123 results about "Training - action" patented technology

Context-aware training systems, apparatuses and systems. The context-aware training systems, apparatuses and systems are computer-implemented and include sensing a user action and, based on a training needs model, estimating a cost or benefit to exposing the user to a training action, selecting a training action from a collection of available training actions and delivering the training action to the user if the user action indicates a need for the user to be trained and the cost or benefit to exposing the user to the training action indicates user exposure to the training action is warranted.

The invention discloses a gesture recognition method. The method comprises the steps that common human body basic actions are captured and are stored as sample actions; according to the sample actions, final training actions are obtained; graphics rendering is conducted on the training actions and a preliminary depth map and a corresponding position identification graph are generated; according to the generated depth map, samples similar to the depth map collected in real time are synthesized; the synthesized samples are used for calculating corresponding depth characteristic vectors, and a random forest model is obtained through training; accurate depth figure outline is extracted through area growth based on smoothness constraining; depth characteristic vectors of each pixel of the depth figure outline are calculated based on the random forest model, and position identification probability of each pixel is determined through the random forest model; noisy points are filtered and identified based on the human body part corresponding to each pixel and the probability of each pixel, and skeleton nodes are generated in a polymerized mode; a time-sequential sequence of the skeleton nodes is recorded and a skeleton motion track is generated; the motion tracks of the nodes of the hands are extracted and are matched with a predefined template, and the gesture action type is recognized. The invention further discloses a gesture recognition device.

The invention discloses a rehabilitation training hand and a rehabilitation training method. The rehabilitation training hand comprises a drive portion and a mechanical portion. The mechanical portion comprises a palm portion body, a finger regulation portion body and a man-machine integration portion body. According to the palm portion body, a palm fixing plate is connected with a palm face of a patient hand through a palm binding belt and connected with a carpal articular face of the patient hand through an arm binding belt. A finger sliding rail and a thumb base are arranged on the palm fixing plate. The finger regulation portion body comprises support rods rotating around coxa pins, knuckle sliders sliding along sliding grooves of the support rods. The coxa pin of the thumb is connected to a thumb base, and coxa pins of the other four fingers are arranged on coxa sliders which are arranged on the finger sliding rail and slide along the finger sliding rail. The man-machine integration portion body comprises a thumb sleeve connected with a remote thumb knuckle of the patient hand and finger sleeves connected with remote knuckles of the other four fingers of the patient. The rehabilitation training hand is small in structure, light and convenient, suitable for being worn on the patient hand and capable of completing hand expanding and finger and palm training actions.

The invention relates to the technical field of mode recognizing, in particular to a real-time human body action recognizing method and device based on depth image sequence. The method comprises the steps of S1, extracting target action sketch from a target depth image sequence and extracting a training action sketch from a training depth image set; S2, performing gesture clustering on the training action sketch and performing action calibrating on the clustered outcome; S3, computing the gesture characteristics of the target action sketch and training action sketch; S4, performing the gesture training based on a Gauss mixing model by combining the gesture characteristics of the training action sketch and constructing a gesture model; S5, computing the transferring probability among all gestures of the clustered outcome in each action and constructing an action image model; and S6, performing action recognizing on the target depth image sequence according to the gesture characteristics of the target action sketch, the gesture model and the action image model. The real-time human body action recognizing method disclosed by the invention has the advantages of improving the efficiency of action recognizing and the accuracy and the robustness of the action recognizing.

An upper limb rehabilitation robot comprises a support, a plurality of traction devices, a supporting plate, a plurality of ropes and a control device. The support comprises an upright column, a top frame and a cross beam, wherein the top frame is located above the forearm of a patient, and the cross beam is located under the forearm of the patient. The supporting plate is used for the patient to place the forearm. One end of each rope is connected with one of the traction devices, and the other end of each rope is fixed on the supporting plate. The control device is used for controlling the angle and the strength of the traction devices during traction of the ropes. According to the upper limb rehabilitation robot, the control device is used for controlling the angle and the strength of the traction devices during the traction of the ropes so that the patient can be assisted in complicated training actions in a three-dimensional space for shoulder joints and elbow joints of the patient. Therefore, a better rehabilitation effect can be achieved for the patient.

The invention discloses a video description generation method based on deep learning and a probabilistic graphical model. The method systematically includes the steps that an existing image data set training rapid region object recognition convolution neural network model is utilized; an existing video data set training action recognition convolution neural network model is utilized; frame-extracting processing is carried out on a video, the rapid region object recognition convolution neural network model and the action recognition convolution neural network model are utilized for recognizing objects and action in the video, and the main content of the video is basically determined; by the utilization of a conditional random field, a subject-verb-object triad <objects, action and objects> with the maximum probability is found, the noise objects in the video are excluded, and a final description result is more accurate; the subject-verb-object triad is input into a long and short period memory network to input appropriate sentences which are namely description of the input video. The video is converted into the description, in this way, people can understand the content of the video more rapidly, and the retrieval speed of the video is increased ether.

The invention relates to an upper limb rehabilitation robot control system. An upper computer transmits instructions to an MCU (micro-programmed control unit) module through a Bluetooth / WIFI, the MCUmodule selects different training modes according to the instructions, a voice module identifies unspecific voice, voice signals are acquired and processed and transmitted to the MCU module by the aidof an SPI (serial peripheral interface) communication protocol, a driving module drives a motor to rotate in a controlled manner, so that training actions are completed, electromyography collection of a healthy side of a patient is implemented by an electromyography control module, electromyography signals are extracted, processed and analyzed, actions of the healthy side of the patient map a mechanical arm, movement intentions are identified, and functional compensation and control is achieved. By the aid of various man-machine interaction modes, targeted rehabilitation training effects canbe effectively achieved, joints of the upper limb of the patient can be accurately positioned when the system is used, and tissue damage and pulled muscle caused by misplaced training are avoided. Thecontrol system is small in size, convenient to use, good in training effect and low in manufacturing cost.

Context-aware training systems, apparatuses and systems. The context-aware training systems, apparatuses and systems are computer-implemented and include sensing a user action and, based on a training needs model, estimating a cost or benefit to exposing the user to a training action, selecting a training action from a collection of available training actions and delivering the training action to the user if the user action indicates a need for the user to be trained and the cost or benefit to exposing the user to the training action indicates user exposure to the training action is warranted.

The invention discloses an upper limb rehabilitation training device based on grouping coupling drive. The upper limb rehabilitation training device based on the grouping coupling drive comprises a shoulder three-degree-of-freedom self-adapting mechanism, a first coupling movement mechanism, a shoulder connecting component, a second coupling movement mechanism and a wrist two-degree-of-freedom self-adapting mechanism, the first coupling movement mechanism is firmly connected with the shoulder three-degree-of-freedom self-adapting mechanism, the shoulder connecting component is rotationally connected with the first coupling movement mechanism, the second coupling movement mechanism is firmly connected with the shoulder connecting component, and the wrist two-degree-of-freedom self-adapting mechanism is firmly connected with the second coupling movement mechanism. The upper limb rehabilitation training device based on the grouping coupling drive is capable of regulating automatically according to the affected limb sizes of different patients and the concrete use situations, the human body can coordinate with the mechanism, the human body comfort and safety coefficient are greatly improved, and the movements transit smoothly and coordinate with each other so that the upper limb rehabilitation training device based on the grouping coupling drive is capable of finishing various complex active and passive training actions.

The invention relates to a breath training system, which is used for aiding a patient with chronic obstructive pulmonary disease in rehabilitation training and comprises a server device and a client device. The client device comprises a client communication part, a client storage part, a client display part, a client input part and a client processing part, wherein the client processing part comprises first to fourth stage guidance and indication modules which guide breathing exercises of first to fourth training actions respectively. The first training action adopts a lip-contracting ventral breathing way; the second training action adopts a breathing way of making a fist, deep breathing and slowly expiring air; the third training action adopts a whole body breathing exercising way; and the fourth training action adopts a deep inspiration-holding breathing-slowly air expiring way.

The invention provides an autonomous motion identification technology based private coach smart band, and particularly provides an integrated private coach smart band fusing three functions of standard library self-definition, motion identification and standard action self-correction. The smart band has three modes which are an action definition mode, a motion identification mode and a training mode respectively. In the action definition mode, a user can self-define standard actions and construct a standard library, equivalently, the user has a private coach; in the motion identification mode, the user can utilize the smart band to obtain a motion result for analysis to perform self-learning and realize self-improvement; and in the training mode, a training action is selected from a terminal device and motion is performed. The smart band is capable of predicting more standard actions by utilizing new motion data and data in the standard library according to a prediction algorithm so as to self-correct the standard actions and update the standard library, and the prediction is more accurate along with continuous increment of training actions.

The invention provides an intelligent vision function training system based on eyeball tracking and a training method thereof and particularly relates to a vision improvement training system and method based on eyeball tracking. The vision improvement training system and method based on eyeball tracking are intelligent vision function training system and training method effectively identifying eyeactions of a trainee through an eyeball tracking technology and scientifically guiding the trainee to perform vision training actions in time and at high precision. The intelligent vision function training system comprises an eyeball tracking device, a display and sound device, an input device and a control device. Compared with other training systems feeding back eyeball tracking effects by relying on subjective narration of trainees and other modes, the intelligent vision function training system has the advantages of being rapid, timely, accurate and objective and plays an important role on appropriate setting of follow-up vision training. The intelligent vision function training system makes training content rich and varied, exercises the vision of trainees and can also make the trainees obtain the effects of intelligence development, entertainment and rest.

The invention discloses a control system for an intelligent health chair. The control system for the intelligent health chair comprises a main control module, an upper computer, an intelligent perception module, a drive module, a mode module and a power source management module, wherein the upper computer is composed of a man-machine interaction module and a wireless intelligent terminal, and the main control module is composed of a micro programming system circuit, a wireless communication module and a connector circuit. A built-in communication module is arranged in the wireless intelligent terminal to conduct wireless communication with the wireless communication module of the main control module. The main control module is a lower machine and receives feed-back information of the intelligent perception module, the drive module and the mode module. The main control module controls the drive module and the mode module to conduct health training actions. Each module in the intelligent health chair is respectively supplied with power through the power source management module. By means of the control system for the intelligent heath chair, the defects of the prior art are overcome, and the system can be controlled in an integration and wireless mode and operated in a humanization mode. Moreover, the control system for the intelligent health chair has the advantages of being capable of evaluating the health effect and facilitating scientific planning of the health training.

The invention provides a writing trainer based on force feedback. The writing trainer comprises a hand writing board for collecting user writing data, a display for displaying a writing track, a writing pen for training a user and force feedback equipment for controlling a writing pen, wherein force feedback equipment comprises a force feedback processor; the force feedback processor stores a standard mode of a writing training action and controls the force feedback equipment through a program; the writing action is controlled by a force arm of the force feedback equipment so as to train the writing of the user; and the track of the writing is displayed by the display. By utilizing the writing trainer, the user can learn in an inactive mode, i.e. the user does not write actively, but just holds the force feedback equipment, the force feedback equipment writes automatically according to the program, then the use follows up to act so as to realize the purpose of learning writing by hand guiding. By utilizing the equipment, the writing track and the information of a teacher can be recoded and the writing skill is taught to students. The writing speed is divided into two types, namely, constant speed and variable speed; by using the two modes in a combinatory analysis mode and with the combination of the force feedback equipment, a method which is beneficial to the writing learning is found.

The invention provides a pelvic floor muscle training system, wherein the pelvic floor muscle is configured to provide at least one optional training setting to users, so that the pelvic floor muscle can guide the users to complete at least one training actions, wherein the training setting is selected from a training setting group, wherein the training setting group comprises training action setting, training time setting, training intensity setting, training frequency setting, training guide setting, training assisting setting and training feedback setting, wherein the pelvic floor muscle system comprises a processor and a client side, wherein the processor is configured to provide at least one optional training setting to users, wherein the training setting can be displayed on the client side.

The invention discloses a method for computing hand rehabilitation indexes based on a sensing technology. The method comprises the steps that a space sensor is arranged in a pair of intelligent gloves and corresponds to a target joint in position, wherein the target joint comprises a thumb interphalangeal joint, a thumb metacarpophalangeal joint, four-finger proximal interphalangeal joints, a metacarpophalangeal joint and a wrist joint. A patient wears the intelligent gloves to complete training actions, the training actions include a hand exercise stretching action, a hand exercise fist making action and a patient fist making action. The space sensor acquires kinematics parameters of the target joint in each training action, and the kinematics parameters include sagittal axis motion angle variation and strength variation. The rehabilitation indexes of the patient are determined according to kinematics parameters. By the adoption of the method, the rehabilitation indexes of the patient can rapidly and accurately calculated, accordingly the rehabilitation situation of the patient is converted into digital quantity, a doctor is effectively guided to assess the rehabilitation situation of the patient, and a rehabilitation plane is further perfected to select hand actions having clinic treatment values.

The invention provides a rehabilitation exercise training method. The method comprises the following steps: acquiring pathological information of a user and selecting corresponding training items based on the pathological information; determining a training action selected by the user in the training items, and acquiring electromyography parameters and gait information of the user when the training action is executed; determining whether the training action is suitable for the user based on the electromyography parameters and the gait information; adding the training action to a training action set to generate a training prescription if the training action is suitable for the user. The invention further provides a rehabilitation exercise training system and a readable storage medium. Whether the training action is suitable for the user can be determined according to the electromyography parameters and the gait information of the user, so that the training action suitable for the user can be selected, adverse effects of the user on the body during training are avoided, and the rehabilitation speed is increased.

The invention discloses an upper limb rehabilitation training machine, which comprises a hand rest for supporting the upper limbs and a movable plate, the hand rest is arranged on the movable plate, and the hand rest can slide on the movable plate. When the upper limb rehabilitation training machine disclosed by the present invention is used, the upper limb is driven to move on the movable plate by the hand rest, and the user follows the upper limb to rotate accordingly during the operation. The trajectory of the movement can be realized according to the preset trajectory, or can be carried out actively by the user. There are various exercise modes and training actions, which can improve the user's interest in using the equipment and solve the problem of inflexibility and single exercise mode of the upper limb rehabilitation training machine. Obviously, the upper limb rehabilitation training machine disclosed in the present invention solves the problem of single motion mode of the upper limb rehabilitation aids by providing a hand rest on the movable board, and the hand rest slides on the movable board.

The invention discloses a multifunctional space body-building device which is characterized by comprising a stress metal frame, various spring exercisers and fixed foot rings. Under a space weightlessness state, the principle of force and reacting force is utilized for strength trainings, and the stress metal frame is of an assembling type and assembled by using a supporting bar, a stress bent bar, a stress cross bar and a fixed base, wherein all the metal bars are connected, and tenons are embedded in pipe cavities to fix the metal bars by combining screws; and the fixed foot rings are matched to bear acting force during strength trainings. An upper limb spring exerciser, a waist and belly spring exerciser and a lower limb spring exerciser are also designed with different lengths and different contractility by further utilizing the expansion force of a high-strength elastic band and various training postures designed according to the ergonomic principle, and three body building modes are designed by combining the influence of weightlessness on human body physiological functions in the aerospace medicine and utilizing sports medicine theories and methods according to the ergonomic principle, thereby successfully realizing various training actions under gravity environment.

The embodiment of the invention provides a human skeleton rehabilitation training method and device, electronic equipment and a storage medium, and relates to the field of rehabilitation training. According to the method, the video data are collected through the electronic equipment, and the coordinates of the human skeleton points are obtained according to the video data, so that the real-time action angle in the user training process is calculated, whether the user completes the training action or not is determined, the actual training condition of the user is effectively monitored in the user training process. And the user can effectively carry out skeleton rehabilitation training, so that the training effect is improved.

A simulative magic ball is a simulative batting device for training actions and skills of a plurality of competitive balls with simulative batting method for the first time. The simulative magic ball belongs to the field of family physical exercise or game toys. A plurality of ball exercises can be enjoyed in an ordinary home. The simulative batting method is easily mastered and the skills of the competitive balls are quickly improved. The simulative magic ball can strengthen the body, train strong basic skills, and provide great joys. Taking simulative table tennis as example, specific skills such as push and block, cut, lift, drive, sidespin, loop drive, and the like, can be practiced. The simulative magic ball can be used for training badminton, foot shuttlecock skills, etc. The device can be extended to play skill games, the batting difficulty and entertainment fun can be increased by a compound pendulum and a conical pendulum. In the simulative batting and games, physical knowledge is combined. Teenagers can deeply understand and research into the corresponding physical phenomena, get relieved from muscle and bone, prevent from short sight, and release pressure. The simulative magic ball has the advantages of physical training, physics learning and eyes exercise, and is applied to family sports and student exercise.

A training action identification method comprises the steps of: measuring motion parameters of motion, wherein the motion parameters comprise an accelerated speed, an angular speed and a rotation attitude angle; according to a preset constraints model, performing denoising processing for the motion parameters to obtain optimal motion parameters; and employing a classification algorithm to performclassification for the optimal motion parameters to identify the action. According to the limb length and the limb connection relation, the constraints model is established, and the constraints modelis employed to perform denoising for the actually measured motion parameters to obtain the optimal motion parameters so as to allow the motion identification robustness to be higher.

The invention discloses an intelligent ball training action recognition system and method. The intelligent ball training action recognition system comprises an intelligent ball which is equipped with a micro control unit, a nine-shaft inertial motion sensing apparatus, a clock apparatus, a storage apparatus, a battery apparatus and Bluetooth equipment, and an intelligent ball training action recognition APP, wherein the intelligent ball training action recognition APP module is arranged in mobile equipment and matched with the intelligent ball in use; the intelligent ball training action recognition APP module specifically comprises a single action recognition module, a sequence action recognition module and a technology recognition module; the intelligent ball training action recognition system also comprises a cloud end server; and the cloud end server is connected with the mobile equipment through the internet and used for receiving combined actions and various grading results transmitted by the technology recognition module to be downloaded by other mobile terminals. According to the technical scheme, by combination of the intelligent ball for collecting motion related data and the mobile equipment, recognition and evaluation on training actions of ball fans can be realized.

The invention discloses a rehabilitation seat for comprehensively autonomous training action capacity of upper limbs. The rehabilitation seat comprises a seat main body, a bracket and two rehabilitation trainers which are symmetrically arranged on the two sides of the seat main body, wherein one end of the bracket is arranged on the seat main body while the other end is connected with the rehabilitation trainers; each of the rehabilitation trainers comprises an arm training mechanism, a palm training mechanism, a shoulder joint training mechanism, an autonomous force measuring mechanism and a control system; the bracket is connected with the arm training mechanism through a universal connecting mechanism; the shoulder joint training mechanism is connected with the bottom of the arm training mechanism; the palm training mechanism is arranged at the tail end of the arm training mechanism. The rehabilitation seat provided by the invention can be used for training the upper limbs and can confirm the degree of assisting the patient according to the degree of initiative participation of the patient so as to fully utilize the initiative of the patient and increase the function recover speed of the patient. The rehabilitation seat has the advantages of large action scope, more action modes, and the like.

The invention discloses a rehabilitation exercise training system and method. The training system comprises a first posture sensor, a second posture sensor, a third posture sensor and a controller, the first posture sensor, the second posture sensor and the third posture sensor acquire a first quaternion, a second quaternion and a third quaternion of human body postures and transmit the first quaternion, the second quaternion and the third quaternion to the controller, the first quaternion, the second quaternion and the third quaternion are transformed into Euler angle information of differentrotation orders under an inertial navigation coordinate system by the controller, and current body postures of human bodies are judged according to the Euler angle information. According to the system, rehabilitation exercises of patients are measured and quantized, and the system evaluates whether training actions are standard or not according to quantitative values and has special significancefor action quality, action errors initial standing habits of the patients. Participation of professional doctors (rehabilitation therapists) is omitted when the patients are trained, the system is notrestrained by time and places, and remote rehabilitation training and guide are achieved.

The invention discloses a body-building exercise flexibility training auxiliary device which comprises two first fixed blocks, wherein the two first fixed blocks are connected through a horizontally arranged connection rod; a vertically arranged supporting sleeve is fixedly connected to the center of the upper side of the connection rod; a supporting rod is inserted into the supporting sleeve; a locking screw is arranged on one side of the supporting sleeve; a locking opening corresponding to the locking screw is formed in the supporting sleeve; multiple locking slots corresponding to the locking screw are formed in the supporting rod; the upper end of the supporting rod is fixedly connected with a handle parallel to the connection rod; one side of the connection rod is fixedly connected with a transverse pipe; every two of the transverse pipe, the supporting sleeve and the connection rod are perpendicularly arranged; the end, which is far away from the connection rod, of the transverse pipe is fixedly connected with a second fixed block. The body-building exercise flexibility training auxiliary device is simple in structure and convenient to use and collect, can guarantee the standard of hip flexibility training action of a user and enhance the hip flexibility training effect, is safer, and can prevent ankles of the user from being hurt during training.

The invention discloses an upper limb rehabilitation training device, and belongs to the technical field of limb rehabilitation training instruments. The training device mainly comprises an arm lifting mechanism, an arm traversing mechanism and a wrist rehabilitation mechanism. The arm traversing mechanism is connected with the arm lifting mechanism through a lifting screw. The wrist rehabilitation mechanism is fixed on a sliding block of the arm traversing mechanism, and an intermediate shaft, a cross horizontal shaft and a cross vertical shaft of the wrist rehabilitation mechanism intersectat one point. The training device is simple and compact in structure, convenient to operate and capable of achieving upper limb rehabilitation training actions of multiple users simultaneously.

The invention relates to the technical field of target search, in particular to a method for training action planning model and target searching method, and the method comprises the steps: obtaining acurrent test image, a target object and a current step number; obtaining a predicted boundary frame and a boundary frame of the target object; determining a reward of the current action plan based onthe predicted size relationship between the bounding box and the bounding box of the target object; inputting the current test image, the target object and the current step number into an action planning model, and predicting the probability distribution of the next action and the reward corresponding to the next action; and carrying out reinforcement learning on the strategy network and the value network according to the reward planned by the current action and the probability distribution of the next action and the corresponding reward. On the basis of the predicted boundary frame and the actual boundary frame of the target object, a reward of the current action plan is determined; and reinforcement learning is carried out on the strategy network and the value network by using a prediction result, so that the strategy network and the value network are optimal, and relatively high search efficiency is achieved.

The invention relates to a human body upper limb function automatic auxiliary recovery apparatus. The human body upper limb function automatic auxiliary recovery apparatus comprises a control system and two opposite human body upper limb auxiliary recovery devices. The human body upper limb auxiliary recovery devices are fixed to a seat, and the control system is used for automatically controlling the operation of the human body upper limb auxiliary recovery devices. Each human body upper limb auxiliary recovery device comprises a shoulder joint side abduction unit, a shoulder joint flexion and extension unit and an elbow joint flexion and extension unit. The human body upper limb function automatic auxiliary recovery apparatus can achieve control, cooperation and use at three degrees of freedom, more professional recovery training actions can be set through a program, and therefore a patient with upper limbs which cannot move freely can be better trained.