Upper limb intelligent rehabilitation training methods and systems

Abstract

This invention provides an intelligent rehabilitation training method and system for the upper limbs, applied to a host computer, comprising: acquiring three-dimensional joint information of the human body through a depth camera; determining whether the target object is in a sitting position based on the three-dimensional joint information; if not, extracting and storing facial information of the target object; if so, and if the target object is in the database, performing Kalman trajectory optimization and fixed-point filtering on the three-dimensional joint information of the human body to obtain an optimized trajectory; performing DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation movement, and sending it to a lower computer so that the lower computer can drive the affected side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation movement; adjusting the spatial position of the motor relative to the affected side based on the three-dimensional joint information of the human body, as well as adjusting the safety space and mechanical position.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to intelligent rehabilitation training methods and systems for the upper limbs. Background Technology

[0002] Currently, the upper limb intelligent rehabilitation training robot collects depth information to obtain human joint information through depth sensors; based on the information of each joint point, it distinguishes the information of the healthy arm that needs to be obtained, calculates the angle information of the joints based on the information of the healthy arm, and realizes the movement of the affected side by changing the angle of each joint of one arm.

[0003] However, this robot requires manual adjustment of its position and configuration based on different people's height and arm length, making the operation process cumbersome; when switching patients, information needs to be confirmed with the patient before switching can proceed; and the mirror image learning of the healthy and affected sides is inaccurate, resulting in a poor patient experience. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an intelligent rehabilitation training method and system for the upper limbs, which can intelligently switch patient information through facial search, and drive the affected side to carry out rehabilitation training through the trajectory of the healthy side of the corresponding rehabilitation movements; it can automatically adjust the spatial position of the motor to the affected side through the three-dimensional joint information of the human body, and automatically adjust the safety space and mechanical position, without the need for manual operation, thus improving the user experience.

[0005] In a first aspect, embodiments of the present invention provide an intelligent rehabilitation training method for the upper limbs, applied to a host computer, the method comprising:

[0006] Acquire 3D joint information of the human body using a depth camera;

[0007] Determine whether the target object is in a seated state based on the aforementioned three-dimensional human joint point information;

[0008] If not, then facial information is extracted and stored from the target object;

[0009] If so, and the target object is determined to be in the database, the human body's three-dimensional joint information is subjected to Kalman trajectory optimization and fixed-point filtering to obtain the optimized trajectory.

[0010] The optimized trajectory is matched with DTW to obtain the healthy side trajectory of the corresponding rehabilitation movement, and then sent to the lower computer so that the lower computer can drive the affected side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation movement.

[0011] The motor is adjusted to the spatial position relative to the affected side based on the three-dimensional joint information of the human body, as well as the safety space and mechanical position are adjusted.

[0012] Furthermore, the three-dimensional joint information of the human body is based on the depth camera as the center point, and includes the three-dimensional coordinates of the shoulder joint, the three-dimensional coordinates of the wrist joint, and the three-dimensional coordinates of the elbow joint.

[0013] Furthermore, the three-dimensional joint information of the human body is subjected to Kalman trajectory optimization and fixed-point filtering to obtain the optimized trajectory, including:

[0014] The three-dimensional coordinates of the shoulder joint and the three-dimensional coordinates of the wrist joint are transformed to obtain the transformed three-dimensional coordinates of the wrist joint with the shoulder joint as the center.

[0015] The transformed three-dimensional coordinates of the wrist joint are optimized using a Kalman trajector to obtain the motion trajectory.

[0016] The motion trajectory is filtered at a fixed point to obtain the optimized trajectory.

[0017] Furthermore, the optimized trajectory is subjected to DTW matching to obtain the unaffected side trajectory of the corresponding rehabilitation movement, including:

[0018] Calculate the angles between the vectors of adjacent frames of the optimized trajectory and the sagittal, horizontal, and coronal planes;

[0019] The vectors of the adjacent frames of the optimized trajectory are matched with the angles between the vectors and the sagittal, horizontal and coronal planes, and with the angles between the vectors of the adjacent frames of the pre-stored standard and the sagittal, horizontal and coronal planes, to obtain the healthy side trajectory of the corresponding rehabilitation action.

[0020] Furthermore, adjusting the motor to the spatial position relative to the affected side based on the three-dimensional joint information of the human body, and adjusting the safety space and mechanical position, includes:

[0021] The motor is adjusted to the spatial position relative to the affected side based on the three-dimensional coordinates of the wrist joint.

[0022] Calculate arm length and shoulder height based on the three-dimensional coordinates of the shoulder joint, the wrist joint, and the elbow joint;

[0023] The safety space is adjusted according to the arm length;

[0024] Adjust the mechanical position according to the shoulder height.

[0025] Furthermore, determining whether the target object is in a seated state based on the aforementioned three-dimensional human joint information includes:

[0026] The target area is set based on the three-dimensional joint point information of the human body;

[0027] Determine whether the target object in the target area is in the sitting state.

[0028] Furthermore, facial information extraction and storage are performed on the target object, including:

[0029] When the target object is a new user, facial information is extracted from the new user to obtain the facial feature information of the new user;

[0030] Obtain the three-dimensional joint information of the new user;

[0031] Calculate the new user's shoulder height and arm length based on the new user's three-dimensional joint point information;

[0032] The facial features, shoulder height, and arm length of the new user are stored.

[0033] Secondly, embodiments of the present invention provide an intelligent rehabilitation training system for the upper limbs, applied to a host computer, the system comprising:

[0034] The acquisition module is used to acquire 3D joint information of the human body through a depth camera.

[0035] The judgment module is used to determine whether the target object is in a sitting state based on the three-dimensional joint point information of the human body;

[0036] The extraction module is used to extract and store facial information of the target object when the target object is not in the sitting state.

[0037] The optimization module is used to perform Kalman trajectory optimization and fixed-point filtering on the human body's three-dimensional joint point information when the target object is in the sitting state, and to obtain the optimized trajectory, provided that the target object is in the database.

[0038] The matching module is used to perform DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation action, and send it to the lower computer so that the lower computer can drive the affected side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation action.

[0039] The adjustment module is used to adjust the motor to the spatial position relative to the affected side based on the three-dimensional joint information of the human body, as well as to adjust the safety space and mechanical position.

[0040] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method described above.

[0041] Fourthly, embodiments of the present invention provide a computer-readable medium having processor-executable non-volatile program code that causes the processor to perform the method described above.

[0042] This invention provides an intelligent rehabilitation training method and system for the upper limbs, applied to a host computer. The method includes: acquiring three-dimensional joint information of the human body using a depth camera; determining whether the target object is in a seated state based on the three-dimensional joint information; if not, extracting and storing facial information of the target object; if the target object is in the database, performing Kalman trajectory optimization and fixed-point filtering on the three-dimensional joint information to obtain an optimized trajectory; performing DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation movement, and sending it to a lower computer so that the lower computer can drive the affected side to perform rehabilitation training based on the healthy side trajectory of the corresponding rehabilitation movement; adjusting the motor to the relative spatial position of the affected side based on the three-dimensional joint information, as well as adjusting the safety space and mechanical position; intelligently switching patient information through face search, and driving the affected side to perform rehabilitation training based on the healthy side trajectory of the corresponding rehabilitation movement; automatically adjusting the motor to the relative spatial position of the affected side based on the three-dimensional joint information, as well as automatically adjusting the safety space and mechanical position, without manual operation, thus improving the user experience.

[0043] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0045] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0046] Figure 1 This is a flowchart of the upper limb intelligent rehabilitation training method provided in Embodiment 1 of the present invention;

[0047] Figure 2 This is a flowchart of step S104 in the upper limb intelligent rehabilitation training method provided in Embodiment 1 of the present invention;

[0048] Figure 3This is a flowchart of step S105 in the upper limb intelligent rehabilitation training method provided in Embodiment 1 of the present invention;

[0049] Figure 4 This is a schematic diagram of a QC intertwined table provided in Embodiment 1 of the present invention;

[0050] Figure 5 This is a schematic diagram illustrating the acquisition of DTW values ​​provided in Embodiment 1 of the present invention;

[0051] Figure 6 This is a schematic diagram of the upper limb intelligent rehabilitation training system provided in Embodiment 2 of the present invention;

[0052] Figure 7 This is a schematic diagram of the upper limb intelligent rehabilitation training system provided in Embodiment 2 of the present invention.

[0053] icon:

[0054] 1-Base; 2-Display screen; 3-Depth camera; 4-Base; 5-Robotic arm; 11-Acquisition module; 12-Judgment module; 13-Extraction module; 14-Optimization module; 15-Matching module; 16-Adjustment module. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] To facilitate understanding of this embodiment, the embodiments of the present invention will be described in detail below.

[0057] Example 1:

[0058] Figure 1 The flowchart is for the upper limb intelligent rehabilitation training method provided in Embodiment 1 of the present invention.

[0059] Reference Figure 1 Applied to a host computer, this method includes the following steps:

[0060] Step S101: Obtain three-dimensional joint information of the human body using a depth camera;

[0061] Step S102: Determine whether the target object is in a sitting position based on the three-dimensional joint information of the human body; if not, proceed to step S103; if yes, proceed to step S104.

[0062] Step S103: Extract and store facial information of the target object;

[0063] Step S104: If the target object is in the database, perform Kalman trajectory optimization and fixed-point filtering on the human body's three-dimensional joint information to obtain the optimized trajectory.

[0064] Specifically, if the target is not seated, it is considered a new user. The system collects the new user's facial information and extracts its facial features. It also calculates the new user's shoulder height and arm length, storing these information. If the target is seated, the system uses a face search function to check if the target is present in the database. If found, the system displays the target's information or training plan.

[0065] Step S105: Perform DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation movement, and send it to the lower computer so that the lower computer can drive the affected side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation movement.

[0066] Here, after the trajectory of the healthy side of the corresponding rehabilitation movement is sent to the lower computer, data is filled in for the points that are not in the movement to make the trajectory of the healthy side more complete, so that the movement of the affected side does not have to wait until the movement is completed.

[0067] Step S106: Adjust the motor to the spatial position relative to the affected side based on the three-dimensional joint information of the human body, and adjust the safety space and mechanical position.

[0068] Furthermore, the 3D human body joint information is centered on the depth camera and includes, but is not limited to, the 3D coordinates of the shoulder joint, wrist joint, and elbow joint, as well as the 3D coordinates of the head and neck. The depth camera can be a nuicrack+ depth camera, acquiring 20 3D human body joint points.

[0069] Furthermore, refer to Figure 2 Step S104 includes the following steps:

[0070] Step S201: Perform coordinate transformation on the three-dimensional coordinates of the shoulder joint and the wrist joint to obtain the transformed three-dimensional coordinates of the wrist joint with the shoulder joint as the center.

[0071] Here, coordinate transformation is performed on the three-dimensional coordinates of the shoulder joint and the wrist joint, so that the three-dimensional coordinates of the wrist joint are switched to the same coordinate system as the mechanical mirror, that is, the transformed three-dimensional coordinates of the wrist joint with the shoulder joint as the center.

[0072] Step S202: Perform Kalman trajectory optimization on the converted wrist joint three-dimensional coordinates to obtain the motion trajectory;

[0073] Here, the transformed wrist joint 3D coordinates are optimized using Kalman trajectories, which can make the movement trajectory smoother and the movement on the affected side more comfortable.

[0074] Step S203: Perform fixed-point filtering on the motion trajectory to obtain the optimized trajectory.

[0075] Specifically, since the motion trajectory obtained after Kalman trajectory optimization when the healthy side is stationary (without rehabilitation exercises) exhibits jitter, an anti-jitter function is added to filter the jitter information within the motion trajectory range, thereby improving the user experience.

[0076] Furthermore, refer to Figure 3 Step S105 includes the following steps:

[0077] Step S301: Calculate the angle Q between the vectors of adjacent frames of the optimized trajectory and the sagittal plane, horizontal plane, and coronal plane;

[0078] Specifically, refer to formulas (1) and (2):

[0079] Q = q1, q2, ... q i …q n (1)

[0080] Where q1 is the angle between the vector formed by the wrist joint in two adjacent frames and the sagittal, horizontal, and coronal planes [α1, α2, α3], and q2 is the angle between the vector formed by the wrist joint in two adjacent frames and the sagittal, horizontal, and coronal planes [α4, α5, α6], q i The angle between the vector formed by the wrist joint in two adjacent frames and the sagittal, horizontal, and coronal planes [α7, α8, α9], q n The angle between the vector formed by the wrist joint in two adjacent frames and the sagittal, horizontal, and coronal planes [α] 10 α 11 α 12 ].

[0081] C = c1, c2, ... c j …c n (2)

[0082] Where c1 is the angle formed by the existing standard adjacent frame with the sagittal plane, horizontal plane, and coronal plane [β1, β2, β3], c2 is the angle formed by the existing standard adjacent frame with the sagittal plane, horizontal plane, and coronal plane [β4, β5, β6], c j The angles [β7, β8, β9] formed by adjacent frames of the existing standard with the sagittal, horizontal, and coronal planes, respectively, are given by c. n The angle [β] formed by adjacent frames of an existing standard with the sagittal, horizontal, and coronal planes. 10,β 11 ,β 12 ].

[0083] Step S302: Perform DTW matching between the vectors of adjacent frames of the optimized trajectory and the angle Q between the vectors of adjacent frames and the sagittal, horizontal and coronal planes, and the angle C between the vectors of adjacent frames of the pre-stored standard and the sagittal, horizontal and coronal planes, to obtain the healthy side trajectory of the corresponding rehabilitation movement.

[0084] By using DTW (Dynamic Time Warping) to match the angle C corresponding to different movements and obtain the healthy side trajectory of the corresponding rehabilitation movement for learning, it is possible to achieve safe learning on the affected side while using the healthy side trajectory of the better rehabilitation movement for rehabilitation training.

[0085] Refer to formulas (3), (4), (5), and (6):

[0086] wk=(i,j)k (3)

[0087] Among them, reference Figure 4 , i, j correspond to the data with indices 1-n in QC respectively, and wk is a table interwoven with QC. The leftmost column of elements is: W(cj,q1)=dis(cj,q1)+W(c(j-1),q1); the bottom row of elements is: W(c1,qi)=dis(c1,qi)+W(c1,q(i-1)); the other grids in the middle are filled with formula (6). W(i,j) is the element in column i and row j of the grid, and dis(cj,qi) is the Euclidean distance from Q to C.

[0088] W = w1, w2, ... w k (4)

[0089] Where W represents all values ​​in the grid in wk, totaling the length of Q * the length of C.

[0090]

[0091] Among them, reference Figure 5 Starting from the first element in the upper right corner, work to the left and add it to the smallest element in the lower left corner, and so on to get the final value.

[0092] W(i,j)=dis(q) i c j )+min{W(i-1,j-1),W(i-1,j),W(i,j-1)} (6)

[0093] The anti-shake function L(xyz) is specifically implemented by real-time monitoring of adjacent consecutive frames, and eliminating shaking by applying a threshold based on the coordinate changes of the wrist joint points in adjacent frames, as detailed in formulas (7) and (8):

[0094] A=[(x1,y1,z1),(x2,y2,z2),...(x n y n , z n (7)

[0095]

[0096] Taking the collection of five consecutive frames of data as an example, if the maximum difference is greater than a threshold, it is considered a motion state and no processing is performed, i.e., {now[x, y, z](max)} A -min A )>T};

[0097] If the difference is less than the threshold, and the difference between the current value and one of the five consecutive frames is less than the threshold, then the starting point is a single value, i.e., {averageA averageA-now[x,y,z]}. <T,(max A -min A ) <T};

[0098] If the difference is less than the threshold, and the difference between the current value and one of the five consecutive frames is greater than the threshold, then no processing is performed, i.e., {now[x,y,z]averageA-now[x,y,z]>T, (max A -min A ) <T}。

[0099] Furthermore, step S106 includes the following steps:

[0100] Step S401: Adjust the motor to the spatial position relative to the affected side based on the three-dimensional coordinates of the wrist joint;

[0101] Here, the motor is adjusted to the relative spatial position of the affected side by adjusting the three-dimensional coordinates of the healthy wrist joint, thereby enabling the affected side to undergo traction learning.

[0102] Step S402: Calculate arm length and shoulder height based on the three-dimensional coordinates of the shoulder joint, wrist joint, and elbow joint.

[0103] Step S403: Adjust the safety space according to the arm length;

[0104] Step S404: Adjust the mechanical position according to shoulder height.

[0105] Specifically, the device is powered on, the motor is enabled, and relevant parameters are configured. Then, the main controller receives the healthy side trajectory of the corresponding rehabilitation movement sent by the host computer. By using the arm length data (length in the straightened state), a physical safety space limit value can be determined, that is, the maximum safety space. The above determination can prevent the patient from being stretched beyond the space range, causing secondary injury.

[0106] The lifting column is adjusted according to shoulder height to keep the mechanical and transmitted data within a mirrored coordinate system. The lower-level computer obtains the three-dimensional coordinates of the healthy wrist joint from the visual sensor via Bluetooth transmission, and then adjusts the motor to the corresponding spatial position on the affected side based on these coordinates.

[0107] Furthermore, step S102 includes the following steps:

[0108] Step S501: Set the target area based on the three-dimensional joint information of the human body;

[0109] Step S502: Determine whether the target object in the target area is in a sitting state.

[0110] Specifically, the target area is set based on the three-dimensional joint information of the human body to avoid interference from outsiders in the accuracy of detection. The target object in the target area is repeatedly judged to see if it is in a sitting state (ready state). If it is, facial features are extracted and the database is searched to see if the target object (patient) exists.

[0111] Furthermore, step S103 includes the following steps:

[0112] Step S601: When the target object is a new user, extract facial information from the new user to obtain the facial feature information of the new user;

[0113] Step S602: Obtain the 3D joint information of the new user;

[0114] Step S603: Calculate the new user's shoulder height and arm length based on the new user's three-dimensional joint point information;

[0115] Step S604: Store the new user's facial features, shoulder height, and arm length.

[0116] Specifically, if the target is not seated, it is considered a new user. The system collects and extracts the new user's facial information to obtain facial feature information. It also calculates the new user's shoulder height and arm length, storing these data. Finally, after locating the new user, the system processes the shoulder and wrist joint 3D coordinates in real-time under a patient-health linkage mode. The existing database consists of standard rehabilitation training movements and the angle sequences between the sagittal, horizontal, and coronal planes; each movement contains three angle sequences.

[0117] This application implements an intelligent patient switching function in an upper limb rehabilitation training device, which can accurately determine the patient's identity information and previous training plans without requiring extensive operations. It obtains optimized trajectory calculation angle information by comparing the angle information formed by the trajectories of pre-stored standard rehabilitation movements, and transmits the corresponding healthy-side trajectory of the rehabilitation movement to the machine. Using end-effector traction and three-dimensional joint information of the human body, and based on the real-time mirroring of the healthy-side movement, it realizes the rehabilitation movement on the affected side.

[0118] Example 2:

[0119] Figure 6 This is a schematic diagram of the upper limb intelligent rehabilitation training system provided in Embodiment 2 of the present invention.

[0120] Reference Figure 6 Applied to a host computer, the system includes:

[0121] Acquisition module 11 is used to acquire three-dimensional joint information of the human body through a depth camera;

[0122] The judgment module 12 is used to determine whether the target object is in a sitting state based on the three-dimensional joint information of the human body;

[0123] Extraction module 13 is used to extract and store facial information of the target object when the target object is not in a sitting state;

[0124] Optimization module 14 is used to optimize the human body's three-dimensional joint information using Kalman trajectory optimization and fixed-point filtering when the target object is in a sitting state and the target object is in the database, so as to obtain the optimized trajectory.

[0125] The matching module 15 is used to perform DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation movement, and send it to the lower computer so that the lower computer can drive the affected side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation movement.

[0126] The adjustment module 16 is used to adjust the spatial position of the motor relative to the affected side based on the three-dimensional joint information of the human body, as well as to adjust the safety space and mechanical position.

[0127] Figure 7 This is a schematic diagram of the upper limb intelligent rehabilitation training system provided in Embodiment 2 of the present invention.

[0128] Reference Figure 7 The system includes a host computer and a slave computer. The host computer includes a base 1, a display screen 2 and a depth camera 3, and the slave computer includes a base 4 and a robotic arm 5.

[0129] When a patient needs rehabilitation training, the patient places the affected side on the robotic arm 5. At this time, the depth camera 3 collects the patient's three-dimensional joint point information. Then, the host computer determines whether the patient is in a sitting position based on the three-dimensional joint point information. If so, the host computer performs Kalman trajectory optimization and fixed-point filtering on the three-dimensional joint point information to obtain the optimized trajectory. The optimized trajectory is then matched with DTW to obtain the healthy side trajectory for the corresponding rehabilitation movement, and sent to the lower computer. The lower computer drives the affected side (the patient's right arm) to perform rehabilitation training based on the healthy side trajectory of the corresponding rehabilitation movement. The host computer can also adjust the motor to the relative spatial position of the affected side based on the three-dimensional joint point information, thereby performing traction learning on the affected side.

[0130] This invention provides an intelligent rehabilitation training method and system for the upper limbs, applied to a host computer. The method includes: acquiring three-dimensional joint information of the human body using a depth camera; determining whether the target object is in a seated state based on the three-dimensional joint information; if not, extracting and storing facial information of the target object; if the target object is in the database, performing Kalman trajectory optimization and fixed-point filtering on the three-dimensional joint information to obtain an optimized trajectory; performing DTW matching on the optimized trajectory to obtain the healthy side trajectory of the corresponding rehabilitation movement, and sending it to a lower computer so that the lower computer can drive the affected side to perform rehabilitation training based on the healthy side trajectory of the corresponding rehabilitation movement; adjusting the motor to the relative spatial position of the affected side based on the three-dimensional joint information, as well as adjusting the safety space and mechanical position; intelligently switching patient information through face search, and driving the affected side to perform rehabilitation training based on the healthy side trajectory of the corresponding rehabilitation movement; automatically adjusting the motor to the relative spatial position of the affected side based on the three-dimensional joint information, as well as automatically adjusting the safety space and mechanical position, without manual operation, thus improving the user experience.

[0131] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the upper limb intelligent rehabilitation training method provided in the above embodiments.

[0132] This invention also provides a computer-readable medium having processor-executable non-volatile program code, on which a computer program is stored, and which, when run by a processor, executes the steps of the upper limb intelligent rehabilitation training method described above.

[0133] The computer program product provided in this embodiment of the invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0134] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0135] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0136] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0137] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0138] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An upper limb intelligent rehabilitation training system, characterized in that, The system is applied to an upper computer, and the system comprises: An acquisition module is configured to acquire human three-dimensional joint information through a depth camera; A judgment module is configured to judge whether a target object is in a sitting state according to the human three-dimensional joint information; An extraction module is configured to perform face information extraction and information storage on the target object when the target object is not in the sitting state; An optimization module is configured to, when the target object is in the sitting state, determine the target object in a database, perform Kalman trajectory optimization and fixed-point filtering on the human three-dimensional joint information, and obtain an optimized trajectory; A matching module is configured to perform DTW matching on the optimized trajectory, obtain a healthy side trajectory of a corresponding rehabilitation action, and send the healthy side trajectory to a lower computer, so that the lower computer drives a sick side to perform rehabilitation training according to the healthy side trajectory of the corresponding rehabilitation action; An adjustment module is configured to adjust a spatial position of a motor relative to the sick side according to the human three-dimensional joint information, and adjust a safety space and a mechanical position; The human three-dimensional joint information is taken as a circle point of the depth camera, and the human three-dimensional joint information comprises shoulder joint three-dimensional coordinates, wrist joint three-dimensional coordinates, and elbow joint three-dimensional coordinates; The optimization module is specifically configured to: perform coordinate conversion on the shoulder joint three-dimensional coordinates and the wrist joint three-dimensional coordinates to obtain converted wrist joint three-dimensional coordinates taken as a circle point of the shoulder joint; perform Kalman trajectory optimization on the converted wrist joint three-dimensional coordinates to obtain a motion trajectory; and perform fixed-point filtering on the motion trajectory to obtain the optimized trajectory.

2. The upper limb intelligent rehabilitation training system according to claim 1, characterized in that, The matching module is specifically configured to: calculate an angle between a vector of adjacent frames of the optimized trajectory and a sagittal plane, a horizontal plane, and a coronal plane; perform DTW matching on the angle between the vector of the adjacent frames of the optimized trajectory and the sagittal plane, the horizontal plane, and the coronal plane, and an angle between a vector of adjacent frames of a pre-stored standard and the sagittal plane, the horizontal plane, and the coronal plane, to obtain the healthy side trajectory of the corresponding rehabilitation action.

3. The upper limb intelligent rehabilitation training system according to claim 1, characterized in that, The adjustment module is specifically configured to: adjust the spatial position of the motor relative to the sick side according to the wrist joint three-dimensional coordinates; calculate an arm length and a shoulder height according to the shoulder joint three-dimensional coordinates, the wrist joint three-dimensional coordinates, and the elbow joint three-dimensional coordinates; adjust the safety space according to the arm length; and adjust the mechanical position according to the shoulder height.

4. The upper limb intelligent rehabilitation training system according to claim 1, characterized in that, The judgment module is specifically configured to: set a target area according to the human three-dimensional joint information; and judge whether the target object in the target area is in the sitting state.

5. The upper limb intelligent rehabilitation training system according to claim 1, characterized in that, The extraction module is specifically configured to: when the target object is a new user, perform face information extraction on the new user to obtain face feature information of the new user; acquire three-dimensional joint information of the new user; calculate a shoulder height of the new user and an arm length of the new user according to the three-dimensional joint information of the new user; and store the face feature information of the new user, the shoulder height of the new user, and the arm length of the new user.

Images