Actuating and testing integrated method, control device and exercising and testing integrated equipment

Abstract

The invention provides a training and testing integrated method, a control device and training and testing integrated equipment, and relates to the technical field of sports equipment.The training and testing integrated method comprises the steps that in response to a creation instruction for a training plan, corresponding training items and testing standards are generated; generating a corresponding training image according to the training item and the test standard, and displaying the training image; acquiring real-time physical sign data and action posture information when the user completes the training action according to the training image; based on the real-time physical sign data and the action posture information, a training effect evaluation result is determined; and if the training effect evaluation result does not meet a preset standard condition, determining a substandard training item, and displaying a decomposition action corresponding to the training item until the user completes all the training items and reaches the preset standard condition. According to the invention, closed-loop integration of training and evaluation can be realized, and training scientificity and adaptivity are improved.

Description

Technical Field

[0001] This application relates to the field of sports equipment technology, and in particular to a method, control device and equipment for integrated training and testing. Background Technology

[0002] Current mainstream training methods rely on in-person guidance from professional coaches, pre-recorded instructional videos, or basic fitness apps, but these methods have fundamental flaws. While professional coaching can provide immediate feedback, its high cost and low accessibility make it difficult to cover everyday home training scenarios. Pre-recorded videos and apps can only output demonstration content in a one-way manner, failing to dynamically perceive the user's actual state during execution. Users lack objective evidence to judge the correctness of movement, the appropriateness of intensity, and the degree of goal achievement during training, often relying solely on subjective feelings for rough assessments. This can easily lead to significantly reduced training effectiveness due to movement deviations, and even induce the risk of sports injuries. Summary of the Invention

[0003] The main objective of this invention is to provide an integrated training and testing method, which aims to improve the scientific nature and adaptability of training, thereby reducing the risk of sports injuries and increasing the achievement rate of training results.

[0004] To achieve the above objectives, the present invention provides an integrated training and testing method, the integrated training and testing method comprising:

[0005] In response to the command to create a training plan, the corresponding training items and test standards are generated;

[0006] According to the training items and the testing standards, generate corresponding training images and display the training images;

[0007] Acquire real-time vital sign data and movement posture information of the user when completing training actions based on the training images;

[0008] The training effect evaluation results are determined based on real-time vital sign data and movement posture information;

[0009] If the training effect evaluation result does not meet the preset achievement conditions, the training items that do not meet the standards are identified, and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions.

[0010] Optionally, determining the training effect evaluation result based on real-time vital sign data and movement posture information includes:

[0011] Extract the current heart rate data, exercise energy consumption value, and blood oxygen saturation data corresponding to the real-time vital signs data;

[0012] The current heart rate data is compared with the target heart rate interval corresponding to the training program to determine the heart rate achievement rate;

[0013] The energy consumption value of the exercise is compared with the target energy consumption threshold corresponding to the training project to determine the energy consumption completion rate;

[0014] The blood oxygen saturation data is compared with the blood oxygen standard range corresponding to the training program to determine whether the blood oxygen level meets the standard.

[0015] Obtain the skeletal key points of the motion posture information, construct a three-dimensional posture model based on the skeletal key points, and generate a user motion trajectory sequence;

[0016] The user's action trajectory sequence is dynamically time-normalized and compared with the standard action trajectory sequence corresponding to the training project to generate an action standardization score.

[0017] The training effect evaluation result is generated by weighted fusion of the heart rate target achievement rate, energy consumption completion rate, blood oxygen target achievement status, and movement standardization score.

[0018] Optionally, obtaining real-time vital sign data and movement posture information of the user when performing training actions based on the training images includes:

[0019] Establish a communication connection with wearable devices and acquire the user's electrocardiogram signals, body temperature changes, and triaxial acceleration data;

[0020] Determine the energy consumption data corresponding to the triaxial acceleration data;

[0021] Feature extraction is performed on the electrocardiogram signal and body temperature changes to generate a first biological sign feature vector;

[0022] The real-time vital sign data is generated based on the first biological vital sign feature vector and the energy consumption data;

[0023] Image data of the user performing training actions is acquired through image acquisition devices;

[0024] Human posture recognition is performed on the image data, the coordinate sequence of skeletal key points is extracted, and the action posture information is generated.

[0025] Optionally, generating corresponding training images according to the training items and the testing standards, and displaying the training images, includes:

[0026] Based on the movement type of the training project and the number of target groups and times in the test standard, retrieve the corresponding standard 3D human animation sequence from the preset movement image library;

[0027] In response to the selection operation of virtual coach image, background scene and rhythmic background music, the selected virtual coach image is fused and rendered with standard three-dimensional human animation sequence to generate the training image;

[0028] The training video and rhythmic background music are then synchronously output to the display component to guide the user in following the beat and completing standardized training movements.

[0029] Optionally, if the training effect evaluation result does not meet the preset achievement conditions, the unqualified training items are identified, and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions, including:

[0030] If the overall score of the training effect evaluation result is lower than the preset target score, the training effect evaluation result is determined to be that the preset target condition is not met.

[0031] Analyze the scores of each sub-indicator in the training effect evaluation results, identify one or more training items with the lowest scores, and mark them as the substandard training items;

[0032] For each training item that fails to meet the standard, retrieve the corresponding multi-angle slow-speed breakdown video of the movement from the teaching video library;

[0033] The display component prioritizes playing the breakdown video explaining the breakdown action;

[0034] During the playback of the decomposed motion video, the user's real-time vital signs data and motion posture information while watching the decomposed motion video are reacquired.

[0035] Based on the reacquired real-time vital signs data and movement posture information, a secondary evaluation result is generated for the substandard training item. If the secondary evaluation result meets the preset pass conditions, the item is marked as passed; otherwise, the breakdown of the movement explanation video is played repeatedly until it is passed.

[0036] Optionally, the step of generating corresponding training items and test standards in response to a command to create a training plan includes:

[0037] In response to the instruction to create a training plan, the system determines the user's selected training objective, available training time, and preferred exercise type, and generates the corresponding training objective.

[0038] The training objective is matched with the knowledge graph of the preset training objective to determine the training parts and ability dimensions corresponding to the training objective;

[0039] The number and combination of training items are determined based on the training area, the ability dimension, and the available training time.

[0040] The training items and the testing standards are determined based on the preferred type of exercise, the number of training items, and the combination of the training items.

[0041] Optionally, the step of generating corresponding training items and test standards in response to a command to create a training plan includes:

[0042] In response to a command to create a training plan, obtain the user's identification information;

[0043] Obtain historical physical fitness test data, past training records, and body morphology indicators corresponding to the identity information;

[0044] Analyze the historical physical fitness test data, past training records, and body shape indicators to assess the user's current overall physical fitness level and weaknesses;

[0045] Based on the overall physical fitness level, select a basic training program template of corresponding difficulty from the preset training program library;

[0046] Based on the aforementioned weaknesses, the corresponding departments in the basic solution template are strengthened, and the corresponding testing standards are adjusted to generate the adjusted training projects and testing standards.

[0047] Optionally, the integrated training and testing method further includes:

[0048] Establish a communication connection with the user's smart wearable device to obtain the user's resting heart rate, sleep duration and quality, body temperature data, and daily steps, and generate corresponding long-term body data;

[0049] The changes in the long-term physical data are analyzed at preset intervals to determine the user's recovery status and training load adaptation.

[0050] If a user's resting heart rate is detected to be abnormally high or their sleep quality to be significantly reduced, a corresponding training load overload warning message is generated and sent to the user terminal and the coach management terminal.

[0051] In addition, to achieve the above objectives, the present invention also provides a control device, the control device comprising: a memory, a processor, and a control program for a training and testing integrated device stored in the memory and executable on the processor, the control program for the training and testing integrated device being configured to implement the training and testing integrated method as described above.

[0052] Furthermore, to achieve the above objectives, the present invention also provides an integrated training and testing device, the integrated training and testing device comprising:

[0053] A display component, wherein the display component is used to present training images and feedback information;

[0054] An image acquisition component, which is used to capture the user's movements and postures during training;

[0055] Vital signs monitoring component, wherein the vital signs monitoring component is used to collect the user's heart rate, blood oxygen saturation and energy consumption data in real time; and

[0056] The control device described above is electrically connected to the display component, the image acquisition component, and the vital sign monitoring component, respectively.

[0057] The control device is used to respond to the creation command for the training plan, generate corresponding training items and test standards, generate corresponding training images according to the training items and test standards, and control the display component to display the training images.

[0058] The control device is also used to control the image acquisition component and the vital sign monitoring component to acquire real-time vital sign data and movement posture information of the user when completing training actions based on the training images, and to determine the training effect evaluation result based on the real-time vital sign data and movement posture information. If the training effect evaluation result does not meet the preset achievement conditions, the unqualified training items are identified and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions.

[0059] This invention, in response to a training plan creation command, generates corresponding training items and testing standards. Then, according to the training items and testing standards, it generates and displays corresponding training videos. Next, it acquires real-time vital sign data and posture information of the user as they complete the training movements based on the training videos. Based on this data, it determines the training effect evaluation result. If the evaluation result does not meet preset achievement conditions, it identifies the substandard training items and displays their corresponding decomposed movements until the user completes all training items and meets the preset achievement conditions. This achieves a closed-loop training process, solving the problem of separation between training and evaluation, and enabling closed-loop integration of training and evaluation, improving the scientific rigor and adaptability of training. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0061] Figure 1 This is a schematic diagram of the integrated training and testing method according to an embodiment of the present invention;

[0062] Figure 2This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0063] Figure 3 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0064] Figure 4 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0065] Figure 5 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0066] Figure 6 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0067] Figure 7 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0068] Figure 8 This is a schematic diagram of the integrated training and testing method according to another embodiment of the present invention;

[0069] Figure 9 This is a schematic diagram of the user interface of the integrated training and testing device of the present invention. Figure 1 ;

[0070] Figure 10 This is a schematic diagram of the user interface of the integrated training and testing device of the present invention. Figure 2 ;

[0071] Figure 11 This is a schematic diagram of the user interface of the integrated training and testing device of the present invention. Figure 3 ;

[0072] Figure 12 This is a schematic diagram of the user interface of the integrated training and testing device of the present invention. Figure 4 .

[0073] The realization of the objective, functional features and advantages of the present invention will be described in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0074] With the popularization of national fitness awareness and the development of sports science, systematic training has become an important means to improve physical fitness, correct posture, and conduct rehabilitation exercises. Current mainstream training methods rely on on-site guidance from professional coaches, pre-recorded instructional videos, or basic fitness apps, but these methods have fundamental flaws. While professional coaching can provide immediate feedback, it is limited by high costs and low accessibility, making it difficult to cover everyday home training scenarios. Pre-recorded videos and apps can only output demonstration content in a one-way manner, unable to dynamically perceive the user's actual state during execution. Users lack objective evidence to judge the correctness of movements, the appropriateness of intensity, and the degree of goal achievement during training, often relying solely on subjective feelings for rough assessments. This easily leads to significantly reduced training effectiveness due to movement deviations, and may even induce sports injuries.

[0075] The core problem with existing technologies lies in the mechanical separation of training and evaluation. Training systems focus on content presentation but fail to simultaneously collect user physiological indicators (such as heart rate fluctuations, changes in blood oxygen saturation, and energy expenditure levels) and movement posture details (such as skeletal key point displacement trajectories and joint angle deviations). Even if some smart devices can monitor individual physiological parameters, this data is not integrated for real-time quantitative analysis of training effectiveness. More importantly, the system lacks a closed-loop response mechanism based on the analysis results: when training effectiveness is detected to be below expectations, it cannot automatically identify weak training items, nor can it dynamically push targeted decomposed movement guidance. This fragmented design renders the training process unscientific and unadaptive, making it difficult for users to continuously optimize their training behavior without professional supervision, ultimately leading to a low rate of training goal achievement and failing to meet the needs of personalized and efficient training.

[0076] The main solution of this application embodiment is as follows: in response to the instruction to create a training plan, corresponding training items and test standards are generated, and corresponding training images are generated and displayed according to the training items and test standards. Real-time vital sign data and movement posture information of the user when completing training actions according to the training images are obtained, and the training effect evaluation result is determined based on the real-time vital sign data and movement posture information. If the training effect evaluation result does not meet the preset achievement conditions, the training items that do not meet the standards are identified, and the decomposed actions corresponding to the training items are displayed, until the user completes all training items and meets the preset achievement conditions.

[0077] In this embodiment, for ease of description, the following description will focus on the control device as the executing entity.

[0078] This application provides a solution that realizes a closed-loop training process, solves the problem of separation between training and evaluation, and has the advantages of realizing closed-loop integration of training and evaluation, improving the scientific nature and adaptability of training.

[0079] Therefore, this invention proposes an integrated training and testing method. It is understood that the integrated training and testing equipment is equipped with a control device for storing and executing the following method. The control device can be implemented using a main controller, such as an MCU (Microcontroller Unit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), or a SOC (System On Chip).

[0080] Current sports training methods largely rely on coaching or pre-recorded videos, generally lacking real-time, objective evaluation and feedback mechanisms. Users struggle to accurately assess the correctness of their movements and the appropriateness of training intensity, leading to poor training results and even sports injuries. Existing technologies separate training and evaluation, failing to collect and analyze physiological indicators and movement quality in real time, and unable to dynamically adjust training content or provide targeted corrections, resulting in a lack of scientific rigor, adaptability, and personalization in training.

[0081] Therefore, referring to Figure 1 In one embodiment of the present invention, the integrated training and testing method includes steps S100-S500, wherein:

[0082] S100: Respond to the command to create a training plan and generate the corresponding training items and test standards;

[0083] S200: Generate and display the corresponding training images according to the training items and testing standards;

[0084] S300: Obtain real-time vital signs data and movement posture information of the user when completing training movements based on training images;

[0085] S400: Based on real-time vital signs data and movement posture information, determine the training effect evaluation results;

[0086] S500: If the training effect evaluation result does not meet the preset achievement conditions, identify the training items that do not meet the standards and display the corresponding decomposed actions of the training items until the user completes all training items and meets the preset achievement conditions.

[0087] A training plan is a holistic exercise program designed for a user, encompassing a series of training goals, training items, training duration, and intensity. This plan can be customized based on user needs or system evaluation results. A training item refers to a specific exercise unit within the training plan, such as a particular physical training movement, a set of strength training exercises, or a segment of aerobic exercise. Each training item typically corresponds to specific movement requirements and training goals. Testing standards are quantitative indicators used to measure whether a user has met the preset requirements when completing a training item. These standards can include multiple dimensions such as movement accuracy, heart rate zones, and energy expenditure. Training videos are visual content used to guide users in their training, and can include elements such as standard movement demonstrations, virtual coaching guidance, and background scenes. These training videos help users understand and imitate training movements through visual presentation.

[0088] Real-time vital signs data refers to physiological indicators such as heart rate, blood oxygen saturation, and energy consumption continuously collected by sensors and other devices during user training. This real-time vital signs data reflects the user's immediate physical state during exercise. Movement posture information refers to the relative positions and movement trajectories of various body parts acquired through image or depth sensors during training. This movement posture information is used to analyze the standardization and accuracy of the user's movements. Training effect evaluation results refer to a comprehensive evaluation of the user's overall performance in completing the training program based on real-time vital signs data and movement posture information. This evaluation result is used to determine whether the user has achieved the training goals. Preset attainment conditions refer to the thresholds or standard set set before the start of the training plan or program to determine whether the training effect evaluation results are satisfactory. When the evaluation results meet these conditions, it indicates that the user has reached the standard. Decomposing movements refers to breaking down a complex training movement into several simple, easy-to-master sub-movements. When a user fails to reach the standard, learning the decomposed movements helps the user understand and correct errors, gradually mastering the complete movement.

[0089] In this embodiment, in response to the instruction to create a training plan, corresponding training programs and testing standards are first generated. The user can input their training goals, desired training duration, and preferred exercise type. Based on this input information, the control device matches and generates a series of specific training programs from a pre-set training program library and sets preliminary testing standards for each program. For example, if the user selects a "fat loss" goal and an "aerobic exercise" type, the control device may generate training programs such as "running" and "jumping rope," and set corresponding duration and intensity standards.

[0090] Subsequently, according to the training program and testing standards, corresponding training videos are generated and displayed. For example, the control device can retrieve instructional video clips or still image sequences that match the current training program from a preset video library. The content of these training videos is presented to the user to intuitively guide the user in understanding and performing training actions.

[0091] As the user completes the training movements based on the training video, the control device acquires the user's real-time vital signs and posture information. For example, the user can wear a basic heart rate monitoring device and manually input their heart rate reading. Simultaneously, a regular camera can record the user's training process for subsequent manual or simple image recognition analysis of the user's movements.

[0092] Next, based on real-time vital signs data and movement posture information, the training effect evaluation result is determined. For example, the control device can perform a simple comparison between the heart rate data manually entered by the user and the heart rate zone set in the training program to determine whether it is within the target range. For movement posture, it can be determined whether the user has completed the main movement elements through manual observation or simple image analysis based on preset rules. Thus, a preliminary evaluation result is generated, such as "meets the standard" or "does not meet the standard".

[0093] Finally, if the training effectiveness evaluation result does not meet the preset achievement criteria, the control device identifies the substandard training item and displays the corresponding breakdown of the training item until the user completes all training items and meets the preset achievement criteria. Specifically, when the evaluation result indicates that the user has failed to meet the criteria, the control device identifies the specific training item that caused the failure. For this substandard item, the control device retrieves a simplified version or step-by-step instructional video from the preset teaching resources and displays it to the user. After the user watches and attempts these breakdown actions, the control device performs another simple monitoring and evaluation until the user can correctly complete the item and meet the preset standard.

[0094] This application's integrated training and assessment method effectively solves the problem of the separation of training and assessment in traditional training methods by organically combining the generation of training plans, guidance from training images, acquisition of users' real-time vital signs and movement postures, evaluation of training effects, and decomposed movement guidance when standards are not met. This integrated training and assessment method can provide users with continuous tracking of training status and accurate judgment of effects, and provide targeted corrective guidance when users perform poorly, thus forming a closed loop of "training-monitoring-evaluation-correction". This significantly improves the safety, effectiveness, and intelligence level of users training at home or alone, and avoids poor training effects or sports injuries caused by subjective judgment.

[0095] This invention, in response to a training plan creation command, generates corresponding training items and testing standards. Then, according to the training items and testing standards, it generates and displays corresponding training videos. Next, it acquires real-time vital sign data and posture information of the user as they complete the training movements based on the training videos. Based on this data, it determines the training effect evaluation result. If the evaluation result does not meet preset achievement conditions, it identifies the substandard training items and displays their corresponding decomposed movements until the user completes all training items and meets the preset achievement conditions. This achieves a closed-loop training process, solving the problem of separation between training and evaluation, and enabling closed-loop integration of training and evaluation, improving the scientific rigor and adaptability of training.

[0096] In implementation Figure 1 If the multi-dimensional data is not processed and comprehensively considered during the implementation of the solution, the evaluation results may not be comprehensive or accurate enough, making it difficult to effectively reflect the user's true training status and action standardization, thereby affecting the subsequent training guidance effect.

[0097] Therefore, referring to Figure 2 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The illustrated embodiment determines the training effect evaluation result based on real-time vital sign data and movement posture information, including steps S410-S470, wherein:

[0098] S410. Extract the current heart rate data, exercise energy consumption value and blood oxygen saturation data corresponding to the real-time vital signs data;

[0099] S420. Compare the current heart rate data with the target heart rate zone corresponding to the training program to determine the heart rate achievement rate.

[0100] S430. Compare the energy consumption value with the target energy consumption threshold corresponding to the training program to determine the energy consumption completion rate.

[0101] S440. Compare the blood oxygen saturation data with the blood oxygen standard range corresponding to the training program to determine whether the blood oxygen level meets the standard.

[0102] S450: Obtain the skeletal key points of the motion posture information, and build a three-dimensional posture model based on the skeletal key points to generate a user motion trajectory sequence;

[0103] S460. Perform dynamic time-normalization comparison between the user's action trajectory sequence and the standard action trajectory sequence corresponding to the training project to generate an action standardization score.

[0104] S470, weighted and integrated with heart rate target achievement rate, energy consumption completion rate, blood oxygen target achievement and movement standardization score, generates training effect evaluation results.

[0105] In extracting real-time vital sign data, including current heart rate, exercise energy expenditure, and blood oxygen saturation, this embodiment aims to separate and identify key indicators directly related to the user's physiological state and exercise intensity from the raw real-time vital sign data. Current heart rate reflects the frequency of the user's cardiac activity and is an important parameter for measuring exercise load and cardiovascular response; exercise energy expenditure quantifies the energy consumed by the user during training and is directly related to the intensity and duration of training; blood oxygen saturation represents the oxygen content in the blood and is a key indicator for assessing respiratory system function and the body's oxygen supply capacity. The extraction of this real-time vital sign data forms the basis for subsequent multi-dimensional assessments. This can be achieved through a pre-set data parsing module that performs data cleaning, format conversion, and feature extraction based on the output formats and data types of different vital sign sensors.

[0106] The evaluation then assesses whether the user's heart rate remained within the expected healthy and effective training range during training. The target heart rate range is pre-defined based on the nature and intensity of the training program, as well as the user's individual physiological characteristics (such as age and maximum heart rate). By comparing real-time heart rate data with this range, the percentage of time the user's heart rate remained within the target range or the degree of deviation can be calculated, thus quantifying the heart rate achievement rate. This helps determine whether the user has achieved the expected aerobic or anaerobic training effect, avoiding training intensity that is too low or too high.

[0107] Next, the degree of match between the user's actual energy expenditure during training and the goals set in the training program is measured. The target energy expenditure threshold is preset based on the training program's goals (such as fat loss, muscle gain, etc.) and duration. By comparing these parameters, the energy expenditure completion rate can be obtained, for example, expressed as a percentage. This helps assess the effectiveness of the training, ensuring that the user has reached the predetermined amount of exercise, thereby achieving the training goals.

[0108] Determining blood oxygen levels is crucial for monitoring a user's blood oxygen saturation during exercise to ensure optimal health. The standard blood oxygen saturation range typically represents a healthy physiological range; excessively low levels may indicate insufficient oxygen supply or respiratory problems. By comparing real-time blood oxygen saturation data to this standard range, it's possible to determine if a user's blood oxygen levels are normal during training, thereby assessing their physical fitness and safety.

[0109] Generating user motion trajectory sequences aims to extract refined motion features from user posture information. Skeletal keypoints refer to specific coordinate points of human joints, torso, and other parts, which are obtained from image data using image recognition technologies (such as deep learning models). Based on these two-dimensional or three-dimensional skeletal keypoints, a three-dimensional posture model of the user in space can be constructed, and then the trajectory of these keypoints changing over time can be tracked to form a user motion trajectory sequence. This user motion trajectory sequence can accurately describe every detail of the user's movements and motion paths during training, providing basic data for subsequent evaluation of motion standardization.

[0110] A motion standardization score is generated to quantify the similarity and standardization between a user's motion and a standard motion. The standard motion trajectory sequence is a pre-recorded or modeled sequence of professionally demonstrated motions. Dynamic Time Warping (DTW) is an algorithm that effectively compares the similarity of two time series, even if they have non-linear deformations along the time axis. Using the DTW algorithm, the distance or similarity between the user's motion trajectory sequence and the standard motion trajectory sequence can be calculated and converted into a motion standardization score. This score intuitively reflects the accuracy and standardization of the user's motion and is a key indicator for evaluating training quality.

[0111] Generating training effectiveness evaluation results aims to synthesize multiple evaluation metrics into a comprehensive and quantitative assessment. Weighted fusion means assigning different weights to different metrics. These weights can be preset or dynamically adjusted based on the training program's objectives, individual user differences, or expert experience. For example, for training emphasizing cardiopulmonary function, the weight of heart rate target achievement rate might be higher; for training emphasizing movement accuracy, the weight of movement standardization score might be higher. In this way, a comprehensive score or grade can be obtained, more comprehensively and objectively reflecting the user's overall training effectiveness.

[0112] This embodiment provides a comprehensive and detailed assessment of a user's training effectiveness from two core dimensions: physiological load and movement execution. By extracting and comparing real-time vital signs such as heart rate, energy expenditure, and blood oxygen saturation, it objectively reflects whether the user's physiological responses and exercise intensity during training have met expectations, ensuring the effectiveness and safety of the training. Furthermore, by acquiring skeletal key points and constructing a three-dimensional posture model, combined with dynamic time warping comparison, it accurately quantifies the standardization and accuracy of the user's movements, overcoming the shortcomings of traditional assessment methods in judging movement details. Finally, these multi-dimensional indicators are weighted and fused to generate a comprehensive training effectiveness assessment result. This not only provides quantitative feedback but also more accurately identifies the user's strengths and weaknesses during training, providing a scientific basis for subsequent personalized training adjustments and guidance, significantly improving the assessment accuracy and guidance effectiveness of the integrated training and testing method.

[0113] Relying solely on a single or vague data collection method may result in insufficient data dimensions or low precision, thus affecting the comprehensiveness and accuracy of subsequent training effect evaluation. For example, relying solely on visual capture of movements and postures may fail to reflect the user's physiological load and physical state during training, leading to a lack of depth and objectivity in the evaluation results.

[0114] Therefore, referring to Figure 3 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The illustrated embodiment acquires real-time vital sign data and movement posture information of the user when performing training actions based on training images, including steps S310-S360, wherein:

[0115] S310. Establish a communication connection with the wearable device and acquire the user's electrocardiogram signal, body temperature changes, and triaxial acceleration data;

[0116] S320. Determine the energy consumption data corresponding to the triaxial acceleration data;

[0117] S330. Extract features from electrocardiogram signals and body temperature changes to generate the first biological sign feature vector;

[0118] S340. Generate real-time vital sign data based on the first biological vital sign feature vector and energy consumption data;

[0119] S350: Acquire image data of the user performing training actions through an image acquisition device;

[0120] S360 performs human posture recognition on image data, extracts the coordinate sequence of skeletal key points, and generates motion posture information.

[0121] Establishing a communication connection with wearable devices and acquiring the user's electrocardiogram (ECG) signals, body temperature changes, and triaxial acceleration data involves pairing the integrated training and testing device with the user's smart wearable device (such as a smartwatch, smart bracelet, or smart clothing) using wireless communication technologies such as Bluetooth Low Energy (BLE) and Wi-Fi. Once the connection is established, the integrated training and testing device can receive raw data in real time from the wearable device's ECG, body temperature, and triaxial acceleration sensors. ECG signals reflect the user's cardiac activity, body temperature changes indicate the user's physiological thermal load, and triaxial acceleration data accurately records the user's body's motion state and intensity in three-dimensional space.

[0122] Determining energy expenditure data involves processing and analyzing triaxial acceleration data acquired during user exercise using a specific algorithm model. This model can convert acceleration data into estimated energy expenditure values ​​based on the relationship between exercise intensity and energy expenditure, combined with the user's basic physiological parameters (such as weight, height, and age). For example, methods based on metabolic equivalents (METs) can be used, or machine learning models can be employed to identify acceleration patterns, thereby quantifying the calories or energy consumed by the user during training.

[0123] Generating a first biometric feature vector involves preprocessing and analyzing raw electrocardiogram (ECG) signals and body temperature change data to extract representative physiological features. For ECG signals, indicators such as heart rate (HR) and heart rate variability (HRV) can be extracted; for body temperature changes, average body temperature, temperature fluctuation range, and rate of temperature rise can be extracted. These extracted features, after quantification and encoding, form a multi-dimensional first biometric feature vector, used to comprehensively characterize the user's physiological state.

[0124] Generating real-time vital sign data refers to integrating the extracted primary biometric feature vector with the calculated energy consumption data. This integration can be achieved through data fusion techniques, such as normalizing various indicators and then performing a weighted sum according to preset weights, or using machine learning models for feature fusion, thereby generating a unified and comprehensive set of real-time vital sign data. This set of real-time vital sign data can reflect the user's cardiovascular load, thermal load, and energy consumption resulting from overall exercise intensity during training.

[0125] Acquiring image data during training movements refers to using image acquisition devices (such as high-definition cameras, depth sensors, etc.) integrated into or connected to training and testing equipment to capture the user's motion images in real time while the user is performing training movements. This image data contains visual information about the user's movements and forms the basis for subsequent movement posture analysis.

[0126] Extracting the coordinate sequence of skeletal key points to generate motion posture information refers to inputting the acquired image data into a pre-trained human posture recognition model. This model can automatically detect and identify various skeletal key points of the human body in the image (such as the head, neck, shoulders, elbows, wrists, hips, knees, and ankles), and output the coordinate sequence of these key points in two-dimensional or three-dimensional space. These skeletal key point coordinate sequences accurately describe the user's body posture at different points in time, thus forming detailed motion posture information.

[0127] This embodiment establishes a communication connection with wearable devices and utilizes image acquisition equipment to synchronously and multidimensionally acquire real-time physiological data and precise movement posture information of the user during training. Specifically, the electrocardiogram (ECG) signals, body temperature changes, and triaxial acceleration data acquired through the wearable device, after feature extraction and energy consumption calculation, can generate real-time vital sign data including heart rate, body temperature characteristics, and energy consumption, comprehensively reflecting the user's physiological load and physical state. Furthermore, the image data acquired through the image acquisition equipment, after human posture recognition, can extract the coordinate sequence of skeletal key points, thereby accurately quantifying the user's movement posture. This comprehensive and refined data acquisition method overcomes the limitations of a single data source, providing a richer, more accurate, and objective data foundation for subsequent training effect evaluation. Based on this high-quality real-time vital sign data and movement posture information, the system can more accurately determine whether the user has achieved the training goals, identify deviations in movement, and provide more targeted feedback and guidance, thereby significantly improving the scientific rigor and effectiveness of training evaluation.

[0128] In practical applications, if the training images are generated in a simplistic or uninterrupted manner, it may be difficult to effectively motivate users and provide an immersive training experience, thereby affecting users' understanding and execution of standardized training movements.

[0129] Therefore, referring to Figure 4 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The illustrated embodiment generates and displays corresponding training images according to training items and testing standards, including steps S210-S230, wherein:

[0130] S210. Based on the movement type of the training project and the number of target groups and repetitions in the test standard, retrieve the corresponding standard three-dimensional human animation sequence from the preset movement image library.

[0131] S220, in response to the selection operation of virtual coach image, background scene and rhythmic background music, merges and renders the selected virtual coach image with the standard three-dimensional human body animation sequence to generate training video;

[0132] The S230 outputs training images and rhythmic background music to the display component in sync, guiding users to follow the beat and complete standardized training movements.

[0133] When generating and displaying training videos according to training programs and testing standards, the control device first retrieves the corresponding standard 3D human animation sequences from a preset action video library based on the movement type of the training program and the target number of sets and repetitions in the testing standards. The movement type refers to the specific classification of the training movement, such as squats, push-ups, yoga postures, etc. The control device selects the corresponding standard movement according to the movement type specified in the training plan. The target number of sets and repetitions are quantitative indicators set in the training plan, such as "3 sets of squats, 10 repetitions per set." This information is used to determine the playback length and repetition count of the animation sequence. The preset action video library is a database storing a large amount of 3D animation data of standard training movements. Each movement may contain animation sequences with multiple perspectives and different difficulty levels. The standard 3D human animation sequences are virtual human movement sequences recorded or modeled by professional coaches or kinesiology experts, conforming to ergonomics and exercise standards. They are presented in a 3D model format, allowing observation from any angle and providing users with accurate movement demonstrations. This step aims to accurately match and extract the most suitable standard movement demonstrations from a rich resource library based on specific training requirements, laying the foundation for subsequent training video generation.

[0134] Subsequently, in response to the selection of the virtual coach avatar, background scene, and rhythmic background music, the control device blends and renders the selected virtual coach avatar with a standard 3D human animation sequence to generate a training video. The virtual coach avatar refers to the virtual character model that the user can choose, which can be a virtual character of different genders, ages, and styles. The background scene refers to the environment in which the virtual coach is located in the training video, such as a gym, outdoor park, or sci-fi space, which the user can choose according to personal preference. The rhythmic background music refers to music with a specific beat and rhythm, used to help users master the rhythm and frequency of training movements, enhancing the fun and immersion of training. The selection operation refers to the user specifying the virtual coach avatar, background scene, and background music through interface interaction (such as clicking or selecting). The blending and rendering is a graphics processing process that combines the user-selected virtual coach avatar model with the previously retrieved standard 3D human animation sequence and places it in the selected background scene. Using computer graphics technology, it generates realistic 3D animation images, ultimately forming a dynamic video or animation that includes virtual coach demonstration movements, a specific background, and possibly visual effects. This step enhances the user's training experience by introducing personalized and immersive elements, transforming standard movement demonstrations into more engaging and instructive training videos.

[0135] Finally, the control device synchronously outputs the training video and rhythmic background music to the display component, guiding the user to follow the beat and complete standardized training movements. Synchronized output ensures precise alignment of the training video playback with the rhythmic background music's timeline, making the virtual coach's movements synchronize with the music's beat. The display component refers to the hardware device used to display the training video, such as a monitor, television screen, or VR / AR headset. Through dual guidance of vision (training video) and hearing (rhythmic background music), the system helps users accurately grasp the rhythm, speed, and standardization of the movements, thereby improving training effectiveness.

[0136] This embodiment accurately retrieves standard 3D human animation sequences from a preset motion image library based on the movement type, target number of sets, and repetitions of the training program, ensuring the professionalism and accuracy of the training guidance. It then responds to the user's selection of a virtual coach avatar, background scene, and rhythmic background music, blending the selected elements with the standard 3D human animation sequences to generate personalized training images, greatly enhancing the fun and immersion of the training. Furthermore, synchronously outputting the training images and rhythmic background music to the display component provides users with dual visual and auditory guidance, helping them better grasp the rhythm and standardization of movements. This effectively improves user compliance and training effectiveness in completing standardized training movements, solving the problem of low user engagement caused by the monotony and lack of interactivity in traditional training images.

[0137] In practical applications, simply displaying the breakdown of actions may not be effective in guiding users to their specific weaknesses. Users may still find it difficult to understand the root cause of their own action deviations, leading to repeated attempts and low efficiency, making it difficult to quickly reach the preset target conditions.

[0138] Therefore, referring to Figure 5 The present invention also provides an integrated training and testing method in one embodiment, based on the above. Figure 1 In the illustrated embodiment, if the training effect evaluation result does not meet the preset achievement conditions, the training items that do not meet the standards are identified, and the corresponding decomposed actions of the training items are displayed until the user completes all training items and meets the preset achievement conditions, including steps S510-S560, wherein:

[0139] S510. When the overall score of the training effect evaluation result is lower than the preset target score, the training effect evaluation result is determined to be that the preset target condition is not met.

[0140] S520. Analyze the scores of each sub-indicator in the training effect evaluation results, identify one or more training items with the lowest scores, and mark them as substandard training items.

[0141] S530: For each training item that fails to meet the standard, retrieve the corresponding multi-angle slow-speed breakdown video of the movement from the teaching video library.

[0142] S540. Prioritize playing the breakdown animation video explaining the breakdown animation in the display component.

[0143] S550: During the playback of the decomposed motion video, reacquire the user's real-time vital signs data and motion posture information while watching the decomposed motion video.

[0144] S560. Based on the reacquired real-time vital sign data and movement posture information, generate a secondary assessment result for the substandard training item. If the secondary assessment result meets the preset pass criteria, mark the item as passed; otherwise, repeatedly play the breakdown video until passed.

[0145] The training effectiveness evaluation result is typically a quantitative indicator, which can be a comprehensive score calculated by weighting multiple sub-indicators. The preset pass / fail score is a threshold set in advance by the control device to determine whether the training effect has met the expected standard. When the comprehensive score falls below this threshold, the control device determines that the user's current training effect is substandard and requires guidance and correction. This mechanism ensures the objectivity and quantifiability of the evaluation.

[0146] Training effectiveness evaluation results typically include scores for multiple sub-indicators, such as posture scores, heart rate performance scores, and energy consumption scores for different movement phases. By meticulously analyzing these sub-indicator scores, the control device can accurately identify one or more training items where the user performed worst and scored lowest. These identified items are marked as substandard training items, thus providing clear targets for subsequent targeted guidance. This fine-grained analysis helps to accurately pinpoint the user's weaknesses.

[0147] Building upon this foundation, the instructional video library is a pre-stored database containing a vast amount of training movement instructional resources. When the control device identifies a substandard training item, it retrieves and displays specially produced multi-angle, slow-motion breakdown videos of the movement from the instructional video library, based on the item's identifier. These videos typically showcase every detail of the movement from different perspectives (such as front, side, and back views), played in slow motion, and may even include text or audio narration to ensure that users can clearly observe and understand the correct posture, point of force application, and rhythm of the movement.

[0148] Furthermore, the display component can be a smart screen, projector, or VR / AR device, etc. After identifying the substandard training items and retrieving the corresponding breakdown of the movements in video, the control device will instruct the display component to prioritize playing these breakdown videos. This means that before the user performs corrective training, the control device will ensure that the user first watches and understands the correct movement techniques, rather than immediately moving on to the next attempt, thus providing the user with clear visual guidance.

[0149] In addition, real-time vital signs data include heart rate, blood oxygen saturation, and energy expenditure, while movement and posture information is obtained by acquiring key skeletal data from the user through image acquisition equipment. The control device continuously reacquires this data while the user views the breakdown of the movement images. This approach aims to monitor the user's physiological responses and initial movement attempts during the learning and imitation phases, providing more comprehensive data support for subsequent secondary assessments, and even making a preliminary judgment on whether the user is learning diligently or engaging in simple imitation exercises.

[0150] Finally, after the user watches the breakdown of the movements and may attempt an initial imitation, the control device will conduct a secondary evaluation of the substandard training item based on the re-acquired real-time vital sign data and movement posture information. This secondary evaluation is similar to the initial evaluation mechanism, but focuses more on the user's performance after understanding and correction. If the secondary evaluation result meets the preset pass criteria, the control device will mark the training item as passed, indicating that the user has mastered the movement. Conversely, if the user still fails to meet the criteria, the control device will repeatedly play the breakdown of the movements explanation video, forming a closed-loop, iterative learning and correction process until the user fully masters and meets the criteria.

[0151] This embodiment enables more refined evaluation and targeted guidance of user training effectiveness. First, by setting a comprehensive score and a preset passing score, the determination of training effectiveness becomes more objective and quantifiable. Second, by analyzing the scores of various sub-indicators, it can accurately identify the user's weaknesses in specific training items, avoiding blind correction. Targeted, multi-angle, slow-motion breakdown videos of the movements are retrieved and prioritized for playback, providing users with clear and easy-to-understand corrective teaching content. More importantly, real-time vital sign data and movement posture information of the user are reacquired during the playback of the breakdown videos, allowing the control device to monitor the user's state during learning and imitation in real time, providing comprehensive data support for subsequent secondary evaluation. Finally, through secondary evaluation based on the reacquired data and an iterative mechanism of replaying the explanation videos when standards are not met, a closed-loop, personalized error-correction learning process is formed. This significantly improves the efficiency and accuracy of users correcting incorrect movements, ensuring that users can truly master and achieve the preset training goals, thus effectively solving the problem of insufficient guidance effect from simply displaying breakdown movements.

[0152] If training programs and testing standards are generic or pre-set, they may not fully take into account the user's individual needs, physical condition, and training preferences, resulting in insufficient targeting of the training plan and affecting training effectiveness and user experience.

[0153] Therefore, referring to Figure 6 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The illustrated embodiment generates corresponding training items and test standards in response to a command to create a training plan, including steps S110-S140, wherein:

[0154] S110. In response to the instruction to create a training plan, determine the training objective selected by the user, the available training time, and the preferred type of exercise, and generate the corresponding training objective.

[0155] S120. Match the training objective with the knowledge graph of the preset training objective to determine the training parts and ability dimensions corresponding to the training objective.

[0156] S130. Determine the number and combination of training items based on the training body parts, ability dimensions, and available training time.

[0157] S140. Based on the preferred type of exercise, the number of training items, and the combination of training items, determine the training items and testing standards.

[0158] In response to a command to create a training plan, the system determines the user's selected training goal, available training time, and preferred exercise type, generating a corresponding training goal. This aims to capture the user's personalized needs and preferences when creating the training plan, serving as the basis for subsequent intelligent plan generation. The control device provides a user interface, allowing users to input or select their current training goal, such as fat loss, muscle gain, or improved endurance; desired available training time, such as 30 minutes or 1 hour; and preferred exercise type, such as yoga, strength training, or aerobic exercise. This input information is received and integrated by the control device to form a clear training goal.

[0159] The process involves matching the training objective with a knowledge graph of preset training objectives to determine the corresponding training body parts and ability dimensions. This step utilizes a structured knowledge graph to transform the user's input natural language or selected training objective into specific training requirements related to body parts and ability dimensions that the control device can understand and process. The knowledge graph of preset training objectives can store various relationships between training objectives and body parts and ability dimensions. For example, "fat loss" may be associated with whole-body aerobic exercise and core strength, while "muscle gain" may be associated with strength training for specific muscle groups. After receiving the user-generated training objective, the control device uses techniques such as semantic matching, keyword extraction, or graph traversal to find the most matching node in the knowledge graph, thereby determining the main body parts involved in the user's training and the ability dimensions that need to be improved.

[0160] Based on the user's personalized training needs (training areas, ability dimensions) and time constraints, the system intelligently plans the macro-structure of the training program, including the number of training programs and their arrangement. The control device can have a built-in rule engine or optimization algorithm. For example, if the training area is the whole body, the ability dimension is endurance, and the available time is 60 minutes, the control device might recommend 3-5 aerobic training programs, considering the transitions and rest periods between programs. If the training area is the upper limbs, the ability dimension is strength, and the available time is 30 minutes, the control device might recommend 2-3 strength training programs targeting upper limb muscles, optimizing the number of sets and repetitions. This process aims to balance training effectiveness with the user's time constraints.

[0161] Based on the user's preferred exercise type, the number of training programs, and the combination of training programs, the training programs and testing standards are determined. This is the final step in generating specific training programs and corresponding testing standards. It refines the previously determined macro structure into executable training content and ensures that it conforms to the user's exercise preferences. The control device selects programs from the yoga movement library if the user prefers yoga, and from the machine or bodyweight training library if the user prefers strength training. Combining the determined number and combination of training programs, specific training movements are selected from the preset movement library. In addition, for each selected training program, the control device automatically generates corresponding testing standards based on its nature, the user's ability dimension, and training goals. For example, the target number of sets, repetitions, completion time, and heart rate zone for a certain movement, to ensure the measurability and attainment of training goals.

[0162] This embodiment responds to commands to create a training plan by first determining the user's selected training objective, available training time, and preferred exercise type, and then generating a corresponding training objective. Subsequently, it matches this training objective with a knowledge graph of preset training objectives to intelligently determine the specific training body parts and ability dimensions. Based on this, the control device rationally determines the number and combination of training items according to these personalized training body parts, ability dimensions, and the user-set available training time. Finally, combining the user's preferred exercise type, the control device accurately generates training items and corresponding testing standards that meet the user's personalized needs. This approach overcomes the limitations of traditional training plan generation methods in terms of generality and lack of personalization, making the generated training plans more targeted, scientific, and user-friendly. Users do not need to manually select complex training parameters; the control device can intelligently recommend based on their simple input, significantly improving the customization of training plans and user experience, thereby helping to improve training adherence and final training effectiveness.

[0163] Compared to Figure 6 In the illustrated embodiment, if the training plan generation process fails to fully consider the user's individual differences and actual physical condition, such as relying solely on the user's subjective input or a general template, the training plan may not match the user's actual needs, thereby affecting the training effect, and may even cause physical injury due to inappropriate training intensity.

[0164] Optionally, refer to Figure 7 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The illustrated embodiment, in response to a command to create a training plan, generates corresponding training items and test standards, including steps S150-S190, wherein:

[0165] S150, In response to the instruction to create a training plan, obtain the user's identity information;

[0166] S160. Obtain historical physical fitness test data, past training records and body shape indicators corresponding to the identity information;

[0167] S170: Analyze historical physical fitness test data, past training records and body shape indicators to assess the user's current overall physical fitness level and weaknesses;

[0168] S180. Based on the overall physical fitness level, select a basic training program template of corresponding difficulty from the preset training program library;

[0169] S190. Based on the weak links, strengthen the corresponding departments in the basic solution template, adjust the corresponding test standards, and generate the adjusted training projects and test standards.

[0170] When the control device receives an instruction from a user to create a training plan, it first needs to verify the identity of the user currently operating. This can be achieved in various ways, such as through account information provided during user login, the identification code of the smart device bound to the user, or directly through biometric technologies (such as fingerprint or facial recognition). The purpose of obtaining identity information is to accurately associate and retrieve the user's unique data subsequently.

[0171] After obtaining the user's identification information, the control device retrieves and obtains historical data related to that user from the storage control device based on this identification information. Historical physical fitness test data may include the user's physical fitness test results at different times, health check reports, etc.; past training records cover the user's previously completed training programs, training duration, training intensity, completion rate, and evaluation results; body morphology indicators may include height, weight, body fat percentage, muscle mass, and circumference. This data is an important basis for comprehensively assessing the user's current physical condition and athletic ability.

[0172] The control device performs in-depth analysis of the acquired multi-dimensional historical data. This analysis can utilize data mining, machine learning algorithms, or expert control device rules to comprehensively evaluate the user's physical performance, exercise habits, and body composition. For example, by comparing historical physical test data, it can determine the user's progress or decline in physical fitness; by analyzing past training records, it can identify the user's strengths and weaknesses in specific training programs; and by using body shape indicators, it can assess whether the user's body composition is balanced. Ultimately, the control device can quantify the user's overall physical fitness level (e.g., beginner, intermediate, advanced) and accurately identify their weaknesses in areas such as strength, endurance, flexibility, and balance.

[0173] The control device has a pre-set training program library containing various basic program templates for different fitness levels and training goals. After assessing the user's overall fitness level, the control device intelligently selects a basic program template from the library that matches the user's current fitness level. For example, for beginners, a lower-intensity, simpler introductory template is selected; for advanced users, a higher-intensity, more complex advanced template is selected. This matching ensures that the starting point of training matches the user's actual ability, avoiding training intensity that is too high or too low.

[0174] After selecting a basic training program template, the control device will personalize the template based on previously identified user weaknesses. For example, if the user is weak in core strength, the control device may add more core training exercises to the basic program, or increase the number of sets and repetitions of existing core training exercises; if the user lacks flexibility, stretching and flexibility training will be added. Furthermore, to ensure the effectiveness of the reinforcement, the control device will also adjust the testing standards associated with these reinforcement programs accordingly, making them more challenging or more focused on improving weaknesses. Ultimately, through this fine-tuning, a fully personalized training program and testing standards are generated.

[0175] This embodiment goes beyond simply relying on the universality of user-created instructions; it delves deeper into the user's individual data, including historical physical fitness test data, past training records, and body morphology indicators. Through intelligent analysis of this data, the control device can accurately assess the user's overall physical fitness level and identify their weaknesses. Based on this, the control device can select the most suitable basic training program template from a preset training program library, specifically strengthening the user's weaknesses and adjusting the corresponding testing standards. This personalized, data-driven training plan generation method ensures that the generated training programs and testing standards are highly matched to the user's actual physical condition, thereby significantly improving the effectiveness and safety of training. Users can receive more targeted training guidance, avoiding the risks of blind training or inappropriate training intensity, and thus achieving training goals more efficiently.

[0176] In actual training, focusing only on the performance of a single training session may not fully reflect the user's physical recovery status and training load adaptability, and there is a risk that the user may suffer physical injury or poor training results due to overtraining.

[0177] Therefore, referring to Figure 8 Another embodiment of the present invention provides a training and testing integrated method, based on the above. Figure 1 The embodiment shown further includes steps S600-S800, wherein:

[0178] S600: Establish a communication connection with the user's smart wearable device, obtain the user's resting heart rate, sleep duration and quality, body temperature data, and daily steps, and generate corresponding long-term body data.

[0179] S700 uses preset interval duration analysis to determine the long-term trend of physical data and the user's recovery status and training load adaptation.

[0180] S800 If it detects that the user's resting heart rate is abnormally high or sleep quality is significantly reduced, it generates a corresponding training load overload warning and sends the training load overload warning to the user terminal and the coach management terminal.

[0181] To comprehensively understand the user's physical condition, this embodiment first establishes a communication connection with the user's smart wearable device. This smart wearable device can be a smartwatch, smart bracelet, or smart ring, and it interacts with the integrated training and testing device via wireless communication protocols such as Bluetooth and Wi-Fi. Through this connection, the control device can continuously acquire key physiological indicators such as the user's resting heart rate, sleep duration and quality, body temperature, and daily steps. Resting heart rate is an important indicator of cardiovascular health and physical recovery status, and its changes reflect the degree of fatigue; sleep duration and quality are directly related to the body's recovery efficiency; body temperature data can indicate the body's stress response or potential health problems; and daily steps reflect the user's daily activity level. The continuous collection and integration of these multi-dimensional data constitutes the user's long-term physical data profile, laying the foundation for subsequent physical condition assessments.

[0182] Based on this, the control device dynamically analyzes the acquired long-term physical data at preset intervals. These preset intervals can be flexibly configured according to the actual application scenario, such as daily, weekly, or monthly. The analysis focuses on the trend of data changes, rather than just the data value at a single point in time. For example, if the resting heart rate shows a continuous upward trend, it may indicate insufficient physical recovery or excessive training load; if the sleep quality score continues to decline, it may indicate fatigue. By using machine learning algorithms or expert control devices to comprehensively evaluate the trends of these long-term data changes, it is possible to accurately determine the user's current physical recovery status (e.g., whether fatigued, well-recovered, etc.) and their adaptation to the current training load (e.g., whether they can withstand the current training intensity, whether there is a risk of overload, etc.).

[0183] To promptly address potential overload risks, the control device will immediately generate a training load overload warning if it detects a sustained abnormal increase in the user's resting heart rate or a significant decrease in sleep quality. The criteria for "sustained abnormal increase" or "significant decrease" can be preset; for example, a resting heart rate exceeding a certain percentage above its baseline average for several consecutive days, or a sleep quality score falling below a certain threshold below its average for several consecutive days. Once these preset conditions are triggered, the control device will automatically generate a warning message containing specific abnormal indicators and suggested adjustments. This warning message will not only be sent to the user's terminal (such as the user's mobile application or the display screen of the integrated training and testing device) to remind the user to pay attention to their physical condition and take appropriate measures, but will also be sent to the coach management terminal, enabling coaches to promptly grasp the trainees' physical status and adjust training plans accordingly, avoiding physical injury or poor training results due to overtraining.

[0184] This embodiment, building upon real-time training assessment, further incorporates the monitoring and analysis of the user's long-term physical data. By establishing a communication connection with the smart wearable device, it continuously acquires key physiological indicators such as the user's resting heart rate, sleep duration and quality, body temperature, and daily steps. At preset intervals, it conducts in-depth analysis of the trends in these long-term physical data, enabling a comprehensive and dynamic assessment of the user's physical recovery status and adaptation to training load. When the control device detects signs of overload, such as a persistently abnormally high resting heart rate or a significant decline in sleep quality, it can promptly generate an overload warning and send it to the user terminal and the coach management terminal. This allows users and coaches to promptly understand the potential risks of overtraining, proactively adjust training plans or arrange rest periods, effectively avoiding physical injury and decreased training effectiveness caused by overtraining, ensuring the scientific, safe, and effective nature of training, and improving the user's overall training experience and health level.

[0185] The present invention also proposes a control device, which includes: a memory, a processor, and a control program for the integrated training and testing device stored in the memory and executable on the processor. The control program for the integrated training and testing device is configured to implement the above-described integrated training and testing method.

[0186] It is worth noting that since the control device of the present invention is based on the above-mentioned integrated training and testing method, the embodiments of the control device of the present invention include all the technical solutions of all embodiments of the above-mentioned integrated training and testing method, and the technical effects achieved are exactly the same, so they will not be repeated here.

[0187] The present invention also proposes an integrated training and testing device, which includes a display component, an image acquisition component, a vital sign monitoring component, and a control device as described in at least one of the above embodiments, wherein:

[0188] The display component is used to present training images and feedback information; the image acquisition component is used to capture the user's movements and postures during training; the vital signs monitoring component is used to collect the user's heart rate, blood oxygen saturation, and energy consumption data in real time; the control device is electrically connected to the display component, the image acquisition component, and the vital signs monitoring component respectively; the control device is used to respond to the creation command for the training plan, generate the corresponding training items and test standards, and then generate the corresponding training images according to the training items and test standards, and control the display component to display the training images; the control device is also used to control the image acquisition component and the vital signs monitoring component to obtain the user's real-time vital signs data and movement and posture information when the user completes the training movements according to the training images, and to determine the training effect evaluation result based on the real-time vital signs data and movement and posture information. If the training effect evaluation result does not meet the preset achievement conditions, the unqualified training items are identified, and the decomposed movements corresponding to the training items are displayed, until the user completes all training items and meets the preset achievement conditions.

[0189] The display component can take various forms, such as LCD screens, OLED screens, projectors, or VR / AR headsets. Its main function is to intuitively display training images to users, including virtual coaches, standard movement demonstrations, and background scenes. It can also provide real-time feedback on training effect evaluation results, heart rate curves, energy consumption data, and breakdown images of movements corresponding to substandard training items. By configuring appropriate size, resolution, and refresh rate, the display component ensures users receive a clear and smooth visual experience, thereby better understanding and following training instructions. The image acquisition component is a key component for capturing the user's movements and postures during training. This component can use depth cameras (e.g., Kinect, Intel RealSense), high frame rate RGB cameras, or infrared camera arrays. Its working principle is to acquire the user's three-dimensional spatial information or two-dimensional image sequence in real time through optical or depth sensing technology, and then extract the coordinates of key skeletal points to generate movement and posture information. The high precision and real-time performance of this component are crucial for accurately identifying user movements and evaluating the standardization of movements. The vital signs monitoring component is responsible for collecting the user's physiological data in real time during training. This component can integrate various sensors, such as photoplethysmography (PPG) sensors or electrocardiogram (ECG) sensors for monitoring heart rate; pulse oximeters for measuring blood oxygen saturation; and inertial measurement units (IMUs) such as accelerometers and gyroscopes for estimating energy expenditure. These sensors can be designed as part of wearable devices (such as smart bracelets, smartwatches, and smart clothing), transmitting data to the control device via wireless communication protocols (such as Bluetooth and Wi-Fi), or directly embedded into the main structure of the integrated training and testing device to ensure continuous and accurate acquisition of vital sign data. The control device is the core of the entire integrated training and testing device, and its function is to coordinate the work of various components and execute the integrated training and testing method. This control device typically includes one or more processors (such as a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), or application-specific integrated circuit (ASIC)) and a memory. The processor is responsible for running the control program stored in the memory, performing complex computational tasks, such as performing human posture recognition on raw image data acquired by the image acquisition component, analyzing physiological data collected by the vital sign monitoring component, generating training effect evaluation results, and dynamically adjusting training content based on the evaluation results (such as displaying decomposed movements). The memory stores control programs, preset training plans, standard movement libraries, instructional video libraries, and users' historical training data. The control device interacts with the display components, image acquisition components, and vital sign monitoring components via electrical connections for high-speed data exchange and command control, ensuring real-time response and efficient operation of the entire control system.

[0190] Through the integrated design described above, this training and testing device provides a physically unified and functionally coordinated platform. The device can display training images in real time, accurately capture the user's posture and physiological data, and provide intelligent analysis and feedback through the control unit. This integrated design avoids the complexity of multi-device collaborative operation, simplifies the user experience, and ensures the synchronization and accuracy of data acquisition, thereby improving the immediacy and effectiveness of training guidance. When training results are unsatisfactory, the device can immediately display decomposed movements and provide targeted corrective guidance, significantly improving the user's training efficiency and achievement rate. Furthermore, the tight electrical connections between components and the unified scheduling of the control unit ensure smooth data flow and low processing latency, enabling users to obtain a seamless and immersive training and testing experience, effectively solving the problems of delayed feedback and inaccurate guidance in traditional training.

[0191] In one implementation, User A wants to improve core strength and flexibility using a smart fitness control device. The device first responds to User A's instruction to create a training plan. During this process, the device matches User A's input (e.g., User A selects "core strength" and "flexibility" as training goals, sets the available training time to 30 minutes, and prefers "yoga-style" exercises) with a knowledge graph of preset training goals to determine the target muscle groups (e.g., rectus abdominis, obliques, hamstrings) and ability dimensions (e.g., endurance, range of motion). Based on this information, the device identifies five training exercises (e.g., plank, Russian twist, cat-cow pose, downward-facing dog, seated forward bend) and their combinations, and generates specific test criteria for each exercise (e.g., plank hold time, number of Russian twists, range of motion for flexibility movements).

[0192] Alternatively, the control device can generate training plans automatically by acquiring data. When user A issues a training plan creation command, the control device obtains user A's identification information and retrieves their historical physical fitness test data, past training records, and body shape indicators. The control device analyzes this data to assess user A's current overall physical fitness level (e.g., moderate core strength, limited hamstring flexibility) and weaknesses. Based on the assessment results, the control device selects a basic training program template of corresponding difficulty from a preset training program library and strengthens the relevant parts of the template based on user A's weaknesses (e.g., adding hamstring stretching exercises). In addition, it adjusts the corresponding testing standards, ultimately generating personalized training programs and testing standards.

[0193] Next, the control device generates and displays corresponding training videos according to the generated training items and testing standards. For example, for the first training item, "plank," the control device retrieves the corresponding standard 3D human animation sequence from a preset motion video library based on the movement type and the target holding time set in the testing standards. User A can respond to the control device's prompts by selecting a virtual coach avatar, background scene (such as "beach"), and rhythmic background music. The control device blends and renders the selected virtual coach avatar with the standard 3D human animation sequence to generate the final training video, and outputs it synchronously with the rhythmic background music to the display component to guide user A to complete standardized training movements in time with the beat. Unlike traditional fitness applications that only provide pre-recorded videos, this control device can dynamically generate customized training videos based on the user's personalized choices and training plans, providing a more immersive training experience.

[0194] During user A's training movements based on training videos, the control device acquires the user's vital signs and posture information in real time. Specifically, the control device establishes a communication connection with a wearable device (such as a smartwatch) worn by user A to acquire the user's electrocardiogram (ECG) signals, body temperature changes, and triaxial acceleration data in real time. The control device processes the triaxial acceleration data to determine energy consumption data and extracts features from the ECG signals and body temperature changes to generate a first biometric feature vector. Combining this feature vector with the energy consumption data, the control device generates real-time vital sign data, including current heart rate, exercise energy consumption, and blood oxygen saturation. Furthermore, through an image acquisition device integrated into the display component, the control device acquires image data of user A performing the training movements, performs human posture recognition on the image data, extracts the coordinate sequence of skeletal key points, and generates posture information. This real-time, multimodal data acquisition method overcomes the limitation of existing technologies that lack objective data support during the training process.

[0195] Subsequently, the control device determines the training effect evaluation result based on real-time vital sign data and movement posture information. The control device extracts current heart rate, exercise energy expenditure, and blood oxygen saturation data from the real-time vital sign data. It compares the current heart rate data (e.g., 130 beats / minute) with the target heart rate range (e.g., 120-140 beats / minute) corresponding to the "plank" exercise to determine the heart rate achievement rate. Furthermore, it compares the exercise energy expenditure value (e.g., 50 kcal) with the target energy expenditure threshold (e.g., 45 kcal) to determine the energy expenditure completion rate. The blood oxygen saturation data (e.g., 98%) is compared with the standard blood oxygen range (e.g., 95-100%) to determine the blood oxygen achievement status. Regarding movement posture information, the control device acquires skeletal key points and constructs a three-dimensional posture model based on these key points, generating a movement trajectory sequence for user A. This trajectory sequence is dynamically time-warped and compared with the standard movement trajectory sequence corresponding to the "plank" exercise to generate a movement standardization score (e.g., 75 points, indicating that user A's hips slightly sank). Finally, the control device weighted and integrated scores from heart rate target achievement rate, energy expenditure completion rate, blood oxygenation target achievement, and movement standardization to generate a comprehensive training effect evaluation result (e.g., a total score of 78 points). This comprehensive and quantitative evaluation method significantly improves the scientific rigor and objectivity of training effect assessment.

[0196] If the training effect evaluation results do not meet the preset attainment criteria, the control device will identify the substandard training items and display the corresponding decomposed movements until user A completes all training items and meets the preset attainment criteria. For example, if the overall score of the "plank" item is 78 points, which is lower than the preset attainment score of 85 points, the control device will determine that this item does not meet the preset attainment criteria. The control device further analyzes the scores of each sub-indicator in the evaluation results, identifies the lowest score for movement standardization, and therefore marks "plank" as a substandard training item. For this substandard item, the control device retrieves the corresponding multi-angle slow-speed decomposed movement explanation video from the teaching video library. The display component prioritizes playing the decomposed movement video, demonstrating in detail how to properly tighten the core and keep the body in a straight line. During the playback of the decomposed movement video, the control device re-acquires the real-time vital sign data and movement posture information of user A while watching and attempting to correct the movements. Based on the re-acquired data, the control device generates a secondary evaluation result for the substandard training item. If the secondary evaluation result (e.g., 88 points) meets the preset passing criteria, the project is marked as passed; otherwise, the control device will repeatedly play the breakdown video explaining the actions and conduct a second evaluation until user A passes the project. This closed-loop mechanism of "training-monitoring-evaluation-correction" effectively solves the problems of separation between training and evaluation, lack of real-time feedback and targeted corrective guidance in existing technologies, significantly improving the effectiveness and safety of training.

[0197] Furthermore, this integrated training and testing method also includes long-term physical data monitoring. The control device establishes a communication connection with the smart wearable device worn by user A, acquiring the user's resting heart rate, sleep duration and quality, body temperature data, and daily steps to generate corresponding long-term physical data. The control device analyzes the changing trends of the long-term physical data at preset intervals (e.g., weekly) to determine user A's recovery status and training load adaptation. If user A's resting heart rate is detected to be abnormally high or sleep quality to be significantly reduced, the control device will generate corresponding training load overload warning information and send it to user A's terminal and the coach management terminal, thereby effectively preventing overtraining and potential sports injuries.

[0198] It is worth noting that since the integrated training and testing equipment of the present invention is based on the above-mentioned control device, the embodiments of the integrated training and testing equipment of the present invention include all the technical solutions of all the embodiments of the above-mentioned control device, and the technical effects achieved are exactly the same, which will not be repeated here.

[0199] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for integrating training and testing, characterized in that, The integrated training and testing method includes: In response to the command to create a training plan, the corresponding training items and test standards are generated; According to the training items and the testing standards, generate corresponding training images and display the training images; Acquire real-time vital sign data and movement posture information of the user when completing training actions based on the training images; The training effect evaluation results are determined based on real-time vital sign data and movement posture information; If the training effect evaluation result does not meet the preset achievement conditions, the training items that do not meet the standards are identified, and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions.

2. The integrated training and testing method as described in claim 1, characterized in that, The determination of training effectiveness evaluation results based on real-time vital sign data and movement posture information includes: Extract the current heart rate data, exercise energy consumption value, and blood oxygen saturation data corresponding to the real-time vital signs data; The current heart rate data is compared with the target heart rate interval corresponding to the training program to determine the heart rate achievement rate; The energy consumption value of the exercise is compared with the target energy consumption threshold corresponding to the training project to determine the energy consumption completion rate; The blood oxygen saturation data is compared with the blood oxygen standard range corresponding to the training program to determine whether the blood oxygen level meets the standard. Obtain the skeletal key points of the motion posture information, construct a three-dimensional posture model based on the skeletal key points, and generate a user motion trajectory sequence; The user's action trajectory sequence is dynamically time-normalized and compared with the standard action trajectory sequence corresponding to the training project to generate an action standardization score. The training effect evaluation result is generated by weighted fusion of the heart rate target achievement rate, energy consumption completion rate, blood oxygen target achievement status, and movement standardization score.

3. The integrated training and testing method as described in claim 1, characterized in that, The acquisition of real-time vital sign data and movement posture information of the user when performing training actions based on the training images includes: Establish a communication connection with wearable devices and acquire the user's electrocardiogram signals, body temperature changes, and triaxial acceleration data; Determine the energy consumption data corresponding to the triaxial acceleration data; Feature extraction is performed on the electrocardiogram signal and the body temperature change to generate a first biological sign feature vector; The real-time vital sign data is generated based on the first biological vital sign feature vector and the energy consumption data; Image data of the user performing training actions is acquired through image acquisition devices; Human posture recognition is performed on the image data, the coordinate sequence of skeletal key points is extracted, and the action posture information is generated.

4. The integrated training and testing method as described in claim 1, characterized in that, The step of generating corresponding training images according to the training items and the testing standards, and displaying the training images, includes: Based on the movement type of the training project and the number of target groups and times in the test standard, retrieve the corresponding standard 3D human animation sequence from the preset movement image library; In response to the selection operation of virtual coach image, background scene and rhythmic background music, the selected virtual coach image is fused and rendered with standard three-dimensional human body animation sequence to generate the training image; The training video and rhythmic background music are then synchronously output to the display component to guide the user in following the beat and completing standardized training movements.

5. The integrated training and testing method as described in claim 1, characterized in that, If the training effect evaluation result does not meet the preset achievement conditions, the training items that do not meet the standards are identified, and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions, including: If the overall score of the training effect evaluation result is lower than the preset target score, the training effect evaluation result is determined to be that the preset target condition is not met. Analyze the scores of each sub-indicator in the training effect evaluation results, identify one or more training items with the lowest scores, and mark them as the substandard training items; For each training item that fails to meet the standard, retrieve the corresponding multi-angle slow-motion breakdown video of the movement from the teaching video library; The display component prioritizes playing the breakdown video explaining the breakdown action; During the playback of the decomposed motion video, the user's real-time vital signs data and motion posture information while watching the decomposed motion video are reacquired. Based on the reacquired real-time vital signs data and movement posture information, a secondary evaluation result is generated for the substandard training item. If the secondary evaluation result meets the preset pass conditions, the item is marked as passed; otherwise, the breakdown video explaining the movements is played repeatedly until it is passed.

6. The integrated training and testing method as described in claim 1, characterized in that, The process of generating corresponding training items and test standards in response to a command to create a training plan includes: In response to the instruction to create a training plan, the system determines the user's selected training objective, available training time, and preferred exercise type, and generates the corresponding training objective. The training objective is matched with the knowledge graph of the preset training objective to determine the training parts and ability dimensions corresponding to the training objective; The number and combination of training items are determined based on the training area, the ability dimension, and the available training time. The training items and the testing standards are determined based on the preferred type of exercise, the number of training items, and the combination of the training items.

7. The integrated training and testing method as described in claim 1, characterized in that, The process of generating corresponding training items and test standards in response to a command to create a training plan includes: In response to a command to create a training plan, obtain the user's identification information; Obtain historical physical fitness test data, past training records, and body morphology indicators corresponding to the identity information; Analyze the historical physical fitness test data, past training records, and body shape indicators to assess the user's current overall physical fitness level and weaknesses; Based on the overall physical fitness level, select a basic training program template of corresponding difficulty from the preset training program library; Based on the aforementioned weaknesses, the corresponding departments in the basic solution template are strengthened, and the corresponding testing standards are adjusted to generate the adjusted training projects and testing standards.

8. The integrated training and testing method as described in claim 1, characterized in that, The integrated training and testing method also includes: Establish a communication connection with the user's smart wearable device to obtain the user's resting heart rate, sleep duration and quality, body temperature data, and daily steps, and generate corresponding long-term body data; The changes in the long-term physical data are analyzed at preset intervals to determine the user's recovery status and training load adaptation. If a user's resting heart rate is detected to be abnormally high or their sleep quality to be significantly reduced, a corresponding training load overload warning message is generated and sent to the user terminal and the coach management terminal.

9. A control device, characterized in that, The control device includes: a memory, a processor, and a control program for the integrated training and testing device stored in the memory and executable on the processor, wherein the control program for the integrated training and testing device is configured to implement the integrated training and testing method as described in any one of claims 1 to 8.

10. A training and testing integrated device, characterized in that, The integrated training and testing equipment includes: A display component, wherein the display component is used to present training images and feedback information; An image acquisition component, which is used to capture the user's movements and postures during training; Vital signs monitoring component, wherein the vital signs monitoring component is used to collect the user's heart rate, blood oxygen saturation and energy consumption data in real time; and The control device as described in claim 9, wherein the control device is electrically connected to the display component, the image acquisition component, and the vital sign monitoring component, respectively; The control device is used to respond to the command to create a training plan, generate corresponding training items and test standards, generate corresponding training images according to the training items and test standards, and control the display component to display the training images. The control device is also used to control the image acquisition component and the vital sign monitoring component to acquire real-time vital sign data and movement posture information of the user when completing training actions based on the training images, and to determine the training effect evaluation result based on the real-time vital sign data and movement posture information. If the training effect evaluation result does not meet the preset achievement conditions, the unqualified training items are identified and the decomposed actions corresponding to the training items are displayed until the user completes all training items and meets the preset achievement conditions.

Images