Target object throwing project site dot matrix identification combination and distance measurement method
By setting up a combination of dot matrix marks on the sports field, using machine vision technology to extract and analyze image features, and fitting the trajectory curve with parabolic rules, the problem of high hardware cost of existing sports measurement solutions has been solved, and a low-cost, universal solution has been achieved. Adaptable throwing distance measurement.
Patent Information
- Application Number
- CN202310889540.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-19
- Publication Date
- 2025-01-21
AI Technical Summary
Existing sports measurement solutions require auxiliary equipment or sensor burial, the hardware cost is high, and it is difficult to achieve low-cost and universal ranging solutions.
Using machine vision-based target object throwing project venue point mark combination, through fixed-point collection of images, extract point mark features and target object features, combined with parabolic rules to fit the trajectory curve, and calculate the throwing distance.
It realizes a low-cost, low-hardware arrangement-based throwing distance measurement based on machine vision. It has strong universality and is suitable for digital statistical motion data and motion process analysis.
Smart Images
Figure CN119338903A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of distance measurement based on machine vision, and in particular to a field array identification combination and a distance measurement method for a target throwing event. Background Art
[0002] In the existing technology, the training or assessment of some sports has been automated or even intelligent. For example, in sports such as shot put and shot put, sports performance needs to be tested. However, existing sports measurement solutions, such as sensor detection, require auxiliary equipment or sensor embedding, and the hardware cost is relatively high. Summary of the invention
[0003] The present invention proposes a combination of dot matrix markers for a target throwing event site, including a first type of dot matrix marker and one or more second type of dot matrix markers, wherein the first type of dot matrix marker forms a project area recognizable to the naked eye on the site, and the one or more second type of dot matrix markers divide the project area into multiple sub-project areas, each sub-project area has at least one starting dot matrix marker, and the trajectory of a target object effectively thrown at any starting dot matrix marker should be located directly above the sub-project area to which the starting dot matrix marker belongs.
[0004] Furthermore, a sub-item area has at least one first-type starting dot matrix mark, a plurality of first-type starting dot matrix marks corresponding to a plurality of sub-item areas are all located on the same straight line, and a portion of the boundary of the item area is located on the straight line.
[0005] Furthermore, a plurality of first type starting point matrix markers are connected end to end.
[0006] Furthermore, it also includes at least one visual dot matrix mark, so that after the target object is effectively thrown, the distance between the landing point of the target object in the sub-project area and the starting dot matrix mark of the sub-project area can be measured and obtained with reference to the dot matrix mark.
[0007] A distance measurement method for a multi-person target throwing event based on the dot matrix mark combination comprises the following steps:
[0008] Collect multiple images containing regional features of the first sub-project at fixed points;
[0009] Acquire a first positional relationship between the dot matrix identification feature and the image, and acquire a first position in the image of a landing point of a target object that falls into a first sub-item area in the image;
[0010] Acquire a second positional relationship between the landing point and the dot matrix marker according to the first positional relationship and the first position, and acquire an actual position of the landing point according to the second positional relationship and an actual position of the pre-stored dot matrix marker;
[0011] The first throwing distance is calculated based on the actual position of the landing point and the pre-stored throwing starting position.
[0012] Furthermore, “obtaining a first positional relationship between the dot matrix identification feature and the image” specifically includes:
[0013] Capturing a first image, the first image including a dot matrix identification feature, wherein the first image and the image are both acquired through fixed-point acquisition by the same acquisition terminal;
[0014] The positional relationship between the dot matrix identification feature extracted from the first image and the first image can be called as the first positional relationship between the dot matrix identification feature and the image.
[0015] Furthermore, “obtaining a first position in the image of a falling point of a target object falling into the first sub-item area in the image” specifically includes:
[0016] Analyzing a plurality of first-type target features belonging to a first target and a first-type second position of the plurality of first-type target features in the images from the target features contained in the plurality of images;
[0017] A landing point of the first target object is analyzed according to the plurality of first type second positions, the landing point falls within the first sub-item area, and a first position of the landing point in the image is calculated.
[0018] Furthermore, “analyzing a plurality of first-type target features belonging to the first target from the target features contained in the plurality of images” specifically includes:
[0019] Extracting multiple second positions of target object features in multiple images;
[0020] The multiple second positions are fitted into multiple trajectory curves according to a preset parabola rule, the multiple trajectory curves include a first trajectory curve corresponding to the first target object, and the multiple second positions fitted with the first trajectory curve are multiple first-type second positions, and the multiple target object features corresponding to the multiple first-type second positions are multiple first-type target object features belonging to the first target object.
[0021] Furthermore, “analyzing the landing point of the first target object according to the plurality of first type second positions” specifically means that the end point of the first trajectory curve is the landing point of the first target object.
[0022] Furthermore, the one or more first-type second positions that do not participate in fitting the first trajectory curve due to not meeting the preset parabola rule include at least one first-type second position that has a preset positional relationship with the landing point of the first target object.
[0023] Furthermore, “analyzing a plurality of first-type target features belonging to the first target and a plurality of first-type second positions of the plurality of first-type target features in the images from the target features contained in the plurality of images” is specifically: when the first position appearing in the X-th frame image is lower than the first position appearing in the X+-th frame image, and the first position appearing in the X-th frame image is lower than the first position appearing in the X-th frame image, the first position appearing in the X-th frame image is the landing point of the first target.
[0024] Furthermore, it also includes:
[0025] Fitting a first trajectory curve from the plurality of first positions according to a parabola rule;
[0026] Obtaining the relative positions of multiple motion individual features contained in the image;
[0027] A correspondence between the first target object and the first moving individual in the multiple moving individual features is established according to the first trajectory curve and the relative positions of the multiple moving individual features in the image, thereby obtaining a correspondence between the first throwing distance and the first moving individual.
[0028] Furthermore, it also includes:
[0029] Acquire multiple third positions of multiple motion individual features contained in the image;
[0030] A corresponding relationship between the first target object and the first moving individual is established according to the first trajectory curve and the plurality of third positions, thereby obtaining a corresponding relationship between the first throwing distance and the first moving individual.
[0031] Beneficial effects:
[0032] In the present invention, the proposed method for measuring the throwing distance of a target object based on machine vision abandons the traditional method of burying sensors or using wearable devices, and estimates the throwing distance during the motion process based on machine vision. It has low requirements for hardware layout, has strong universality, and is adaptable to aspects such as digital statistical motion data and analysis of the motion process. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 This is a schematic diagram of a target throwing event site array identification combination proposed in Example 1. DETAILED DESCRIPTION
[0034] Example 1
[0035] refer to Figure 1Embodiment 1 proposes a target throwing event site dot matrix marking combination, including a first type dot matrix marking 1 and one or more second type dot matrix markings, the first type dot matrix marking 1 forms a project area 2 recognizable to the naked eye on the site, and the one or more second type dot matrix markings divide the project area 2 into multiple sub-project areas 21, and a sub-project area 21 has at least one starting dot matrix marking 3.
[0036] Preferably, the trajectory of the target object effectively thrown at any starting point matrix mark 3 should be located directly above the sub-item area 21 to which the starting point matrix mark 3 belongs.
[0037] The actual scene can be multi-person shot put or shot put distance measurement, and the target object is specifically a shot put or a shot put. The dot matrix mark combination is used for distance measurement based on machine vision. Since the trajectory of the target object of the effective throw should be located directly above the sub-item area 21 to which the starting dot matrix mark 3 belongs, it is easier to judge the corresponding relationship between the throwing results and the sports individual.
[0038] Furthermore, a sub-item area 21 has at least one first-type starting dot matrix mark 3, and multiple first-type starting dot matrix marks 3 corresponding to multiple sub-item areas 21 are all located on the same straight line, and part of the boundary of the item area 2 is located on the straight line.
[0039] Furthermore, a plurality of first type starting point matrix markers 3 are connected end to end.
[0040] Furthermore, it also includes at least one visualized dot matrix mark 4, so that after the target object is effectively thrown, the distance between the landing point of the target object in the sub-project area 21 and the starting dot matrix mark 3 of the sub-project area 21 can be measured and obtained with reference to the dot matrix mark 4.
[0041] Example 2
[0042] Embodiment 2 proposes a distance measurement method for a multi-person target throwing event based on the dot matrix mark combination proposed in Embodiment 1, comprising the following steps:
[0043] S1. Collect multiple images containing regional features of the first sub-item at fixed points.
[0044] The meaning of fixed-point acquisition is that a plurality of images are acquired by the same terminal device at different times or in continuous frames, so that the position of the first sub-item regional feature in the plurality of images remains consistent.
[0045] S2. Obtain a first positional relationship between the dot matrix identification feature and the image, and obtain a first position of the target object in the image that falls into the first sub-item area.
[0046] The target object is specifically a shot put or a solid ball.
[0047] Specifically include:
[0048] S21, capturing a first image, wherein the first image contains a dot matrix identification feature, and the first image and the image are both acquired through fixed-point capture by the same acquisition terminal.
[0049] Wherein, a dot matrix logo composed of a plurality of visual dot matrix logos is arranged on the site, and the actual positions of the visual dot matrix logos are pre-stored.
[0050] Since the dot matrix markings on the site are visualized, they can be collected. The first image and the image are both obtained through fixed-point collection by the same collection terminal, so that no matter whether it is the dot matrix markings, the target features or the first sub-project area features, their relative positions in the image and the first image can always be consistent.
[0051] S22: extracting the positional relationship between the dot matrix identification feature and the first image from the first image, which can be called as the first positional relationship between the dot matrix identification feature and the image.
[0052] Even if the dot matrix identification features become unclear or even disappear on the field, it will not affect.
[0053] S23: Analyze, from the target features contained in the multiple images, a plurality of first-type target features belonging to the first target and the first-type second positions of the plurality of first-type target features in the images.
[0054] Specifically include:
[0055] S231, extracting multiple second positions of target object features in multiple images.
[0056] S232, fitting the plurality of second positions into a plurality of trajectory curves according to a preset parabola rule, wherein the plurality of trajectory curves include a first trajectory curve corresponding to the first target object, and the plurality of second positions fitted with the first trajectory curve are a plurality of first-type second positions, and the plurality of target object features corresponding to the plurality of first-type second positions are a plurality of first-type target object features belonging to the first target object.
[0057] The trajectory of the target from the initial throw to the rest after landing can be divided into two stages:
[0058] The first stage is from the initial throw to the landing point, and the second stage is from the landing point to the stopping point.
[0059] Based on common sense, the first stage of natural flight in the air basically conforms to a certain existing parabolic rule; while the second stage, as a bounce from touching the ground or rolling to a standstill, does not follow any rules and is not applicable to any parabolic rule.
[0060] If you want to get the final distance, the landing point is the key.
[0061] Since the targets fly freely with each other and have their own starting points and end points, as well as their own flight trajectories, multiple trajectory curves can be fitted according to the preset parabola rules based on multiple relative positions.
[0062] Any frame image contains and only contains one first position, and two adjacent ones of the multiple first positions belong to adjacent frame images. Then, there is a preset second position relationship between the multiple first positions of the fitted first trajectory line. The second relative position can be a preset distance between the two, a position relationship between the two, etc. The first target object is one of the multiple targets, and the first type of target feature also belongs to one of the multiple target features. The multiple first positions naturally include the landing point of the first target object in the image.
[0063] S233, analyzing a plurality of first-type target object features at a first-type second position in the image.
[0064] S24. Analyze the landing point of the first target object according to the multiple first-type second positions, and the landing point falls within the first sub-item area, and calculate the first position of the landing point in the image.
[0065] The end point of the first stage is the landing point, and the last first position of the first target object participating in the fitting in the image is the landing point. Although the multiple first positions of the second stage do not participate in the fitting of the first trajectory curve because they do not conform to the parabola rule, the first stage is adjacent to the second stage and the landing point is used as the dividing point. Then, at least one of the multiple first positions of the second stage is adjacent to the landing point. The accuracy of the determined landing point can be verified by analyzing whether the multiple first positions of the second stage include a first position adjacent to the landing point. Therefore, the first position corresponding to the end point of the first trajectory curve is the landing point of the first target object, and one or more first-type second positions that do not conform to the preset parabola rule and do not participate in the fitting of the first trajectory curve also include at least one first-type second position having a preset position relationship with the landing point of the first target object.
[0066] S3. Acquire a second position relationship between the first position and the dot matrix marker according to the first position relationship, and acquire the actual position of the landing point according to the second position relationship and the actual position of the pre-stored dot matrix marker.
[0067] After obtaining the second positional relationship between the landing point and the dot matrix identification feature, it is equivalent to knowing between which points in the dot matrix the landing point is located or which point it coincides with. The actual position of the pre-stored dot matrix identification feature refers to the actual positional relationship between multiple point features contained in the pre-stored dot matrix identification feature and / or the actual positions of multiple point features. When it is known between which points in the dot matrix the landing point is located or which point it coincides with, and the actual positional relationship between the multiple point features contained in the dot matrix identification feature and / or the actual positions of multiple point features are known, the actual position of the landing point can naturally be calculated within the allowable error range.
[0068] S4, calculating the first throwing distance according to the actual position and the pre-stored throwing starting position.
[0069] In multi-person shot put or multi-person shot put events, the actual starting position of the throw is preset.
[0070] S5. Establish a correspondence between the first target object and the first moving individual in the multiple moving individual features according to the first trajectory curve and the relative positions of the multiple moving individual features in the image, so as to obtain a correspondence between the first throwing distance and the first moving individual.
[0071] Example 3
[0072] Embodiment 3 proposes a distance measurement method for a multi-person target throwing event based on the dot matrix mark combination proposed in Embodiment 1, comprising the following steps:
[0073] S1. Collect multiple images containing regional features of the first sub-item at fixed points.
[0074] The meaning of fixed-point acquisition is that a plurality of images are acquired by the same terminal device at different times or in continuous frames, so that the position of the first sub-item regional feature in the plurality of images remains consistent.
[0075] S2. Obtain a first positional relationship between the dot matrix identification feature and the image, and obtain a first position in the image of a landing point of a target object that falls into a first sub-item area in the image.
[0076] The target object is specifically a shot put or a solid ball.
[0077] Specifically include:
[0078] S21, capturing a first image, wherein the first image contains a dot matrix identification feature, and the first image and the image are both acquired through fixed-point capture by the same acquisition terminal.
[0079] Wherein, a dot matrix logo composed of a plurality of visual dot matrix logos is arranged on the site, and the actual positions of the visual dot matrix logos are pre-stored.
[0080] Since the dot matrix markings on the site are visualized, they can be collected. The first image and the image are both obtained through fixed-point collection by the same collection terminal, so that no matter whether it is the dot matrix markings, the target features or the first sub-project area features, their relative positions in the image and the first image can always be consistent.
[0081] S22: extracting the positional relationship between the dot matrix identification feature and the first image from the first image, which can be called as the first positional relationship between the dot matrix identification feature and the image.
[0082] Even if the dot matrix identification features become unclear or even disappear on the field, it will not affect.
[0083] S23: Acquire a first position in the image of a landing point of a target object that falls into the first sub-item area in the image.
[0084] Specifically include:
[0085] S231. Analyze, from the target features contained in the multiple images, a plurality of first-type target features belonging to the first target and the first-type second positions of the plurality of first-type target features in the images.
[0086] Among them, when the first position appearing in the Xth frame image is lower than the first position appearing in the X+1th frame image, and the first position appearing in the Xth frame image is lower than the first position appearing in the X-1th frame image, the first position appearing in the Xth frame image is the landing point of the first target object.
[0087] S232: Analyze the landing point of the first target object according to the multiple first type second positions, the landing point falls within the first sub-item area, and calculate the first position of the landing point in the image.
[0088] S3. Acquire a second positional relationship between the landing point and the dot matrix marker according to the first positional relationship and the first position, and acquire an actual position of the landing point according to the second positional relationship and an actual position of the pre-stored dot matrix marker.
[0089] After obtaining the second positional relationship between the landing point and the dot matrix identification feature, it is equivalent to knowing between which points in the dot matrix the landing point is located or which point it coincides with. The actual position of the pre-stored dot matrix identification feature refers to the actual positional relationship between multiple point features contained in the pre-stored dot matrix identification feature and / or the actual positions of multiple point features. When it is known between which points in the dot matrix the landing point is located or which point it coincides with, and the actual positional relationship between the multiple point features contained in the dot matrix identification feature and / or the actual positions of multiple point features are known, the actual position of the landing point can naturally be calculated within the allowable error range.
[0090] S4. Calculate the first throwing distance according to the actual position of the landing point and the pre-stored throwing starting position.
[0091] In multi-person shot put or multi-person shot put events, the actual starting position of the throw is preset.
[0092] S5. Fitting the multiple first positions into a first trajectory curve according to a parabola rule.
[0093] S6. Obtain the relative positions of multiple motion individual features contained in the image.
[0094] S7. Establish a correspondence between the first target object and the first moving individual in the plurality of moving individual features according to the first trajectory curve and the relative positions of the plurality of moving individual features in the image, thereby obtaining a correspondence between the first throwing distance and the first moving individual.
[0095] In layman's terms, we can get the corresponding relationship between athletes and results.
[0096] Example 4
[0097] Embodiment 4 proposes a distance measurement method for a multi-person target throwing event based on the dot matrix mark combination proposed in Embodiment 1, comprising the following steps:
[0098] S1. Collect multiple images containing regional features of the first sub-item at fixed points.
[0099] S2. Obtain a first positional relationship between the dot matrix identification feature and the image, and obtain a first position in the image of a landing point of a target object that falls into a first sub-item area in the image.
[0100] Specifically include:
[0101] S21, capturing a first image, wherein the first image contains a dot matrix identification feature, and the first image and the image are both acquired through fixed-point capture by the same acquisition terminal.
[0102] S22: extracting the positional relationship between the dot matrix identification feature and the first image from the first image, which can be called as the first positional relationship between the dot matrix identification feature and the image.
[0103] S23: Acquire a first position in the image of a landing point of a target object that falls into the first sub-item area in the image.
[0104] S3. Acquire a second positional relationship between the landing point and the dot matrix marker according to the first positional relationship and the first position, and acquire an actual position of the landing point according to the second positional relationship and an actual position of the pre-stored dot matrix marker.
[0105] S4. Calculate the first throwing distance according to the actual position of the landing point and the pre-stored throwing starting position.
[0106] S5. Acquire multiple third positions of multiple motion individual features contained in the image.
[0107] S6. Establish a corresponding relationship between the first target object and the first moving individual according to the first trajectory curve and the plurality of third positions, thereby obtaining a corresponding relationship between the first throwing distance and the first moving individual.
[0108] In layman's terms, we can also obtain the corresponding relationship between individual athletes and their performance.
[0109] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A target throwing event site array identification combination, characterized in that: It includes a first type of dot matrix marker and one or more second type of dot matrix markers. The first type of dot matrix marker forms a project area identifiable to the naked eye on the site. The one or more second type of dot matrix markers divide the project area into multiple sub-project areas. A sub-project area has at least one starting dot matrix marker.
2. The dot matrix marking combination according to claim 1, characterized in that: The trajectory of the target object effectively thrown at any starting point matrix mark should be located directly above the sub-item area to which the starting point matrix mark belongs.
3. The dot matrix marking combination according to claim 1, characterized in that: A sub-project area has at least one first-type starting dot matrix mark, and a plurality of first-type starting dot matrix marks corresponding to a plurality of sub-project areas are all located on the same straight line, and a portion of the boundary of the project area is located on the straight line. Preferably, a plurality of first type starting point matrix markers are connected end to end.
4. The dot matrix marking combination according to claim 1, characterized in that: It also includes at least one visual dot matrix mark (4), so that after the target object is effectively thrown, the distance between the landing point of the target object in the sub-project area and the starting dot matrix mark of the sub-project area can be measured and obtained by referring to the dot matrix mark (4).
5. A distance measurement method for a multi-person target throwing event based on the dot matrix mark combination as claimed in claim 4, characterized in that: The steps include: Collect multiple images containing regional features of the first sub-project at fixed points; Acquire a first positional relationship between the dot matrix mark (4) feature and the image, and acquire a first position in the image of a point where a target object that falls into the first sub-item area in the image falls; Acquire a second positional relationship between the landing point and the dot matrix marker (4) based on the first positional relationship and the first position, and acquire the actual position of the landing point based on the second positional relationship and the actual position of the pre-stored dot matrix marker (4); The first throwing distance is calculated based on the actual position of the landing point and the pre-stored throwing starting position.
6. The distance measurement method according to claim 5, characterized in that: “Obtaining a first positional relationship between the dot matrix marker (4) feature and the image” specifically includes: Capturing a first image, the first image containing a dot matrix mark (4) feature, the first image and the image being acquired through fixed-point acquisition by the same acquisition terminal; The positional relationship between the dot matrix mark (4) feature and the first image is extracted from the first image and can be called as the first positional relationship between the dot matrix mark (4) feature and the image.
7. The distance measurement method according to claim 4, characterized in that: “Obtaining a first position in the image of a falling point of a target object falling into a first sub-item area in the image” specifically includes: Analyzing a plurality of first-type target features belonging to a first target and a first-type second position of the plurality of first-type target features in the images from the target features contained in the plurality of images; A landing point of the first target object is analyzed according to the plurality of first type second positions, the landing point falls within the first sub-item area, and a first position of the landing point in the image is calculated.
8. The distance measurement method according to claim 7, characterized in that: “Analyzing a plurality of first-type target features belonging to a first target from target features contained in a plurality of images” specifically includes: Extracting multiple second positions of target object features in multiple images; The multiple second positions are fitted into multiple trajectory curves according to a preset parabola rule, the multiple trajectory curves include a first trajectory curve corresponding to the first target object, and the multiple second positions fitted with the first trajectory curve are multiple first-type second positions, and the multiple target object features corresponding to the multiple first-type second positions are multiple first-type target object features belonging to the first target object.
9. The distance measurement method according to claim 8, characterized in that: “Analyzing the landing point of the first target object based on multiple first-type second positions” specifically means: the end point of the first trajectory curve is the landing point of the first target object.
10. The distance measurement method according to claim 9, characterized in that: The one or more first-type second positions that do not participate in fitting the first trajectory curve due to not meeting the preset parabola rule include at least one first-type second position that has a preset position relationship with the landing point of the first target object.
11. The distance measurement method according to claim 7, characterized in that: "Analyzing multiple first-type target features belonging to the first target and first-type second positions of multiple first-type target features in the images from target features contained in multiple images" specifically means: when the first position appearing in the Xth frame image is lower than the first position appearing in the X+1th frame image, and the first position appearing in the Xth frame image is lower than the first position appearing in the X-1th frame image, the first position appearing in the Xth frame image is the landing point of the first target.
12. The distance measurement method according to claim 11, characterized in that: Also includes: Fitting a first trajectory curve from the plurality of first positions according to a parabola rule; Obtaining the relative positions of multiple motion individual features contained in the image; A correspondence between the first target object and the first moving individual in the multiple moving individual features is established according to the first trajectory curve and the relative positions of the multiple moving individual features in the image, thereby obtaining a correspondence between the first throwing distance and the first moving individual.
13. The distance measurement method according to claim 5, characterized in that: Also includes: Acquire multiple third positions of multiple motion individual features contained in the image; A corresponding relationship between the first target object and the first moving individual is established according to the first trajectory curve and the plurality of third positions, thereby obtaining a corresponding relationship between the first throwing distance and the first moving individual.