An AI-based basketball detection method, system and device

Abstract

The present invention discloses a basketball detection method, system and device based on artificial intelligence, belonging to the technical field of basketball recognition. The method includes the following steps: obtaining site environment data, device performance data and basketball image data; comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjusting the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; comprehensively analyzing the device performance data to obtain the device performance index at each monitoring time point, comprehensively analyzing the corrected basketball image at each monitoring time point, the device performance index at each monitoring time point and the site environment evaluation value at each monitoring time point to obtain the image accuracy evaluation value at each monitoring time point, and performing adjustment feedback according to the image accuracy evaluation value, so as to improve the image calibration accuracy and detection accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of basketball recognition, and particularly to a basketball detection method, system and device based on artificial intelligence. Background Art

[0002] The principle of artificial intelligence is the technology and theory that simulate human intelligence, enabling machines to think, make decisions and act autonomously, and achieve intelligent behaviors like humans. Artificial intelligence has experienced multiple development waves, from the initial symbolicism to the subsequent connectionism, and then to the current deep learning, with each step marking the continuous progress of artificial intelligence technology.

[0003] Existing basketball motion trajectory detection methods are mainly achieved through various technical means such as GPS technology, camera capture technology, sensor technology, and data analysis and visualization technology. At the same time, there are certain defects in aspects such as signal reception, occlusion, imaging quality, data processing complexity, privacy protection, sensor self-defects, data fusion, wearing comfort, and algorithm limitations.

[0004] For example, the ball target tracking method and system based on re-detection disclosed in the patent application with the publication number: CN112446333A includes: initializing the constructed tracker, validator and detector to obtain a position blacklist; obtaining a detection image and inputting it into the tracker, and using a target tracking algorithm to identify and output the predicted position of the target ball; transmitting the output of the tracker to the validator, and using a visual tracking algorithm to score the output of the tracker. If the score is qualified, the detection of the next frame of image is executed, otherwise the next step is executed; using a target detection algorithm for global detection, and obtaining the position of the target ball after filtering the blacklist positions from the global detection results.

[0005] For example, an algorithm for tracking a basketball in a ball game video in panoramic mode disclosed in the invention patent announcement with the announcement number: CN106991359B includes: 1) generating a scene annotation; 2) detecting candidate balls; 3) generating short trajectories; 4) correcting and filtering short trajectories; 5) connecting short trajectories into long trajectories. The algorithm for tracking a basketball in a ball game video in panoramic mode provided by the present invention utilizes some prior information of the basketball court in the basketball movement (such as the position of the basketball court sports field, the position of the basketball hoop, the average height of players, etc.), and several main movement forms of the basketball (such as holding the ball, dribbling, passing, shooting), extracts the corresponding movement characteristics, and adds these movement characteristics to the tracking of the basketball.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, it is found that the above technologies have at least the following technical problems:

[0007] In the prior art, when detecting the basketball movement trajectory outdoors, due to the poor image calibration accuracy, there is a problem of low accuracy in detecting the basketball in the basketball image. Summary of the Invention

[0008] The present invention provides a basketball detection method, system and device based on artificial intelligence, which solves the problem of low accuracy in detecting the basketball when detecting the basketball movement trajectory outdoors in the prior art due to the poor image calibration accuracy, and realizes the improvement of the image calibration accuracy and detection accuracy.

[0009] The present invention provides a basketball detection method based on artificial intelligence, including the following steps: obtaining site environment data, device performance data and basketball image data; comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjusting the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; comprehensively analyzing the device performance data to obtain the device performance index at each monitoring time point, comprehensively analyzing the corrected basketball image, the device performance index and the site environment evaluation value at each monitoring time point to obtain the image accuracy evaluation value at each monitoring time point, and performing adjustment feedback according to the image accuracy evaluation value.

[0010] Further, the step of comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point includes: the site environment data includes light intensity, illuminance data and light coverage rate, wherein the illuminance data includes vertical illuminance and horizontal illuminance; monitoring the vertical illuminance and horizontal illuminance at each position point, and obtaining the reference light intensity, allowable deviation light intensity, allowable deviation vertical illuminance, allowable deviation horizontal illuminance and critical light coverage rate from the basketball detection database; comprehensively analyzing to obtain the site environment evaluation value at each monitoring time point.

[0011] Further, the step of adjusting the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point includes: the basketball image data includes the basketball image at each monitoring time point; obtaining the site environment evaluation threshold from the basketball detection database, comparing the site environment evaluation value at each monitoring time point with the site environment evaluation threshold, if the site environment evaluation value at a certain monitoring time point is less than the site environment evaluation threshold, then subtracting the site environment evaluation value from the site environment evaluation threshold to obtain the deviation site environment evaluation value, and adjusting the basketball image at the monitoring time point according to the deviation site environment evaluation value to obtain the corrected basketball image at the monitoring time point; if the site environment evaluation value at a certain monitoring time point is greater than or equal to the site environment evaluation threshold, no additional processing is performed; statistically obtaining the corrected basketball image at each monitoring time point.

[0012] Further, the steps of adjusting the basketball image at the monitoring time point according to the deviation site environment evaluation value include: matching the deviation site environment evaluation value with the contrast adjustment amplitude levels corresponding to the preset deviation site environment evaluation values in the basketball detection database to obtain the contrast adjustment amplitude level of the basketball image, and adjusting the contrast of the basketball image according to the contrast adjustment amplitude level of the basketball image.

[0013] Further, the steps of comprehensively analyzing the device performance data to obtain the device performance index at each monitoring time point include: the device performance data including battery power, resolution, and frame rate; obtaining the critical battery power, critical resolution, reference frame rate, and allowable deviation frame rate from the basketball detection database; and comprehensively analyzing to obtain the device performance index at each monitoring time point.

[0014] Further, the steps of comprehensively analyzing the corrected basketball images at each monitoring time point, the device performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point to obtain the image accuracy evaluation value at each monitoring time point include: extracting the basketball position, basketball movement angle, and basketball movement speed at each monitoring time point from the corrected basketball images at each monitoring time point, and jointly marking them as the basketball movement data at each monitoring time point; comparing the basketball movement data at each monitoring time point with the basketball movement data at the adjacent previous monitoring time point respectively to obtain the position change parameter, movement angle change parameter, and movement speed change parameter at each monitoring time point; matching the reference position change parameter and reference movement angle change parameter at each monitoring time point according to the basketball movement speed at each monitoring time point, and obtaining the allowable deviation position change parameter, allowable deviation movement angle change parameter, allowable deviation movement speed change parameter, reference movement speed change parameter, critical device performance index, and critical site environment evaluation value from the basketball detection database; and comprehensively analyzing to obtain the image accuracy evaluation value at each monitoring time point.

[0015] Further, the acquisition method of the image accuracy evaluation value at each monitoring time point is as follows:

[0016] ;

[0017] wherein, represents the image accuracy evaluation value at the i-th monitoring time point, represents the image accuracy evaluation influence factor corresponding to the basketball position, represents the image accuracy evaluation influence factor corresponding to the basketball movement angle, represents the image accuracy evaluation influence factor corresponding to the basketball movement speed, represents the image accuracy evaluation influence factor corresponding to the device performance index, Represents the influencing factor of image accuracy evaluation corresponding to the site environment evaluation value, Represents the position change parameter at the i-th monitoring time point, Represents the reference position change parameter, Represents the allowable deviation position change parameter, Represents the motion angle change parameter at the i-th monitoring time point, Represents the reference motion angle change parameter, Represents the allowable deviation motion angle change parameter, Represents the motion speed change parameter at the i-th monitoring time point, Represents the reference motion speed change parameter, Represents the allowable deviation motion speed change parameter, Represents the equipment performance index at the i-th monitoring time point, Represents the critical equipment performance index, Represents the site environment evaluation value at the i-th monitoring time point, Represents the critical site environment evaluation value, where i is the number of each monitoring time point, i = 1, 2, 3,..., N, and N is the total number of monitoring time points.

[0018] Furthermore, the steps of adjusting and feedback according to the image accuracy evaluation value include: obtaining the image accuracy evaluation threshold from the basketball detection database; comparing the image accuracy evaluation value with the image accuracy evaluation threshold. If the image accuracy evaluation value is less than the image accuracy evaluation threshold, further perform image correction processing on the corrected basketball image. If the image accuracy evaluation value is greater than or equal to the image accuracy evaluation threshold, no additional processing is performed.

[0019] The embodiment of the present application provides an artificial intelligence-based basketball detection system, including a data acquisition module, an image correction module, an image adjustment module, and a basketball detection database; wherein, the data acquisition module is used to acquire site environment data, equipment performance data, and basketball image data; the image correction module is used to comprehensively analyze the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjust the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; the image adjustment module is used to comprehensively analyze the equipment performance data to obtain the equipment performance index at each monitoring time point, comprehensively analyze the image accuracy evaluation value at each monitoring time point according to the corrected basketball image at each monitoring time point, the equipment performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point, and perform adjustment and feedback according to the image accuracy evaluation value.

[0020] An embodiment of the present application provides an electronic device for basketball detection based on artificial intelligence, including: a processor and a memory for storing executable instructions of the processor; when the processor is configured to execute the instructions, the electronic device implements a basketball detection method based on artificial intelligence.

[0021] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0022] 1. By providing a basketball detection method, system and device based on artificial intelligence, the present invention can comprehensively analyze the venue environment data, device performance data and basketball image data, and perform adjustment feedback according to the evaluation results, thereby effectively improving the accuracy and reliability of basketball detection.

[0023] 2. By accurately adjusting the basketball image data according to the venue environment evaluation values at each monitoring time point, the present invention effectively compensates for the negative impact of venue light changes on the quality of basketball images, and thus realizes the correction and optimization of basketball images at each monitoring time point, providing a more reliable basis for subsequent image accuracy evaluation and adjustment feedback.

[0024] 3. By performing adjustment feedback according to the image accuracy evaluation value, the present invention can timely identify and correct errors caused by factors such as venue environment changes and device performance fluctuations during the basketball image detection process, and thus realizes the continuous optimization and improvement of the basketball motion trajectory detection system, providing more accurate data support for basketball games or training. Description of the Drawings

[0025] Figure 1 It is a flowchart of a basketball detection method based on artificial intelligence provided by an embodiment of the present application.

[0026] Figure 2 It is a graph showing the change of the device performance index of the basketball detection method based on artificial intelligence provided by an embodiment of the present application.

[0027] Figure 3 It is a schematic structural diagram of a basketball detection system based on artificial intelligence provided by an embodiment of the present application. Detailed Embodiments

[0028] Embodiments of the present application provide a basketball detection method, system and device based on artificial intelligence, which solve the problem of low accuracy of basketball detection in the prior art when detecting the basketball movement trajectory outdoors due to poor image calibration accuracy. By comprehensively analyzing the site environment data, the site environment evaluation value at each monitoring time point is obtained, and the basketball image data is adjusted according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; the device performance index at each monitoring time point is obtained by comprehensively analyzing the device performance data, and the image accuracy evaluation value at each monitoring time point is obtained by comprehensively analyzing the corrected basketball image at each monitoring time point, the device performance index at each monitoring time point and the site environment evaluation value at each monitoring time point, and adjustment feedback is performed according to the image accuracy evaluation value, so as to improve the image calibration accuracy and detection accuracy.

[0029] The technical solution in the embodiments of the present application aims to solve the problem of low accuracy of basketball detection in the prior art when detecting the basketball movement trajectory outdoors due to poor image calibration accuracy. The general idea is as follows:

[0030] By obtaining site environment data, device performance data and basketball image data; comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjusting the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; comprehensively analyzing the device performance data to obtain the device performance index at each monitoring time point, and comprehensively analyzing the corrected basketball image at each monitoring time point, the device performance index at each monitoring time point and the site environment evaluation value at each monitoring time point to obtain the image accuracy evaluation value at each monitoring time point, and performing adjustment feedback according to the image accuracy evaluation value, so as to achieve comprehensive monitoring and accurate evaluation of the basketball movement scenario.

[0031] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0032] Such as Figure 1As shown in the figure, it is a flowchart of a basketball detection method based on artificial intelligence provided by an embodiment of the present application. The method includes the following steps: obtaining site environment data, device performance data, and basketball image data; comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjusting the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; comprehensively analyzing the device performance data to obtain the device performance index at each monitoring time point, comprehensively analyzing the corrected basketball image at each monitoring time point, the device performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point to obtain the image accuracy evaluation value at each monitoring time point, and performing adjustment feedback according to the image accuracy evaluation value.

[0033] In this embodiment, the present invention can comprehensively obtain various data related to basketball movement, including site environment, device performance, and basketball images, so as to realize the all-round monitoring of the basketball movement scene; at the same time, it can provide accurate and reliable basketball movement data for coaches, referees, athletes, etc., helping them better understand the situation of the game or training, formulate more scientific and reasonable tactics and training plans, and provide strong data support for subsequent decision-making.

[0034] In addition, the basketball detection database is used to store relevant data of the basketball detection method based on artificial intelligence, including: reference light intensity, allowable deviation light intensity, allowable deviation vertical illuminance, allowable deviation horizontal illuminance, critical light coverage rate, site environment evaluation impact factor corresponding to the light intensity, site environment evaluation threshold, and contrast adjustment amplitude level corresponding to each deviation site environment evaluation value, etc. The data in the basketball detection database can be obtained from professional basketball databases or comprehensive sports databases such as BasketballReference and Synergy Sports, or can be obtained through cooperation with basketball clubs, associations or media organizations.

[0035] Further, the step of comprehensively analyzing the site environment data to obtain the site environment evaluation value at each monitoring time point includes: the site environment data includes light intensity, illuminance data, and light coverage rate, where the illuminance data includes vertical illuminance and horizontal illuminance; monitoring the vertical illuminance and horizontal illuminance at each position point, and obtaining the reference light intensity, allowable deviation light intensity, allowable deviation vertical illuminance, allowable deviation horizontal illuminance, and critical light coverage rate from the basketball detection database; comprehensively analyzing to obtain the site environment evaluation value at each monitoring time point.

[0036] Among them, the obtaining method of the site environment evaluation value at each monitoring time point is as follows:

[0037] ;

[0038] In the formula, represents the site environmental assessment value at the i-th monitoring time point, represents the influencing factor of the site environmental assessment corresponding to the light intensity, represents the influencing factor of the site environmental assessment corresponding to the vertical illuminance, represents the influencing factor of the site environmental assessment corresponding to the horizontal illuminance, represents the influencing factor of the site environmental assessment corresponding to the light coverage rate, represents the light intensity at the i-th monitoring time point, represents the reference light intensity, represents the allowable deviation of light intensity, represents the vertical illuminance at the j-th position point at the i-th monitoring time point, represents the allowable deviation of vertical illuminance, represents the horizontal illuminance at the j-th position point at the i-th monitoring time point, represents the allowable deviation of horizontal illuminance, represents the light coverage rate at the i-th monitoring time point, represents the critical light coverage rate, where i is the number of each monitoring time point, i = 1, 2, 3,..., N, N is the total number of monitoring time points, j is the number of each position point, j = 1, 2, 3,..., M, and M is the total number of position points.

[0039] , , and are respectively the influencing factors of the site environmental assessment corresponding to the light intensity, vertical illuminance, horizontal illuminance and light coverage rate preset in the basketball detection database. These influencing factors are numerical indicators for measuring the influence of the above light factors on the site environmental assessment value. Specifically, there is a mapping relation table for each of the light intensity, vertical illuminance, horizontal illuminance and light coverage rate. The table records each possible value of the light factor and its corresponding influencing factor of the site environmental assessment. These mapping relations can be one-to-one or many-to-one. For example, when it is necessary to evaluate the site environmental assessment value at a certain monitoring time point, the measured light intensity, vertical illuminance, horizontal illuminance and light coverage rate can be respectively input into their corresponding mapping relation tables, and the influencing factors of the site environmental assessment corresponding to these values can be quickly found. The value range of the influencing factor is between 0 and 1.

[0040] In this embodiment, the light intensity can be measured by using a lux meter; the illuminance data can be obtained by measuring at each position point with professional equipment such as a lux meter; the light coverage rate can be calculated by measuring the ratio of the area irradiated by the light source to the total area. Among them, the change of the illuminance data will affect the distribution of the light intensity. For example, the central light intensity of a light source with a narrow beam angle is prominent, and the spot range is small, which is suitable for accent lighting; the higher the light intensity, usually the larger the area that the light source can irradiate (when the light source power and type remain unchanged), thereby increasing the light coverage rate; the optimization of the illuminance data can ensure that the light better covers the target area, reduce light waste, and increase the light coverage rate. The lighting evaluation value of the site environment obtained through comprehensive analysis helps to judge whether the lighting in the current image meets the site and image detection requirements, and whether there are problems of insufficient lighting or over-illumination.

[0041] Further, the steps of adjusting the basketball image data according to the site environment evaluation values at each monitoring time point to obtain the corrected basketball images at each monitoring time point include: the basketball image data includes the basketball images at each monitoring time point; obtaining the site environment evaluation threshold from the basketball detection database, comparing the site environment evaluation values at each monitoring time point with the site environment evaluation threshold. If the site environment evaluation value at a certain monitoring time point is less than the site environment evaluation threshold, then subtract the site environment evaluation value at this monitoring time point from the site environment evaluation threshold to obtain the deviation site environment evaluation value, and adjust the basketball image at this monitoring time point according to the deviation site environment evaluation value to obtain the corrected basketball image at this monitoring time point; if the site environment evaluation value at a certain monitoring time point is greater than or equal to the site environment evaluation threshold, no additional processing is performed; and the corrected basketball images at each monitoring time point are obtained through statistics.

[0042] In this embodiment, the site environment evaluation threshold is a standard value used to judge whether the site environment evaluation value reaches or exceeds the degree that significantly affects the quality of the basketball image. By comparing the site environment evaluation value with the threshold and adjusting the basketball image according to the deviation value, the influence of site environment factors on the image quality, such as light changes and shadow interference, can be eliminated or reduced, thereby improving the accuracy and clarity of the image. When the site environment evaluation value is less than the threshold, it is considered that the basketball image needs to be adjusted; when the site environment evaluation value is greater than or equal to the threshold, it is considered that the influence of the site environment on the quality of the basketball image is within an acceptable range and no additional processing is required.

[0043] Further, the steps of adjusting the basketball image at the monitoring time point according to the deviation site environment evaluation value include: matching the deviation site environment evaluation value with the contrast adjustment amplitude level corresponding to each preset deviation site environment evaluation value in the basketball detection database, where the deviation site environment evaluation value and the contrast adjustment amplitude level corresponding to each preset deviation site environment evaluation value in the basketball detection database form a mapping set, inputting the deviation site environment evaluation value into the mapping set to obtain the basketball image contrast adjustment amplitude level corresponding to the deviation site environment evaluation value, and adjusting the contrast of the basketball image according to the basketball image contrast adjustment amplitude level.

[0044] In this embodiment, the deviation site environment evaluation value refers to the difference between the site environment evaluation value and the site environment evaluation threshold, which is used to quantify the specific impact degree of the site environment on the quality of basketball images. When adjusting the contrast of basketball images according to the contrast adjustment amplitude level of basketball images, the contrast adjustment amplitude level of basketball images includes a first adjustment level, a second adjustment level, and a third adjustment level. Among them, when in the first adjustment level, it indicates that the environment indicated by the deviation site environment evaluation value has a small impact on the contrast of basketball images, but a certain degree of adjustment is still required to ensure the image quality. Contrast enhancement methods such as power-law transformation can be used, but a small transformation intensity should be maintained during application, slightly increasing or decreasing the contrast of the image to fine-tune the difference between bright and dark areas in the image. For example, for power-law transformation, the fluctuation range of its exponent is [0.95, 1.05]; when in the second adjustment level, it indicates that the environment has a more significant impact on the contrast of basketball images and a larger adjustment amplitude is required. For power-law transformation, the fluctuation range of its exponent is [0.8, 1.2]. If there are over-bright or over-dark areas in the image, the level adjustment method can be used to stretch the brightness of these areas to a more appropriate range; when in the third adjustment level, it indicates that the environment has a very large impact on the contrast of basketball images and a large adjustment is required to improve the image quality. For power-law transformation, the fluctuation range of its exponent is [0.6, 1.4]. If there is serious noise or distortion in the image, noise removal or image smoothing processing can be performed before or after contrast adjustment. For example, the value range of the deviation site environment evaluation value in the first adjustment level is (0, 0.1], the value range of the deviation site environment evaluation value in the second adjustment level is (0.1, 0.2], and the value range of the deviation site environment evaluation value in the third adjustment level is greater than 0.2. The deviation site environment evaluation value at a certain monitoring time point is 0.15. Therefore, the contrast adjustment amplitude level of the basketball image corresponding to this deviation site environment evaluation value is the second adjustment level. Currently, by adjusting the contrast of basketball images according to the deviation site environment evaluation value, the image quality can be significantly improved, making the image clearer and more vivid. This helps coaches, athletes, or audiences better observe and analyze the game situation; the corrected basketball images have higher data availability and can be used for more accurate game analysis, tactical formulation, and athlete training evaluation.

[0045] Further, the steps of comprehensively analyzing the equipment performance data to obtain the equipment performance index at each monitoring time point include: the equipment performance data includes battery power, resolution, and frame rate; obtaining the critical battery power, critical resolution, reference frame rate, and allowable deviation frame rate from the basketball detection database; and comprehensively analyzing to obtain the equipment performance index at each monitoring time point.

[0046] Among them, the acquisition method of the equipment performance index at each monitoring time point is as follows:

[0047] ;

[0048] wherein, represents the device performance index at the i-th monitoring time point, represents the influencing factor of the device performance index corresponding to the battery power, represents the influencing factor of the device performance index corresponding to the resolution, represents the influencing factor of the device performance index corresponding to the frame rate, represents the battery power at the i-th monitoring time point, represents the critical battery power, represents the resolution at the i-th monitoring time point, represents the critical resolution, represents the frame rate at the i-th monitoring time point, represents the reference frame rate, represents the allowable deviation frame rate.

[0049] , and are respectively the influencing factors of the device performance index corresponding to the battery power, resolution, and frame rate preset in the basketball detection database. These influencing factors are numerical indicators for measuring the influence of the above parameters on the device performance index. Specifically, there is a mapping relation table for each of the battery power, resolution, and frame rate. The table records each possible parameter value and its corresponding influencing factor of the device performance index. These mapping relations can be one-to-one or many-to-one. In practical applications, when it is necessary to evaluate the device performance index at a certain monitoring time point, the measured battery power, resolution, and frame rate can be respectively input into their corresponding mapping relation tables, and the influencing factors of the device performance index corresponding to these values can be quickly found. The value range of the influencing factor is from 0 to 1.

[0050] In this embodiment, the battery power can be obtained through the battery status indicator built in the device or a third-party battery management software; the resolution can be directly queried using tools such as ImageMagick and TinyPNG; the frame rate can be measured through professional performance testing software such as Fraps. There is an indirect association between the battery power and the resolution and frame rate. High resolution and high frame rate usually require more electrical energy to drive, so they will cause greater consumption of the battery power. By comprehensively analyzing the obtained device performance index, the performance of the device can be comprehensively evaluated, and suggestions for performance optimization can be put forward, such as reducing the resolution or frame rate to extend the battery life, or upgrading the hardware configuration to improve the performance.

[0051] Set the device performance index impact factor corresponding to the battery power to 0.3, the device performance index impact factor corresponding to the resolution to 0.3, the device performance index impact factor corresponding to the frame rate to 0.4, the battery power to 80%, the critical battery power to 20%, the resolution to 1080P, the critical resolution to 1080P, and the reference frame rate to 30 The allowable deviation frame rate is set to 5 When the frame rate is continuously increasing, calculate the device performance index. The device performance index data table of the basketball detection method based on artificial intelligence is shown in Table 1

[0052] Table 1 Device performance index data table of the basketball detection method based on artificial intelligence

[0053]

[0054] As Figure 2 shown, it is the device performance index change diagram of the basketball detection method based on artificial intelligence provided by the embodiment of the present application. As shown in Table 1 and Figure 2 It can be seen that when the device performance index impact factor corresponding to the battery power, the device performance index impact factor corresponding to the resolution, the device performance index impact factor corresponding to the frame rate, the battery power, the critical battery power, the resolution, the critical resolution, the reference frame rate, and the allowable deviation frame rate remain unchanged and the frame rate is continuously increasing, the closer the frame rate is to the reference frame rate, the greater the device performance index, and vice versa

[0055] Furthermore, the steps of comprehensively analyzing the image accuracy evaluation value of each monitoring time point according to the corrected basketball images of each monitoring time point, the device performance index of each monitoring time point, and the site environment evaluation value of each monitoring time point include: extracting the basketball position, basketball movement angle, and basketball movement speed of each monitoring time point from the corrected basketball images of each monitoring time point, and jointly marking them as the basketball movement data of each monitoring time point; comparing the basketball movement data of each monitoring time point with the basketball movement data of the adjacent previous monitoring time point respectively to obtain the position change parameter, movement angle change parameter, and movement speed change parameter of each monitoring time point; matching the reference position change parameter and reference movement angle change parameter of each monitoring time point according to the basketball movement speed of each monitoring time point, and obtaining the allowable deviation position change parameter, allowable deviation movement angle change parameter, allowable deviation movement speed change parameter, reference movement speed change parameter, critical device performance index, and critical site environment evaluation value from the basketball detection database; comprehensively analyzing to obtain the image accuracy evaluation value of each monitoring time point

[0056] In this embodiment, the basketball position is a three-dimensional vector. The center point of the basketball court can be set as the origin, the direction perpendicular to the ground is set as the z-axis, the game dividing line of the basketball court is set as the x-axis, and the line perpendicular to the x-axis on the ground is set as the y-axis. The center point of the basketball is regarded as the basketball position point; the basketball movement angle can be obtained by installing a sensor on the basketball to extract the basketball tilt angle information; the basketball movement speed can be directly extracted from the sensor data. The position change parameter at each monitoring time point can be obtained by extracting the straight-line distance between the basketball position points at each monitoring time point and the basketball position point at the previous adjacent monitoring time point; the movement angle change parameter and the movement speed change parameter at each monitoring time point are obtained by directly taking the difference between the basketball movement angle and the basketball movement speed at a certain monitoring time point and the basketball movement angle and the basketball movement speed at the previous adjacent monitoring time point, and then taking the absolute value. The corrected basketball image is obtained by adjusting the contrast in the original basketball image to eliminate or reduce the influence of the site environment and equipment performance on the image quality; the equipment performance index reflects the performance status of the shooting equipment, and the quality of the equipment performance directly affects the clarity and stability of the captured image; the site environment evaluation value is a comprehensive evaluation of the site light, and these environmental factors will also affect the quality of the captured image. The image accuracy evaluation value obtained through comprehensive analysis is a quantitative representation of the accuracy of the basketball image at each monitoring time point, reflecting the reliability and accuracy of the basketball movement data in the image. Based on the accurate image data, the movement characteristics of the basketball, such as the movement trajectory and speed change, can be further analyzed, providing strong support for game analysis, tactical formulation, etc.

[0057] Furthermore, the acquisition method of the image accuracy evaluation value at each monitoring time point is as follows:

[0058] ;

[0059] where, represents the image accuracy evaluation value at the i-th monitoring time point, represents the image accuracy evaluation influence factor corresponding to the basketball position, represents the image accuracy evaluation influence factor corresponding to the basketball movement angle, represents the image accuracy evaluation influence factor corresponding to the basketball movement speed, represents the image accuracy evaluation influence factor corresponding to the equipment performance index, represents the image accuracy evaluation influence factor corresponding to the site environment evaluation value, represents the position change parameter at the i-th monitoring time point, represents the reference position change parameter, represents the allowable deviation position change parameter, represents the movement angle change parameter at the i-th monitoring time point, Represents the reference motion angle variation parameter, Represents the allowable deviation motion angle variation parameter, Represents the motion speed variation parameter at the i-th monitoring time point, Represents the reference motion speed variation parameter, Represents the allowable deviation motion speed variation parameter, Represents the equipment performance index at the i-th monitoring time point, Represents the critical equipment performance index, Represents the site environment evaluation value at the i-th monitoring time point, Represents the critical site environment evaluation value.

[0060] In this embodiment, 、 、 、 and are respectively the image accuracy evaluation influence factors corresponding to the preset basketball position, basketball motion angle, basketball motion speed, equipment performance index and site environment evaluation value in the basketball detection database. These influence factors are numerical indicators that measure the influence of the above parameters on the image accuracy evaluation value. Specifically, there is a mapping relationship table for each of the basketball position, basketball motion angle, basketball motion speed, equipment performance index and site environment evaluation value. The table records each possible parameter value and its corresponding image accuracy evaluation influence factor. These mapping relationships can be one-to-one or many-to-one. In practical applications, when it is necessary to evaluate the image accuracy evaluation value at a certain monitoring time point, the measured basketball position, basketball motion angle, basketball motion speed, equipment performance index and site environment evaluation value can be respectively input into their corresponding mapping relationship tables, and the image accuracy evaluation influence factors corresponding to these values can be quickly found. The value range of the influence factor is between 0 and 1.

[0061] Furthermore, the steps of adjusting and feedback according to the image accuracy evaluation value include: obtaining the image accuracy evaluation threshold from the basketball detection database; comparing the image accuracy evaluation value with the image accuracy evaluation threshold. If the image accuracy evaluation value is less than the image accuracy evaluation threshold, further perform image correction processing on the corrected basketball image. If the image accuracy evaluation value is greater than or equal to the image accuracy evaluation threshold, no additional processing is performed.

[0062] In this embodiment, the image accuracy evaluation threshold is a set value used to determine whether the current image accuracy evaluation value meets the quality standard. If the image accuracy evaluation value is less than the image accuracy evaluation threshold, it indicates that the accuracy of the current image is low, and further image correction processing needs to be performed on the corrected basketball image to improve the image quality. If the image accuracy evaluation value is greater than or equal to the image accuracy evaluation threshold, it indicates that the accuracy of the current image already meets the requirements and no additional processing is required. When performing image correction processing on the corrected basketball image, manual annotation can be introduced to assist in identifying and correcting the position and motion state of the basketball in the image or adjusting the parameters of the target tracking and recognition algorithm, such as the detection threshold, the size of the tracking window, etc., to improve the accuracy and stability of the algorithm.

[0063] As Figure 3 shown, it is a schematic structural diagram of a basketball detection system based on artificial intelligence provided by an embodiment of the present application. The basketball detection system based on artificial intelligence provided by an embodiment of the present application includes: a data acquisition module, an image correction module, an image adjustment module, and a basketball detection database. Among them, the data acquisition module is used to acquire site environment data, device performance data, and basketball image data. The image correction module is used to comprehensively analyze the site environment data to obtain the site environment evaluation value at each monitoring time point, and adjust the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point. The image adjustment module is used to comprehensively analyze the device performance data to obtain the device performance index at each monitoring time point, comprehensively analyze the image accuracy evaluation value at each monitoring time point according to the corrected basketball image at each monitoring time point, the device performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point, and perform adjustment feedback according to the image accuracy evaluation value.

[0064] An embodiment of the present application also provides a basketball detection electronic device based on artificial intelligence, including: a processor and a memory for storing executable instructions of the processor. When the processor is configured to execute the instructions, the electronic device implements a basketball detection method based on artificial intelligence.

[0065] In summary, the embodiments of the present application obtain site environment data, equipment performance data, and basketball image data; comprehensively analyze the site environment data to obtain the site environment evaluation values at each monitoring time point, and adjust the basketball image data according to the site environment evaluation values at each monitoring time point to obtain the corrected basketball images at each monitoring time point; comprehensively analyze the equipment performance data to obtain the equipment performance indices at each monitoring time point, comprehensively analyze the corrected basketball images at each monitoring time point, the equipment performance indices at each monitoring time point, and the site environment evaluation values at each monitoring time point to obtain the image accuracy evaluation values at each monitoring time point, and perform adjustment feedback according to the image accuracy evaluation values, thereby improving the image calibration accuracy and detection accuracy.

[0066] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0067] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the specified functions in Figure 1 one or more of the flows Figure 1 or multiple flows and / or blocks

[0068] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means for implementing the specified functions in Figure 1 one or more of the flows Figure 1 or multiple flows and / or blocks

[0069] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, causing a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 or steps for implementing the functions specified in one block or a plurality of blocks.

[0070] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0071] It is obvious that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A basketball detection method based on artificial intelligence, characterized in that: The following steps are involved: Obtain venue environment data, equipment performance data, and basketball image data; The site environment evaluation value at each monitoring time point is obtained based on the comprehensive analysis of the site environment data, and the basketball image data is adjusted based on the site environment evaluation value at each monitoring time point to obtain a corrected basketball image at each monitoring time point; The equipment performance index at each monitoring time point is obtained through comprehensive analysis of the equipment performance data, and the image accuracy evaluation value at each monitoring time point is obtained through comprehensive analysis of the corrected basketball image at each monitoring time point, the equipment performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point, and adjustment feedback is performed based on the image accuracy evaluation value; The steps of obtaining the image accuracy evaluation value at each monitoring time point by comprehensive analysis based on the corrected basketball image at each monitoring time point, the equipment performance index at each monitoring time point, and the site environment evaluation value at each monitoring time point include: Extract the basketball position, basketball movement angle and basketball movement speed at each monitoring time point from the corrected basketball images at each monitoring time point, and mark them together as basketball movement data at each monitoring time point; The basketball motion data at each monitoring time point are compared with the basketball motion data at the adjacent previous monitoring time point to obtain the position change parameter, motion angle change parameter and motion speed change parameter at each monitoring time point; According to the basketball movement rate matching at each monitoring time point, the reference position change parameter and the reference movement angle change parameter at each monitoring time point are obtained, and the allowable deviation position change parameter, the allowable deviation movement angle change parameter, the allowable deviation movement rate change parameter, the reference movement rate change parameter, the critical equipment performance index and the critical site environment assessment value are obtained from the basketball detection database; Obtaining image accuracy assessment influencing factors corresponding to preset basketball positions, basketball movement angles, basketball movement speeds, equipment performance indexes, and venue environment assessment values ​​from a basketball detection database; The image accuracy assessment value at each monitoring time point is obtained by analyzing the position change parameters at each monitoring time point, the movement angle change parameters at each monitoring time point, the movement rate change parameters at each monitoring time point, the reference position change parameters at each monitoring time point, the reference movement angle change parameters at each monitoring time point, the reference movement rate change parameters, the allowable deviation position change parameters, the allowable deviation movement angle change parameters, the allowable deviation movement rate change parameters, the critical equipment performance index and the critical site environment assessment value, as well as the image accuracy assessment influencing factors corresponding to the preset basketball position, basketball movement angle, basketball movement rate, equipment performance index and site environment assessment values.

2. The basketball detection method based on artificial intelligence as claimed in claim 1, characterized in that: The step of obtaining the site environment assessment value at each monitoring time point based on comprehensive analysis of the site environment data comprises: The site environment data includes light intensity, illumination data and light coverage, wherein the illumination data includes vertical illumination and horizontal illumination; Monitor the vertical and horizontal illumination at each location, and obtain the reference illumination intensity, allowable deviation illumination intensity, allowable deviation vertical illumination, allowable deviation horizontal illumination and critical illumination coverage from the basketball detection database; Comprehensive analysis was performed to obtain the site environmental assessment value at each monitoring time point.

3. The basketball detection method based on artificial intelligence as claimed in claim 1, characterized in that: The step of adjusting the basketball image data according to the field environment evaluation value at each monitoring time point to obtain a corrected basketball image at each monitoring time point comprises: The basketball image data includes basketball images at each monitoring time point; Obtaining a site environment assessment threshold from a basketball detection database, comparing the site environment assessment value at each monitoring time point with the site environment assessment threshold, and if the site environment assessment value at a certain monitoring time point is less than the site environment assessment threshold, subtracting the site environment assessment threshold from the site environment assessment value at the monitoring time point to obtain a deviation site environment assessment value, and adjusting the basketball image at the monitoring time point according to the deviation site environment assessment value to obtain a corrected basketball image at the monitoring time point; If the site environmental assessment value at a certain monitoring time point is greater than or equal to the site environmental assessment threshold, no additional processing is performed; The corrected basketball images at each monitoring time point were obtained by statistics.

4. The basketball detection method based on artificial intelligence as claimed in claim 3, characterized in that: The step of adjusting the basketball image at the monitoring time point according to the deviation field environment assessment value comprises: The deviation venue environment assessment value is matched with the contrast adjustment amplitude level corresponding to each deviation venue environment assessment value preset in the basketball detection database to obtain the basketball image contrast adjustment amplitude level, and the contrast of the basketball image is adjusted according to the basketball image contrast adjustment amplitude level.

5. The basketball detection method based on artificial intelligence as claimed in claim 1, characterized in that: The step of obtaining the equipment performance index at each monitoring time point based on comprehensive analysis of the equipment performance data includes: Device performance data, including battery level, resolution, and frame rate; Obtaining critical battery power, critical resolution, reference frame rate, and allowable deviation frame rate from a basketball detection database; The equipment performance index at each monitoring time point is obtained through comprehensive analysis.

6. The basketball detection method based on artificial intelligence as claimed in claim 1, characterized in that: The method for obtaining the image accuracy evaluation value at each monitoring time point is as follows: ； In the formula, represents the image accuracy evaluation value at the i-th monitoring time point, Indicates the impact factor of image accuracy evaluation corresponding to the basketball position, Indicates the impact factor of image accuracy evaluation corresponding to basketball motion angle, Indicates the impact factor of image accuracy evaluation corresponding to basketball movement speed, Indicates the image accuracy assessment impact factor corresponding to the device performance index, Indicates the image accuracy assessment impact factor corresponding to the site environment assessment value, represents the position change parameter at the i-th monitoring time point, represents the reference position variation parameter, Indicates the allowable deviation position change parameter, represents the motion angle variation parameter at the i-th monitoring time point, represents the reference motion angle variation parameter, Indicates the allowable deviation motion angle variation parameter, represents the motion rate variation parameter at the i-th monitoring time point, represents the reference motion rate variation parameter, Indicates the allowable deviation motion rate change parameter, represents the equipment performance index at the ith monitoring time point, represents the critical equipment performance index, represents the site environmental assessment value at the ith monitoring time point, Represents the critical site environmental assessment value, where i is the number of each monitoring time point, i=1, 2, 3, ..., N, and N is the total number of monitoring time points.

7. The basketball detection method based on artificial intelligence as claimed in claim 1, characterized in that: The step of performing adjustment feedback according to the image accuracy evaluation value comprises: Get image accuracy assessment thresholds from the basketball detection database; The image accuracy assessment value is compared with the image accuracy assessment threshold. If the image accuracy assessment value is less than the image accuracy assessment threshold, the corrected basketball image is further subjected to image correction processing. If the image accuracy assessment value is greater than or equal to the image accuracy assessment threshold, no additional processing is performed.

8. An artificial intelligence-based basketball detection system, using an artificial intelligence-based basketball detection method as claimed in any one of claims 1 to 7, characterized in that: It includes a data acquisition module, an image correction module, an image adjustment module and a basketball detection database; Wherein, the data acquisition module is used to acquire venue environment data, equipment performance data and basketball image data; The image correction module is used to obtain the site environment evaluation value at each monitoring time point according to the comprehensive analysis of the site environment data, and adjust the basketball image data according to the site environment evaluation value at each monitoring time point to obtain the corrected basketball image at each monitoring time point; The image adjustment module is used to obtain the equipment performance index at each monitoring time point based on a comprehensive analysis of the equipment performance data, obtain the image accuracy evaluation value at each monitoring time point based on a comprehensive analysis of the corrected basketball image at each monitoring time point, the equipment performance index at each monitoring time point and the venue environment evaluation value at each monitoring time point, and perform adjustment feedback based on the image accuracy evaluation value.

9. An artificial intelligence-based basketball detection electronic device, characterized in that: include: a processor, a memory for storing instructions executable by the processor; When the processor is configured to execute the instructions, the electronic device implements the basketball detection method based on artificial intelligence as described in any one of claims 1-7.

Images