TWI844438B - An intervention method for promoting cardiorespiratory fitness training and its system thereof - Google Patents

Abstract

The invention provides an intervention method for promoting cardiorespiratory fitness training and its system thereof. The method involves the following steps: Inputting a cardiorespiratory fitness instructional video and the corresponding actual exercise video separately. Analyzing the cardiorespiratory fitness instructional video to generate default skeletal tracking results, obtaining multiple default joint angle values, and corresponding default weight scores. Analyzing the actual exercise video to generate skeletal tracking results and obtaining multiple joint angle values. Finally, computing the joint angle values from the actual exercise video, the default joint angle values, and the default weight scores to produce and display assessment information and corresponding intervention information. This system executes the method and designs appropriate intervention information for different users to enhance training effectiveness.

Description

Intervention methods and systems for improving cardiorespiratory fitness training

The present invention relates to a training assistance method, and more particularly to an intervention method and system for promoting cardiopulmonary fitness training.

The diagnosis of frailty can be used as a clinical manifestation to assess whether a person is facing functional decline and loss of autonomy. If appropriate intervention is given to elderly people living in the community before they become frail, there is a chance to prevent disability, thereby maintaining the health of the elderly and achieving the goal of successful aging. For this reason, purposeful, planned, and repetitive exercise training has always been considered a necessary condition for a healthy lifestyle. It has a positive effect on improving and maintaining physical and functional abilities, and at the same time, it can also improve cognitive status. Generally speaking, when the elderly exercise in the community, they are limited by too many spatial uncertainties in different community environments, so they cannot ensure the safety of walking or exercising.

The Republic of China's invention patent certificate No. I220387 discloses a virtual reality fitness equipment with a specific physical coaching auxiliary function. It allows the body's large muscle groups to continuously exercise for a long time in a rhythmic manner in an aerobic exercise manner under conditions suitable for individual physical fitness. The fitness equipment can be safely and effectively operated to achieve the purpose of exercise and fitness, avoid excessive fatigue, and achieve the purpose of strengthening the cardiopulmonary circulation function. However, this equipment cannot give different weights to different movements, and may ignore the specific parts that are mainly intended to be trained, so that the training effect may be greatly reduced.

Therefore, the present invention proposes an intervention method and system for promoting cardiopulmonary fitness training. The intervention method is designed for the elderly based on the principles of neuroscience and body perception. The elderly can use the system to import the video and audio they want to practice into the system for analysis. The system then analyzes the focus of each movement and gives different weights according to their importance. This allows users to better understand the focus of each movement and strengthen the training of their specific parts. In addition, the elderly can increase, decrease or adjust their training plans according to their personal needs to improve the training effect.

The main purpose of the present invention is to provide an intervention method for promoting cardiopulmonary fitness training. Different preset weight scores are calculated according to the angle values of the joints emphasized by different cardiopulmonary fitness teaching images, so as to provide appropriate intervention information accordingly and improve the accuracy of training assistance.

Another object of the present invention is to provide an intervention system for promoting cardiopulmonary fitness training. The aforementioned intervention method is used to identify cardiopulmonary fitness teaching images and actual sports images by an image recognition module, and the analysis module is used to analyze them to obtain the corresponding joint angle values of the two. The evaluation module gives different weight scores to different movements and outputs corresponding scoring information. Finally, the guidance module outputs the intervention information, so that the user can be more immersed in the situation and at the same time, better grasp the specific degree of posture correctness.

In order to achieve the above-mentioned purpose, one embodiment of the present invention discloses an intervention method for promoting cardiopulmonary fitness training, the steps comprising: Input a cardiopulmonary fitness teaching image; analyze the cardiopulmonary fitness teaching image to generate a default bone tracking result; calculate the default bone tracking result to obtain a corresponding plurality of default joint angle values; calculate the default joint angle values to obtain a corresponding default weight score; input an actual exercise image corresponding to the cardiopulmonary fitness teaching image; analyze the actual exercise image to generate a bone tracking result; calculate the bone tracking result to obtain a corresponding plurality of joint angle values; calculate the joint angle values, the default joint angle values and the default weight score to generate and display scoring information; and generate and display intervention information using the scoring information.

In a preferred embodiment, the actual motion image is selected from dynamic images or static images.

In a preferred embodiment, in the step of calculating the default joint angle values and obtaining a corresponding default weight score, the default joint angle values include the default joint angle values of a default action information at a first time and the default joint angle values of the default action information at a second time. The default joint angle values at the first time are compared with the default joint angle values at the second time to obtain the corresponding default weight score.

In a preferred embodiment, in the step of calculating the joint angle values, the default joint angle values and the default weight scores to generate and display a scoring information, the joint angle values corresponding to each of the default action information are compared with the default joint angles to obtain a corresponding degree of matching, and the degree of matching and the corresponding default weight score are calculated to generate the scoring information.

In a preferred embodiment, the intervening information is text, voice, image, audio and video, or a combination thereof.

In order to achieve the above-mentioned another purpose, an embodiment of the present invention discloses an intervention system for promoting cardiopulmonary fitness training, comprising: an input unit, which inputs a cardiopulmonary fitness teaching image and a corresponding actual sports image; an image recognition module, which is connected to the input unit signal, and is used to recognize the cardiopulmonary fitness teaching image and the actual sports image to obtain a preset bone tracking result corresponding to the cardiopulmonary fitness teaching image, and a bone tracking result corresponding to the actual sports image; an analysis module, which is connected to the image recognition module signal, and is used to analyze the preset bone tracking result and the bone tracking result to obtain the preset bone tracking result. A plurality of preset joint angle values of the bone tracking results, and a plurality of joint angle values corresponding to the bone tracking results; an evaluation module, connected to the analysis module signal, for calculating according to the preset joint angle values to obtain a preset weight score, and generating and displaying scoring information based on the joint angle values, the preset joint angle values and the preset weight score; and a guidance module, connected to the evaluation module signal, for generating and displaying intervention information according to the scoring information.

In a preferred embodiment, the actual motion image is selected from dynamic images or static images.

In a preferred embodiment, a depth photography module is included, which is connected to the input unit signal to capture the actual motion image of a human body.

In a preferred embodiment, a three-dimensional image display is included, which is connected to the input unit and the guidance module signal to display the cardiopulmonary fitness teaching image and the intervention information.

In a preferred embodiment, the intervention information is text, voice, image, audio and video, or a combination thereof.

The beneficial effect of the present invention is that it can further calculate the corresponding weight score according to the cardiopulmonary fitness teaching image to be learned at random, and then calculate according to the user's actual sports image and weight score to weight the parts emphasized by different movements, providing accurate intervention information to achieve the best training effect.

The related patent application features and technical contents of the present invention will be clearly presented in the following detailed description of the preferred embodiments with reference to the drawings.

Please refer to the first figure, the second figure A and the second figure B, which are the method flow charts of one embodiment and another embodiment of the present invention. As shown in the figure, the method includes the following steps.

Step S1: input a cardiopulmonary fitness teaching image;

Step S2: analyzing the cardiopulmonary fitness teaching image to generate a preset bone tracking result;

Step S3: Calculate the default skeleton tracking result to obtain a plurality of corresponding default joint angle values;

Step S4: Calculate the preset joint angle values to obtain a corresponding preset weight score;

Step S5: inputting an actual exercise image corresponding to the cardiopulmonary fitness teaching image;

Step S6: analyzing the actual motion image to generate a bone tracking result;

Step S7: Calculate the skeleton tracking result to obtain the corresponding multiple joint angle values;

Step S8: Calculate the joint angle values, the preset joint angle values and the preset weight score to generate and display a rating information; and

Step S9: Generate and display intervention information based on the rating information.

In one embodiment, the aforementioned steps S1-S9 may also be a continuous process. In another embodiment, steps S1-S4 may be an independent process, such as the second FIG. A, and steps S5-S9 may also be an independent process, such as the second FIG. B.

Please also refer to the third figure, which is a schematic diagram of a system of an embodiment of the present invention. As shown in the figure, the intervention system for promoting cardiopulmonary fitness training includes: an input unit 1, an image recognition module 2, an analysis module 3, an evaluation module 4, a guidance module 5, a depth photography module 6 and a three-dimensional image display 7, wherein the input unit 1 is signal-connected to the image recognition module 2, the depth photography module 6 and the three-dimensional image display 7 respectively; the analysis module 3 is signal-connected to the image recognition module 2 and the evaluation module 4 respectively; and the evaluation module 4 is signal-connected to the guidance module 5 and the three-dimensional image display 7.

Input unit 1 is used to input cardiopulmonary fitness teaching images, wherein the cardiopulmonary fitness teaching images are demonstration images corresponding to sports items designed for the elderly. The source of the teaching images can be images shot by professional teachers, or any teaching images on YouTube, as long as they are helpful for training, such as: muscle strength training, cardiopulmonary training, balance training and flexibility training, and training given according to individual physiological conditions In one embodiment, the input unit 1 further includes a setting unit, which can be based on physiological conditions such as age, health level and exercise habits, wherein the health level is such as whether the user has high blood pressure, high blood lipids, high blood sugar, metabolic syndrome or degenerative arthritis, and the exercise habit is whether the user has an original exercise habit and how many times a week, so as to set a more personal exercise prescription. Furthermore, the corresponding training intensity can be adjusted according to the appropriate heart rate target. For example: the user may suffer from degenerative arthritis and should not have large-scale leg movements, then a second weight score may be generated to gradually correct the score according to the actual user situation, so that the user can be more confident in completing the corresponding action.

The image recognition module 2 receives the cardiopulmonary fitness teaching images and the corresponding actual sports images, and analyzes the preset bone information and bone information corresponding to each action to generate the corresponding preset bone tracking results and bone tracking results, respectively. The actual sports images can be dynamic images or static images, but are not limited to this.

In one embodiment, the human posture recognition method adopted by the image recognition module 2 can be selected from any method such as OpenPose Model, AlphaPose, Single-Stage Multi-Person Pose Machines (SPM) or Regional Multi-Person Pose Estimation (RMPE), among which human posture recognition can also be divided into two methods: Top-down and Bottom-up. Top-down is to first identify the human body in the image, and then identify the key points in the human body. On the contrary, Bottom-up is to first find each key point in the image, and then combine the discrete key points into human limbs according to the relationship between the key points. The human posture recognition adopted by the embodiment of the present invention is not limited to this, as long as the human posture can be recognized.

The image recognition module 2 analyzes the obtained bone tracking results of the user to generate corresponding multiple joint angle values, and compares them in real time with the multiple preset joint angle values analyzed by the instructor's preset bone tracking results in the cardiopulmonary fitness teaching image. When the two are not completely consistent, the preset weight score of each action is measured according to the size of the action, and the scores are calculated in sequence to evaluate the effect of this exercise.

The analysis module 3 obtains a plurality of corresponding preset joint angle values according to the preset bone tracking results of the cardiopulmonary fitness teaching image, and obtains a plurality of corresponding joint angle values according to the bone tracking results of the actual sports image.

In one embodiment, please refer to the fourth figure, which is a schematic diagram of the action before and after of one embodiment of the present invention. As shown in the figure, the skeleton tracking result includes various joints, such as: nose J0, neck J1, left shoulder J2, right shoulder J3, left elbow J4, right elbow J5, left wrist J6, right wrist J7, left hand J8, right hand J9, left hip J10, right hip J11, central hip J12, vertebra J13, left knee J14, right knee J15, left ankle J16, right ankle J17, left foot J18, right foot J19, but not limited thereto.

In one embodiment, please refer to FIG. 5, which is a schematic diagram of the calculation of joint angle values of one embodiment of the present invention. As shown in the figure, the joint angle value, such as the elbow angle value θ, is calculated by using the first vector from the shoulder coordinate value S to the elbow coordinate value E , and a second vector from the elbow coordinate value E to the wrist coordinate value W , and according to the cosine theorem formula in vector space, the following formula can be obtained:

In this way, the angle value of each joint point corresponding to each action can be calculated for subsequent judgment.

The evaluation module 4 is used to calculate the preset weight scores according to the preset joint angle values, and to generate scoring information by calculating the joint angle values, the preset joint angle values and the corresponding preset weight scores. In this way, the weight of each action score can be measured according to the size of the action. The joints with different preset weight scores can be the shoulder joint, elbow joint, wrist joint, hip joint, knee joint and ankle joint. As shown in the figure, the shoulder joint angle changes the most when the left side movement is compared with the right side movement, so the default weight score here is relatively high. For example, the aforementioned joint angle value can be recorded at a speed of every 100 milliseconds, and after comparing the previous values, the change in the joint angle value is used as the weighting basis. If the change is higher, the weighting is heavier, and vice versa. In this way, users can learn how to grasp the key points of the movement more accurately, but not limited to this.

In one embodiment, the evaluation module 4 can compare the preset joint angle values of each preset action information with the corresponding joint angle values. When the two are completely consistent, the scoring information can be directly generated, for example: displaying 100 points or displaying complete compliance. On the contrary, when the two are not completely consistent, the scoring information can be generated as described above by performing calculations based on the joint angle values and the corresponding preset weight scores, but is not limited to this.

Preferably, the guidance module 5 can generate corresponding intervention information based on the scoring information. The intervention information can be text, voice, image, audio or video, or a combination of the above. For example, in the displayed bone tracking results, when the knee is not lifted to the specified position, an upward arrow is displayed at the knee position, and the user is reminded to adjust the movement by dynamic images and voice, but not limited to this.

The depth camera module 6 detects the posture image of the human body, that is, when the user performs the same action according to the cardiopulmonary fitness teaching image that is suitable for him, the posture image at this time can be detected by the depth camera module 6. Preferably, the cardiopulmonary fitness teaching image can also be obtained by photographing the depth camera module 6. In one embodiment, the depth camera module 6 includes at least one depth camera, and the depth camera module 6 can adopt the Kinect somatosensory device, and the number of depth cameras can be set according to needs, but it is not limited to this.

The three-dimensional image display 7 is used to display various information, such as cardiopulmonary fitness teaching images and intervention information, and can also display scoring information, etc., wherein the three-dimensional image display 7 includes a screen device, a head-mounted display device, a projection device or a combination of the above. In one embodiment, in order to present various information more realistically, the following 3D technologies can be used for presentation, such as: augmented reality (AR), virtual reality (VR), mixed reality (MR) or extended reality (X-Reality, XR). In this way, the user can be provided with the best visual training assistance system, but not limited to this.

With the advancement of technology, virtual reality has also begun to be widely used in medical fields. At the same time, there are many empirical studies showing the benefits of using virtual reality for the elderly, which are as follows:

First, in the study of the application of virtual reality in the cognition of the elderly, scholars randomly assigned elderly people aged 65-85 to a group that practiced in virtual reality and a group that only performed general physical exercises. Both groups received eight weeks of training and used the Montreal Cognitive Assessment to assess cognitive ability. The results showed that the group that used virtual reality training significantly improved cognitive and physical abilities compared to the group that performed general physical exercises. That is, through the stimulation of the virtual reality environment, the operation of input activation of the neural network mechanism was significantly enhanced, the volume of the hippocampus and entorhinal cortex was significantly increased, and the performance of spatial cognition was improved.

In addition, virtual reality is used in research on the balance of the elderly. For patients with balance disorders and elderly people with poor balance living in the community, virtual reality system training courses are used twice a week for six weeks. Stability is significantly increased and the center of pressure is significantly reduced, which can improve balance ability and prevent falls in the elderly.

At the same time, research on the application of virtual reality in preventing falls in the elderly also shows that the elderly are prone to falls due to the deterioration of balance, slower movements and poorer coordination. The human body's posture control requires the coordination and cooperation of multiple systems to maintain good movement control and reduce the number of falls. Therefore, scholars compared virtual reality activities with and without the use of video games for female elderly people who are prone to falls. The training time is 30 minutes, twice a week. The research results also show that users who use virtual reality for training significantly increase their walking speed. At the same time, the time for walking 3 meters is reduced, which also means that the risk of falling is reduced.

Finally, in the study on the application of virtual reality to frail elderly people, it was pointed out that giving elderly people interactive video games combined with virtual reality dual tasks can effectively improve their mobility. It can combine movement and cognition to reproduce daily life situations, and the user acceptance rate is high.

Therefore, it is understandable that when the body functions of the elderly begin to deteriorate and lead to symptoms of weakness, it is very easy to affect various activities, including physical and social, and may also affect cognitive functions, resulting in dementia, reducing daily living ability and affecting the quality of life. For this reason, one embodiment of the present invention uses a three-dimensional image display 7 to specifically combine virtual reality with sports training to encourage the elderly in the community to actively participate in cardiopulmonary fitness activities and slow down aging.

As described in step S1, a cardiopulmonary fitness teaching image is input. In one embodiment, the cardiopulmonary fitness teaching image can be a demonstration of recommended exercise items generated according to different physiological conditions of an individual, but it is not limited to this. The user can also input the image he wants to learn.

As described in step S2, the image recognition module 2 recognizes the default skeletal tracking results of each default action information in the cardiopulmonary fitness teaching image, and then proceeds to step S3 to calculate the corresponding multiple default joint angle values with the default skeletal tracking results. Preferably, the three-dimensional image display 7 can be used to display all the default skeletal tracking results of the cardiopulmonary fitness teaching image, and the details of the action can be rotated 360 degrees for the user to refer to and respond accordingly.

As described in step S4, the preset joint angle values are calculated to obtain the preset weight score in each preset action information, wherein the preset joint angle values include the preset joint angle values of the default action information at the first time and the preset joint angle values of the default action information at the second time. The preset joint angle values at the first time are compared with the preset joint angle values at the second time to obtain the corresponding preset weight score. In other words, the joint angle values of actions at different times are compared with each other. The greater the difference between the two, the more important the action of this part is, and the higher the preset weight score is.

As described in step S5, an actual motion image corresponding to the cardiopulmonary fitness teaching image is input, wherein the actual motion image can be obtained by the depth photography module 6, or is just a dynamic image or a static image after shooting. Then, as described in step S6, the image recognition module 2 recognizes the bone tracking result in the actual motion image. Preferably, the bone tracking result can be displayed through the three-dimensional image display 7, that is, the user's own three-dimensional virtual image can be output synchronously, so that the user can see his own movements more specifically.

As described in step S7, please refer to FIG6, which is a partial flow diagram of an embodiment of the present invention. As shown in the figure, the analysis module 3 further calculates the skeletal tracking result of the actual motion image to obtain a plurality of corresponding joint angle values, and then as described in step S8, the evaluation module 4 calculates the joint angle values, the preset joint angle values and the corresponding preset weight scores to generate and display the scoring information.

Preferably, please refer to FIG. 7, which is a partial flow diagram of an embodiment of the present invention. As shown in the figure, first, the joint angle values are compared with the preset joint angle values to obtain the corresponding matching degree. When the joint angle values are not completely consistent with the preset joint angle values, the scoring information is generated after the calculation based on the joint angle values and the preset weight score. This scoring information can have a corresponding preset weight score according to the action proportion, that is, according to the scoring weight of different actions, so that the scoring information can be closer to the actual situation.

In one embodiment, the rating information can be displayed through a three-dimensional image display 7 for user reference, or through an application APP in the user's mobile device P. Of course, the rating information can also be displayed through any electronic product that can install the application APP, but it is not limited to this.

As described in step S9, the guidance module 5 generates corresponding intervention information based on the aforementioned scoring information obtained. The intervention information can be presented in the form of text, graphics, voice, image or video to remind the user that some movements or force application points must be adjusted, etc. The intervention information can be displayed through the three-dimensional image display 7 or by the interface of the application APP.

To more clearly illustrate the actual operation method, please refer to FIG. 8, FIG. 9, FIG. 10A and FIG. 10B, which are schematic diagrams of the operation process of one embodiment of the present invention. As shown in the figure, first, the user can search for teaching videos in the electronic device C, and of course, multiple teaching videos can be generated by other methods and uploaded to the server Server, where the server Server provides the image recognition module 2, the analysis module 3, the evaluation module 4 and the guidance module 5 for operation. After uploading to the server Server, a default weight score corresponding to the teaching video will be generated, which is steps S1-S4. In one embodiment, steps S1 to S4 can be independent processes, that is, multiple teaching videos can be uploaded to the server Server first, and each teaching video will correspond to a default weight score. When the user User opens the application APP through the mobile device P, it can connect to the server Server and start the subsequent scoring process, as shown in the eighth figure.

Next, when the user User wants to perform training, the depth camera module 6 can capture the actual sports images of the user performing cardiopulmonary fitness training, and upload the actual sports images to the server Server for scoring. In one embodiment, the user User can also wear a smart wearable device SW to monitor the heartbeat status and blood oxygen status of the user User at the same time, as shown in Figure 9, but not limited to this, and can also be used with other wired or wireless physiological monitoring devices to further grasp other physiological states of the user User, such as: breathing frequency and number. In one embodiment, the smart wearable device SW can be a smart watch, and the smart watch can be selected from Garmin or other brands of smart watches.

Finally, the application APP can be opened through the mobile device P of the user User, and all the processes of the user User's actual motion image can be displayed on the interface of the application APP, and a virtual skeleton V can be corresponding to the interface. This virtual skeleton V corresponds to the user User's actual motion image, and is marked with different colors to indicate correctness or error, for example: correct is marked with a blue line segment, and error is marked with an orange line segment. The color can be replaced according to needs, so that the user User can understand whether his or her own motion status is correct.

In one embodiment, as shown in FIG. 10A , the upper portion is an example of an embodiment of an interface displayed by the application APP, that is, the virtual skeleton V overlaps the actual motion image; the lower portion is an example of another embodiment of the displayed interface, that is, the virtual skeleton V can be shrunk and displayed on one side of the actual motion image without overlapping it. Of course, there can also be other display methods, and when the user User wears the smart wearable device SW, in the left frame of the application APP interface, the rating information, heart rate information and blood oxygen information can be displayed in sequence from top to bottom.

In another embodiment, as shown in FIG. 10B, the upper portion is an example of an embodiment of an interface displayed by the application APP, and the displayed interface is divided into two screens, namely screen A on the left and screen B on the right, wherein screen A is the actual motion image of the user User, and its virtual skeleton V overlaps with the actual motion image, and at the same time, screen B is a synchronized teaching image; the lower portion is an example of another embodiment of the displayed interface, similarly, screen A is the actual motion image of the user User, and its virtual skeleton V can be reduced and displayed on one side of the actual motion image without overlapping with it, and at the same time, screen B is also a synchronized teaching image, and the rating information, heart rate information and blood oxygen information are displayed in the middle of the interface respectively.

The difference between Figure 10A and Figure 10B is that the interface layout and information displayed by the application APP are different. Of course, the function options of the application APP interface can also be set according to needs, and the interface layout can be set according to needs, which is not limited to this.

Figure 9, Figure 10A and Figure 10B are schematic diagrams of the operation process of steps S5-S9. Steps S5-S9 can be a process independent of steps S1-S4, that is, the server Server may already store many teaching videos and their default weight scores, and the user User can execute steps S5-S9 at any time. In one embodiment, the final score and actual motion image can be stored in the server Server, and the user User can have a personal independent storage space in the server Server to store personal scores and actual motion images, and can allow the user User to repeatedly view personal scores and actual motion images.

In summary, an embodiment of the present invention is an intervention method and system for promoting cardiopulmonary fitness training, which can provide corresponding cardiopulmonary fitness teaching images for training content based on the elderly's own conditions, including balance ability, coordination ability, etc., and further provide users with three-dimensional images for viewing reference, and can provide real-time feedback on the corresponding training results to increase the user's learning motivation and sense of accomplishment, thereby meeting the needs of users and improving the lack of learning due to venue and safety. Therefore, the purpose of the present invention can be achieved.

However, the above is only a preferred embodiment of the present invention, and should not be used to limit the scope of implementation of the present invention. That is, all simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the content of the invention description are still within the scope of the present patent.

1: Input unit 2: Image recognition module 3: Analysis module 4: Evaluation module 5: Guidance module 6: Depth photography module 7: 3D image display A: Screen APP: Application B: Screen C: Electronic device E: Elbow coordinate value J0: Nose J1: Neck J2: Left shoulder J3: Right shoulder J4: Left elbow J5: Right elbow J6: Left wrist J7: Right wrist J8: Left hand J9: Right hand J10: Left hip J11: Right hip J12: Central hip J13: Vertebrae J14: Left knee J15: Right knee J16: Left ankle J17: Right ankle J18: Left foot J19: right foot S: shoulder coordinate value S1: step S2: step S3: step S4: step S5: step S6: step S7: step S8: step S9: step Server: server SW: smart wearable device P: mobile device User: user V: virtual skeleton W: wrist coordinate value

Figure 1: It is a method flow chart of one embodiment of the present invention; Figure 2A: It is a method flow chart of another embodiment of the present invention; Figure 2B: It is a method flow chart of another embodiment of the present invention; Figure 3: It is a system schematic diagram of one embodiment of the present invention; Figure 4: It is a schematic diagram of the action before and after of one embodiment of the present invention; Figure 5: It is a schematic diagram of the joint angle value calculation of one embodiment of the present invention; Figure 6: It is a partial process schematic diagram of one embodiment of the present invention; Figure 7: It is a partial process schematic diagram of one embodiment of the present invention; Figure 8: It is a schematic diagram of the action process of one embodiment of the present invention; Figure 9: It is a schematic diagram of the action process of one embodiment of the present invention; Figure 10A: It is a schematic diagram of the action process of one embodiment of the present invention; and Figure 10B: This is a schematic diagram of the operation process of another embodiment of the present invention.

S1-S9: Steps

Claims (10)

An intervention method for promoting cardiopulmonary fitness training, comprising the following steps: Inputting a cardiopulmonary fitness teaching image; Analyzing the cardiopulmonary fitness teaching image to generate a preset bone tracking result; Calculating the preset bone tracking result to obtain a plurality of corresponding preset joint angle values; Calculating the preset joint angle values to obtain a corresponding preset weight score; Inputting an actual sports image corresponding to the cardiopulmonary fitness teaching image; Analyzing the actual sports image to generate a bone tracking result; Calculating the bone tracking result to obtain a plurality of corresponding joint angle values; Calculating the joint angle values, the preset joint angle values and the preset weight score to generate and display a rating information; and Generate and display intervention information based on the rating information. According to the intervention method for promoting cardiopulmonary fitness training described in claim 1, the actual motion image is selected from dynamic images or static images. According to the intervention method for promoting cardiopulmonary fitness training described in claim 1, in the step of calculating the preset joint angle values and obtaining a corresponding preset weight score, the preset joint angle values include the preset joint angle values corresponding to a preset action information at a first time and the preset joint angle values corresponding to the default action information at a second time. The preset joint angle values at the first time are compared with the preset joint angle values at the second time to obtain the corresponding preset weight score. According to the intervention method for promoting cardiopulmonary fitness training described in claim 3, in the step of calculating the joint angle values, the default joint angle values and the default weight scores to generate and display a scoring information, the joint angle values corresponding to each of the default action information are compared with the default joint angles to obtain a corresponding degree of matching, and the degree of matching and the corresponding default weight score are calculated to generate the scoring information. An intervention method for promoting cardiopulmonary fitness training according to claim 1, wherein the intervention information is text, voice, image, video or a combination thereof. An intervention system for promoting cardiopulmonary fitness training comprises: An input unit for inputting a cardiopulmonary fitness teaching image and a corresponding actual sports image; An image recognition module connected to the input unit signal for recognizing the cardiopulmonary fitness teaching image and the actual sports image to obtain a preset bone tracking result corresponding to the cardiopulmonary fitness teaching image and a bone tracking result corresponding to the actual sports image; An analysis module connected to the image recognition module signal for analyzing the preset bone tracking result and the bone tracking result to obtain a plurality of preset joint angle values corresponding to the preset bone tracking result and a plurality of joint angle values corresponding to the bone tracking result; An evaluation module, connected to the analysis module signal, calculates according to the preset joint angle values, obtains a preset weight score, and generates and displays a rating information based on the joint angle values, the preset joint angle values and the preset weight score; and A guidance module, connected to the evaluation module signal, generates and displays an intervention information based on the rating information. An interventional system for promoting cardiopulmonary fitness training according to claim 6, wherein the actual motion image is selected from a dynamic image or a static image. The intervention system for promoting cardiopulmonary fitness training according to claim 6 includes a depth photography module connected to the input unit signal to capture the actual movement image of a human body. The intervention system for promoting cardiopulmonary fitness training according to claim 6 includes a three-dimensional image display connected to the input unit and the guidance module signal to display the cardiopulmonary fitness teaching image and the intervention information. An intervention system for promoting cardiopulmonary fitness training according to claim 6, wherein the intervention information is text, voice, image, video or a combination of the above.