the target area, and uses a vertex of the image as an origin, and two edges of the image connected to the origin are respectively the originxShaft andyaxes, thereby establishing a planar coordinate system; adjusting the video-acquisition equipment so that the markers on both sides follow a planar coordinate systemxArranged in the axial direction;
the calibration module detects the markers in the image of the target area to obtain the central coordinates { (1 { (B) { (R) of the markers at one side of the ranging areax i,0 ,y i,0 )|i=1,2,…,mCentral coordinates of the marker on the other side { (S) } and the marker on the other side { (S)x i,1 ,y i,1 )|i=1,2,…,m}; wherein (A), (B), (C), (D), (C), (B), (C)x i,0 ,y i,0 ) Indicates one side ofiThe center coordinates of each marker, ((ii))x i,1 ,y i,1 ) Indicates the other side isiThe center coordinates of each marker; a plurality of algorithms can detect the markers in the image, such as the algorithms of Yolo, fast rcnn and the like based on the deep learning algorithm, and the algorithms of template matching and the like of the traditional target detection can also be used. By using the deep learning technology, the calibration module can detect the marker by itself, so that the calibration module can calibrate the target area by itself, manual calibration work of a sensor scheme is not needed, and manpower is saved.
In specific implementation, the method for detecting a Yolo target by using the Yolo target detection algorithm includes: the system comprises an input layer, a backbone network, a feature extraction layer and a prediction layer;
the input layer includes: the device comprises a data enhancement module, a self-adaptive anchor frame calculation module and a self-adaptive picture scaling module;
the backbone network includes: attention structure, cross-level partial networks;
the feature extraction layer includes: a characteristic pyramid network, a pyramid attention network;
the prediction layer includes: a loss function calculation layer, a non-maximum suppression structure;
processing an input image by an input module to obtain an image characteristic diagram;
the image feature map passes through an attention structure in a backbone network to obtain a double-sampling feature map, and passes through a cross-level partial network to obtain a splicing feature map;
splicing the feature map to obtain a prediction feature map through a feature pyramid network and a pyramid attention network of a feature extraction layer;
and selecting a prediction frame with the highest confidence coefficient from the prediction characteristic graph through a non-maximum suppression structure of the prediction layer as a final prediction frame, and calculating the width and height of the boundary frame of the final prediction frame and the center coordinate of the boundary frame.
Assuming the target object toxThe positive direction of the axis is taken as the moving direction, and the video acquisition equipment acquires each frame of moving image in the moving process of the target object and sends the moving image to the floor frame calculation module; the object may also have the negative x-axis direction as the direction of motion. The target area is now shown with reference to fig. 4. The object refers to different sports items, the object refers to the foot or the outline of the foot of a player in standing long jump, and the solid ball and shot item objects refer to the ball bodies of the solid ball and the shot, and the like.
The landing frame calculation module detects each frame of moving image to obtain a boundary frame containing a target object; comparing whether the displacement of the boundary box containing the target object in the two adjacent frames of moving images is less than a threshold valueδIf the number of the frame is less than the preset threshold value, judging that the previous frame of the two adjacent frames is a landing frame image, and detecting a boundary frame containing a target object in the corresponding landing frame image; otherwise, continuously comparing two adjacent frames; and calculating the displacement of the boundary frame of the target object of the previous frame and the target object of the next frame by adopting different calculation methods according to different items. Assuming that the target bounding box of the previous and subsequent frames is displaced byαTwo points corresponding to the boundary frame of the target object of the previous and subsequent frames are taken (x box,0 ,y box,0 ) And (a)x box,1 ,y box,1 ). Such as calculating the standing jump distance of the athlete,α=|(x box,1 -x box,0 |(ii) a The distance of the solid ball or the shot is calculated,α=|(y box,1 -y box,0 |calculating the distance of throwing the javelin project,α 2 =(x box,1 -x box,0 ) 2 +(y box,1 -y box,0 ) 2 。
the distance calculation module calculates the coordinate of any point on the edge of the landing frame image, which contains the closest distance between the boundary frame of the target object and the start line: (x’ end ,y’ end ) And then (a)x’ end ,y’ end ) In a strip and range areamLines having intersections with virtual lines of markers, obtainedmThe coordinates of the intersection point, which is expressed as { (x j , y j ),j=1,2,…,mIn which (are)x j , y j ) The lines and the secondjCoordinates of intersection points of virtual connecting lines of the markers are determined;
distance calculation module pairmAn intersection coordinate { (x j , y j ),j=1,2,…,mGo through the traversal, find out so thatx’ end >x j Maximum subscript value of established intersection coordinatesj max (ii) a If the set plane coordinate system is opposite to the normal movement direction of the target object, namely the target object is in the plane coordinate systemxThe negative direction of the axis is taken as the moving direction, then the direction is found outx’ end ＜x j Maximum subscript value of established intersection point coordinatesj max (ii) a Thereby calculating the maximum intersection point coordinate (1)x jmax , y jmax ) Distance from the dividing lined 0 ：
d 0 =∆× (j max -1)+L (1)
Calculating a compensation distance using equation (2)d 1 ：
d 1 =Δ×(x’ end -x jmax-1 )/(x jmax -x jmax-1 ) (2)
Calculating the final movement distance of the target object by using the formula (3)dThat is, the actual moving distance of the target object measured by the moving distance measuring system:
d= d 0 +d 1(3)
the algorithm has strong universality, and can meet the distance measurement requirements in most sports, such as sports items of standing long jump, solid ball, shot and the like.
In a specific implementation, the system for measuring a movement distance further comprises: a wire pressing detection module; after the preparation signal is sent out, video data of athletes in the preparation area are collected by the video collecting equipment and sent to the line pressing detection module;
the line pressing detection module detects the boundary frame of the two feet of the athlete in the video data and the width of the boundary framew f And heighth f And center coordinates (x f , y f ) And calculating the toe coordinates by using the formula (4) <x foot , y foot )：
x foot =x f +0.5w f ；y foot =y f (4)
The two end coordinates of the dividing line detected by the line pressing detection module are respectively (x start,0 , y start,0 ),(x start,1 ,y start,1 ) Passing through the tiptoe coordinate: (x foot , y foot ) To be parallel toxThe coordinates of the intersection points of the straight lines in the axial direction and the dividing lines are expressed as (A)x t , y t ) Calculating the coordinates of the intersection point by using the formula (5) ((x t , y t )：
x t =x start,1 +((y start,1 - y t )/(y start,1 - y start,0 ))×(x start,1 -x start,0 )；y t =y foot (5)
If it is notx foot >x t If not, the line is not pressed.
Detecting both ends of the dividing line by using a commonly used detection algorithm such as a Yolo or fast rcnn target detection algorithm to obtain a central coordinate of the target object ((x start,0 , y start,0 ),(x start,1 , y start,1 )。
If the set plane coordinate system is opposite to the normal movement direction of the target object, namely the target object is in the plane coordinate systemxThe negative direction of the axis is taken as the motion direction, and the coordinate of the toe is passed (x foot , y foot ) Are made in parallel withxCoordinates of intersection points of straight lines in the axial direction: (x t , y t ) If, ifx foot ＜xIf not, the line is pressed, otherwise, the line is not pressed. The algorithm effectively solves the problem that the existing pressure sensor scheme cannot accurately detect the line pressing violation of the athlete through a machine vision calculation method.
Claims (4)
1. A machine vision-based moving distance measuring system, comprising: the device comprises a calibration device, video acquisition equipment and a calculation server, and is used for measuring the movement distance of a target object in a target area; the computing server includes: the system comprises a calibration module, a landing frame calculation module and a distance calculation module;
the target area is divided into a preparation area and a ranging area by a dividing line, and two sides of the ranging area are symmetrically distributed with a distance measuring areamSignMaterial, 2 in totalmA marker; the distance between two adjacent markers on any side is equal, a pair of markers are formed by the markers on the symmetrical positions on two sides, and a virtual connecting line formed by the 1 st pair of markers closest to the dividing line is used as a starting line; the distance between the dividing line and the starting line is set asL；
The video acquisition equipment acquires the image of the target area, and takes one vertex of the image as an origin point, and two edges of the image connected with the origin point are respectivelyxShaft andyaxes, thereby establishing a planar coordinate system; adjusting the video capture device so that the markers on both sides follow the planar coordinate systemxArranged in the axial direction;
the calibration module detects the markers in the image of the target area to obtain the center coordinates { (A) of the markers on one side of the ranging areax i,0 ,y i,0 )|i=1,2,…,mCentral coordinates of the marker on the other side { (S) } and the marker on the other side { (S)x i,1 ,y i,1 )|i=1,2,…,m}; wherein (A), (B), (C) and Cx i,0 ,y i,0 ) Indicates one side ofiThe center coordinates of each marker, ((ii))x i,1 ,y i,1 ) Indicates the other side isiThe center coordinates of each marker;
assuming the target object toxThe forward direction of the axis is taken as the moving direction, and the video acquisition equipment acquires each frame of moving image in the moving process of the target object and sends the moving image to the floor frame calculation module;
the landing frame calculation module detects each frame of moving image to obtain a boundary frame containing a target object; comparing whether the displacement of the boundary box containing the target object in the two adjacent frames of moving images is less than a threshold valueδIf the number of the frame is less than the preset threshold value, judging that the previous frame of the two adjacent frames is a landing frame image, and detecting a boundary frame containing a target object in the corresponding landing frame image; otherwise, continuously comparing two adjacent frames;
the distance calculation module calculates the starting point and the position in a boundary box containing a target object in the image in the landing frameCoordinates of any point on the edge closest to the line distance: (x’ end ,y’ end ) And pass throughx’ end ,y’ end ) Making a strip in the distance measuring areamLines having intersections with virtual lines of markers, obtainedmThe coordinates of the intersection point, which is expressed as { (x j , y j ),j=1,2,…,mIn which (are)x j , y j ) Represents the line and the secondjCoordinates of the intersection point of the virtual connecting lines of the markers are calculated;
the distance calculation module pairmAn intersection coordinate { (x j , y j ),j=1,2,…,mGo through the traversal, find out so thatx’ end >x j Maximum subscript value of established intersection coordinatesj max (ii) a Thereby calculating the maximum intersection coordinates by the equation (1) ((x jmax , y jmax ) Distance from the dividing lined 0 ：
d 0 =∆× (j max -1)+L (1)
Calculating the compensation distance using equation (2)d 1 ：
d 1 =Δ×(x’ end -x jmax-1 )/(x jmax -x jmax-1 ) (2)
Calculating the distance from the target object to the dividing line by using the formula (3)d：
d= d 0 +d 1 (3)。
2. The machine vision based moving distance measuring system of claim 1, further comprising: a wire pressing detection module; after sending a preparation signal, the video acquisition equipment acquires video data of athletes in the preparation area and sends the video data to the line pressing detection module;
the line pressing detection module detects the boundary frames of the two feet of the athlete in the video data and the width of the boundary framesw f And heighth f And center coordinates (x f , y f ) And the coordinates of the toe are calculated by the formula (4) < CHEM >x foot , y foot )：
x foot =x f +0.5w f ；y foot =y f (4)
The line pressing detection module detects that the coordinates of two ends of the dividing line are respectively (x start,0 , y start,0 ),(x start,1 ,y start,1 ) Passing through the tiptoe coordinate (x foot , y foot ) Are made in parallel withxThe coordinates of the intersection points of the straight lines in the axial direction and the dividing lines are expressed as (x t , y t ) Calculating the coordinates of the intersection point by using the formula (5) ((x t , y t )：
x t =x start,1 +((y start,1 - y t )/(y start,1 - y start,0 ))×(x start,1 -x start,0 )；y t =y foot (5)
If it is notx foot >x t If not, the line is not pressed.
3. An electronic device comprising a memory and a processor, wherein the memory is configured to store a program that enables the processor to execute the system of claim 1 or 2, and the processor is configured to execute the program stored in the memory.
4. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the system according to claim 1 or 2.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210928996.7A CN114998439A (en) | 2022-08-03 | 2022-08-03 | Movement distance measuring system based on machine vision and application |
| CN202211089460.7A CN115457143A (en) | 2022-08-03 | 2022-09-07 | Calibration distance measuring device, movement distance measurement method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210928996.7A CN114998439A (en) | 2022-08-03 | 2022-08-03 | Movement distance measuring system based on machine vision and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114998439A true CN114998439A (en) | 2022-09-02 |
Family
ID=83021072
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210928996.7A Pending CN114998439A (en) | 2022-08-03 | 2022-08-03 | Movement distance measuring system based on machine vision and application |
| CN202211089460.7A Pending CN115457143A (en) | 2022-08-03 | 2022-09-07 | Calibration distance measuring device, movement distance measurement method and system |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211089460.7A Pending CN115457143A (en) | 2022-08-03 | 2022-09-07 | Calibration distance measuring device, movement distance measurement method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN114998439A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116510274A (en) * | 2023-05-08 | 2023-08-01 | 深圳市培林体育科技有限公司 | Ball game ranging system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102706319A (en) * | 2012-06-13 | 2012-10-03 | 深圳泰山在线科技有限公司 | Distance calibration and measurement method and system based on image shoot |
| CN110624203A (en) * | 2019-10-14 | 2019-12-31 | 上海淡竹体育科技有限公司 | Method and device for testing free standing long jump |
| KR102065337B1 (en) * | 2019-07-19 | 2020-01-14 | 대한민국 | Apparatus and method for measuring movement information of an object using a cross-ratio |
| US20210125360A1 (en) * | 2019-10-25 | 2021-04-29 | 7-Eleven, Inc. | Contour-based detection of closely spaced objects |
| CN113137923A (en) * | 2020-01-17 | 2021-07-20 | 上海淡竹体育科技有限公司 | Measuring method of standing long jump performance |
| CN114432685A (en) * | 2022-01-19 | 2022-05-06 | 北京女娲补天科技信息技术有限公司 | Standing long jump testing method and device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103245336B (en) * | 2013-05-20 | 2015-10-28 | 苏州大学 | For the distance-measuring device of length jump event |
| CN112562005A (en) * | 2019-09-26 | 2021-03-26 | 华为技术有限公司 | Space calibration method and system |
| CN111627043B (en) * | 2020-04-13 | 2023-09-19 | 浙江工业大学 | Simple human body curve acquisition method based on markers and feature screeners |
| KR102544972B1 (en) * | 2020-11-16 | 2023-06-20 | 주식회사 핏투게더 | Method for tracking sports participants, device for tracking sports participants, and system for tracking sports participants |
| CN114693785A (en) * | 2020-12-31 | 2022-07-01 | 华为技术有限公司 | A method, system and related equipment for target positioning |
| CN114501306B (en) * | 2021-12-29 | 2024-05-14 | 中国电信股份有限公司 | Sampling method, sampling device, electronic equipment and storage medium |
-
2022
- 2022-08-03 CN CN202210928996.7A patent/CN114998439A/en active Pending
- 2022-09-07 CN CN202211089460.7A patent/CN115457143A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102706319A (en) * | 2012-06-13 | 2012-10-03 | 深圳泰山在线科技有限公司 | Distance calibration and measurement method and system based on image shoot |
| KR102065337B1 (en) * | 2019-07-19 | 2020-01-14 | 대한민국 | Apparatus and method for measuring movement information of an object using a cross-ratio |
| CN110624203A (en) * | 2019-10-14 | 2019-12-31 | 上海淡竹体育科技有限公司 | Method and device for testing free standing long jump |
| US20210125360A1 (en) * | 2019-10-25 | 2021-04-29 | 7-Eleven, Inc. | Contour-based detection of closely spaced objects |
| CN113137923A (en) * | 2020-01-17 | 2021-07-20 | 上海淡竹体育科技有限公司 | Measuring method of standing long jump performance |
| CN114432685A (en) * | 2022-01-19 | 2022-05-06 | 北京女娲补天科技信息技术有限公司 | Standing long jump testing method and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116510274A (en) * | 2023-05-08 | 2023-08-01 | 深圳市培林体育科技有限公司 | Ball game ranging system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115457143A (en) | 2022-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109903312B (en) | Football player running distance statistical method based on video multi-target tracking | |
| US11624825B2 (en) | Object tracking | |
| CN107543530B (en) | Method, system, and non-transitory computer-readable recording medium for measuring rotation of ball | |
| Ohayon et al. | Robust 3d head tracking using camera pose estimation | |
| CN103458977A (en) | Sensing device and sensing method used for virtual golf simulation device | |
| CN113239797B (en) | Human body action recognition method, device and system | |
| CN110648362B (en) | A Binocular Stereo Vision Badminton Positioning Recognition and Attitude Calculation Method | |
| CN107392157B (en) | Intelligent virtual live broadcasting method for Chinese chess game based on machine vision | |
| CN103223236B (en) | Intelligent evaluation system for table tennis training machine | |
| CN101996311A (en) | Yoga stance recognition method and system | |
| CN109646924A (en) | A kind of visualization distance measuring method and device | |
| CN110433471A (en) | A kind of badminton track monitoring analysis system and method | |
| US20190299058A1 (en) | Camera system for filming golf game and the method for the same | |
| CN115690150A (en) | A video-based multi-target displacement tracking monitoring method and device | |
| CN116392800A (en) | A standing long jump ranging method and system based on target detection and image processing | |
| CN113198166B (en) | Long jump detection method and detection device | |
| CN102012213B (en) | New method for measuring foreground height through single image | |
| CN114998439A (en) | Movement distance measuring system based on machine vision and application | |
| CN111754549B (en) | A Badminton Player Trajectory Extraction Method Based on Deep Learning | |
| CN102148919B (en) | Method and system for detecting balls | |
| CN110910489B (en) | Monocular vision-based intelligent court sports information acquisition system and method | |
| CN116012415A (en) | Ping-pong ball rotation estimation method and rotation estimation system based on vision measurement | |
| US20230364468A1 (en) | Deep learning method of determining golf swing and golf ball parameters from radar signal and image data | |
| Douret et al. | A multi-cameras 3d volumetric method for outdoor scenes: a road traffic monitoring application | |
| CN115272919A (en) | Basketball court highlight moment recognition method and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220902 |
|
| WD01 | Invention patent application deemed withdrawn after publication |