Fitness behavior intelligent monitoring method and system based on Internet of Things

Abstract

The invention discloses an intelligent body-building behavior monitoring method and system based on the Internet of Things, and relates to the technical field of body-building monitoring, and the method comprises the steps: carrying out the first classification of body-building equipment, calculating a comprehensive body-building standard degree according to the body-building duration and the action standard degree in historical monitoring data, and carrying out the second classification; constructing a two-dimensional classification matrix based on the two categories, setting a reference monitoring interval and a period length, and determining initial monitoring parameters of six categories of instruments in the matrix; acquiring associated parts and standard motion data of all fitness equipment, constructing an associated part set, identifying and positioning the parts and the equipment of the fitness personnel in the video, analyzing the motion speed and the track of the associated parts, and calculating a motion track deviation value; calculating a speed mean value and a trajectory deviation mean value, and calculating a speed coincidence ratio and a unit fitness standard degree so as to obtain a unit comprehensive fitness standard degree and adjust monitoring parameters; intelligent monitoring of fitness equipment classification, fitness personnel behavior monitoring and dynamic monitoring parameter adjustment based on the Internet of Things is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of fitness monitoring, and in particular, to an intelligent monitoring method and system for fitness behavior based on the Internet of Things. Background Art

[0002] In today's society, with the improvement of people's living standards and the enhancement of health awareness, fitness has become an important part of many people's daily lives. However, traditional fitness methods often lack scientific guidance and personalized monitoring, resulting in poor fitness effects and even possible sports injuries due to improper exercise methods. At the same time, the rapid development of the Internet of Things technology has provided new possibilities for the intelligent monitoring of fitness behavior.

[0003] The Internet of Things technology connects various intelligent devices and sensors to the Internet to build a large and efficient monitoring network. In the field of fitness, this technology can be applied to the comprehensive monitoring of fitness equipment, fitness personnel, and fitness environments. For example, wearable devices such as smart bracelets and smart watches can real-time monitor physiological indicators such as the user's heart rate, blood pressure, and steps; intelligent fitness equipment can record the user's movement trajectory, movement speed, calories consumed, and other exercise data.

[0004] However, the existing fitness monitoring technology still has some deficiencies. Traditional monitoring methods may adopt fixed monitoring intervals and cycle lengths and cannot dynamically adjust according to the actual performance of fitness personnel. Secondly, most traditional monitoring methods do not conduct in-depth analysis on the associated parts of fitness equipment and cannot accurately evaluate the movement standard degree of fitness personnel. Summary of the Invention

[0005] (1) Technical Problems to be Solved

[0006] In view of the technical problems in the background art, the present invention proposes an intelligent monitoring method and system for fitness behavior based on the Internet of Things. By obtaining the standard movement data of the associated parts of each fitness equipment and real-time analyzing the actual movement trajectory and speed of fitness personnel, and by calculating the unit comprehensive fitness standard degree of the current fitness personnel using the corresponding fitness equipment, the initial monitoring interval and the initial monitoring cycle length are adjusted in real time, thereby solving the technical problems recorded in the background art.

[0007] (2) Technical Solutions

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0009] An intelligent monitoring method for fitness behavior based on the Internet of Things, comprising:

[0010] Classify fitness equipment into a first category; obtain the monitoring record form of each fitness equipment, calculate the comprehensive fitness standard of each fitness equipment based on the fitness duration and movement standard in the historical monitoring data, and classify the fitness equipment into a second category;

[0011] Construct a two-dimensional classification matrix based on the first and second classifications of fitness equipment; set a reference monitoring interval and a reference monitoring cycle length, and determine the initial monitoring interval and the initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix respectively;

[0012] Obtain the associated parts of each fitness equipment and the standard movement data of each associated part, and construct the set of associated parts of each fitness equipment; identify and locate the parts of the fitness personnel and the fitness equipment in the detection video based on computer vision and deep learning algorithms; analyze the movement speed and movement trajectory of each associated part in the video, and calculate the movement trajectory deviation value of each associated part respectively;

[0013] Calculate the average speed and the average movement trajectory deviation of each associated part; calculate the speed compliance ratio of each associated part according to the average speed and the standard movement speed range; calculate the unit fitness standard of the associated part based on the speed compliance ratio and the average movement trajectory deviation; combine the unit fitness standards of all associated parts to obtain the unit comprehensive fitness standard of the current fitness personnel using the corresponding fitness equipment, and adjust the initial monitoring interval and the initial monitoring cycle length; complete the calculation of the fitness standard of the current fitness record when using the fitness equipment.

[0014] Specifically, obtain all the fitness durations t in the historical monitoring data from the monitoring record forms of each fitness equipment i and the corresponding fitness standards Fs i , and combine the movement standard data in all the monitoring data of the same fitness equipment to obtain the comprehensive fitness standard CFs of each fitness equipment i , and the expression is: where i represents the i-th historical monitoring data of each fitness equipment, and n represents the total number of historical monitoring data of each fitness equipment.

[0015] Specifically, the first classification is primary equipment, intermediate equipment, and advanced equipment;

[0016] The second classification is: mark the fitness equipment with a comprehensive fitness standard CFs i lower than the fitness standard threshold CFs0 as error-prone equipment; mark the fitness equipment with a comprehensive fitness standard CFs i not lower than the fitness standard threshold CFs0 as non-error-prone equipment.

[0017] Specifically, obtain the associated parts of each fitness device and form an associated part set {P1, P2, …, P m} for the corresponding fitness device, where P j represents the j-th associated part in the associated part set, and m represents the total number of associated parts in the associated part set;

[0018] Obtain the standard motion data for each associated part, including the motion trajectory and motion speed data.

[0019] Specifically, automatically record the real-time process video of the fitness personnel using the fitness device according to the preset initial monitoring interval and initial monitoring cycle length;

[0020] Identify each associated part in the associated part set from the video data. By tracking the associated part in the video for consecutive frames, record its position in each frame and analyze to obtain the motion trajectory; calculate the real-time motion speed of the associated part according to the position change of the associated part in consecutive frames.

[0021] Furthermore, during the process of using the fitness device, perform multiple repeated cycle trainings, and record the motion trajectory of the associated part in each repeated cycle training;

[0022] Align the actual motion trajectory of the associated part when the fitness personnel uses the fitness device with the standard motion trajectory of the corresponding associated part according to each complete repeated cycle training, and calculate the Euclidean distance between the actual trajectory point and the corresponding standard trajectory point to obtain the motion trajectory deviation value of one repeated cycle training;

[0023] The alignment according to each complete repeated cycle training is specifically achieved by aligning the time of one repeated cycle training in the standard motion trajectory with the time when the current fitness personnel actually completes one repeated cycle training, adjusting the standard motion trajectory frame by frame with the same time ratio based on the time ratio, and then calculating based on the Euclidean distance between the actual trajectory point and the corresponding standard trajectory point.

[0024] Specifically, based on the real-time monitoring of the motion speed of the associated part within the monitoring cycle, obtain the speed mean Rap; when it is monitored that the fitness personnel completes one repeated cycle training, obtain the motion trajectory deviation value of this repeated cycle training, and perform a mean operation with the motion trajectory deviation values of several previous repeated cycle trainings to obtain the motion trajectory deviation mean Mdt.

[0025] Furthermore, based on the speed mean Rap of each associated part and the standard motion speed range [Rap1, Rap2], calculate the speed compliance ratio Sir of each associated part. Specifically, when the speed mean Rap of the associated part is within the standard motion speed range [Rap1, Rap2], the value of the speed compliance ratio Sir is 1;

[0026] When the average speed Rap of the associated part is not within the standard motion speed range [Rap1, Rap2], if Rap < Rap1, the value of the speed compliance ratio Sir is If Rap > Rap2, the value of the speed compliance ratio Sir is

[0027] Furthermore, calculate the unit fitness standard UFs of the associated part, and the expression is: Where Mdt0 represents the preset motion trajectory deviation threshold of the associated part;

[0028] Combine the unit fitness standard UFs of all associated parts in the associated part set j , and calculate the unit comprehensive fitness standard UCFs of the current fitness person using the corresponding fitness equipment. The expression is: Where β j represents the preset weight coefficient of the jth associated part in the associated part set, and

[0029] Adjust the initial monitoring interval T3 to T3 * UCFs, and adjust the initial monitoring cycle length T4 to

[0030] An intelligent monitoring system for fitness behavior based on the Internet of Things, including:

[0031] A fitness equipment classification module, which is used to classify the fitness equipment for the first time; obtain the monitoring record form of each fitness equipment, calculate the comprehensive fitness standard of each fitness equipment based on the fitness duration and action standard degree in the historical monitoring data, and classify the fitness equipment for the second time;

[0032] A monitoring parameter setting module, which is used to construct a two-dimensional classification matrix based on the first and second classifications of the fitness equipment; set the reference monitoring interval and the reference monitoring cycle length, and determine the initial monitoring interval and the initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix respectively;

[0033] A motion data acquisition module, which is used to obtain the associated part of each fitness equipment and the standard motion data of each associated part, and construct an associated part set of each fitness equipment; identify and locate the fitness person's part and the fitness equipment in the detection video based on computer vision and deep learning algorithms; analyze the motion speed and motion trajectory of each associated part in the video, and calculate the motion trajectory deviation value of each associated part respectively;

[0034] The standard degree evaluation and monitoring adjustment module is used to calculate the average speed and the average deviation of the movement trajectory of each associated part; calculate the speed compliance ratio of each associated part according to the average speed and the standard movement speed range; calculate the unit fitness standard degree of the associated part based on the speed compliance ratio and the average deviation of the movement trajectory; combine the unit fitness standard degrees of all associated parts to obtain the unit comprehensive fitness standard degree of the current fitness person using the corresponding fitness equipment, and adjust the initial monitoring interval and the initial monitoring cycle length; calculate the fitness standard degree of the current fitness record when using the fitness equipment.

[0035] (III) Beneficial effects

[0036] The present invention provides an intelligent monitoring method and system for fitness behavior based on the Internet of Things, having the following beneficial effects:

[0037] 1. Through the first classification (primary, intermediate, and advanced equipment) and the second classification (error-prone equipment, non-error-prone equipment) of fitness equipment, refined management of fitness equipment is achieved; the second classification calculates the comprehensive fitness standard degree based on the fitness duration and action standard degree in historical monitoring data, and identifies error-prone equipment, which helps to increase the monitoring intensity of equipment with more non-standard operations;

[0038] 2. The method of determining monitoring parameters based on the diagonal principle makes the allocation of monitoring resources more scientific and reasonable; for different types and characteristics of fitness equipment, according to factors such as their usage difficulty and error-prone degree, flexibly adjust the monitoring interval and cycle length, which can not only avoid waste of resources caused by over-monitoring while ensuring effective monitoring of fitness personnel, but also strengthen monitoring for key equipment and error-prone situations to improve monitoring efficiency;

[0039] 3. Accurate monitoring of the process of fitness personnel using fitness equipment is achieved. Through advanced computer vision and deep learning technologies, fitness personnel and equipment can be accurately identified, the movement information of associated parts can be obtained in real time, and compared with standard movement data, providing detailed and accurate data support for subsequent evaluation of the operation specification degree of fitness personnel, helping to timely discover non-standard actions of fitness personnel during the use of equipment, and providing a strong basis for fitness guidance and safety guarantee;

[0040] 4. It can comprehensively and objectively evaluate the specification degree of fitness personnel using fitness equipment. By comprehensively considering the speed compliance ratio and the average deviation of the movement trajectory, the evaluation result is more accurate and reliable; dynamically adjust the monitoring parameters according to the unit comprehensive fitness standard degree, realizing the dynamic optimal allocation of monitoring resources, improving the pertinence and effectiveness of monitoring; recording the fitness standard degree of the current fitness record helps fitness personnel understand their own fitness situation and also provides data support for fitness coaches to formulate more personalized fitness guidance plans. Description of the drawings

[0041] Figure 1 Schematic diagram of the steps of an intelligent monitoring method for fitness behaviors based on the Internet of Things provided by the present invention;

[0042] Figure 2 Schematic diagram of the structure of an intelligent monitoring system for fitness behaviors based on the Internet of Things provided by the present invention. Detailed implementation manners

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Refer to Figure 1 , the present invention provides an intelligent monitoring method for fitness behaviors based on the Internet of Things, including:

[0045] Step 1: Classify fitness equipment for the first time; obtain a monitoring record form for each fitness equipment, and calculate the comprehensive fitness standard degree of each fitness equipment based on the fitness duration and action standard degree in the historical monitoring data, and classify the fitness equipment for the second time;

[0046] The following steps are included in Step 1:

[0047] Step 101: Obtain various fitness equipment in the gymnasium and obtain the first classification results of all fitness equipment, specifically divided into primary equipment, intermediate equipment, and advanced equipment;

[0048] Among them, the primary equipment refers to fitness equipment with a single function, simple operation, and mainly used for basic physical exercises, such as dumbbells, treadmills, fitness balls, and other fitness equipment; these equipment are simple to use and have high safety, so only fitness personnel using these fitness equipment need to be monitored for fitness at longer time intervals;

[0049] The intermediate equipment refers to fitness equipment with diverse functions, adjustable parameters, and can meet certain exercise needs, but the operation is relatively not complicated, such as elliptical machines, rowing machines, and other equipment; the use of these fitness equipment requires certain operation skills and the safety is general, so fitness personnel using these fitness equipment need to be monitored for fitness at medium time intervals;

[0050] Advanced equipment refers to fitness equipment with complex functions, combining multiple exercise modes, and having relatively high requirements for physical fitness and skills, such as Smith machines, gantry racks, bench presses, etc.; the use of these fitness equipment is difficult and requires high physical fitness and skills, and strict safety measures are needed. Therefore, fitness monitoring of fitness personnel using these fitness equipment needs to be carried out at short time intervals;

[0051] The classification of primary equipment, intermediate equipment, and advanced equipment is specifically preset by professional fitness coaches;

[0052] Step 102: Obtain the monitoring record forms of each fitness equipment from the gym management system. The monitoring record forms of fitness equipment record the historical monitoring data of the fitness equipment, including the fitness duration, motion data of related parts, and fitness standard data; among them, the motion data of related parts represents data such as the motion speed and motion trajectory when exercising the body parts corresponding to the use of the fitness equipment; the fitness standard represents the degree of motion standard of the related parts of the fitness personnel during a complete use of the corresponding fitness equipment, and is specifically calculated by combining the actual motion data and standard motion data of multiple related parts; one monitoring data of the fitness equipment corresponds to one use record;

[0053] Obtain all the fitness durations t in the historical monitoring data from the monitoring record forms of each fitness equipment i and the corresponding fitness standard Fs i , and combine the motion standard data in all the monitoring data of the same fitness equipment to obtain the comprehensive fitness standard CFs of each fitness equipment i , and the expression is: where i represents the i-th historical monitoring data of each fitness equipment, and n represents the total number of historical monitoring data of each fitness equipment;

[0054] Step 103: Conduct a second classification of all fitness equipment, specifically: preset a fitness standard threshold CFs0 by a professional fitness coach, and mark the fitness equipment with a comprehensive fitness standard CFs i lower than the fitness standard threshold CFs0 as error-prone equipment, indicating that the operation methods of fitness personnel when using these fitness equipment are mostly not standardized, and greater monitoring efforts need to be made on these equipment;

[0055] Mark the fitness equipment with a comprehensive fitness standard CFs i not lower than the fitness standard threshold CFs0 as non-error-prone equipment, indicating that most fitness personnel can use and operate these fitness equipment in a standardized manner, and greater monitoring efforts do not need to be made on these equipment.

[0056] When in use, combine the content in Steps 101 to 103:

[0057] By classifying fitness equipment into a first classification (beginner, intermediate, and advanced equipment) and a second classification (error-prone equipment and non-error-prone equipment), refined management of fitness equipment is achieved; the second classification calculates the comprehensive fitness standard based on the fitness duration and movement standard in historical monitoring data to identify error-prone equipment, which helps to increase the monitoring intensity for equipment with more non-standard operations.

[0058] Step 2: Construct a two-dimensional classification matrix based on the first and second classifications of fitness equipment; set a reference monitoring interval and a reference monitoring cycle length, and determine the initial monitoring interval and the initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix respectively;

[0059] The steps in Step 2 include the following steps:

[0060] Step 201: Based on the first and second classifications of fitness equipment, construct a two-dimensional classification matrix of fitness equipment; obtain the six elements in the two-dimensional classification matrix, which are beginner non-error-prone equipment, beginner error-prone equipment, intermediate non-error-prone equipment, intermediate error-prone equipment, advanced non-error-prone equipment, and advanced error-prone equipment;

[0061] Step 202: Set a reference monitoring interval T1 and a reference monitoring cycle length T2. Among them, the monitoring cycle length represents the actual monitoring time length, that is, during this period, the fitness operations of fitness personnel are monitored in real time; the monitoring interval represents the time when the fitness operations of fitness personnel are not monitored in real time; the specific monitoring process is to monitor the fitness operations of fitness personnel in real time for a monitoring cycle length, and then stop for a monitoring interval time, and then monitor the fitness operations of fitness personnel in real time for a monitoring cycle length again; the reference monitoring interval and the reference monitoring cycle length are fixed preset values;

[0062] Step 203: Based on the diagonal principle, determine the initial monitoring interval and the initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix respectively. The diagonal principle means that the initial monitoring interval and the initial monitoring cycle length of several types of fitness equipment located on the diagonal in the two-dimensional classification matrix are the same; specifically: adjust the initial monitoring interval of the beginner non-error-prone equipment to twice the reference monitoring interval T1, that is, 2T1, and adjust the initial monitoring cycle length to half of the reference monitoring cycle length T2, that is

[0063] Adjust the initial monitoring intervals of the beginner error-prone equipment and the intermediate non-error-prone equipment to the reference monitoring interval T1, and adjust the initial monitoring cycle length to half of the reference monitoring cycle length T2, that is

[0064] Adjust the initial monitoring interval of intermediate error-prone devices and advanced non-error-prone devices to twice the reference monitoring interval T1, i.e., 2T1, and adjust the initial monitoring cycle length to the reference monitoring cycle length T2;

[0065] Adjust the initial monitoring interval of advanced error-prone devices to the reference monitoring interval T1, and adjust the initial monitoring cycle length to the reference monitoring cycle length T2;

[0066] During use, combine the content in steps 201 to 203:

[0067] The method of determining monitoring parameters based on the diagonal principle makes the allocation of monitoring resources more scientific and reasonable; for fitness devices of different categories and characteristics, flexibly adjust the monitoring interval and cycle length according to factors such as their usage difficulty and error-proneness. This can not only ensure the effective monitoring of fitness personnel while avoiding resource waste caused by over-monitoring, but also strengthen the monitoring of key devices and error-prone situations to improve the monitoring efficiency.

[0068] Step 3: Obtain the associated parts of each fitness device, the standard motion data of each associated part, and construct the set of associated parts of each fitness device; identify and locate the parts of the fitness personnel and the fitness device in the detection video based on computer vision and deep learning algorithms; analyze the motion speed and motion trajectory of each associated part in the video, and calculate the motion trajectory deviation value of each associated part respectively;

[0069] The steps in step 3 include the following steps:

[0070] Step 301: Obtain the structure, function, and correct usage method of each fitness device based on the official instruction manual of the fitness device, specifically including the associated parts of each fitness device, and the motion trajectory and motion speed data of the corresponding associated parts when using each fitness device, and record them as the standard motion data of each associated part; the associated part of a fitness device refers to the specific body part on the body that directly interacts with the device, bears the motion load, or produces a motion effect when using a specific fitness device for exercise; for example, when doing bicep curls with dumbbells, the biceps brachii is the body part directly associated with the dumbbells because the weight and motion trajectory of the dumbbells directly act on the biceps brachii, so the front side of the upper arm where the biceps brachii is located is an associated part of the dumbbells;

[0071] Step 302: Obtain the associated parts of each fitness device and form the set of associated parts {P1, P2…, P m} of the corresponding fitness device, where P j represents the jth associated part in the set of associated parts, and m represents the total number of associated parts in the set of associated parts;

[0072] Step 303: Based on the binocular cameras installed in the gym, identify and locate the fitness personnel in the video by recognizing computer vision and deep learning algorithms, identify whether the fitness personnel are using fitness equipment, and periodically capture the usage process of the fitness personnel on the fitness equipment. Specifically:

[0073] Calibrate the binocular cameras to obtain accurate spatial information, and then train a model to distinguish the interaction action features between the human body and the equipment. When it is recognized that the fitness personnel are using fitness equipment, activate the shooting function of the binocular cameras, and automatically record the real-time process of the fitness personnel using the fitness equipment according to the preset initial monitoring interval and initial monitoring cycle length.

[0074] Use the trained OpenPose deep learning model to identify each part of the fitness personnel in the recorded real-time video data. Identify the fitness equipment used by the fitness personnel based on the pre-trained fitness equipment recognition model. The OpenPose deep learning model is an open-source computer vision model for human pose estimation, which can monitor and identify the body joints and limb contours of multiple people in real time and output the positions of each part in the form of two-dimensional coordinates. The fitness equipment recognition model realizes the model for identifying the type of fitness equipment based on image recognition by pre-entering the image data of each fitness equipment at multiple angles.

[0075] Step 304: Based on the associated part set corresponding to the fitness equipment, identify each associated part in the associated part set from the video data, and obtain the corresponding standard motion data for each associated part, including the standard motion speed range and the standard motion trajectory.

[0076] Analyze the motion trajectory and motion speed of each associated part during the real-time process of the fitness personnel using the fitness equipment recorded. Among them, the motion trajectory is obtained by tracking the associated part in the video for consecutive frames and recording its position in each frame. Calculate the real-time motion speed of the associated part according to the position change of the associated part in consecutive frames.

[0077] During the process of using fitness equipment, multiple repeated cycle trainings will be carried out, and the movement trajectories of related parts are recorded each time. When a fitness person uses the fitness equipment, the actual movement trajectories of related parts are aligned with the standard movement trajectories of corresponding related parts according to each complete repeated cycle training, and the Euclidean distance between the actual trajectory points and the corresponding standard trajectory points is calculated to obtain the movement trajectory deviation value of one repeated cycle training. The alignment according to each complete repeated cycle training is specifically achieved by aligning the time of one repeated cycle training in the standard movement trajectory with the time when the current fitness person actually completes one repeated cycle training, adjusting the standard movement trajectory frame by frame with the same time ratio based on the time ratio, and then calculating based on the Euclidean distance between the actual trajectory points and the corresponding standard trajectory points.

[0078] During use, combine the content in steps 301 to 304:

[0079] It realizes the precise monitoring of the process of a fitness person using fitness equipment. Through advanced computer vision and deep learning technologies, it can accurately identify the fitness person and the equipment, obtain the movement information of related parts in real time, and compare it with the standard movement data, providing detailed and accurate data support for subsequent evaluation of the operation specification degree of the fitness person, helping to timely discover the non-standard actions of the fitness person during the use of the equipment, and providing a strong basis for fitness guidance and safety guarantee.

[0080] Step 4: Calculate the average speed and the average movement trajectory deviation of each related part; calculate the speed compliance ratio of each related part according to the average speed and the standard movement speed range; calculate the unit fitness standard degree of the related part based on the speed compliance ratio and the average movement trajectory deviation; combine the unit fitness standard degrees of all related parts to obtain the unit comprehensive fitness standard degree of the current fitness person using the corresponding fitness equipment, and adjust the initial monitoring interval and the initial monitoring cycle length; calculate the fitness standard degree of the current fitness record when the use of the fitness equipment is completed.

[0081] The step 4 includes the following steps:

[0082] Step 401: Based on the real-time monitoring of the movement speed of the related part within the monitoring cycle, obtain the average speed Rap; when it is monitored that the fitness person completes one repeated cycle training, obtain the movement trajectory deviation value of this repeated cycle training, and perform an average operation with the movement trajectory deviation values of several previous repeated cycle trainings to obtain the average movement trajectory deviation.

[0083] If, at the end of a monitoring cycle length, the fitness person has not completed a repeated cycle of training, then based on the ratio of the average time for the fitness person to complete several repeated cycles of training before and the time for the fitness person to perform the current repeated cycle of training at the end of the current monitoring cycle length, in the previous aligned standard motion trajectory, obtain a partial standard motion trajectory for the corresponding time period based on the time ratio, starting from the beginning of the standard motion trajectory and taking it backward, and calculate the motion trajectory deviation value between the partial standard motion trajectory and the current uncompleted part of the repeated cycle of training; similarly, after the start of a monitoring cycle length, when the time for the fitness person to complete the repeated cycle of training for the first time is detected, perform the same analysis, but when obtaining the partial standard motion trajectory for the corresponding time period based on the time ratio, take it forward from the end of the standard motion trajectory;

[0084] Step 402: Calculate the speed compliance ratio Sir of each associated part based on the average speed Rap of each associated part and the standard motion speed range [Rap1, Rap2]. Specifically, when the average speed Rap of the associated part is within the standard motion speed range [Rap1, Rap2], the value of the speed compliance ratio is 1;

[0085] When the average speed Rap of the associated part is not within the standard motion speed range [Rap1, Rap2], if Rap < Rap1, then the value of the speed compliance ratio Sir is If Rap > Rap2, then the value of the speed compliance ratio Sir is

[0086] Step 403: Calculate the unit fitness standard degree UFs of the associated part based on the speed compliance ratio Sir of each associated part and the average motion trajectory deviation Mdt. The expression is:

[0087] where Mdt0 represents the preset motion trajectory deviation threshold of the associated part;

[0088] Step 404: Combine the unit fitness standard degrees UFs of all the associated parts in the associated part set j , to obtain the unit comprehensive fitness standard degree UCFs of the current fitness person using the corresponding fitness equipment. The expression is: where β j represents the weight coefficient of the jth associated part in the associated part set, which is specifically preset by a professional fitness coach, and

[0089] Based on the unit comprehensive fitness standard degree UCFs, adjust the initial monitoring interval T3 and the initial monitoring cycle length T4 of the current fitness equipment. Adjust the initial monitoring interval T3 to T3 * UCFs, and adjust the initial monitoring cycle length T4 to

[0090] When the fitness personnel complete the use of the current fitness equipment, the average value of all unit comprehensive fitness standard degrees is calculated to obtain the fitness standard degree Fs of the current fitness record, which is recorded in the monitoring record form of the current fitness equipment.

[0091] During use, combine the content in steps 401 to 404:

[0092] It can comprehensively and objectively evaluate the standard degree of fitness personnel using fitness equipment. By comprehensively considering the speed compliance ratio and the average value of motion trajectory deviation, the evaluation result is more accurate and reliable; dynamically adjust the monitoring parameters according to the unit comprehensive fitness standard degree, realizing the dynamic optimal allocation of monitoring resources, improving the pertinence and effectiveness of monitoring; recording the fitness standard degree of the current fitness record helps fitness personnel understand their own fitness situation and also provides data support for fitness coaches to formulate more personalized fitness guidance plans.

[0093] Reference Figure 2 , the present invention also provides an intelligent monitoring system for fitness behavior based on the Internet of Things, including:

[0094] A fitness equipment classification module, used to classify fitness equipment for the first time; obtain the monitoring record form of each fitness equipment, calculate the comprehensive fitness standard degree of each fitness equipment based on the fitness duration and motion standard degree in the historical monitoring data, and classify the fitness equipment for the second time;

[0095] A monitoring parameter setting module, used to construct a two-dimensional classification matrix based on the first and second classifications of fitness equipment; set the reference monitoring interval and the reference monitoring cycle length, and determine the initial monitoring interval and the initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix respectively;

[0096] A motion data acquisition module, used to obtain the associated parts of each fitness equipment and the standard motion data of each associated part, and construct the associated part set of each fitness equipment; identify and locate the fitness personnel parts and fitness equipment in the detection video based on computer vision and deep learning algorithms; analyze the motion speed and motion trajectory of each associated part in the video, and calculate the motion trajectory deviation value of each associated part respectively;

[0097] The standard degree evaluation and monitoring adjustment module is used to calculate the average speed and the average deviation of the motion trajectory of each associated part; calculate the speed compliance ratio of each associated part according to the average speed and the standard motion speed range; calculate the unit fitness standard degree of the associated part based on the speed compliance ratio and the average deviation of the motion trajectory; combine the unit fitness standard degrees of all associated parts to obtain the unit comprehensive fitness standard degree of the current fitness person using the corresponding fitness equipment, and adjust the initial monitoring interval and the initial monitoring cycle length; calculate the fitness standard degree of the current fitness record when using the fitness equipment.

[0098] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer storage medium or transmitted through the computer storage medium.

[0099] The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0100] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An intelligent monitoring method for fitness behavior based on the Internet of Things, characterized in that: The steps include: The fitness equipment is first classified; a monitoring record sheet of each fitness equipment is obtained, and based on the fitness duration and movement standard in the historical monitoring data, the comprehensive fitness standard of each fitness equipment is calculated, and the fitness equipment is second classified; Construct a two-dimensional classification matrix based on the first and second classifications of fitness equipment; Set the benchmark monitoring interval and benchmark monitoring cycle length, and determine the initial monitoring interval and initial monitoring cycle length for the device types corresponding to the six elements in the two-dimensional classification matrix; Obtain the associated parts of each fitness equipment and the standard motion data of each associated part, and construct the associated parts set of each fitness equipment; identify and locate the parts of the fitness personnel and fitness equipment in the detection video based on computer vision and deep learning algorithms; analyze the movement speed and movement trajectory of each associated part in the video, and calculate the movement trajectory deviation value of each associated part respectively; Calculate the mean speed of each associated part and the mean deviation of the motion trajectory; calculate the speed compliance ratio of each associated part according to the mean speed and the standard motion speed range; calculate the unit fitness standard of the associated part based on the speed compliance ratio and the mean deviation of the motion trajectory; combine the unit fitness standard of all associated parts to obtain the unit comprehensive fitness standard of the current fitness person using the corresponding fitness equipment, adjust the initial monitoring interval and the initial monitoring cycle length; calculate the fitness standard of the current fitness record when the fitness equipment is used.

2. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 1, characterized in that: Obtain all fitness durations t in the historical monitoring data from the monitoring record table of each fitness equipment i And the corresponding fitness standard Fs i , and combine the action standard data of all monitoring data of the same fitness equipment to obtain the comprehensive fitness standard CFs of each fitness equipment i , the expression is: Wherein, i represents the i-th historical monitoring data of each fitness equipment, and n represents the total number of historical monitoring data of each fitness equipment.

3. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 2, characterized in that: The first classification is elementary equipment, intermediate equipment and advanced equipment; The second classification is: comprehensive fitness standard CFs i Fitness equipment with a fitness standard lower than the threshold value CFs0 is marked as error-prone equipment; the comprehensive fitness standard CFs i Fitness equipment that is not less than the fitness standard threshold CFs0 is marked as non-error-prone equipment.

4. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 3, characterized in that: Get the associated parts of each fitness equipment and form the associated parts set {P1, P2…, P m }, where P j represents the jth associated part in the associated part set, and m represents the total number of associated parts in the associated part set; Obtain standard motion data of each associated part, including motion trajectory and motion speed data.

5. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 4, characterized in that: Automatically record the real-time process video of fitness personnel using fitness equipment according to the preset initial monitoring interval and initial monitoring cycle length; Identify the associated parts in the associated parts set from the video data, track the associated parts in the video in consecutive frames, record their positions in each frame and analyze the motion trajectory; calculate the real-time motion speed of the associated parts based on the position changes of the associated parts in consecutive frames.

6. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 5, characterized in that: Perform multiple repetitive cycle training while using fitness equipment, and record the movement trajectory of related parts in each repetitive cycle training; When the fitness personnel use the fitness equipment, the actual motion trajectory of the associated parts is aligned with the standard motion trajectory of the corresponding associated parts according to each complete repeated cycle training, and the Euclidean distance between the actual trajectory point and the corresponding standard trajectory point is calculated to obtain the motion trajectory deviation value of one repeated cycle training; The alignment according to each complete repetitive cycle training is specifically performed by aligning the time of one repetitive cycle training in the standard motion trajectory with the time when the current fitness person actually completes one repetitive cycle training, adjusting the standard motion trajectory frame by frame with the same time ratio based on the time ratio, and then calculating it based on the Euclidean distance between the actual trajectory point and the corresponding standard trajectory point.

7. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 6, characterized in that: Based on the real-time monitoring of the movement speed of the relevant parts during the monitoring period, the speed average Rap is obtained; when the fitness person is monitored to complete a repeated cycle training, the motion trajectory deviation value of the repeated cycle training is obtained, and the average is calculated with the motion trajectory deviation values ​​of several previous repeated cycle trainings to obtain the motion trajectory deviation mean Mdt.

8. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 7, characterized in that: The speed compliance ratio Sir of each associated part is calculated based on the speed mean Rap of each associated part and the standard motion speed range [Rap1, Rap2]. Specifically, when the speed mean Rap of the associated part is within the standard motion speed range [Rap1, Rap2], the speed compliance ratio Sir takes a value of 1; When the speed mean Rap of the associated part is not within the standard motion speed range [Rap1, Rap2], if Rap<Rap1, the speed compliance ratio Sir is If Rap>Rap2, the speed ratio Sir is 9. The method for intelligent monitoring of fitness behavior based on the Internet of Things as claimed in claim 8, characterized in that: Calculate the unit fitness standard UFs of the associated parts, the expression is: Wherein, Mdt0 represents the preset motion trajectory deviation threshold of the associated part; Combine the unit fitness standard UFs of all related parts in the related parts set j , calculate the unit comprehensive fitness standard UCFs of the current fitness personnel using the corresponding fitness equipment, the expression is: Among them, β j represents the preset weight coefficient of the jth associated part in the associated part set, and Adjust the initial monitoring interval T3 to T3*UCFs, and adjust the initial monitoring cycle length T4 to 10. An intelligent fitness behavior monitoring system based on the Internet of Things, characterized in that: include: The fitness equipment classification module is used to perform a first classification on the fitness equipment; obtain a monitoring record sheet for each fitness equipment, calculate the comprehensive fitness standard of each fitness equipment based on the fitness duration and movement standard in the historical monitoring data, and perform a second classification on the fitness equipment; A monitoring parameter setting module is used to construct a two-dimensional classification matrix based on the first and second classifications of fitness equipment; set a reference monitoring interval and a reference monitoring cycle length, and determine an initial monitoring interval and an initial monitoring cycle length for the equipment types corresponding to the six elements in the two-dimensional classification matrix; The motion data acquisition module is used to obtain the associated parts of each fitness equipment and the standard motion data of each associated part, and to construct the associated parts set of each fitness equipment; to identify and locate the parts of the fitness personnel and fitness equipment in the detection video based on computer vision and deep learning algorithms; to analyze the motion speed and motion trajectory of each associated part in the video, and to calculate the motion trajectory deviation value of each associated part respectively; The standard evaluation and monitoring adjustment module is used to calculate the speed mean and motion trajectory deviation mean of each associated part; calculate the speed compliance ratio of each associated part based on the speed mean and the standard motion speed range; calculate the unit fitness standard of the associated part based on the speed compliance ratio and the motion trajectory deviation mean; combine the unit fitness standard of all associated parts to obtain the unit comprehensive fitness standard of the current fitness person using the corresponding fitness equipment, adjust the initial monitoring interval and the initial monitoring cycle length; calculate the fitness standard of the current fitness record when the fitness equipment is used.

Images