Standing long jump testing device capable of automatic measurement and method thereof

Abstract

The invention discloses a standing long jump testing device and method capable of automatic measurement, and belongs to the technical field of measurement. According to the method, environmental information is collected through a binocular camera, a laser radar and a pressure sensing pad, a reliability evaluation model of each sensor is constructed, and a reliability coefficient of each sensor is calculated; and based on a dynamic weight distribution model, fusing the detection result and the weight of each sensor, and outputting a final long jump distance. Wherein weight distribution is dynamically adjusted according to environmental factors (such as illumination, visibility and ground flatness), and data fusion is realized through a distance determination model. The problem that a single sensor is easily interfered by the environment is solved, the measurement precision and the environmental adaptability are remarkably improved, and the method is particularly suitable for complex scenes such as sandpits.

Description

Technical Field

[0001] The invention belongs to the field of measurement technology, and in particular relates to a standing long jump test device capable of automatic measurement and a method thereof. Background Art

[0002] Existing standing long jump test devices mostly rely on a single sensor (such as a camera, lidar, or pressure-sensitive mat) to measure distance, but a single sensor is easily interfered with by environmental factors. For example, the recognition accuracy of a binocular camera decreases in low light or poor visibility; the measurement error of a lidar increases when there is noise interference or insufficient point cloud density; and the pressure-sensitive mat completely fails on uneven or soft surfaces such as sand pits. These problems result in low measurement accuracy and poor environmental adaptability of traditional devices, making it difficult to meet the testing needs in complex scenarios. Therefore, there is an urgent need for an automatic measurement solution that can integrate multi-sensor data and dynamically adapt to environmental changes to improve the accuracy and reliability of standing long jump tests. Summary of the Invention

[0003] In view of the deficiencies in the prior art, the present invention provides an automatic measuring standing long jump test device and method thereof, which solve the above problems.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A standing long jump test method, comprising:

[0005] Obtaining binocular camera detection environment information, lidar detection environment information, and pressure sensing pad detection environment information;

[0006] Based on the binocular camera environment detection information, a binocular camera reliability model is constructed to output the binocular camera reliability evaluation coefficient;

[0007] Based on the lidar detection environment information, a lidar reliability model is constructed to output the lidar reliability evaluation coefficient;

[0008] Based on the pressure sensing pad detection environment information, a pressure sensing pad reliability model is constructed to output the pressure sensing pad reliability evaluation coefficient;

[0009] Based on the reliability evaluation coefficients of the binocular camera, the laser radar, and the pressure sensing pad, a dynamic weight distribution model is constructed to output the weights of the detection results of the binocular camera, the laser radar, and the pressure sensing pad in the pre-built distance determination model;

[0010] The detection results of the binocular camera, lidar, and pressure sensing pad, as well as their respective weights in the distance determination model, are input into the distance determination model to output the final detection distance.

[0011] On the basis of the above technical solutions, the present invention also provides the following optional technical solutions:

[0012] Further technical solution: The distance determination model is expressed as:

[0013] d final =w cam d cam +w lidar d lidar +w pad d pad

[0014] Among them, d final Indicates the final detection distance, d cam Indicates the detection distance of the binocular camera, d lidar Represents the laser radar detection distance, d pad Indicates the detection distance of the pressure sensor pad, w cam 、w lidar 、w pad represents the weight coefficient and w cam +w lidar +w pad =1.

[0015] Further technical solution: The dynamic weight allocation model is expressed as:

[0016]

[0017] Among them, w cam Represents the weight of the binocular camera detection result in the pre-built distance determination model, w lidar represents the weight of the lidar detection result in the pre-built distance determination model, w pad represents the weight of the pressure sensing pad detection result in the pre-built distance determination model, R cam Represents the reliability evaluation coefficient of the binocular camera, R lidar Represents the reliability evaluation coefficient of the laser radar, R pad Represents the reliability evaluation coefficient of the pressure sensing pad.

[0018] Further technical solution: The binocular camera detects environmental information including light intensity, visibility and heel recognition accuracy; the laser radar detects environmental information including visibility, point cloud density and noise ratio; the pressure sensing pad detects environmental information including the effective contact area ratio, ground flatness and ground hardness; the visibility in the binocular camera environmental detection information and the laser radar detection environmental information is the same visibility.

[0019] Further technical solution: The specific method of constructing a binocular camera reliability model based on binocular camera environment detection information and outputting a binocular camera reliability evaluation coefficient is as follows:

[0020] Ratio the current visibility and heel recognition accuracy with the corresponding standard values ​​to obtain the visibility index and heel recognition accuracy index;

[0021] Importing light intensity into the formula Output light intensity index, where I light Indicates the current light intensity, I0 indicates the reference light intensity, I b It represents the maximum allowable deviation from the light intensity, and k represents the light intensity sensitivity coefficient;

[0022] Import the visibility index, light intensity index, and heel recognition accuracy index into the constructed binocular camera reliability model and output the binocular camera reliability evaluation coefficient;

[0023] The binocular camera reliability model is expressed as:

[0024]

[0025] Among them, R cam Represents the reliability evaluation coefficient of the binocular camera, f(I light ) represents the light intensity index, g(V vist ) represents the visibility index, A heel represents the heel recognition accuracy index, C norm Represents the theoretical maximum reliability evaluation coefficient.

[0026] Further technical solution: The steps of constructing a lidar reliability model based on lidar detection environment information and outputting a lidar reliability evaluation coefficient are as follows:

[0027] The visibility and point cloud density are compared with the corresponding optimal values ​​to obtain the visibility index and point cloud density index;

[0028] The visibility index, point cloud density index and noise ratio are imported into the constructed lidar reliability model to output the lidar reliability evaluation coefficient;

[0029] The LiDAR reliability model is expressed as:

[0030] R lidar =D ind (1-N ind )V ind

[0031] Among them, R lidar Denotes the reliability evaluation coefficient of the laser radar, D ind Represents the point cloud density index, N ind Indicates the noise ratio, V ind Represents the visibility index.

[0032] Further technical solutions: the steps of constructing a pressure-sensitive mat reliability model based on the detection of environmental information and outputting a pressure-sensitive mat reliability evaluation coefficient are:

[0033] The standard flatness is compared with the current ground flatness to obtain a flatness index, and the current ground hardness is compared with the ideal ground hardness to obtain a ground hardness index.

[0034] The effective touch area ratio, the flatness index and the ground hardness index are introduced into the constructed pressure-sensitive reliability model to output a pressure-sensitive mat reliability evaluation coefficient.

[0035] The pressure-sensitive mat reliability model is represented as:

[0036] R pad =S act (1-F ind )H ind

[0037] Wherein, R pad represents the pressure-sensitive mat reliability evaluation coefficient, S act represents the effective touch area ratio, F ind represents the flatness index, and H ind represents the ground hardness index.

[0038] Further technical solutions: if the long jump field is a sand pit, R pad = 0.

[0039] An automatic measurement standing long jump test device adopts the above standing long jump test method, and comprises a box body, a binocular camera and a laser radar mounted on the box body, and a pressure-sensitive mat.

[0040] The present application provides a kind of automatic measurement standing long jump test device and method, compared with prior art has the following beneficial effects:

[0041] 1、The present application cooperates binocular camera, laser radar and pressure-sensitive mat, integrates the advantages of each sensor, makes up for the limitations of single sensor, simultaneously based on environmental information constructs reliability evaluation model, dynamically adjusts the weight of each sensor, ensures that high-precision results can still be output under complex conditions such as light change, noise interference or uneven ground, for special scenes such as sand pit, automatically shield pressure-sensitive mat data (weight is zero), avoid invalid data interference, improve system robustness. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is flowchart of the present application.

[0043] Figure 23D schematic diagram of the device of the present invention.

[0044] Notes on the accompanying figures: 1. Box body; 2. Binocular camera; 3. LiDAR; 4. Pressure sensing pad. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] The specific implementation of the present invention is described in detail below with reference to specific embodiments.

[0047] See also Figure 1 as well as Figure 2 , provided in one embodiment of the present invention, is a standing long jump test method, comprising the following steps:

[0048] Obtaining binocular camera detection environment information, lidar detection environment information, and pressure sensing pad detection environment information;

[0049] Based on the binocular camera environment detection information, a binocular camera reliability model is constructed to output the binocular camera reliability evaluation coefficient;

[0050] Based on the lidar detection environment information, a lidar reliability model is constructed to output the lidar reliability evaluation coefficient;

[0051] Based on the pressure sensing pad detection environment information, a pressure sensing pad reliability model is constructed to output the pressure sensing pad reliability evaluation coefficient;

[0052] Based on the reliability evaluation coefficients of the binocular camera, the laser radar, and the pressure sensing pad, a dynamic weight distribution model is constructed to output the weights of the detection results of the binocular camera 2, the laser radar 3, and the pressure sensing pad 4 in the pre-constructed distance determination model;

[0053] The detection results of the binocular camera 2, the laser radar 3 and the pressure sensing pad 4 and their respective weights in the distance determination model are input into the distance determination model to output the final detection distance.

[0054] Specifically, the method establishes a reliability evaluation model matched with the characteristics of each sensor by collecting environmental parameters of three types of sensors in parallel. For example, in the binocular camera reliability model, the light intensity is converted into an index that adapts to different brightness scenes through an exponential function, and the visual visibility and heel recognition rate reflect the degree of environmental interference through ratio calculation; in the laser radar reliability model, the point cloud density and noise ratio directly determine the signal quality score; in the pressure-sensitive pad reliability model, the ground hardness and flatness parameters are used to determine whether the touch area meets the measurement conditions. Then, the three types of reliability scores are input into a dynamic weight model to automatically assign the weight proportion of each sensor in distance calculation. When the reliability of a sensor decreases due to environmental interference, its weight coefficient decreases accordingly, thereby reducing the impact of error data on the final result.

[0055] Compared with the prior art, the traditional scheme directly outputs measurement results using only single sensor data without considering the dynamic influence of environmental factors on sensor performance. The present scheme realizes adaptive fusion of multi-sensor data by introducing an environmental information-driven reliability evaluation mechanism. For example, in a sandpit site scene, the pressure-sensitive pad 4 reliability score is automatically reset to zero, and the system only fuses binocular camera 2 and laser radar 3 data; in a foggy environment, the weight of laser radar 3 is increased to compensate for the deficiency of binocular camera 2 visibility.

[0056] Preferably, the binocular camera detects environmental information including light intensity, visibility, and heel recognition accuracy, the laser radar detects environmental information including visibility, point cloud density, and noise ratio, the pressure-sensitive pad detects environmental information including the proportion of effective touch area, ground flatness, and ground hardness, and the visibility in the binocular camera environmental detection information and the laser radar detection environmental information is the same visibility.

[0057] The light intensity refers to the visible light radiation intensity value collected by the light sensor, which can be implemented by a photoresistor or a photodiode, and is used to quantify the light conditions of the working environment of the visual sensor. The visibility refers to the degree of hindering of particulate matter in the air to the propagation of light, which can be implemented by a transmittance detector or a scattered light measuring device, and is used to represent the interference level of the environment to the optical sensor. The heel recognition accuracy refers to the probability of correctly recognizing the foot touch point when the binocular camera 2 captures the foot movement, which can be calculated by taking the complement of the misjudgment rate calculated by an image recognition algorithm, and is used to reflect the reliability of the action capture of the visual system. The point cloud density refers to the number of three-dimensional coordinate points generated by the laser radar 3 scanning in a unit volume, which can be implemented by counting the number of effective point clouds in a unit scanning area, and is used to evaluate the data integrity of the laser radar 3. The noise ratio refers to the proportion of abnormal outliers in the point cloud data of the laser radar 3, which can be calculated by separating the effective points and noise points by a clustering algorithm, and is used to represent the credibility of the measurement data of the laser radar 3. The effective touch area area ratio refers to the ratio of the effective area of the pressure sensing mat 4 that can trigger an effective signal to the total area, which can be calculated by the trigger state statistics of the pressure sensor array, and is used to determine whether the device deployment meets the measurement requirements. The ground flatness refers to the degree of fluctuation of the touch area surface, which can be implemented by calculating the standard deviation of multiple point scanning by a laser range finder, and is used to evaluate the fitting state of the pressure sensing mat 4 and the ground. The ground hardness refers to the compressive strength of the touch area material, which can be implemented by measuring the Shore hardness value by a hardness tester, and is used to determine whether the ground material is suitable for pressure sensing measurement. The visibility detection value refers to the data output of the same visibility measurement module shared by the binocular camera 2 and the laser radar 3, which can be implemented by the central processor to synchronize the distribution of environmental parameter data, and is used to eliminate the system error of different sensors in detecting the same environmental parameter.

[0058] Specifically, binocular camera 2 is configured with detection parameters for light intensity, visibility, and heel recognition accuracy. Light intensity directly affects image acquisition quality, visibility reflects the impact of ambient light transmittance on visual recognition, and heel recognition accuracy quantifies a key performance indicator for motion capture. These three parameters are combined to form a multi-dimensional reliability assessment benchmark for the visual system. LiDAR 3 uses visibility, point cloud density, and noise ratio as detection parameters. Visibility shares data with binocular camera 2 to prevent detection bias, point cloud density reflects scan data integrity, and noise ratio indicates data reliability, forming a point cloud quality assessment system. Pressure sensor pad 4 uses effective contact area percentage, ground flatness, and ground hardness as detection parameters. The effective contact area percentage determines device deployment rationality, ground flatness measures device-ground contact status, and ground hardness assesses material suitability. These three parameters work together to identify failure scenarios on soft surfaces. Binocular camera 2 and LiDAR 3 share visibility parameters. By using a unified environmental detection benchmark, these parameters eliminate measurement discrepancies between different sensors for the same parameter, ensuring cross-comparability in reliability assessments. When it is detected that the effective contact area ratio is lower than the threshold, the ground flatness exceeds the tolerance range, or the ground hardness does not meet the standard, it is determined that the pressure sensing pad 4 is in an invalid environment such as a sand pit and the shielding mechanism is triggered.

[0059] Compared to existing technologies, the traditional solution relies solely on a single illumination parameter to assess reliability, failing to reflect the impact of visibility changes on stereo vision and lacking a quantitative indicator for motion capture accuracy. The lidar (LiDAR) lacks a correlation assessment model between point cloud density and noise ratio, making it difficult to accurately identify data quality degradation. The pressure-sensing mat (Pressure Sensor) lacks a ground hardness detection mechanism, making it ineffective in identifying special scenarios like sand pits. This solution, through multi-dimensional parameter combination and cross-sensor parameter sharing, enables precise correlation analysis between environmental interference factors and sensor performance degradation, resolving the issue of inconsistent reliability assessment benchmarks across multiple sensors.

[0060] Through the above technical solutions, this application has constructed an environmental parameter detection system for multiple sensors, and achieved standardization of the input parameters of the reliability assessment model by selecting key indicators that are strongly correlated with the physical characteristics of each sensor. Among them, the visibility parameter sharing mechanism eliminates the system error of cross-sensor environmental detection, the ground hardness detection parameters of the pressure sensing pad 4 provide a direct criterion for sand pit scene recognition, and the heel recognition accuracy quantitative index makes up for the lack of reliability assessment of motion capture in traditional visual systems. This solution completes the environmental adaptability pre-judgment before multi-sensor data fusion, provides accurate reliability input data for the dynamic weight allocation model, and avoids the interference of failed sensor data on the final measurement results from the source.

[0061] Preferably, the specific method of constructing a binocular camera reliability model based on binocular camera environment detection information and outputting a binocular camera reliability evaluation coefficient is:

[0062] Ratio the current visibility and heel recognition accuracy with the corresponding standard values ​​to obtain the visibility index and heel recognition accuracy index;

[0063] Importing light intensity into the formula Output light intensity index, where I light Indicates the current light intensity, I0 indicates the reference light intensity, I b It represents the maximum allowable deviation from the light intensity, and k represents the light intensity sensitivity coefficient;

[0064] Import the visibility index, light intensity index, and heel recognition accuracy index into the constructed binocular camera reliability model and output the binocular camera reliability evaluation coefficient;

[0065] The binocular camera reliability model is expressed as:

[0066]

[0067] Among them, R cam Represents the reliability evaluation coefficient of the binocular camera, f(I light ) represents the light intensity index, g(V vist ) represents the visibility index, A heel represents the heel recognition accuracy index, C norm Represents the theoretical maximum reliability evaluation coefficient.

[0068] Compared to existing technologies, traditional methods typically use only a single threshold to determine camera usability. For example, they directly deem a camera inoperable when light intensity falls below a fixed value. Existing technologies fail to account for nonlinear relationships between multiple factors. For example, they may erroneously assign high weights even in moderate light conditions with extremely low visibility. Existing technologies also lack a quantitative evaluation mechanism for heel recognition accuracy, making it impossible to dynamically reflect the impact of fluctuations in detection algorithm performance on reliability.

[0069] Through the above-mentioned technical solution, this application solves the problem of inaccurate reliability assessment of the binocular camera 2 in complex environments. Through multi-parameter fusion calculation and nonlinear mapping, it avoids misjudgments caused by sudden changes in a single environmental factor. For example, at dusk, when light intensity decreases but visibility remains acceptable, the synergistic effect of the visibility index and the light intensity index can accurately reflect the camera's usability, rather than directly disabling the camera. When ground reflections cause a temporary decrease in heel recognition accuracy, the index proportionally reduces the reliability evaluation coefficient, prompting the system to automatically reduce the weight of the camera data, preventing erroneous detection results from affecting the final distance calculation.

[0070] Preferably, the steps of constructing a lidar reliability model based on lidar detection environment information and outputting a lidar reliability evaluation coefficient are as follows:

[0071] The visibility and the point cloud density are respectively processed by ratio processing corresponding to the best value, and then the visibility index and the point cloud density index are obtained;

[0072] The visibility index, the point cloud density index and the noise ratio are introduced into the constructed laser radar reliability model to output a laser radar reliability evaluation coefficient;

[0073] The laser radar reliability model is represented as:

[0074] R lidar =D ind (1-N ind )V ind

[0075] Wherein, R lidar represents the laser radar reliability evaluation coefficient, D ind represents the point cloud density index, N ind represents the noise ratio, and V ind represents the visibility index.

[0076] Through the above technical scheme, the performance of the laser radar 3 in the complex environment can be quantified in real time, the visibility, the point cloud density and the noise ratio are used as detection parameters, the point cloud density reflects the integrity of the scanning data, the noise ratio represents the data credibility, a point cloud quality evaluation system is formed, and the misjudgment of a single threshold is avoided.

[0077] Preferably, the step of constructing a pressure-sensitive mat reliability model based on the pressure-sensitive mat detection environment information to output a pressure-sensitive mat reliability evaluation coefficient is:

[0078] The standard flatness and the current ground flatness are processed by ratio processing to obtain a flatness index, and the current ground hardness and the ideal ground hardness are processed by ratio processing to obtain a ground hardness index;

[0079] The effective touch area proportion, the flatness index and the ground hardness index are introduced into the constructed pressure-sensitive reliability model to output a pressure-sensitive mat reliability evaluation coefficient;

[0080] The pressure-sensitive mat reliability model is represented as:

[0081] R pad =S act (1-F ina )H ind

[0082] Wherein, R pad represents the pressure-sensitive mat reliability evaluation coefficient, S act represents the effective touch area proportion, F ind represents the flatness index, and H indIndicates the ground hardness index. If the long jump site is a sand pit, R pad =0.

[0083] Specifically, the effect of actual field flatness on the contact area is quantified by processing the ratio of ground flatness to the standard value. For example, when the ratio is greater than 1, it indicates that the flatness is below the standard, resulting in a reduction in the effective touchdown area. The ratio of ground hardness to the ideal value is used to evaluate whether the material hardness is suitable for the mechanical detection threshold of the pressure sensing pad 4. For example, the hardness ratio of a sand pit is close to zero, indicating that it cannot meet the detection requirements. The calculation of the effective touchdown area ratio further limits the range of the currently available detection area. The three are multiplied together to generate a reliability coefficient. When the coefficient is lower than the preset threshold, it indicates that the detection result of the pressure sensing pad 4 is unreliable. In particular, when a sand pit is detected, the coefficient is directly reset to zero, thereby completely excluding the pressure sensing pad 4 data in the subsequent dynamic weight distribution, and avoiding interference with the measurement results caused by the touchdown area drift caused by the flow of sand.

[0084] Compared to existing technologies, traditional pressure-sensing mats (4) fail completely in challenging environments like sand pits due to unstable or insufficiently hard ground contact areas, resulting in erroneous measurement data that cannot be corrected. This solution dynamically assesses the reliability of the pressure-sensing mat (4) by integrating multi-dimensional parameters such as ground flatness, hardness, and effective ground contact area. It also forcibly blocks its data in sand pit scenarios to prevent erroneous data from participating in the fusion calculations.

[0085] Through the above technical solution, the present application solves the problem of measurement failure caused by the unstable contact area of ​​the pressure sensing pad 4 in a sand pit or soft ground. Through dynamic reliability evaluation and scene adaptation mechanism, invalid sensor data is automatically identified and eliminated in complex environments, thereby improving the overall accuracy of the standing long jump test results.

[0086] Preferably, the dynamic weight allocation model is expressed as:

[0087]

[0088] Among them, w cam represents the weight of the detection result of binocular camera 2 in the pre-built distance determination model, w lidar represents the weight of the LiDAR 3 detection result in the pre-built distance determination model, w pad represents the weight of the pressure sensing pad 4 detection result in the pre-built distance determination model, R can Represents the reliability evaluation coefficient of the binocular camera, R lidar Represents the reliability evaluation coefficient of the laser radar, R pad Represents the reliability evaluation coefficient of the pressure sensing pad.

[0089] The reliability evaluation coefficient is a quantitative parameter calculated from environmental information detected by sensors, reflecting the sensor's reliability in the current environment. The dynamic weight allocation model is an algorithm that squares and normalizes the reliability evaluation coefficients of each sensor. This can be achieved by calculating the ratio of the squared value of each coefficient to the total sum of squares in real time. This is used to amplify the decision weight of high-reliability sensors and suppress the influence of low-reliability sensors. Squaring is a nonlinear process that quadratically calculates the reliability evaluation coefficients. This can be implemented through a mathematical operation module, enhancing the advantages of high-reliability sensors and mitigating the interference of low-reliability sensors. Normalization limits the weight coefficients to a sum of 1. This can be achieved through division or a normalization algorithm to ensure the mathematical stability of the distance determination model.

[0090] Specifically, when the reliability evaluation coefficient of binocular camera 2 decreases due to insufficient lighting, its squared value's contribution to the total sum of squares decreases significantly, reducing its weight in the final distance calculation. If the reliability evaluation coefficient of pressure sensor mat 4 is zero due to a bunker failure, its squared value contributes to zero, completely eliminating its weight coefficient. LiDAR 3 has a higher reliability evaluation coefficient when the point cloud density is sufficient and the noise ratio is low, and the squared operation gives it a dominant weight. By calculating the squared value of each sensor's reliability evaluation coefficient in real time and normalizing it, the model can dynamically adjust the weight distribution, prioritizing data from sensors with strong environmental adaptability while automatically blocking failed or interfered sensors.

[0091] Compared to existing technologies, traditional methods use fixed weights or linearly weighted data fusion. For example, assigning a fixed weight to pressure sensor pad 4 can still affect measurement results in a sand pit scenario. Alternatively, they directly assign weights linearly based on reliability coefficients, failing to effectively distinguish between high-reliability and low-reliability sensors. This solution strengthens the decision weights of high-reliability sensors through squaring operations, achieving more reasonable weight allocation when sensor performance varies significantly. Furthermore, normalization prevents model crashes caused by single sensor failures.

[0092] Through the above technical solution, this application solves the problem of unreasonable weight distribution caused by reliability differences of multiple sensors in dynamic environments. For example, in the sand pit scene, the pressure sensor pad 4 data is automatically excluded, and in the low-light scene, the binocular camera 2 weight is reduced and the lidar 3 weight is increased, thereby reducing the impact of environmental interference on the measurement results and improving the accuracy and environmental adaptability of the standing long jump test.

[0093] Preferably, the distance determination model is expressed as:

[0094] d final =w cam d camw lidar d lidar w pad d pad

[0095] wherein d final represents the final detection distance, d cam represents the binocular camera 2 detection distance, d lidar represents the laser radar 3 detection distance, d pad represents the pressure-sensitive pad 4 detection distance, w cam , w lidar , w pad represents the weight coefficient and w cam w lidar w pad = 1.

[0096] Referring to Figure 2 , an automatic measurement standing long jump test device using the standing long jump test method described above, comprising a box body 1, a binocular camera 2 and a laser radar 3 and a pressure-sensitive pad 4 installed on the box body 1.

[0097] It should be noted that in this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0098] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, alternatives, and variations can be made in the embodiments without departing from the spirit and scope of the present application as defined by the appended claims and their equivalents.

Claims

1. A standing long jump test method, characterized in that: The following steps are involved: Obtaining binocular camera detection environment information, lidar detection environment information, and pressure sensing pad detection environment information; Based on the binocular camera environment detection information, a binocular camera reliability model is constructed to output the binocular camera reliability evaluation coefficient; Based on the lidar detection environment information, a lidar reliability model is constructed to output the lidar reliability evaluation coefficient; Based on the pressure sensing pad detection environment information, a pressure sensing pad reliability model is constructed to output the pressure sensing pad reliability evaluation coefficient; Based on the reliability evaluation coefficients of the binocular camera, the laser radar, and the pressure sensing pad, a dynamic weight distribution model is constructed to output the weights of the detection results of the binocular camera, the laser radar, and the pressure sensing pad in the pre-built distance determination model; The detection results of the binocular camera, lidar, and pressure sensing pad, as well as their respective weights in the distance determination model, are input into the distance determination model to output the final detection distance.

2. The standing long jump test method according to claim 1, wherein: The distance determination model is expressed as: d final =w cam d cam +w lidar d lidar +w pad d pad Among them, d final Indicates the final detection distance, d cam Indicates the detection distance of the binocular camera, d lidar Represents the laser radar detection distance, d pad Indicates the detection distance of the pressure sensor pad, w cam 、w lidar 、w pad represents the weight coefficient and w cam +w lidar +w pad =1.

3. The standing long jump test method according to claim 2, wherein: The dynamic weight allocation model is expressed as: Among them, w cam Represents the weight of the binocular camera detection result in the pre-built distance determination model, w lidar represents the weight of the lidar detection result in the pre-built distance determination model, w pad represents the weight of the pressure sensing pad detection result in the pre-built distance determination model, R cam Represents the reliability evaluation coefficient of the binocular camera, R lidar Represents the reliability evaluation coefficient of the laser radar, R pad Represents the reliability evaluation coefficient of the pressure sensing pad.

4. The standing long jump test method according to claim 3, wherein: The binocular camera detection environment information includes light intensity, visibility and heel recognition accuracy; the laser radar detection environment information includes visibility, point cloud density and noise ratio; the pressure sensing pad detection environment information includes the effective contact area ratio, ground flatness and ground hardness. The visibility in the binocular camera environment detection information and the laser radar detection environment information is the same visibility.

5. The standing long jump test method according to claim 3, wherein: The specific method of constructing a binocular camera reliability model based on binocular camera environment detection information and outputting the binocular camera reliability evaluation coefficient is as follows: Ratio the current visibility and heel recognition accuracy with the corresponding standard values ​​to obtain the visibility index and heel recognition accuracy index; Importing light intensity into the formula Output light intensity index, where I light Indicates the current light intensity, I0 indicates the reference light intensity, I b It represents the maximum allowable deviation from the light intensity, and k represents the light intensity sensitivity coefficient; Import the visibility index, light intensity index, and heel recognition accuracy index into the constructed binocular camera reliability model and output the binocular camera reliability evaluation coefficient; The binocular camera reliability model is expressed as: Among them, R cam Represents the reliability evaluation coefficient of the binocular camera, f(I light ) represents the light intensity index, g(V vist ) represents the visibility index, A heel represents the heel recognition accuracy index, C norm Represents the theoretical maximum reliability evaluation coefficient.

6. The standing long jump test method according to claim 3, wherein: The steps for constructing a lidar reliability model based on lidar detection environment information and outputting the lidar reliability evaluation coefficient are as follows: The visibility and point cloud density are compared with the corresponding optimal values ​​to obtain the visibility index and point cloud density index; The visibility index, point cloud density index and noise ratio are imported into the constructed lidar reliability model to output the lidar reliability evaluation coefficient; The LiDAR reliability model is expressed as: R lidar =D ind (1-N ind )V ind Among them, R lidar Denotes the reliability evaluation coefficient of the laser radar, D ind Represents the point cloud density index, N ind Indicates the noise ratio, V ind Represents the visibility index.

7. The standing long jump test method according to claim 3, wherein: The steps for constructing a pressure sensing pad reliability model based on the pressure sensing pad detection environment information and outputting the pressure sensing pad reliability evaluation coefficient are as follows: The flatness index is obtained by comparing the standard flatness with the current ground flatness, and the ground hardness index is obtained by comparing the current ground hardness with the ideal ground hardness; The effective ground contact area ratio, flatness index and ground hardness index are imported into the constructed pressure sensing reliability model to output the reliability evaluation coefficient of the pressure sensing pad; The reliability model of the pressure sensing pad is expressed as: R pad =S act (1-F ind )H ind Among them, R pad Represents the reliability evaluation coefficient of the pressure sensing pad, S act Indicates the effective contact area ratio, F ind Indicates the flatness index, H ind Indicates the ground hardness index.

8. The standing long jump test method according to claim 7, wherein: If the long jump site is a sand pit, then R pad =0.

9. A standing long jump test device capable of automatic measurement, using the standing long jump test method according to claims 1-8, characterized in that: It includes a box body, a binocular camera installed on the box body, a laser radar and a pressure sensing pad.

Images