Golf grade test scoring method, storage medium and system
Through multi-angle shooting and image analysis, the swing movement and ball trajectory are identified, and the subjectivity and tracking problems of manual rating in the golf grade exam are solved, achieving more accurate and fast ratings.
Patent Information
- Application Number
- CN202210165921.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-23
- Publication Date
- 2025-01-14
- Estimated Expiration
- 2042-02-23
AI Technical Summary
In the golf grade test, the swing hit posture and stopping point have subjective intervention and difficulty in tracking the ball speed, resulting in inaccurate scoring and slow speed.
The golf trajectory and swing movements were photographed by multi-angle angles, and the hit score was obtained through the identification and similarity comparison of human key points, and the expected dropping point and the actual stopping point were analyzed using continuous images to calculate the score.
It improves the scoring accuracy and speed of the golf level test, reduces manual subjective intervention, and can quickly and accurately find the golf stop point.
Smart Images
Figure CN114550048B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image analysis, and in particular to a golf grade test scoring method, storage medium and system. Background Art
[0002] Golf level tests mainly assess whether a golfer's swing and hitting posture are standard and whether the golf ball stops accurately. Currently, golf level tests mainly rely on manual assessment and scoring, which has the following problems: (1) Manual assessment of a golfer's swing and hitting posture is subjective and can easily lead to inaccurate scoring results; (2) After a golf ball is hit, its trajectory and stopping point need to be observed by the human eye. However, since the speed of a golf ball is very fast after it is hit, it is difficult for the human eye to track the trajectory of the golf ball, resulting in the target being easily lost and difficulty in finding the stopping point of the golf ball quickly and accurately. Summary of the invention
[0003] The technical problem to be solved by the present invention is how to improve the accuracy and speed of scoring in a golf grade test.
[0004] In order to solve the above technical problems, the present invention provides a golf grade test scoring method, comprising the following steps:
[0005] A1. For a single swing action of a golf player, multiple swing action images are obtained by shooting from multiple angles;
[0006] A2. Identify key points of the human body in each swing and hitting action image;
[0007] A3. Compare the key points of the human body in each swing action image with the key points of the human body in the standard golf swing action image, and obtain the swing action score according to the similarity comparison result;
[0008] B1. Continuously photographing the golf ball at a plurality of angles at a preset time interval to obtain a plurality of sets of continuous images, wherein each angle corresponds to a set of continuous images;
[0009] B2. analyzing the expected landing position of the hit golf ball according to the golf ball positions displayed in the plurality of sets of continuous images and the preset time interval;
[0010] B3. Among the multiple cameras that respectively capture different ball landing areas, call the camera that can capture the expected ball landing area to capture, analyze the actual landing area of the golf ball according to the captured content, and call the camera that can capture the actual landing area to capture and obtain the actual landing point of the golf ball;
[0011] B4. Obtaining a stopping score based on the difference between the actual stopping point and the preset stopping point;
[0012] C. Combine the hitting action score and the stopping score to get the final score.
[0013] Preferably, in step B3, if it is determined that the expected ball landing place is not located in the actual ball stopping area, the camera that can capture the expected ball landing place is stopped from being called for shooting, and the camera that can capture the actual ball stopping area is instead called for shooting.
[0014] Preferably, the step B2 specifically calculates the acceleration and movement angle of the golf ball after being hit according to the positions of the golf ball in the multiple groups of continuous images and the preset time interval, and then analyzes the expected landing position of the hit golf ball according to the acceleration and the movement angle.
[0015] Preferably, in step B1, the angles of the plurality of groups of continuous images include angles capable of capturing changes in the longitudinal movement of a hit golf ball and angles capable of capturing changes in the lateral movement of a hit golf ball.
[0016] Preferably, in step A1, the angles of the plurality of swing and hitting action images include an angle of looking straight ahead at the golf player in a swing and hitting preparation state and an angle of looking sideways at the golf player in a swing and hitting preparation state.
[0017] Preferably, the step C specifically performs weighted calculation on the batting action score and the stopping score according to preset weights to obtain a final score.
[0018] Preferably, in the step A1, a golf player's swing and hitting video is filmed from multiple angles, and / or in the step B1, a golf ball stopping video is filmed; comprising step D: outputting and displaying the swing and hitting video and / or the stopping video.
[0019] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the golf grade test scoring method described above are implemented.
[0020] The present invention also provides a golf grade test scoring system, comprising a control module, a first camera for photographing a golf ball's motion trajectory, a second camera for photographing a golf ball's stopping point, and a third camera for photographing a golf player, wherein there are multiple first cameras, multiple second cameras, and multiple third cameras, each of which is used to photograph different ball stopping areas, and the control module is electrically connected to each first camera, the second camera, and the third camera; the control module comprises a computer-readable storage medium and a processor connected to each other, and the computer-readable storage medium is as described above; in the step B1, the multiple first cameras are used to continuously photograph the hit golf ball at different angles at preset time intervals; in the step B3, among the multiple second cameras, the camera that can photograph the expected landing point is called to photograph to obtain the actual stopping point of the hit golf ball; in the step A1, the multiple third cameras are used to photograph from multiple angles to obtain multiple swing and hitting action images.
[0021] Preferably, each second camera only captures a corresponding ball stopping area and does not capture other ball stopping areas.
[0022] The present invention has the following beneficial effects: by comparing the human body key points in each swing and hitting action image with the human body key points in the golf standard hitting action image for similarity, subjective intervention caused by manual assessment and scoring can be avoided, so the hitting action score obtained according to the similarity comparison result is more accurate than the manual assessment and scoring; after analyzing the expected landing point of the golf ball according to the golf ball position and preset time displayed in multiple groups of continuous images, a camera that can capture the expected landing point is called to capture the actual stopping point of the golf ball, which can quickly and accurately find the actual stopping point of the golf ball compared to the human eye observing the movement trajectory and stopping point of the golf ball, that is, the stopping score can be obtained relatively quickly according to the difference between the actual stopping point and the preset stopping point. In summary, the present invention can improve the accuracy and speed of golf grade test scoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 Schematic diagram of a golf course according to the present invention.
[0024] Figure 2 The present invention is a circuit connection diagram of a golf grade test scoring system.
[0025] Figure 3 It is a flow chart of the golf grade test scoring method of the present invention.
[0026] Figure 4It is a top view of the positional relationship between a golf player who is in a swing and hitting preparation state and two third cameras. DETAILED DESCRIPTION
[0027] The present invention is further described in detail below in conjunction with specific implementation methods.
[0028] Golf courses such as Figure 1 As shown, including the golf course green 1 and the golf course hitting position 2, see Figure 4 , golf player 8 uses a golf club to hit golf ball 9 to the left onto golf course green 1 at hitting position 2 of the golf course. In this embodiment, golf course green 1 is divided into nine stopping areas, namely stopping area a, stopping area b, stopping area c, stopping area d, stopping area e, stopping area f, stopping area g, stopping area h and stopping area i. The golf grade test mainly examines whether the swing and hitting posture of golf player 8 is standard and whether the stopping point of the hit golf ball 9 is accurate. In order to analyze whether the stopping point of the hit golf ball 9 is accurate, a preset stopping point 3 for the golf grade test needs to be set. The closer the golf ball 9 hit by golf player 8 is to the preset stopping point 3, the higher the stopping score of golf player 8. Figure 1 The preset ball stopping point 3 is specifically set in the ball stopping area b of the green 1 of the golf course.
[0029] Two first cameras 5 and two third cameras 7 are installed at the hitting position 2 of the golf course, and a second camera 6 is installed in each of the nine ball stopping areas of the golf green 1. The two third cameras 7 have different shooting angles and are used to shoot the swing and hitting movements of the golf player 8 from different angles; the two first cameras 5 have different shooting angles and are used to shoot the movement trajectory of the golf ball 9 after being hit from different angles; the nine second cameras 6 are used to shoot the actual stopping conditions of the golf balls in the nine ball stopping areas, that is, each second camera 6 only shoots the corresponding ball stopping area and does not shoot other ball stopping areas. The control module 4 is electrically connected to each first camera 5, the second camera 6 and the third camera 7, respectively, to form a control module 4 as shown in the following figure. Figure 2 The golf grade test scoring system shown.
[0030] It should be noted that the first camera 5, the second camera 6 and the third camera 7 can be set to other numbers, depending on the actual situation. It is only necessary to ensure that there are multiple first cameras 5, the second camera 6 and the third camera 7 to ensure that the hit golf ball 9 can be photographed from multiple angles, the actual ball stopping conditions of multiple ball stopping areas can be photographed respectively, and the swing and hitting movements of the golf player 8 can be photographed from multiple angles.
[0031] The control module 4 includes a computer-readable storage medium and a processor connected to each other. A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the following is realized: Figure 3 The golf grade test scoring method shown includes the following steps A1, A2, A3, B1, B2, B3, B4, and C.
[0032] First, the golf grade test scoring system performs the following steps A1, A2, and A3 to obtain a shot action score:
[0033] A1. For a single swing action of a golf player, multiple swing action images are obtained by shooting from multiple angles.
[0034] In this embodiment, the control module 4 uses the golf player human body key point detection model to identify the golf player 8 who is swinging and hitting the ball and his human body key points. It should be noted that the golf player human body key point detection model is a neural network model constructed using a deep learning algorithm, which is trained using a data set containing 50,000 golf player swinging and hitting pictures. Before training, the data set is annotated in the following two ways: (1) the golf player in the picture is marked in the form of a rectangular frame; (2) the human body key points in the rectangular frame are marked in the form of a circular frame. Then, the golf player human body key point detection model is trained using the data set annotated in different ways, so that it can execute the target person detection mode and the human body key point detection mode. Specifically: when the golf player human body key point detection model executes the target person detection mode, it can identify the golf player who is swinging and hitting the ball; when the golf player human body key point detection model executes the human body key point detection mode, it can identify the human body key points of the golf player.
[0035] In order to shoot the golf player 8's swing and hitting action from different angles, the two third cameras 7 need to shoot at different angles. Figure 4: The golf player 8 who is ready to swing and hit the ball faces the golf ball 9, one of the third cameras 7 faces the golf player 8 who is ready to swing and hit the ball directly, and the third camera 7 can capture the front swing and hitting action image of the golf player 8; the other third camera 7 faces the golf player 8 who is ready to swing and hit the ball sideways, and the third camera 7 can capture the side swing and hitting action image of the golf player 8. When the golf player 8 swings and hits the ball at the hitting position 2 on the court, the two third cameras 7 capture the image of the hitting position 2 on the court in real time. At this time, the control module 4 uses the golf player's human body key point detection model to identify the golf player 8 who is swinging and hitting the ball, and then for the single swing and hitting action of the golf player 8, the two third cameras 7 are respectively used to capture two swing and hitting action images from the above two shooting angles. The angles of the two swing and hitting action images are respectively the angle of looking straight at the golf player 8 in the swing and hitting preparation state and the angle of looking sideways at the golf player 8 in the swing and hitting preparation state. In this way, the control module 4 can use the two third cameras 7 to capture the front swing and hitting action images and the side swing and hitting action images of the golf player 8.
[0036] A2. Identify key points of the human body in each swing and hitting action image.
[0037] After obtaining the front swing action image and the side swing action image of the golf player 8, the control module 4 uses the golf player human body key point detection model to identify the human body key points of the golf player 8 in the two swing action images. It should be noted that the human body key points to be identified in this embodiment include the left eye, right eye, left ear, right ear, nose, neck, left shoulder, right shoulder, left elbow, right elbow, left hand, right hand, hip, left knee, right knee, left foot and right foot. The human body key points in the front swing action image can reflect the head twisting degree, hand posture and stance width of the golf player 8, and the human body key points in the side swing action image can reflect the distance between the hand and body of the golf player 8 and the degree of knee bending.
[0038] A3. Perform a similarity comparison between the key points of the human body in each swing hitting action image and the key points of the human body in the standard golf hitting action image, and obtain a hitting action score based on the similarity comparison result.
[0039] The golf standard hitting action image includes a front standard image and a side standard image. The control module 4 uses the golf player human body key point detection model to identify the human body key points in the front standard image and the side standard image respectively. Then, the control module 4 compares the human body key points in the front swing hitting action image with the human body key points in the front standard image, and compares the human body key points in the side swing hitting action image with the human body key points in the side standard image, and then obtains the hitting action score according to the front similarity comparison result and the side similarity comparison result.
[0040] Specifically, the control module 4 can calculate the head movement score of the golf player 8 according to the key points of the human body such as the left eye, right eye, left ear, right ear, nose, neck, left shoulder, and right shoulder; calculate the hand movement score of the golf player 8 according to the key points of the human body such as the left shoulder, right shoulder, left elbow, right elbow, left hand, right hand, and hips; and calculate the leg movement score of the golf player 8 according to the key points of the human body such as the hips, left knee, right knee, left foot, and right foot.
[0041] Take the front side as an example:
[0042] When calculating the head action score, firstly, the standard twist degree of the head in the standard golf hitting action is analyzed based on the distances between the left eye, right eye, left ear, right ear, nose and other key human body points and the neck with the neck as the center in the standard frontal image, and then the twist degree of the head of the golf player 8 when swinging is analyzed based on the distances between the left eye, right eye, left ear, right ear, nose and other key human body points and the neck with the neck as the center in the frontal swing hitting action image, and then the twist degree of the head of the golf player 8 when swinging is compared with the standard twist degree of the head in the standard golf hitting action for similarity, and the head action score is obtained based on the similarity comparison result;
[0043] When calculating the front hand action score, firstly, the midpoint between the two shoulders is taken as the center in the front standard image, and the standard hand posture of the standard golf hitting action is analyzed according to the distances between the left elbow, the right elbow, the left hand, and the right hand, which are the key points of the human body, and the center. Then, the midpoint between the two shoulders is taken as the center in the front swing hitting action image, and the hand posture of the golf player 8 when swinging the club is analyzed according to the distances between the left elbow, the right elbow, the left hand, and the right hand, which are the key points of the human body, and the center. Then, the hand posture of the golf player 8 when swinging the club is compared with the standard hand posture of the standard golf hitting action for similarity, and the front hand action score is obtained according to the similarity comparison result.
[0044] When calculating the front leg action score, first, in the front standard image, the standard stance width of the standard golf hitting action is analyzed based on the distance between the left knee and the right knee, and the distance between the left foot and the right foot. Then, in the front swing hitting action image, the stance width of the golf player 8 when swinging the club is analyzed based on the distance between the left knee and the right knee, and the distance between the left foot and the right foot. Then, the stance width of the golf player 8 when swinging the club is compared with the standard stance width of the standard golf hitting action for similarity, and the front leg action score is obtained based on the similarity comparison result.
[0045] Take the side view as an example:
[0046] When calculating the side hand action score, firstly, in the side standard image, the standard distance between the hand and the body of the golf standard hitting action is analyzed according to the distance between the left elbow, the right elbow, the left hand, the right hand and the hip, and then in the side swing hitting action image, the actual distance between the hand and the body of the golf player 8 when swinging the club is analyzed according to the distance between the left elbow, the right elbow, the left hand, the right hand and the hip, and then the actual distance is compared with the standard distance for similarity, and the side hand action score is obtained according to the similarity comparison result;
[0047] When calculating the lateral leg action score, first, in the lateral standard image, the standard knee flexion degree of the standard golf hitting action is analyzed according to the positional relationship between the left knee, right knee, left foot, right foot and the hips, and then in the lateral swing hitting action image, the knee flexion degree of the golf player 8 when swinging the club is analyzed according to the positional relationship between the left knee, right knee, left foot, right foot and the hips, and then the knee flexion program of the golf player 8 when swinging the club is compared with the standard knee flexion degree of the standard golf hitting action for similarity, and the lateral leg action score is obtained according to the similarity comparison result.
[0048] It should be noted that the similarity comparison result is proportional to the action score. For example, if the similarity between the degree of head twisting of golf player 8 when swinging and the standard degree of head twisting of standard golf hitting action is 80%, then the head action score is 80 points; if the similarity between the hand posture of golf player 8 when swinging and the standard hand posture of standard golf hitting action is 75%, then the front hand action score is 75 points; if the similarity between the stance width of golf player 8 when swinging and the standard stance width of standard golf hitting action is 90%, then the front leg action score is 90 points; if the similarity between the actual hand-body distance of golf player 8 when swinging and the standard hand-body distance of standard golf hitting action is 85%, then the side hand action score is 85 points; if the similarity between the knee flexion procedure of golf player 8 when swinging and the standard knee flexion degree of standard golf hitting action is 80%, then the side leg action score is 80 points.
[0049] After obtaining the head action score, the front hand score, the front leg score, the side hand score, and the side leg score as described above, first calculate the average hand score between the front hand score and the side hand score as (75+85) / 2=80 points, and the average leg score between the front leg score and the side leg score as (90+80) / 2=85 points, and then add the head action score, the hand score average and the leg score average to get the batting action score of 80+80+85=245 points.
[0050] After executing the above steps A1, A2, and A3 to obtain the hitting action score, the golf grade test scoring system executes the following steps B1, B2, B3, and B4 to obtain the stopping score:
[0051] B1. Continuously photographing the hit golf ball at multiple angles and at preset time intervals to obtain multiple groups of continuous images, wherein each angle corresponds to a group of continuous images.
[0052] In this embodiment, the two first cameras 5 have different shooting angles. Specifically, one of the first cameras 5 faces the hitting direction of the golf ball 9, and the first camera 5 can capture the longitudinal movement change of the hit golf ball 9; the other first camera 5 looks down at the hitting direction of the golf ball 9, and the first camera 5 can capture the lateral movement change of the hit golf ball 9. After the golf ball 9 is hit, the control module 4 uses the two first cameras 5 to continuously capture the hit golf ball 9 at the two shooting angles mentioned above at a preset time interval (for example, 0.1 seconds), wherein each shooting angle corresponds to capturing a set of continuous images, thereby obtaining a set of continuous images of the hit golf ball 9 from the side and a set of continuous images of the hit golf ball 9 from the top.
[0053] B2. Analyze the expected landing position of the hit golf ball based on the positions of the golf ball displayed in the multiple sets of continuous images and the preset time interval.
[0054] In this embodiment, a set of continuous images of the hit golf ball 9 from the side includes three images, which are image one, image two and image three in order of shooting time; a set of continuous images of the hit golf ball 9 from the top also includes three images, which are image four, image five and image six in order of shooting time.
[0055] Take a set of continuous images of a golf ball 9 being hit from the side as an example:
[0056] First, the control module 4 compares the position of the golf ball 9 shown in the first image with the position of the golf ball 9 shown in the second image, obtains the first movement distance in the striking direction and the first longitudinal movement distance of the golf ball 9, and then calculates the first movement speed in the striking direction of the golf ball 9 according to the first movement distance in the striking direction and the above-mentioned preset time interval, and calculates the first longitudinal movement speed of the golf ball 9 according to the first longitudinal movement distance and the above-mentioned preset time interval; then, the control module 4 compares the position of the golf ball 9 shown in the second image with the position of the golf ball 9 shown in the third image, obtains the second movement distance in the striking direction and the second longitudinal movement distance of the golf ball 9, and then calculates the second movement speed in the striking direction of the golf ball 9 according to the second movement distance in the striking direction and the above-mentioned preset time interval, and calculates the second longitudinal movement speed of the golf ball 9 according to the second longitudinal movement distance and the above-mentioned preset time interval; then, the striking direction acceleration of the golf ball 9 is calculated according to the first movement speed in the striking direction, the second movement speed in the striking direction and the above-mentioned preset time interval, and the longitudinal acceleration and longitudinal movement angle of the golf ball 9 are calculated according to the first longitudinal movement speed, the second longitudinal movement speed and the above-mentioned preset time interval.
[0057] Take a set of consecutive images of a golf ball 9 being hit from above as an example:
[0058] First, the control module 4 compares the position of the golf ball 9 shown in the fourth image with the position of the golf ball 9 shown in the fifth image to obtain a first lateral movement distance of the golf ball 9, and then calculates a first lateral movement speed of the golf ball 9 according to the first lateral movement distance and the above-mentioned preset time interval; then, the control module 4 compares the position of the golf ball 9 shown in the fifth image with the position of the golf ball 9 shown in the sixth image to obtain a second lateral movement distance of the golf ball 9, and then calculates a second lateral movement speed of the golf ball 9 according to the second lateral movement distance and the above-mentioned preset time interval; then, the lateral acceleration and lateral movement angle of the golf ball 9 are calculated according to the first lateral movement speed, the second lateral movement speed and the above-mentioned preset time interval.
[0059] After calculating the hitting direction acceleration, longitudinal acceleration, longitudinal movement angle, lateral acceleration and lateral movement angle of the golf ball 9, the control module 4 combines the hitting direction acceleration, longitudinal acceleration and lateral acceleration to obtain the acceleration of the golf ball 9, and combines the longitudinal movement angle and the lateral movement angle to obtain the movement angle of the golf ball 9, and then analyzes the expected landing point of the hit golf ball 9 based on the acceleration and movement angle of the golf ball 9.
[0060] B3. Among the multiple cameras that respectively shoot different ball landing areas, call the camera that can shoot the expected landing point of the ball, analyze the actual landing area of the hit golf ball based on the shooting content, and call the camera that can shoot the actual landing area to obtain the actual landing point of the hit golf ball.
[0061] In this embodiment, the control module 4 analyzes that the expected landing place of the golf ball 9 is located in the stopping area e. Since the golf ball 9 will roll a certain distance after landing in the stopping area e, the control module 4 calls the second camera 6 (i.e., the second camera 6 used to shoot the stopping area e) that can shoot the expected landing place among the nine second cameras 6 that shoot the nine stopping areas respectively to capture the golf ball 9 that rolls after landing in the stopping area e, and then analyzes the actual stopping area of the golf ball 9 according to the shooting content. If the golf ball 9 stops in the stopping area e, that is, the actual stopping area of the golf ball 9 is still the stopping area e, the control module 4 uses the second camera 6 used to shoot the stopping area e to shoot and obtain the actual stopping point of the golf ball 9, which is still located in the stopping area e. If the golf ball 9 falls on the ball stopping area e and then rolls to another ball stopping area, for example, rolls to the ball stopping area f and stops, that is, the actual ball stopping area of the golf ball 9 is the ball stopping area f, the control module 4 first calls the second camera 6 that can photograph the expected landing place (that is, the second camera 6 used to photograph the ball stopping area e) to capture the golf ball 9 that rolls after falling on the ball stopping area e. When it is determined that the golf ball 9 rolls into the ball stopping area f, the control module 4 stops calling the second camera 6 that can photograph the ball stopping area e to take pictures, and instead calls the second camera 6 that can photograph the ball stopping area f to capture the golf ball 9 that continues to roll after entering the ball stopping area f. When the golf ball 9 stops, the second camera 6 that can photograph the ball stopping area f is used to photograph and obtain the actual stopping point of the hit golf ball 9.
[0062] B4. The stopping score is obtained based on the difference between the actual stopping point and the preset stopping point.
[0063] After the control module 4 obtains the actual stopping point of the golf ball 9 by shooting, it calculates the difference between the actual stopping point and the preset stopping point 3, and then obtains the stopping score according to the difference. Specifically, each stopping area is provided with a basic stopping score and an additional stopping score, wherein: the basic stopping score of the stopping area b is 90 points, the basic stopping score of the stopping areas a, c, e is 80 points, the basic stopping score of the stopping areas d, f, h is 70 points, and the basic stopping score of the stopping areas g, i is 60 points; the additional stopping score has 10 intervals, namely 1 point, 2 points, 3 points, 4 points, 5 points, 6 points, 7 points, 8 points, 9 points and 10 points. If the actual landing point of the golf ball 9 is located in the landing area b, the 10 additional landing scores are determined according to the distance between the actual landing point and the expected landing point 3. The smaller the distance, the higher the additional landing score. If the actual landing point of the golf ball 9 is not located in the landing area b, the 10 landing scores are determined according to the distance between the actual landing point and the landing area b. The smaller the distance, the higher the additional landing score.
[0064] In this embodiment, because the actual stopping point of the golf ball 9 is located in the stopping area e, the basic stopping score of the golf ball 9 is 80 points. Then, the control module 4 calculates the distance between the actual stopping point of the golf ball 9 and the stopping area b according to the actual stopping point of the golf ball 9 obtained by the shooting, and then obtains the additional stopping score of 5 points according to the distance analysis. Therefore, the stopping score of the golf ball 9 in this embodiment is 80+5=85 points.
[0065] After obtaining the stroke score and the stop score, the golf grade test scoring system performs the following step C to obtain the final score:
[0066] C. Combine the hitting action score and the stopping score to get the final score.
[0067] The control module 4 performs weighted calculation on the hitting action score and the stopping score according to the preset weights to obtain a final score. For example, the hitting action score is weighted to 25%, and the stopping score is weighted to 75%, so that the final score is 245*25%+85*75%=125 points.
[0068] In this embodiment, the golf grade test scoring system is provided with a display screen, and the two third cameras 7 also shoot the swing and hitting videos of the golf player 9 from two angles respectively, and the second camera 6 corresponding to the stopping area e also shoots the stopping video of the golf ball 9. Among them, the second camera 6 and the third camera 7 use multi-eye linkage focus detection and tracking cameras, which include a fixed-point sub-camera and a moving-point sub-camera; the fixed-point sub-camera has an embedded detection and tracking algorithm to detect the golf player 8 and the key points of the human body and track the golf ball 9; the moving-point sub-camera is responsible for capturing and amplifying the details of the key parts. After obtaining the final score, the control module 4 also performs the following step D: using the display screen to output and display the swing and hitting videos and the stopping videos. In this way, the golf player 8 and the referee can dynamically observe the swing and hitting actions and the stopping situation of the golf player 8 on the display screen.
[0069] In this embodiment, the control module 4 is electrically connected to the two first cameras 5 and the nine second cameras 6, respectively, to form a system for obtaining the actual stopping point of the golf ball. When the computer program stored on the computer-readable storage medium of the control module 4 is executed by the processor, a method for obtaining the actual stopping point of the golf ball can be implemented. The method for obtaining the actual stopping point of the golf ball specifically includes the above steps B1, B2, and B3.
[0070] In addition, the control module 4 is electrically connected to the two first cameras 5 and the nine second cameras 6, respectively, and also forms a golf ball stop scoring system. When the computer program stored on the computer-readable storage medium of the control module 4 is executed by the processor, a golf ball stop scoring method can be implemented. The golf ball stop scoring method specifically first executes the above-mentioned golf ball actual stop point acquisition method to obtain the actual stop point, and then executes the above-mentioned step B4 to obtain the stop score of the golf ball 9.
[0071] The above is only an implementation method of the invention, and does not limit the scope of patent protection. Those skilled in the art can make non-substantial changes or substitutions based on the invention, which still fall within the scope of patent protection.
Claims
1. The golf grade test scoring method is characterized by: The steps include: A1. For a single swing action of a golf player, multiple swing action images are obtained by shooting from multiple angles; A2. Identify key points of the human body in each swing and hitting action image; A3. Compare the key points of the human body in each swing action image with the key points of the human body in the standard golf swing action image, and obtain the swing action score according to the similarity comparison result; B1. Continuously photographing the golf ball at a plurality of angles at a preset time interval to obtain a plurality of sets of continuous images, wherein each angle corresponds to a set of continuous images; B2. analyzing the expected landing position of the hit golf ball according to the golf ball positions displayed in the plurality of sets of continuous images and the preset time interval; B3. Among the multiple cameras that respectively capture different ball landing areas, call the camera that can capture the expected ball landing area to capture, analyze the actual landing area of the golf ball according to the captured content, and call the camera that can capture the actual landing area to capture and obtain the actual landing point of the golf ball; B4. Obtaining a stopping score based on the difference between the actual stopping point and the preset stopping point; C. Combine the hitting action score and the stopping score to get the final score.
2. The golf grade test scoring method according to claim 1, characterized in that: In step B3, if it is determined that the expected ball landing place is not located in the actual ball stopping area, the camera that can capture the expected ball landing place is stopped from being used for shooting, and the camera that can capture the actual ball stopping area is used for shooting instead.
3. The golf grade test scoring method according to claim 1, characterized in that: Specifically, step B2 first calculates the acceleration and movement angle of the golf ball after being hit according to the position of the golf ball in the multiple sets of continuous images and the preset time interval, and then analyzes the expected landing position of the hit golf ball according to the acceleration and the movement angle.
4. The golf grade test scoring method according to claim 1, characterized in that: In the step B1, the angles of the plurality of groups of continuous images include angles capable of capturing changes in the longitudinal movement of a hit golf ball and angles capable of capturing changes in the lateral movement of a hit golf ball.
5. The golf grade test scoring method according to claim 1, characterized in that: In the step A1, the angles of the plurality of swing action images include an angle of viewing the golfer in a swing and hitting preparation state from the front and an angle of viewing the golfer in a swing and hitting preparation state from the side.
6. The golf grade test scoring method according to claim 1, characterized in that: Specifically, the step C performs weighted calculation on the hitting action score and the stopping score according to preset weights to obtain a final score.
7. The golf grade test scoring method according to claim 1 is characterized by: exist In the step A1, a golf player's swing and hitting video is shot from multiple angles, and / or in the step B1, a golf ball stopping video is shot; The method comprises step D: outputting and displaying the swing and ball hitting video and / or the ball stopping video.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the golf grade test scoring method according to any one of claims 1 to 7 are implemented.
9. The golf grade test scoring system is characterized by: The invention comprises a control module, a first camera for photographing a golf ball's motion trajectory, a second camera for photographing a golf ball's stopping point, and a third camera for photographing a golf player, wherein there are a plurality of first cameras, a plurality of second cameras, and a third camera, each of which is used to photograph different stopping areas, and the control module is electrically connected to each of the first cameras, the second cameras, and the third cameras; the control module comprises a computer-readable storage medium and a processor connected to each other, and the computer-readable storage medium is as described in claim 8; in the step B1, the plurality of first cameras are used to continuously photograph the hit golf ball at different angles at preset time intervals; in the step B3, among the plurality of second cameras, a camera that can photograph the expected landing point is called to photograph, so as to obtain the actual stopping point of the hit golf ball; in the step A1, the plurality of third cameras are used to photograph from multiple angles to obtain multiple swing action images.
10. The golf grade test scoring system according to claim 9, characterized in that: Each second camera only captures a corresponding ball stopping area and does not capture other ball stopping areas.
Citation Information
Patent Citations
Golf scoring method and system
CN105879350A
Golf play information processing system and golf play information utilization system
JP2008102625A