TWI784243B - Method of taekwondo poomsae movement detection and comparison - Google Patents

Abstract

The present invention provides a method of taekwondo poomsae movement detection and comparison. First, detecting user’s body detecting positions and using the body detecting positions to build a standard poomsae model of each poomsae. At a start time point, detecting the user’s start location in a place and the user starting doing poomsae. Then, detecting user’s body detecting positions in several time points. At an end time point, detecting the user’s end location in the place. Finally, to obtain a result of complete a poomsae via comparing the body detecting positions in several time points with the standard poomsae model and comparing the start location with the end location.

Description

Taekwondo poomsae motion detection and comparison method

The present invention is a motion detection and comparison method, in particular a Taekwondo poomsae motion detection and comparison method that establishes a personal standard poomsae model based on body recognition positions and compares the detected body recognition positions. method.

In Taekwondo, Poomsae is an item that needs to be tested in the process of promotion and promotion. In addition, practicing Poomsae can increase the accuracy and fluency of basic movements, making Poomsae an extremely important part of Taekwondo training. part of the At present, all Taekwondo poomsae competitions in the world are based on the judgment of the referees. That is to say, during the competition, the referees need to judge the correctness of the contestants' poomsae movements according to their personal experience and standards. , manual judgment will inevitably lead to errors, and the requirements for poomsae movements not only include whether the movements are in place, but also require the strength of the movements, which increases the difficulty of manual judgment.

In the known motion detection comparison technology, a common method is to use sensors worn on the body parts of the contestants to collect the positions of the sensing devices at different times when the contestants perform actions, and to use different The position obtained under position and time is compared with the standard action template to confirm the correctness of the contestant's action. Using this method, in the poomsae competition or training, it is not necessary to rely solely on the judgment of the referee or the coach through the naked eye, but also refer to the results collected by the sensor as one of the basis for the judgment, and even provide the contestants themselves Practice using, however, Adding sensors to competitors or users will undoubtedly increase the fluency of the participants or users' movements, and the sensors must be worn every time they compete or practice, which not only increases transportation The burden of the sensor will also consume a lot of time during each wearing and adjustment process.

In the known motion detection comparison technology, another common method is to use an imaging device to track specific body parts of the user, take images of the user at different times when the user is performing an action, and then combine the images obtained at different times The images are compared with exemplary images of standard actions to provide feedback to the user. In this method, there is no need to install additional sensors on the user's body, which reduces the burden on the user's body. However, this method uses the user's action image and the standard action performed by the standard action demonstrator. For image comparison, if there is a difference in body shape between the user and the standard action demonstrator, it will be necessary to adjust the parameters such as the ratio, which will undoubtedly increase the error value in the comparison. In addition, when doing Taekwondo poomsae movements, different movement angles will be produced according to the position of the body parts. For example, there are upper, middle and lower parts in poomsae movements. These parts will be used according to the different body shapes of the users. Because of the different positions of the user's body, when making poomana movements such as forward kicking, the user needs to kick to the corresponding position due to the different positions of the upper end, middle end, and lower end of the user. If the above-mentioned method is used, the user can only trace with reference to the angle of action of the performer of the standard action, and cannot know whether the user's own action is in place.

Based on the above, the conventional technology is not suitable for the detection and comparison of Taekwondo Poomsae movements. Based on the bottleneck of the above-mentioned conventional technology, the present invention overcomes the above shortcomings and proposes a method for detection and comparison of Taekwondo Poomsae movements. .

The purpose of the present invention is to propose a taekwondo poomsae motion detection and comparison The method uses direct detection of the body recognition position on the user's body to replace the conventional method of collecting information from sensors worn on the user's body, so as to reduce the burden on the user's movement and reduce the time spent on wearing The time it takes for the sensor to run.

The purpose of the present invention is to propose a method for detecting and comparing poomsae movements in Taekwondo. When comparing poomsae movements, the user's own body recognition position is used to establish a personal standard poomsae model for each poomsae. When the user starts to perform poomsae During poomsae, by comparing with the user's own personal standard poomsae model, it replaces the method of comparing with other demonstrators' standard poomsae movements in the conventional technology, so as to improve the accuracy of judging that poomsae is in place.

The purpose of the present invention is to propose a Taekwondo poomsae motion detection and comparison method. In addition to detecting the user's body recognition position, it can also directly detect the user's position on the field and the user's gestures when performing poomsae. The sound and the time it takes to calculate the user's poomsae replace the conventional way of using the referee's manual judgment or calculation to improve the accuracy and efficiency of judgment.

The present invention proposes a taekwondo poomsae motion detection and comparison method, which is suitable for detecting at least one set of poomsae performed by a user in a field, and the steps of the method will be described continuously. In one embodiment of the present invention, firstly, each body recognition position on the user's body is detected, and a plurality of personal standard poem potential models are established for each body recognition position according to each set of pooms; The user is at the origin of the venue, and the user starts to perform Poomsae; then, at multiple detection time points, detect each body recognition position corresponding to the user at each time point; and, at the end time point , to detect the end position of the user on the field. Among them, comparing each body recognition position corresponding to each detection time point and each personal standard poomsae model, and comparing the origin position and the end position, so as to obtain the completion result of the user's poomsae.

In addition, before the start time point, the user's voice is captured and integrated into a specific personal standard poinsema model, and the user's voice is detected at each detection time point, and the obtained Before completing the results, compare the user's voice at each detection time point with each individual standard Poomsae model.

In addition, each set of poomsae has a plurality of standard poomsae times and each set of poomsae contains a plurality of standard action times. The multiple standard action times contained in the set poomsae are integrated into a specific personal standard poomsae model. Before the completion result is obtained, calculate the time elapsed from the start time point to the end time point, calculate the multiple action times of the user performing the at least one set of poomsae, and compare the elapsed time from the first time point to the end time point The time is compared with the specific personal standard poomera model and the time of the actions is compared with the specific at least one personal standard poud potential model.

Among them, each body recognition position on the user's body is detected, personal body characteristics are calculated according to the body recognition position, and then plural personal standard poomsae models are established by each body recognition position and personal body characteristics according to each set of poomsae. Personal physical characteristics include at least foot length, shoulder width, upper end position, middle end position, lower end position, and gait.

Wherein, the detection of the body recognition position on the user's body is initiated after the user poses a standing human anatomical posture.

In addition, the body identification position system includes at least the head, shoulder joint, xiphoid process, hip joint, wrist joint, elbow joint, finger joint, knee joint, ankle joint, sole, and toe.

Furthermore, in an embodiment of the present invention, the start time point and the end time point correspond to a start indication and an end indication respectively, and the start indication and the end indication may be gesture indications and/or sound indications.

In addition, in an embodiment of the present invention, when there are multiple users, the completion results corresponding to the multiple users can be analyzed to confirm the consistency of the completion results.

Furthermore, in one embodiment of the present invention, the completion degree of poomsae is judged and/or scored according to the completion result.

According to the above, the present invention proposes a taekwondo poomsae movement detection and comparison method, by detecting the body recognition position on the user's body, the body recognition position is established to the personal standard poomsae model according to each set of poomsae, When the user starts to move, the real-time body recognition position is compared with the personal standard Posei model to overcome the technical bottleneck in the conventional technology. Moreover, different from the conventional technology, by detecting the The original position and the end position of the user in the field, as well as the comparison of the sound made by the user when performing Poomsa and the time spent in Poomsae, greatly improved the accuracy of judging or scoring in Taekwondo Poomsae competitions or exercises accuracy and efficiency.

a: head

a1: among people

b: shoulder joint

c:xiphoid process

d: hip joint

e, e1, e2: wrist joint

f: Elbow joint

g: finger joints

h: knee joint

i: ankle joint

j: soles of feet

k: toes

S01, S02, S03, S04, S05, S06, S07: steps

S21, S22, S23, S24, S25, S26, S27: steps

S31, S32, S33, S34, S35, S36, S37: steps

S41, S42, S43, S44, S45, S46, S47: steps

S51, S52, S53, S54, S55, S56, S57: steps

Fig. 1 is a schematic flow chart of an embodiment of the present invention.

Figure 2 is a schematic diagram of the body recognition position of the present invention.

Fig. 3 is a schematic flow chart of another embodiment of the present invention.

Fig. 4 is a schematic flow chart of another embodiment of the present invention.

Fig. 5 is a schematic flow chart of another embodiment of the present invention.

Fig. 6 is a schematic flow chart of another embodiment of the present invention.

The advantages and spirit of the present invention can be further understood through the following detailed description and the accompanying drawings. The structure and use of the embodiment of the present invention are described in detail as follows. must understand What is remarkable is that the present invention provides many applicable innovative concepts, which can be widely implemented under a specific background technology. This specific embodiment is only shown in a specific way to make and use the invention, but does not limit the scope of the invention.

The present invention proposes a taekwondo poomsae movement detection and comparison method, which is suitable for detecting at least one set of poomsae performed by a user in a field. Movement detection and comparison are performed during promotion, poomsa competition, or poomsa practice. That is to say, the venue of this method can be any venue, not limited to the competition venue. Among them, the content of Poomsae includes but not limited to Chapters 1 to 8 of Taichi Poomsae, Goryeo type, King Kong type, Taebaek type, Plain type, Ten-job type, Ditae type, Tianquan type, Hanshui type, Yiru type, etc.

Please refer to FIG. 1, which is a schematic flow chart of an embodiment of the present invention. In the embodiment shown in Figure 1, the Taekwondo poomsae movement detection and comparison method includes the following steps: First, in step S01, each body recognition position on the user's body is detected, and according to each set of poomsae, the Each body recognition position establishes a plurality of individual standard poomsae models; in step S02, at the start time point, detect the origin position of the user on the field, and the user starts to perform poomsae; in step S03, in the complex detection At the time point, detect each body recognition position corresponding to the user at each detection time point; then, in step S04, compare each body recognition position corresponding to each detection time point with the personal standard position model; in step S05, at the end time point, detect the end position of the user in the field; in step S06, compare the origin position and end position; finally, in step S07, obtain the user Carry out the completion result of Poomsae.

Taekwondo’s poomsae movements are based on simple, neat and solid movements. Based on the body shape of the human body, there are tall, short, fat, and thin. The proportions of each party are not the same. Shape and make the most appropriate action. Because of this, during the poomsa competition, Competitors are also required to make appropriate Poom posture based on their own body shape. Therefore, in step S01, based on the above-mentioned requirements, the specific body parts of the contestants are detected to establish a standard poomera model belonging to the contestants. However, the specific body parts to be detected are the following The body recognition position will be illustrated in Figure 2. Please refer to Figure 2. The body recognition position includes at least head a (such as Renzhong a1), shoulder joint b, xiphoid process c, hip joint d, wrist joint e, hand Elbow joint f, finger joint g, knee joint h, ankle joint i, sole j, toe k. Wherein, the toe refers to the front end of the foot, and if the user wears shoes, the front end of the shoe is regarded as the toe. In addition, when doing Poomsae, some specific action parts will have different angles during movement. Taking boxing as an example, it can be divided into positive fist, back fist, mallet fist, nail fist, and palm fist. Therefore, in order to distinguish Depending on the action, when detecting the body recognition position, a specific body recognition position will have more than one detection position. For example, in order to distinguish between positive punches and back punches, the thumb can be detected on the part of the wrist joint e For the position e1 below and the position e2 below the palm, when punching, the position of the wrist joint e1 below the thumb, the position of the wrist joint e2 below the palm, and the relative positions of the finger joints h can be used to determine the type of punch. Therefore, detecting the user's body recognition position to establish a personal poomsai standard model is an indispensable part of judging the correctness of the user's poomsai action, and it is also the first step that must be taken in the method.

It should be noted that the detection of the body recognition position on the user's body will only be activated after the user poses in an anatomical pose. For example, the detection of the body recognition position on the user's body can be performed by the user turning around To obtain the body recognition position corresponding to its front, side, and back, or obtain the body recognition position corresponding to the user's front, side, and back through one or more detection devices, or simultaneously use the user's own Turn around and multiple detection devices to obtain the user's body recognition position. In the scoring standard of Taekwondo poomsae competition, In addition to the correctness and smoothness of movements and steps, it also includes the requirement of returning to the origin. For example, in general poomsai competitions, the contestants are required to start and stop within one foot. Therefore, in step S02, at the start time point, that is, when the user performs the preparation posture, the origin position of the user on the field is detected, and in step S05, at the end time point, that is, when the user returns to the preparation posture , detecting the end position of the user on the field, and in step S06 , comparing the position difference between the origin position and the end position. The above-mentioned origin position and end position are to detect the position of the user's heel on the field.

After step S02, that is, after the user starts to move, step S03 is performed to detect the corresponding body recognition positions of the user at each detection time point at multiple detection time points, that is to say, here In the step, the location of each body recognition location is detected in real time along with the user's actions. Next, in step S04, compare each body recognition position corresponding to each detection time point with the personal standard poomsae model. The position and the personal standard poomsae model are compared. Of course, this step can be performed at the same time when the user starts to perform the action, or after the user finishes the poomsae action.

In the final step S07 of this embodiment, the completion result of the user's Poomsae is obtained according to the comparison results of each item with the standard Poomsae model. The completion result described in this step can be used to judge the completion of Poomsa, or to score Poomsa according to the scoring standard.

Wherein, the start time point and the end time point correspond to the start instruction and the end instruction respectively, and the start instruction and the end instruction can be gesture instructions and/or sound instructions.

Please refer to FIG. 3, which is a schematic flow chart of another embodiment of the present invention. First of all, it should be explained that the body recognition position refers to the movement part and body part of poomsae. The settings, in terms of action parts, are divided into fists, palms, elbows, knees, feet and other parts, while body parts are divided into upper, middle and lower parts. In principle, the attack action of the upper hand is aimed at the middle of the person, the attack action of the foot is to kick to the upper 10 cm, the attack action of the middle end is mainly the heart, and the attack action of the lower end is to attack the abdomen under the Dao belt Furthermore, the requirement of returning to the original point is also calculated based on the length of the sole of the user's own foot. It can be seen that Taekwondo Poomsae is determined by whether the user's movements reach the user's own body parts. Whether the action is standard or not. Therefore, the difference between the flow chart in Fig. 3 and Fig. 1 is that in step S21, each body identification position on the user's body is detected, the personal body characteristics are calculated according to the body identification position, and then each body is classified according to each set of products. Identify the location and personal physical characteristics to establish a complex personal standard poinsema model. The personal physical characteristics include at least the length of the soles of the feet, the width of the shoulders, the upper end position, the middle end position, the lower end position, and the pace. For example, steps include lunges, horse steps, three-seven steps, etc. The parts that are the same as those in the process in Fig. 1 will not be repeated here.

In addition, in the scoring standard of Taekwondo poomsa competition, it also includes shouting under appropriate movements. The shouting part is based on the requirements of each poomsa content, and needs to be performed under different movements. Based on the above, please refer to FIG. 4, which is a schematic flow chart of another embodiment of the present invention. The parts in Fig. 4 that are the same as those in Fig. 1 will not be repeated here. The difference from the process in Fig. 1 is that in step S31, each body recognition position on the user's body is detected, and according to each Set up poom potential to establish multiple personal standard poom potential models for each body recognition position, and further capture the user's voice to integrate the user's voice into a specific personal standard poom potential model; in step S33, in the complex detection At the time point, detect each body recognition position corresponding to the user at each detection time point and the sound made by the user; in step S34, compare each body recognition position and the corresponding body recognition position at each detection time point Personal standard poompotential model, compared with each detection time The user's voice and personal standard poomsae model at the time point. According to the above, the difference between the flow method in Figure 4 and that in Figure 1 is that the user's voice is extracted and compared, and when Poomsae obtains the completed result, the voice comparison result is also referred to.

In addition, there are also time requirements in poomsa scoring standards. The time requirements of each poomsa will have different specifications due to the different number of movements or frets. In poomsa, each movement also has a time requirement. Based on the above, please refer to FIG. 5, which is a schematic flow chart of another embodiment of the present invention. The parts in Fig. 5 that are the same as those in Fig. 1 will not be repeated here. The difference from the process in Fig. 1 is that in step S41, each body recognition position on the user's body is detected, and according to each A set of poomsae establishes multiple personal standard poomsae models for each body recognition position, and integrates each standard poomsae time corresponding to each set of poomsae and the standard action time included in each set of poomsae into a specific personal standard poomsae model ; In step S43, under the multiple detection time points, detect each corresponding body recognition position of the user at each detection time point and the sound issued by the user, and calculate each action time; in step S44, compare For each body recognition position corresponding to each detection time point and the specific personal standard poompotence model, compare the action time with the specific personal standard poompotence model; in step S47, in addition to comparing the origin position and the end point position, compare the elapsed time from the start time point to the end time point with the specific personal standard poomsae model and compare the action time with the specific personal standard poomsae model. According to the above, the difference between the process method in Figure 5 and Figure 1 lies in the integration of standard Poomsa time and standard action time in the standard Poomsa model, and the addition of a comparison between Poomsae time and action time. When the completion result is obtained, the time comparison result will also be referred to.

Based on the above embodiments, please refer to FIG. 6, which is a schematic flow chart of another embodiment of the present invention. The embodiment of Fig. 6 is a combination of Fig. 1, Fig. 3, Fig. 4 and Fig. 5 The embodiment. The taekwondo poomsae movement detection and comparison method includes the following steps: first, in step S51, detect each body identification position on the user's body, calculate the personal body characteristics according to the body identification position, and then calculate the individual body characteristics according to each set of poomsae Body recognition position and personal physical characteristics Establish multiple personal standard poomsa models, capture the user's voice to integrate the user's voice into a specific personal standard poomsa model, and set the standard poomsa time corresponding to each set of poomsa And the standard action time included in each set of poomsae is integrated into a specific personal standard poomsae model; in step S52, at the start time point, the origin position of the user on the field is detected, and the user begins to perform poomsae; In step S53, at the plurality of detection time points, detect each body recognition position of the user corresponding to each detection time point and the sound made by the user, and calculate the time of each action; in step S54, compare Each body recognition position corresponding to each detection time point and the specific personal standard poomsae model, compare the user's voice at each detection time point with the specific personal standard poosera model, compare the action time with the specific In step S55, at the end time point, detect the end position of the user on the field; in step S56, compare the origin position and end position, and calculate the distance from the start time point to the end time point The elapsed time, and compare the elapsed time from the start time point to the end time point with the specific personal standard poomsae model; finally, in step S57, the completion result of the user's poomsae is obtained according to the comparison result.

In order to present the technical content and spirit of this embodiment more clearly, the following will take the first chapter of Taiji Pinshi as an example to describe the steps of the embodiment in Figure 6 in detail. There are 20 movements included in the Tai Chi chapter. First, when performing step S51, detect each body recognition position of the user, and calculate personal body characteristics such as shoulder width, foot length, upper end position, and middle end position according to the body recognition position , lower end position, steps, etc., to establish a standard model for each body recognition position and the above 20 movements. For example, the second movement in Chapter 1 of Tai Chi is "close the right foot Step forward, and at the same time make a middle-end opposing right arm attack with the right arm", the detected body recognition position is combined with personal body characteristics to establish a standard model that conforms to the action description, that is, according to the action requirements in poomsae. Establish standard models for each set of poomsae. It should be noted that due to the different body shapes of each user, the body recognition positions such as the upper and middle positions are also different, resulting in differences in the speed and angle of the user's movements. The personal standard Poomsae models will not be exactly the same. In step S1, the user's voice is further captured to integrate into the personal standard poomsae model. In the chapter of Taiji, the eighteenth movement is an action that requires shouting, so in this step, the user's voice is The sound is integrated into the 18th movement in Chapter 1 of Taiji "Put your right foot forward and step forward to form a right lunge, and at the same time make a middle-end opposition to the right fist with your right arm". It should be explained that due to the need for shouting For example, in the third chapter of Taiji, it is the 20th movement "retracting the right foot forward and then stepping forward to form a right forward step. Therefore, in this step, the user's voice will be integrated into a specific personal standard Poomsae model according to the requirements of Poomsae. In step S51, the time specified by each set of poomsae and the standard action time contained in each set of poomsae are integrated into the personal standard poomsae model. In terms of the time specified by poomsae, for example, the requirements in the chapter of Tai Chi 17 seconds to 21 seconds to complete, then the personal standard poomsae model of the first chapter of Taiji is set to be completed in 17 seconds to 21 seconds, and the standard action time, for example, the two-handed knife on both sides of the defense needs to be completed in 4 seconds to 21 seconds. It takes 5 seconds to complete, clenching and raising the fist needs to be completed within 4 to 5 seconds, and the mid-range outer wrist side defense and bow cloth outer wrist upper end defense need to be completed within 7 seconds to 8 seconds. By analogy, the time required for each action is integrated into a specific personal standard poomsae model.

Next, proceed to step S52, at the start time point, that is, when preparing for the parallel preparation posture of the first chapter of Tai Chi, it is detected that the user is located at the origin position of the field. And go to step S53 At multiple detection time points, that is, when the user starts the Tai Chi movement, continuously detect the user's various body recognition positions, and detect the user's cry, and calculate the user's Tai Chi movement. Time spent on individual actions in the chapter.

Next, in step S54, compare the detected body recognition position, the voice and the 20 action times of the user with the personal standard poomsae model of Tai Chi chapter established in step S51.

When performing step S55, at the end time point, that is, when the parallel preparation posture is restored, it is detected that the user is located at the end position of the field, and then step S56 is performed to compare the user's origin position and end position, that is, the start of comparison and the end of the parallel preparation posture, the position of the heel on the field and the time elapsed from the start time point to the end time point of the user are calculated, and the elapsed time from the first time point to the end time point is compared with the personal standard of Chapter 1 of Tai Chi The Poomsae model is compared, that is, whether the elapsed time of the comparison falls between 17 seconds and 21 seconds. Finally, step S57 is performed to obtain the completion result of the user's Tai Chi chapter according to all the above comparison results.

In Taekwondo poomsae competitions, there are also events in which many people participate. This method can also establish multiple personal standard poomsae models corresponding to multiple users at the same time, and simultaneously detect the location of multiple users and conduct Poomsae time, after obtaining the completion results of each user, can analyze the completion results of each user to confirm the consistency of the completion results.

Based on the above, it can be known that this method first establishes the user's personal standard poomsae model according to the user's body recognition position, and integrates the user's voice and the time required by each poomsae into the specific personal standard poomsae model, Detect the origin position of the user on the field before the action starts, and continuously detect specific body parts and use The voice of the user, and compare the detected information and the information obtained after calculation with the personal standard poomsae model, and compare the end position at the end of poomsae with the origin position at the beginning, so as to Judge the degree of completion of the contestant's Poomsae, or give a score. Using this method can not only reduce possible referee misjudgments in Poomsae competitions, but also make it more efficient and accurate to know the results in Poomsae competitions or exercises. When there are many people participating in Poomsae competitions, it can be used for plural Analyze the corresponding completion results of each operator to confirm the consistency of each completion result.

Although the present invention is disclosed above with the aforementioned embodiments, it is not intended to limit the present invention. Any person familiar with similar skills may make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of patent protection shall be subject to what is defined in the scope of patent application attached to this manual.

S01, S02, S03, S04, S05, S06, S07: steps

Claims (13)

A taekwondo poomsae motion detection and comparison method, suitable for detecting at least one user performing at least one set of poomsae in a taekwondo poomsae venue, the method includes the following steps: detecting multiple body identifications on the user's body position, and based on each set of poomsae to establish multiple personal standard poomsae models for these body recognition positions; from the start time of a taekwondo poomsae, the user's toes are detected at a position of the taekwondo poomsae field The origin position, and the user starts to perform the set of poomsae; at multiple detection time points, detect the corresponding body recognition positions of the user at each of the time points; and end at the end of a taekwondo poomsae From the point in time, the user’s toes are detected at an end position of the Taekwondo poomsae field; wherein, the body recognition positions and the personal standard poomsae models corresponding to the time points are compared, and Comparing the origin position and the end position to obtain a completion result of the user performing the set of poses; where the body recognition positions on the user's body are detected by the user posing a standing human body Start detection after dissecting the pose. The Taekwondo poomsa movement detection and comparison method as described in Claim 1, wherein before the Taekwondo poomsa start time point, the user’s voice is further captured, and the user’s voice is integrated into a specific at least A model of the individual standard poompotential. The taekwondo poomsae movement detection and comparison method described in claim 2 further detects the user's voice at each detection time point. The taekwondo poomsae motion detection and comparison method as described in claim 2, wherein after obtaining the completion result Before the results, compare the voices uttered by the user at the detected time points with at least one specific personal standard poinsema model. The taekwondo poomsae motion detection and comparison method as described in Claim 1, wherein the sets of poomsae have a plurality of standard poomsae times and each set of poomsae contains a plurality of standard action times, at the start time Before the point, the standard poomsae times corresponding to these sets of poomsae and the multiple standard action times included in each set of poomsae are integrated into at least one specific personal standard poomsae model. The taekwondo poomsae motion detection and comparison method as described in claim item 5, wherein before the completion result is obtained, the time elapsed from the start time point to the end time point is calculated, and compared with the start time point to the The elapsed time of the end time point is related to at least one personal standard poinsema model. The taekwondo poomsae movement detection and comparison method as described in claim 5, wherein before obtaining the completion result, the time for the user to perform the plurality of movements of the at least one set of poomsae is further calculated, and these movements are compared Time is associated with at least one standard poomsae model specific to the individual. The taekwondo poomsae movement detection and comparison method as described in Claim 1, wherein the body recognition positions include at least the head, shoulder joint, xiphoid process, hip joint, wrist joint, elbow joint, finger joint, and knee Joints, ankle joints, balls of feet, toes. The Taekwondo poomsae motion detection and comparison method as described in Claim 1, wherein when detecting the body identification positions on the user’s body, multiple personal body characteristics are calculated based on the body identification positions, and then based on each The set of poomsae establishes a plurality of personal standard poomsae models for these body identification positions and these personal physical characteristics. The taekwondo poomsae movement detection and comparison method as described in Claim 9, wherein the personal physical characteristics at least include foot length, shoulder width, upper end position, middle end position, lower end position, and pace. The Taekwondo Poomsae motion detection and comparison method as described in Claim 1, wherein the Taekwondo Poomsae start time point and the Taekwondo Poomsae end time point correspond to a start instruction and an end instruction respectively, the start instruction and the The end indication can be gesture indication and/or sound indication. The taekwondo poomsae motion detection and comparison method as described in Claim 1, wherein when there are plural users, the plural completion results corresponding to the plural users can be analyzed to confirm the plural completion results consistency. The taekwondo poomsae movement detection and comparison method as described in Claim 1 further judges the degree of completion and/or scores the set of poomsae according to the completion result.

Images