Head acupuncture point recognition method, device, acupuncture robot and digital human

Abstract

The present invention provides a method, device, acupuncture robot and digital human for head acupoint recognition, belonging to the technical field of acupoint recognition, and the method comprises: determining the front hairline of a head image to be recognized based on a face detection algorithm; determining the Yintang acupoint, the left frontal hair corner point and the right frontal hair corner point of the head image to be recognized based on a key point detection algorithm; determining the longitudinal distance between the Yintang acupoint and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point; determining the unit straight inch and the unit horizontal inch based on the longitudinal distance, the transverse distance and the head bone measurement method; determining the positional relationship between the head acupoint and the front hairline, and determining the head acupoint based on the positional relationship, the unit straight inch and the unit horizontal inch. The present invention improves the accuracy of head acupoint recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of acupoint recognition, and specifically to a head acupoint recognition method, device, acupuncture robot and digital human. Background Art

[0002] As people pay more and more attention to their health, more and more people are choosing acupuncture as a treatment for physical discomfort. The premise of acupuncture is to find accurate acupoint identification.

[0003] There are two mainstream acupoint recognition methods at present. One is to recognize acupoints based on deep learning as disclosed in CN111598949B. It needs to obtain multiple training samples based on artificial marking, and then train the initial model based on multiple training samples to obtain an acupoint recognition model that can be used for acupoint recognition after training. The other is to realize the positioning of acupoints based on bone measurement scale as disclosed in CN114041992A. There are some problems in the above two methods when identifying acupoints on the head. Specifically, they rely on artificial marking to obtain training samples. The recognition accuracy of acupoints depends entirely on the accuracy of samples. The characteristics of the head are not obvious, and the sample accuracy is difficult to guarantee. When the sample accuracy is low, the recognition accuracy of head acupoints is low. The bone measurement scale is composed of multiple telescopic rods, which cannot identify and locate acupoints on the head with a large curvature.

[0004] Therefore, there is an urgent need to provide a head acupuncture point recognition method, device, acupuncture robot and digital human to improve the recognition efficiency of head acupuncture points. Summary of the invention

[0005] In view of this, it is necessary to provide a head acupuncture point recognition method, device, acupuncture robot and digital human to solve the technical problem that the head acupuncture points cannot be accurately recognized in the prior art.

[0006] In one aspect, the present invention provides a method for identifying acupuncture points on the head, comprising:

[0007] Determine the front hairline of the head image to be identified based on the face detection algorithm;

[0008] Determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be identified based on the key point detection algorithm;

[0009] Determine the longitudinal distance between the Yintang point and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point;

[0010] Determine the unit vertical inch and the unit horizontal inch based on the longitudinal distance, the transverse distance and the head bone measurement method;

[0011] The positional relationship between the head acupuncture point and the front hairline is determined, and the head acupuncture point is determined based on the positional relationship, the unit vertical inch and the unit horizontal inch.

[0012] In some possible implementations, determining the front hairline of the head image to be identified based on a face detection algorithm includes:

[0013] Determine a face rectangular frame of the head image to be identified based on the face detection algorithm;

[0014] The frame close to the top of the head in the face rectangular frame is used as the front hairline.

[0015] In some possible implementations, determining the Yintang acupoint of the head image to be identified based on a key point detection algorithm includes:

[0016] Determine a first key point and a second key point symmetrical about the Yintang acupoint based on the key point detection algorithm;

[0017] The midpoint between the first key point and the second key point is used as the Yintang acupoint.

[0018] In some possible implementations, the head bone measurement method includes the longitudinal bone measurement between the Yintang point and the front hairline and the transverse bone measurement between the left frontal hair corner point and the right frontal hair corner point;

[0019] Then the unit vertical inch and the unit horizontal inch are determined based on the longitudinal distance, the transverse distance and the head bone measurement method;

[0020] The ratio of the longitudinal distance to the longitudinal bone inch is taken as the unit vertical inch, and the ratio of the transverse distance to the transverse bone inch is taken as the unit transverse inch.

[0021] In some possible implementations, the head acupuncture points include at least one first acupuncture point and at least one second acupuncture point, the first acupuncture point is an acupuncture point associated with the front hairline, the second acupuncture point is an acupuncture point associated with the back hairline, and the positional relationship includes a first positional sub-relationship and a second positional sub-relationship;

[0022] Then the step of determining the positional relationship between the head acupoints and the front hairline includes:

[0023] Acquire a first positional sub-relationship between the at least one first acupuncture point and the front hairline;

[0024] Obtaining the longitudinal bone inch relationship between the front hairline and the back hairline;

[0025] The second position sub-relationship is determined based on the position association relationship between the at least one second acupuncture point and the rear hairline and the longitudinal bone-inch relationship.

[0026] In some possible implementations, the first acupuncture points include Houding, Baihui, Qianding, Xinhui, Shangxing, Shenting, Qucha, Meichong, Wuchu, Chengguang, Tongtian and Luoque;

[0027] The first position sub-relationship is:

[0028] The back vertex is located at the midline of the front hairline and is 5.5 straight inches away from the front hairline;

[0029] The Baihui point is located on the midline of the front hairline and is 5 inches away from the front hairline;

[0030] The front vertex is located at the midline of the front hairline and is 3.5 straight inches away from the front hairline;

[0031] The fontanelle is located at the midline of the anterior hairline and is 2 inches away from the anterior hairline;

[0032] The upper star is located on the midline of the front hairline and is 1 inch away from the front hairline;

[0033] The Shenting is located at the midline of the front hairline and is 0.5 straight inches away from the front hairline;

[0034] The curve is 1.5 horizontal inches away from the median line of the front hairline and 0.5 vertical inches away from the front hairline;

[0035] The meichong is located at the midpoint of the line connecting the shenting and the qucha;

[0036] The five points are 1.5 horizontal inches away from the midline of the front hairline and 1 vertical inch away from the front hairline;

[0037] The said light-carrying part is 1.5 horizontal inches away from the midline of the said front hairline, and 2.5 vertical inches away from the said front hairline;

[0038] The Tongtian point is 1.5 horizontal inches away from the midline of the front hairline and 4 vertical inches away from the front hairline;

[0039] The Luoque point is 1.5 horizontal inches away from the midline of the front hairline and 5.5 vertical inches away from the front hairline;

[0040] The front hairline midline passes through the Yintang acupoint, is perpendicular to the line between the left forehead hair corner point and the right forehead hair corner point, and fits the head curve of the head arc surface.

[0041] In some possible implementations, the second acupuncture points include Yamen, Fengfu, Naohu, Qiangjian and Yuzhen;

[0042] The position association relationship is:

[0043] The dummy gate is located at the midline of the posterior hairline and is 0.5 inches away from the posterior hairline;

[0044] The Fengfu is located on the midline of the posterior hairline and is 1 inch away from the posterior hairline;

[0045] The Naohou is located at the midline of the posterior hairline and is 2.5 inches away from the posterior hairline;

[0046] The strong space is located at the midline of the posterior hairline and is 4 inches away from the posterior hairline;

[0047] The occipital bone is 1.3 horizontal inches away from the midline of the posterior hairline and 2.5 vertical inches away from the posterior hairline;

[0048] Wherein, the posterior hairline midline is the extension line of the anterior hairline midline along the curved surface of the head.

[0049] On the other hand, the present invention also provides a head acupuncture point recognition device, comprising:

[0050] A front hairline recognition unit, used to determine the front hairline of the head image to be recognized based on a face detection algorithm;

[0051] A forehead hair corner point recognition unit, used to determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be recognized based on a key point detection algorithm;

[0052] A longitudinal and transverse distance determination unit, used to determine the longitudinal distance between the Yintang acupoint and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point;

[0053] A unit vertical and horizontal inch determination unit, configured to determine a unit vertical inch and a unit horizontal inch based on the longitudinal distance, the horizontal distance, and a head bone measurement method;

[0054] The head acupuncture point recognition unit is used to determine the positional relationship between the head acupuncture point and the front hairline, and determine the head acupuncture point based on the positional relationship, the unit vertical inch and the unit horizontal inch.

[0055] On the other hand, the present invention also provides an acupuncture robot, comprising an acupuncture point recognition subsystem for determining acupuncture points on the head and a moxibustion subsystem for performing acupuncture based on the acupuncture points on the head, wherein the acupuncture point recognition subsystem comprises a memory and a processor, wherein:

[0056] The memory is used to store programs;

[0057] The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the head acupoint recognition method in any one of the possible implementations described above.

[0058] On the other hand, the present invention also provides a digital human, including a virtual head model and head acupoints marked in the virtual head model, wherein the head acupoints are determined based on a head acupoint recognition method, and the head acupoint recognition method is the head acupoint recognition method described in any one of the possible implementation methods mentioned above.

[0059] The beneficial effect of adopting the above embodiment is: the head acupoint recognition method provided by the present invention, by setting the front hairline determined based on the face detection algorithm, and determining the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point based on the key point detection algorithm, does not need to directly perform head acupoint recognition through the face detection algorithm or the key point detection algorithm. Compared with the head acupoints, the hairline, Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point have obvious features, and the four key features of the front hairline, Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point can be accurately identified. Then, the head acupoints can be recognized based on the accurately identified four key features and the positional relationship, the unit horizontal inch and the unit vertical inch. There is no need to mark the head acupoints, which eliminates the marking process, avoids the problem of inaccurate head acupoint recognition caused by inaccurate marking, and improves the recognition accuracy of the head acupoints.

[0060] Furthermore, by determining the unit vertical inch and unit horizontal inch by the head bone measurement method, and then determining the head acupuncture points based on the positional relationship, unit vertical inch and unit horizontal inch, the influence of individual differences can be eliminated, and the recognition accuracy of head acupuncture points can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0062] Figure 1 A schematic diagram of a flow chart of an embodiment of a method for identifying acupuncture points on the head provided by the present invention;

[0063] Figure 2 For the present invention Figure 1 A schematic flow chart of an embodiment of S101;

[0064] Figure 3 A schematic diagram of an embodiment of a face rectangular frame recognition result provided by the present invention;

[0065] Figure 4 For the present invention Figure 1 A schematic diagram of an embodiment of determining the Yintang acupoint in step S102;

[0066] Figure 5For the present invention Figure 1 A schematic diagram of an embodiment of determining the position relationship in step S105;

[0067] Figure 6 A schematic diagram of the structure of an embodiment of the head acupuncture point identification device provided by the present invention;

[0068] Figure 7 This is a schematic structural diagram of an embodiment of the acupuncture robot provided by the present invention. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0070] It should be understood that the schematic drawings are not drawn to scale. The flowchart used in the present invention shows the operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present invention, can add one or more other operations to the flowchart, and can also remove one or more operations from the flowchart. Some of the block diagrams shown in the accompanying drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0071] The descriptions of "first", "second", etc. involved in the embodiments of the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0072] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0073] The present invention provides a head acupuncture point recognition method, device, acupuncture robot and digital human, which are respectively described below.

[0074] Figure 1 FIG. 1 is a flow chart of an embodiment of a method for identifying acupuncture points on the head according to an embodiment of the present invention. Figure 1 As shown, the head acupoint identification method includes:

[0075] S101, determining the front hairline of the head image to be identified based on a face detection algorithm;

[0076] S102, determining the Yintang acupoint, the left forehead hair corner point, and the right forehead hair corner point of the head image to be identified based on a key point detection algorithm;

[0077] S103, determining the longitudinal distance between the Yintang point and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point;

[0078] S104, determining the unit vertical inch and the unit horizontal inch based on the longitudinal distance, the transverse distance and the head bone measurement method;

[0079] S105, determining the positional relationship between the head acupuncture points and the front hairline, and determining the head acupuncture points based on the positional relationship, the unit vertical inch, and the unit horizontal inch.

[0080] To avoid time consumption during algorithm training, in a specific embodiment of the present invention, the face detection algorithm in step S101 is an existing multi-task cascaded convolutional neural network (MTCNN) algorithm. The key point detection algorithm in step S102 is a Mediapipe algorithm.

[0081] The embodiment of the present invention sets the face detection algorithm and the key point detection algorithm as open source, existing algorithms, which can save the algorithm training process and improve the recognition efficiency of head acupoints. Furthermore, the head acupoint recognition is realized by using the existing algorithm, which improves the compatibility of the head acupoint recognition method proposed in the embodiment of the present invention with the existing algorithm.

[0082] Specifically, the MTCNN algorithm is a face detection and face alignment method based on deep learning. It can complete the tasks of face detection and face alignment at the same time. Compared with traditional algorithms, it has better performance and faster detection speed. The Mediapipe algorithm can extract 478 key points of the head and face to be identified.

[0083] Compared with the prior art, the head acupoint recognition method provided by the embodiment of the present invention determines the front hairline based on a face detection algorithm, and determines the Yintang acupoint, the left frontal hair corner point and the right frontal hair corner point based on a key point detection algorithm. There is no need to directly perform head acupoint recognition through a face detection algorithm or a key point detection algorithm. Compared with the head acupoints, the hairline, the Yintang acupoint, the left frontal hair corner point and the right frontal hair corner point have obvious features, and the four key features of the front hairline, the Yintang acupoint, the left frontal hair corner point and the right frontal hair corner point can be accurately recognized. Then, the head acupoints can be recognized based on the accurately recognized four key features and their positional relationship, unit horizontal inch and unit vertical inch. There is no need to mark the head acupoints, which eliminates the marking process, avoids the problem of inaccurate head acupoint recognition caused by inaccurate marking, and improves the recognition accuracy of the head acupoints.

[0084] Furthermore, by determining the unit vertical inch and unit horizontal inch by the head bone measurement method, and then determining the head acupoints based on the positional relationship, unit vertical inch and unit horizontal inch, the influence of individual differences can be eliminated and the recognition accuracy of head acupoints can be further improved.

[0085] In some embodiments of the present invention, Figure 2 As shown, step S101 includes:

[0086] S201, determining a face rectangular frame of a head image to be identified based on a face detection algorithm;

[0087] S202: Use the border close to the top of the head in the face rectangular frame as the front hairline.

[0088] The embodiment of the present invention uses the frame close to the top of the head in the facial rectangular frame as the front hairline, without the need for complex processes such as labeling and training the front hairline, thereby further improving the efficiency of determining the front hairline and further improving the efficiency of recognizing head acupuncture points.

[0089] In a specific embodiment of the present invention, Figure 3 As shown, the rectangular frame in the figure is the face rectangular frame. The line segment AB in the figure is the longitudinal distance, and the line segment CD is the transverse distance.

[0090] In some embodiments of the present invention, Figure 4 As shown, determining the Yintang acupoint in step S102 includes:

[0091] Based on the key point detection algorithm, determine the first key point and the second key point symmetrical about the Yintang point; and take the midpoint of the first key point and the second key point as the Yintang point.

[0092] When the key point detection algorithm is the Mediapipe algorithm, such as Figure 4As shown, key points No. 103 and No. 332 are the left forehead hair corner point and the right forehead hair corner point respectively, and key points No. 8 and No. 9 are the first key point and the second key point respectively.

[0093] In a specific embodiment of the present invention, the head bone measurement method includes the longitudinal bone measurement between the Yintang point and the front hairline and the transverse bone measurement between the left frontal hair corner point and the right frontal hair corner point.

[0094] Specifically, the longitudinal bone inch is 3 inches, and the transverse bone inch is 9 inches, that is: Figure 3 The vertical length is 3 inches and the horizontal length is 9 inches.

[0095] It should be noted that the "inch" in the embodiment of the present invention is not a commonly used unit of measurement in life, but an equal division unit. For different people, the length of the equal division unit is different.

[0096] In a specific embodiment of the present invention, step S104 is specifically:

[0097] The ratio of the longitudinal distance to the longitudinal bone inch is taken as the unit vertical inch, and the ratio of the transverse distance to the transverse bone inch is taken as the unit transverse inch.

[0098] Step S104 is specifically as follows: determining the bone position of the head acupuncture point based on the position relationship, and then taking the product of the bone position and the unit vertical inch and the unit horizontal inch as the position of the head acupuncture point in the head image to be identified.

[0099] In a specific embodiment of the present invention, the head acupoints include Yamen, Fengfu, Naohu, Qiangjian, Houding, Baihui, Qianding, Xinhui, Shangxing, Shenting, Meichong, Qucha, Wuchu, Chengguang, Tongtian, Luoque and Yuzhen.

[0100] Since the head acupuncture points are distributed on the front and back of the head, if all the head acupuncture points are located through the front hairline, the error will be accumulated as the distance between the acupuncture points and the front hairline increases. In order to eliminate this accumulation problem, in some embodiments of the present invention, the head acupuncture points include at least one first acupuncture point and at least one second acupuncture point, the first acupuncture point is an acupuncture point associated with the front hairline, the second acupuncture point is an acupuncture point associated with the back hairline, and the position relationship includes a first position sub-relationship and a second position sub-relationship;

[0101] If Figure 5 As shown, the step S105 of determining the positional relationship between the head acupoints and the front hairline includes:

[0102] S501, obtaining a first position sub-relationship between at least one first acupuncture point and the front hairline;

[0103] S502, obtaining the longitudinal bone inch relationship between the front hairline and the back hairline;

[0104] S503: Determine a second position sub-relationship based on the position association relationship and the longitudinal bone-inch relationship between at least one second acupuncture point and the posterior hairline.

[0105] Among them, the second position sub-relationship can be obtained by superimposing the longitudinal bone-inch relationship and the position association relationship.

[0106] The embodiment of the present invention divides the head acupoints into a first acupoint associated with the front hairline and a second acupoint associated with the back hairline, and then determines the second positional relationship based on the accurate longitudinal bone-inch relationship between the front hairline and the back hairline. This can avoid the technical problem of inaccurate positioning of the head acupoints when the head acupoints are far away from the front hairline, further improves the accuracy of the positional relationship, and thus further improves the accuracy of head acupoint recognition.

[0107] Among them, the longitudinal bone relationship is: the distance between the front hairline and the back hairline is 12 inches.

[0108] Specifically, the longitudinal bone cun between the Yintang point and the front hairline is 3 inches, and the distance between the front hairline and the back hairline is 4 times the longitudinal bone cun.

[0109] In a specific embodiment of the present invention, the first position sub-relationship is:

[0110] The back vertex is located at the midline of the front hairline and is 5.5 straight inches away from the front hairline;

[0111] Baihui is located on the midline of the front hairline and is 5 inches away from the front hairline;

[0112] The front vertex is located at the midline of the front hairline and is 3.5 straight inches away from the front hairline;

[0113] The fontanelle is located on the midline of the anterior hairline and is 2 inches away from the anterior hairline.

[0114] The upper star is located on the midline of the anterior hairline and is 1 inch away from the anterior hairline;

[0115] Shenting is located at the midline of the front hairline and is 0.5 inches away from the front hairline;

[0116] The curve is 1.5 inches horizontally away from the median line of the front hairline and 0.5 inches vertically away from the front hairline;

[0117] Meichong is located at the midpoint of the line connecting Shenting and Qucha;

[0118] Five points are 1.5 horizontal inches away from the midline of the front hairline and 1 vertical inch away from the front hairline;

[0119] Chengguang is 1.5 horizontal inches away from the midline of the front hairline and 2.5 vertical inches away from the front hairline;

[0120] Tongtian acupoint is 1.5 horizontal inches away from the midline of the front hairline and 4 vertical inches away from the front hairline;

[0121] Luoque point is 1.5 horizontal inches away from the midline of the front hairline and 5.5 vertical inches away from the front hairline;

[0122] Among them, the midline of the front hairline passes through the Yintang point, is perpendicular to the line between the left forehead hair corner point and the right forehead hair corner point, and fits the head curve of the arc surface of the head.

[0123] In a specific embodiment of the present invention, the position association relationship is:

[0124] The dummy gate is located at the midline of the posterior hairline and is 0.5 inches away from the posterior hairline;

[0125] Fengfu is located on the midline of the posterior hairline, and is 1 inch away from the posterior hairline;

[0126] Naohou is located at the midline of the posterior hairline and is 2.5 inches away from the posterior hairline;

[0127] The strong point is located on the midline of the posterior hairline and is 4 inches away from the posterior hairline;

[0128] The occipital bone is 1.3 horizontal inches away from the midline of the posterior hairline and 2.5 vertical inches away from the posterior hairline;

[0129] Among them, the posterior hairline midline is the extension line of the anterior hairline midline along the arc surface of the head.

[0130] It should be noted that in order to verify the accuracy of the identified head acupoints, the head acupoints can be verified. For example, if Yamen is located below the first cervical vertebra, then after Yamen is identified, the identification result can be verified based on the first cervical vertebra. Another example: Fengfu is located in the depression between the trapezius muscles on both sides, then after Fengfu is identified, the identification result can be verified based on the two trapezius muscles to ensure the accuracy of the identified head acupoints.

[0131] In order to better implement the head acupoint recognition method in the embodiment of the present invention, based on the head acupoint recognition method, the embodiment of the present invention also provides a head acupoint recognition device, such as Figure 6 As shown, the head acupoint recognition device 600 includes:

[0132] A front hairline recognition unit 601 is used to determine the front hairline of the head image to be recognized based on a face detection algorithm;

[0133] A forehead hair corner point recognition unit 602 is used to determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be recognized based on a key point detection algorithm;

[0134] A longitudinal and transverse distance determination unit 603 is used to determine the longitudinal distance between the Yintang point and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point;

[0135] A unit vertical and horizontal dimension determination unit 604, configured to determine a unit vertical dimension and a unit horizontal dimension based on the longitudinal distance, the transverse distance and the head bone measurement method;

[0136] The head acupuncture point recognition unit 605 is used to determine the positional relationship between the head acupuncture points and the front hairline, and determine the head acupuncture points based on the positional relationship, unit vertical inch and unit horizontal inch.

[0137] The head acupoint recognition device 600 provided in the above embodiment can implement the technical solution described in the above head acupoint recognition method embodiment. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above head acupoint recognition method embodiment, which will not be repeated here.

[0138] like Figure 7 As shown, the present invention also provides an acupuncture robot 700. The acupuncture robot 700 includes an acupuncture point identification subsystem 710 for determining acupuncture points on the head and a moxibustion subsystem 720 for performing acupuncture based on acupuncture points on the head. The acupuncture point identification subsystem 710 includes a processor 711, a memory 712 and a display 713. Figure 7 Only some of the components of the acupuncture robot 700 are shown, but it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.

[0139] In some embodiments, the memory 712 may be an internal storage unit of the acupoint identification subsystem 710, such as a hard disk or memory of the acupoint identification subsystem 710. In other embodiments, the memory 712 may also be an external storage device of the acupoint identification subsystem 710, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the acupoint identification subsystem 710.

[0140] Furthermore, the memory 712 may include both an internal storage unit of the acupoint identification subsystem 710 and an external storage device. The memory 712 is used to store application software installed in the acupoint identification subsystem 710 and various data.

[0141] In some embodiments, the processor 711 may be a central processing unit (CPU), a microprocessor or other data processing chip, and is used to run program codes or process data stored in the memory 712, such as the head acupuncture point recognition method of the present invention.

[0142] In some embodiments, the display 713 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 713 is used to display information of the acupoint identification subsystem 710 and to display a visual user interface. The components 711-713 of the acupoint identification subsystem 710 communicate with each other through a system bus.

[0143] In some embodiments of the present invention, when the processor 711 executes the head acupuncture point recognition program in the memory 712, the following steps may be implemented:

[0144] Determine the front hairline of the head image to be identified based on the face detection algorithm;

[0145] Determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be identified based on the key point detection algorithm;

[0146] Determine the longitudinal distance between the Yintang point and the front hairline, and the lateral distance between the left and right frontal hair corners;

[0147] Determine the unit vertical inch and unit horizontal inch based on the longitudinal distance, transverse distance and head bone measurement method;

[0148] Determine the positional relationship between the head acupuncture points and the front hairline, and determine the head acupuncture points based on the positional relationship, unit vertical inch and unit horizontal inch.

[0149] It should be understood that: when the processor 711 executes the head acupuncture point recognition program in the memory 712, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the previous related method embodiments.

[0150] Among them, Figure 7 As shown, the moxibustion subsystem 720 includes a path planning unit 721 and a moxibustion unit 722. The path planning unit 721 is used to perform path planning according to the moxibustion needle positions and acupoint coordinates in the moxibustion unit 722 to generate a planned path. The moxibustion unit 722 moves the moxibustion needle to the location of the acupoint according to the planned path.

[0151] The embodiment of the present invention can replace doctors to complete long-term acupuncture treatment work by providing an acupuncture robot 700, thereby alleviating the burden on doctors and reducing medical expenses.

[0152] On the other hand, an embodiment of the present invention further provides a digital human, including a virtual head model and head acupoints marked in the virtual head model, wherein the head acupoints are determined based on a head acupoint recognition method, and the head acupoint recognition method is the head acupoint recognition method in any one of the above embodiments.

[0153] The embodiment of the present invention can achieve training objectives through the provision of a digital human, such as training on the recognition of acupuncture points on the head.

[0154] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0155] The head acupoint recognition method, device, acupuncture robot and digital human provided by the present invention are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for identifying acupuncture points on the head, characterized in that: include: Determine the front hairline of the head image to be identified based on the face detection algorithm; The face detection algorithm is a multi-task cascade convolutional neural network; Determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be identified based on the key point detection algorithm; The key point detection algorithm is the Mediapipe algorithm; Determine the longitudinal distance between the Yintang point and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point; Determine the unit vertical inch and the unit horizontal inch based on the longitudinal distance, the transverse distance and the head bone measurement method; Determine the positional relationship between the head acupuncture point and the front hairline, and determine the head acupuncture point based on the positional relationship, the unit vertical inch and the unit horizontal inch; The head acupuncture points include at least one first acupuncture point and at least one second acupuncture point, the first acupuncture point is an acupuncture point associated with the front hairline, the second acupuncture point is an acupuncture point associated with the back hairline, and the position relationship includes a first position sub-relationship and a second position sub-relationship; Then the step of determining the positional relationship between the head acupuncture points and the front hairline includes: Acquire a first positional sub-relationship between the at least one first acupuncture point and the front hairline; Obtaining the longitudinal bone inch relationship between the front hairline and the back hairline; Determining the second position sub-relationship based on the position association relationship between the at least one second acupuncture point and the back hairline and the longitudinal bone-inch relationship; The first acupuncture points include Houding, Baihui, Qianding, Xinhui, Shangxing, Shenting, Qucha, Meichong, Wuchu, Chengguang, Tongtian and Luoque; The first position sub-relationship is: The back vertex is located at the midline of the front hairline and is 5.5 straight inches away from the front hairline; The Baihui point is located on the midline of the front hairline and is 5 inches away from the front hairline; The front vertex is located at the midline of the front hairline and is 3.5 straight inches away from the front hairline; The fontanelle is located at the midline of the anterior hairline and is 2 inches away from the anterior hairline; The upper star is located on the midline of the front hairline and is 1 inch away from the front hairline; The Shenting is located at the midline of the front hairline and is 0.5 straight inches away from the front hairline; The curve is 1.5 horizontal inches away from the median line of the front hairline and 0.5 vertical inches away from the front hairline; The meichong is located at the midpoint of the line connecting the shenting and the qucha; The five points are 1.5 horizontal inches away from the midline of the front hairline and 1 vertical inch away from the front hairline; The said light-carrying part is 1.5 horizontal inches away from the midline of the said front hairline, and 2.5 vertical inches away from the said front hairline; The Tongtian point is 1.5 horizontal inches away from the midline of the front hairline and 4 vertical inches away from the front hairline; The Luoque point is 1.5 horizontal inches away from the midline of the front hairline and 5.5 vertical inches away from the front hairline; The front hairline midline passes through the Yintang acupoint, is perpendicular to the line between the left frontal hair corner point and the right frontal hair corner point, and fits the head curve of the head arc surface; The second acupuncture points include Yamen, Fengfu, Naohu, Qiangjian and Yuzhen; The position association relationship is: The dummy gate is located at the midline of the posterior hairline and is 0.5 inches away from the posterior hairline; The Fengfu is located on the midline of the posterior hairline and is 1 inch away from the posterior hairline; The Naohou is located at the midline of the posterior hairline and is 2.5 inches away from the posterior hairline; The strong space is located at the midline of the posterior hairline and is 4 inches away from the posterior hairline; The occipital bone is 1.3 horizontal inches away from the midline of the posterior hairline, and 2.5 vertical inches away from the posterior hairline; Wherein, the posterior hairline midline is the extension line of the anterior hairline midline along the curved surface of the head.

2. The head acupuncture point identification method according to claim 1, characterized in that: The method of determining the front hairline of the head image to be identified based on the face detection algorithm includes: Determine a face rectangular frame of the head image to be identified based on the face detection algorithm; The frame close to the top of the head in the face rectangular frame is used as the front hairline.

3. The head acupuncture point identification method according to claim 1, characterized in that: The method of determining the Yintang acupoint of the head image to be identified based on the key point detection algorithm includes: Determine a first key point and a second key point symmetrical about the Yintang acupoint based on the key point detection algorithm; The midpoint between the first key point and the second key point is used as the Yintang acupoint.

4. The head acupuncture point identification method according to claim 1, characterized in that: The head bone measurement method includes the longitudinal bone measurement between the Yintang point and the front hairline and the transverse bone measurement between the left frontal hair corner point and the right frontal hair corner point; Then the unit vertical inch and the unit horizontal inch are determined based on the longitudinal distance, the transverse distance and the head bone measurement method; The ratio of the longitudinal distance to the longitudinal bone inch is taken as the unit vertical inch, and the ratio of the transverse distance to the transverse bone inch is taken as the unit transverse inch.

5. A head acupuncture point recognition device, characterized in that: include: A front hairline recognition unit, used to determine the front hairline of the head image to be recognized based on a face detection algorithm; The face detection algorithm is a multi-task cascade convolutional neural network; A forehead hair corner point recognition unit, used to determine the Yintang acupoint, the left forehead hair corner point and the right forehead hair corner point of the head image to be recognized based on a key point detection algorithm; The key point detection algorithm is the Mediapipe algorithm; A longitudinal and transverse distance determination unit, used to determine the longitudinal distance between the Yintang acupoint and the front hairline and the transverse distance between the left frontal hair corner point and the right frontal hair corner point; A unit vertical and horizontal inch determination unit, configured to determine a unit vertical inch and a unit horizontal inch based on the longitudinal distance, the horizontal distance, and a head bone measurement method; A head acupuncture point recognition unit, used to determine the positional relationship between the head acupuncture point and the front hairline, and determine the head acupuncture point based on the positional relationship, the unit vertical inch and the unit horizontal inch; The head acupuncture points include at least one first acupuncture point and at least one second acupuncture point, the first acupuncture point is an acupuncture point associated with the front hairline, the second acupuncture point is an acupuncture point associated with the back hairline, and the position relationship includes a first position sub-relationship and a second position sub-relationship; Then the step of determining the positional relationship between the head acupuncture points and the front hairline includes: Acquire a first positional sub-relationship between the at least one first acupuncture point and the front hairline; Obtaining the longitudinal bone inch relationship between the front hairline and the back hairline; Determining the second position sub-relationship based on the position association relationship between the at least one second acupuncture point and the back hairline and the longitudinal bone-inch relationship; The first acupuncture points include Houding, Baihui, Qianding, Xinhui, Shangxing, Shenting, Qucha, Meichong, Wuchu, Chengguang, Tongtian and Luoque; The first position sub-relationship is: The back vertex is located at the midline of the front hairline and is 5.5 straight inches away from the front hairline; The Baihui point is located on the midline of the front hairline and is 5 inches away from the front hairline; The front vertex is located at the midline of the front hairline and is 3.5 straight inches away from the front hairline; The fontanelle is located at the midline of the anterior hairline and is 2 inches away from the anterior hairline; The upper star is located on the midline of the front hairline and is 1 inch away from the front hairline; The Shenting is located at the midline of the front hairline and is 0.5 straight inches away from the front hairline; The curve is 1.5 horizontal inches away from the median line of the front hairline and 0.5 vertical inches away from the front hairline; The meichong is located at the midpoint of the line connecting the shenting and the qucha; The five points are 1.5 horizontal inches away from the midline of the front hairline and 1 vertical inch away from the front hairline; The said light-carrying part is 1.5 horizontal inches away from the midline of the said front hairline, and 2.5 vertical inches away from the said front hairline; The Tongtian point is 1.5 horizontal inches away from the midline of the front hairline and 4 vertical inches away from the front hairline; The Luoque point is 1.5 horizontal inches away from the midline of the front hairline and 5.5 vertical inches away from the front hairline; The front hairline midline passes through the Yintang acupoint, is perpendicular to the line between the left frontal hair corner point and the right frontal hair corner point, and fits the head curve of the head arc surface; The second acupuncture points include Yamen, Fengfu, Naohu, Qiangjian and Yuzhen; The position association relationship is: The dummy gate is located at the midline of the posterior hairline and is 0.5 inches away from the posterior hairline; The Fengfu is located on the midline of the posterior hairline and is 1 inch away from the posterior hairline; The Naohou is located at the midline of the posterior hairline and is 2.5 inches away from the posterior hairline; The strong space is located at the midline of the posterior hairline and is 4 inches away from the posterior hairline; The occipital bone is 1.3 horizontal inches away from the midline of the posterior hairline, and 2.5 vertical inches away from the posterior hairline; Wherein, the posterior hairline midline is the extension line of the anterior hairline midline along the curved surface of the head.

6. An acupuncture robot, characterized in that: The invention comprises an acupuncture point identification subsystem for determining acupuncture points on the head and a moxibustion subsystem for performing acupuncture based on the acupuncture points on the head, wherein the acupuncture point identification subsystem comprises a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the head acupoint identification method as described in any one of claims 1 to 4 above.

7. A digital human, characterized in that: It comprises a virtual head model and head acupoints marked in the virtual head model, wherein the head acupoints are determined based on a head acupoint recognition method, and the head acupoint recognition method is the head acupoint recognition method according to any one of claims 1 to 4 above.

Images