217 results about "Hand joint" patented technology

The invention discloses a three-dimensional human body posture estimation method and device. The method comprises the following steps of S1, using a monocular camera to collect the depth images and the RGB color images of a human body at different angles; S2, constructing a human body skeleton key point detection nerve network based on the RGB color images, and obtaining a key point labeling image; S3, constructing hand joint node 2D-3D mapping network; S4, calibrating the depth image of the human body at the same angle and the key point labeling image, and then carrying out three-dimensionalpoint cloud color conversion on the corresponding depth image so as to obtain a colored depth image; S5, based on the key point labeling image and the colored depth image, using a preset learning network to predict the corresponding position of the labeled human body skeleton key point in the depth image; and S6, combining the outputs of the step S3 and the step S5 to achieve the refined estimation of three-dimensional human body posture estimation.

A virtual skeleton includes a plurality of joints and provides a machine readable representation of a human target observed with a three dimensional depth camera. A relative position of a hand joint of the virtual skeleton is translated as a gestured aiming vector control, and a virtual weapon is aimed in proportion to the gestured aiming vector control.

The invention provides a virtual reality integrated system based on action capture. The system comprises an inertial action capture device, an indoor positioning device, a virtual reality device, a data glove device, an electronic imitation gun device and a backpack computer device; and a first Calman filter is built in the backpack computer device. The inertial action capture device comprises a plurality of action capture sensors; the indoor positioning device is a UWB indoor positioning system; the data glove device comprises a glove body and a plurality of hand joint posture sensors; the electronic imitation gun device comprises an electronic imitation gun and an electronic imitation gun data collection sensor; and each of the action capture sensor, the hand joint posture sensors and the electronic gun posture sensor comprises a three-axis MEMS acceleration sensor, a three-axis MEMS angular velocity sensor, a three-axis MEMS magnetometer, a data filter sensor and a microprocessor with a built-in second Calman filter. The virtual reality integrated system provided by the invention can realize omnibearing interaction of people in the real world with the virtual environment and improve the immersion and the interactivity of the virtual reality system.

A joint fixation system particularly useful for joints of the hand is disclosed. The joint fixation system includes a joint fixation plate having a distal section, a proximal section, and an intermediate section connecting the distal section and the proximal section. The distal section includes a distal section screw hole, and the proximal section includes a proximal section screw hole. The joint fixation plate is pre-bent such that the distal section and the proximal section form an included angle of less than 180 degrees. The joint fixation system includes a first screw dimensioned to engage an inner surface of the distal section screw hole, and a second screw dimensioned to engage an inner surface of the proximal section screw hole. The first screw is angulated proximally when inserted in the distal screw hole, and the second screw is angulated distally when inserted in the proximal screw hole.

This new invention is a frame and servo assembly for attaching to massage chair seats manufactured without an existing servo assembly. It includes a frame body, on which there is sliding rack, a screw drive, a positioning stabilizer bar, and a connecting plate. The sliding rack is coupled to various motor brackets, the screw drive, a driving block, a screw drive stop block, and a plurality of wave shaped massaging wheels. The various motor brackets are coupled to two motors. One of the motors utilizes an axle with an off-center axis on both ends. There are multiple connecting brackets installed on the off-center axle, with wheel brackets installed at the end of the connecting brackets. Various “wave” shaped wheels at the end of the wheel brackets aid to simulate the natural feel of human hand joint movements to a user. At the two ends of the additional motor assemblies are bases for joint assemblies. These joint bases are connected to the connecting brackets via connecting joints. The wheels are located on the two ends of the sliding rack. The positioning stabilizer bar is affixed inside the frame body for stability.

The invention describes a skeletal control of a three-dimensional virtual world. The virtual skeleton includes a plurality of joints and provides a machine readable representation of a human target observed with a three-dimensional depth camera. A relative position of a hand joint of the virtual skeleton is translated as a gestured control, and a three-dimensional virtual world is controlled responsive to the gestured control.

The invention discloses a three-dimensional gesture recognition method based on a depth image and an interaction system. The method comprises steps: a depth information map is obtained by using a depth sensor; an AOI is recognized for the obtained depth information map by using a CNN, and a to-be-recognized hand depth information map is cut; the hand depth information map acquired from the CNN isinputted to a gesture recognition network for recognition, and three-dimensional coordinates of a recognized hand key node are acquired; the hand key node coordinates are subjected to coordinate transformation, which is adapted to world coordinates in the interaction system; by using Socket communication, a hand key node coordinate sequence after coordinate transformation is transmitted to the interaction system; and the interaction system acquires the hand key node information, a semantic judgment condition set by a user is combined, and a physical engine is used to display an interaction result. The three-dimensional coordinates of a hand joint point can be recognized excellently, a gesture semantic action can also be judged, and the application prospect is wide.

The invention discloses a pneumatic hand joint rehabilitation system. The system comprises a glove with five finger sleeves, an air pump and a control device; a back surface of each finger sleeve is provided with a set of pneumatic telescopic adjustment mechanisms for fingers inserted into the finger sleeves along the length thereof, the pneumatic telescopic adjustment mechanisms on each finger sleeve are inflated and deflated by the control device through the air pump, so that the fingers inserted into the finger sleeves passively generate flexion and extension actions, and the rehabilitationtraining is completed; through the control device, it is possible to select the fingers which need to be subjected to rehabilitation training, and a rehabilitation training time and a finger flexionand extension switching time are set; it is possible to perform the rehabilitation training for all the fingers and set the rehabilitation training time and a finger switching time; mirror rehabilitation training that healthy hand positive flexion and extension drives the passive flexion and extension of diseased hands is completed; the degree of grasping of the hands is adjusted. The system can automatically drive fingers of patients to perform passive training such as grasping, gripping, stretching and the like, and effectively restore finger flexibility.

The invention discloses a manipulator with the functions of receiving and stacking. The battery, the lithium battery is a rectangle with equal lengths on both sides, the multi-functional claw is composed of claws, magnetic plates, sensors, connecting rods, axles, and telescopic arms. The present invention is a manipulator with functions of collecting and stacking, Connect and fix the multi-functional gripper in the center of the bottom of the bracket through the connecting rod, and the gripper will work by energizing the touch sensor of the actuator, and the three claws on the multi-functional gripper will be glued together to form a rectangular template, and push back and forth for material collection or Open the claws to match the magnetic plate to pick up items, and the telescopic arm can be driven by the wheel shaft to elongate the claws to stack items, so that it is convenient, fast and time-saving to replace manual work.

The invention discloses a multi-dimensional vision hand function rehabilitation quantitative evaluation system and an evaluation method. The system comprises an optical motion capturing device, a video capture device, a hand function quantitative evaluation system, an interactive touch screen and a collection platform, wherein the optical motion capture device acquires three-dimensional space data and motion vector information of each joint of fingers, a palms and each joints of a hand wrists in real time by an intelligent algorithm; the optical motion capture device outputs motion parameters which are then input into the hand function quantitative evaluation system, the hand function quantitative evaluation system preprocesses data, preprocessing results are screened and stored in a database, and by aiming at evaluation schemes of different evaluated motions, an evaluation result is provided. According to the multi-dimensional vision hand function rehabilitation quantitative evaluation system and the evaluation method, three-dimensional space parameter computation precision of hand joints of a patient can be improved; by combining a computer vision technology with optical intelligent motion capture, more precise quantitative rehabilitation evaluation data is provided, and assists to the a doctor to perform rehabilitation diagnosis on the patients.

The invention relates to a medical rehabilitation training instrument. An upper-limb rehabilitation training robot solves the problems that existing equipment is complex in structure and hand joints cannot move comprehensively. The upper-limb rehabilitation training robot comprises a base (1), a support frame (3), a seat (2), a cantilever beam (4) and an upper limb training mechanism, wherein the upper limb training mechanism comprises a shoulder joint expanding and shrinking mechanism, an arm bending and stretching mechanism, a shoulder joint internal and external rotating mechanism connected and fixed with rotary rods (17 and 18), a front arm forward and backward rotating mechanism the same as the shoulder joint internal and external rotating mechanism, an elbow joint expanding mechanism fixedly connected with the shoulder joint internal and external rotating mechanism and the front arm forward and backward rotating mechanism through two pairs of rotating blocks (32 and 41), and a hand expanding mechanism fixedly connected with the shoulder joint internal and external rotating mechanism through slide bars (44 and 45). The upper-limb rehabilitation training robot is simple to operate and enables the joints to move comprehensively.

The invention discloses a coal mine blasting insecurity action identification and judgment method based on GMM (Gaussian Mixture Model) representation. The method comprises the following specific processes that: (1) preprocessing human body skeleton sequence data, obtaining the human body skeleton sequence data through kinect, obtaining hand joint data through a data glove, and carrying out denoising and normalized preprocessing on the human body skeleton sequence data and the hand joint data; (2) carrying out human body skeleton sequence data representation: 1) obtaining the GMM feature of action, 2) representing a skeleton sequence time sequence, and 3) extracting action representation GMM features; and (4) on the basis of the behavior identification of a Bayes classifier. By use of themethod, one action is comprehended as the sequential combination of multiple key gestures, a method for representing an action key gesture by the GMM is used for simplifying problem processing, so that the method exhibits certain effectiveness, and the problems of long time sequence, complex action and difficulty in judgment of coal mine blasting action can be accurately solved through identification and judgment.

A continuous casting constant-weight cutting device comprises a cutter, a casting roller track, a casting blank flow and a control system. The invention is characterized in that a track beam vertical to the casting blank track is arranged above the casting blank flow of the continuous casting machine, the track beam is arranged with a movable cart disposed with a weight support and a weight sensor, the weight support is disposed with an electric pushing rod connected with a flexible rod whose end is fixed with a beam, two ends of the beam are mounted with a lifter and a hand joint. The invention mounts weight device on the weight cart covering N casting blank flows, therefore, the continuous casting blanks share one weight, to avoid error, and the control system can correct the size via the weight of each flow, to control the weight. The invention can cut the continuous casting blank in constant length and weight, to improve efficiency.

The invention discloses a manual alphabet identification method based on Leap Motion. The method comprises steps of acquiring palm and wrist bone articulation point 3D coordinate information through a depth camera Leap Motion device; performing correlative calculation on the 3D coordinate information so as to obtain bending angle information of hand joints; acquiring hand type feature information and wrist bending degree feature information through feature processing on the angle information; calculating the Euclidean distance between the acquired feature information and templates; and identifying manual alphabets according to a maximum probability criterion, a nearest neighbor criterion and a successive frame flow result consistent principle. By employing the method, the Chinese manual alphabets can be effectively and rapidly identified, manual alphabet elements are relatively independent, and sign language can be identified in a real time manner through the identification of manual alphabet continuous sequences. By employing the method, the sign language based on the manual alphabets can be identified in a real time manner, so that the deaf can effectively communicate with others by the use of a wearable device.

The invention provides a 3D gesture recognition method and system based on a deep convolutional neural network. The method comprises the steps that firstly, using a first deep convolutional neural network or conducting pre-segmentation on a large number of color images containing hands, and extracting hand action parts; secondly, performing hand joint node detection on the extracted hand by usinga second deep convolutional neural network; Performing gesture 3D reconstruction on the detected joint nodes by using a double-flow deep convolutional network; And finally, constructing a softmax network comprising three full connection layers to identify the 3D reconstructed gesture. The technical scheme provided by the invention has the beneficial effects that the gesture recognition precision can be effectively improved. From the perspective of application range, an object of the method is an RGB image collected by a monocular camera, required equipment is simple and cheap, and the application scene is wider.

The invention is suitable for the technical field of computers, and provides a vehicle-mounted man-machine interaction method and vehicle-mounted equipment, and the method comprises the steps: obtaining a target image containing hand information in a preset region; according to a pre-trained gesture recognition model, conducting hand joint point recognition on the hand information, and obtaining three-dimensional space coordinates of all joint points of the hand relative to a preset original point; generating gesture recognition information according to the three-dimensional space coordinates;and determining a vehicle control instruction corresponding to the current gesture recognition information according to a pre-stored mapping relationship between the gesture recognition information and the vehicle control instruction, and outputting the vehicle control instruction. As the image acquisition can be completed only by the camera and the processing process of the image is relatively simple, the human-vehicle interaction cost in the prior art can be reduced, and the gesture is further recognized from the hand information by virtue of the gesture recognition model, and the human-vehicle interaction is carried out through the gesture, so that the influence of the surrounding environment is not easily caused, and the user experience effect can be improved.

The invention provides a gesture recognition method based on a convolutional neural network and 3D estimation, and the method comprises the steps: processing a to-be-recognized image through employinga SegNet-base network model, so as to extract a hand mask feature map in the to-be-recognized image; constructing a deep convolutional network DetectNet based on supervised learning, and performing network positioning on hand joint points in the hand mask feature map to obtain a gesture recognition result of 2D estimation; and processing the 2D estimated gesture recognition result by adopting a PoseNormNet model based on the standard frame and viewpoint estimation to obtain a 3D estimated gesture recognition result.

The invention discloses a three-dimensional virtual-real fusion experiment system, and the system comprises: an object tracking device which is used for recognizing the type of a real experiment object and positioning the three-dimensional space coordinates of the real experiment object in real time, and obtaining the type of the real experiment object and the corresponding real-time three-dimensional space coordinates; a hand tracking device which is used for identifying the joint points of the operating hand and positioning the three-dimensional space coordinates of the joint points of the operating hand in real time to obtain the real-time three-dimensional space coordinates of the joint points of the operating hand; and a virtual-real fusion processing device which is used for generating a corresponding virtual-real fusion image according to the type of the real experimental object, the corresponding real-time three-dimensional space coordinates of the real experimental object andthe real-time three-dimensional space coordinates of the operating hand joint points. According to the invention, the advantages of the virtual-real fusion technology are fully utilized, the virtual object and the real experimental object are fused, the real experimental object can be tracked and cultivated in real time while a user naturally moves the real experimental object, the experimental operation ability and learning interest of the user are cultivated, and the experimental risk is reduced.