36 results about "Gait database" patented technology

The invention relates to the technical field of gait analysis in biomedical engineering and provides a human gait analyzing method and system based on multi-sensor fusion. The method comprises the steps of filtering sensor signals to eliminate the signal noise error according to human motion features, and eliminating the integral error by means of the improved zero velocity updating algorithm, so that the method and system can be adapted to different walking scenes; fusing multiple sensor data with the Denavit-Hartenberg method, and reducing the leg position calculation error; calculating the step speed, step length, step frequency, walking period and walking track of a tested person accurately during walking after error correction is conducted; establishing a gait database, and conducting statistic analysis on gait data of different tested persons with the quantile regression analysis method. By the adoption of the method and system, gait parameter measurement precision can be improved, and gait parameters of different tested persons are comparable through standardization.

The invention discloses a gait recognition method based on the optical flow field, which obtains the gait optical flow field and detects the gait cycle according to the individual gait contour image sequence, and multiple gait cycles in the gait contour image sequence The individual gait optical flow fields are superimposed to obtain the individual gait optical flow map, and the horizontal component and vector of the gait optical flow map line optical flow of each gait optical flow map of the individual, and the line light flow of the gait optical flow map are extracted. The four gait feature vectors are the longitudinal component and vector of the flow, the horizontal component and vector of the optical flow of the gait optical flow diagram, and the longitudinal component and vector of the optical flow of the gait optical flow diagram. The above four gait feature vectors of the test individual are matched with the corresponding four gait feature vectors of the training individual in the gait database, and the test individual is identified in the gait database. The invention decomposes the optical flow components to obtain the gait features to identify the identity, can well extract the gait movement information in the process of people walking, and has strong robustness to clothing, carrying objects and the like.

The invention discloses a gait cycle recognition method based on the fusion of torque and angle feedback, which is characterized in that, according to the axial force and bending moment of the prosthetic leg extension rod of the prosthetic wearer, the ground support reaction force model is established, and according to the installation The pressure sensor at the end of the prosthetic leg extension rod obtains the ground support reaction force data and the support reaction force distribution of the prosthetic foot, and then obtains the knee joint rotation angle data according to the angle sensor installed on the knee joint of the prosthetic leg ; According to the support reaction force data and the knee joint rotation angle data, compare the data in the gait database to determine the phase of the current gait cycle. The invention overcomes the defect that the measurement data of the method of identifying the gait cycle by the electromyographic signal is easily disturbed by the skin surface state, and the gait cycle cannot be accurately judged when the skin is sweating or unclean, and the test result of the sensor is reliable. There is a risk of the sensor falling off.

The invention discloses a prosthetic knee joint control method, which is characterized in that the processor establishes a simplified model of the prosthetic leg, collects ground support reaction force data according to a pressure sensor installed on the prosthetic foot, and then according to the angle of the knee joint installed on the prosthetic leg The sensor obtains knee joint rotation angle data; according to the support reaction force data and the knee joint rotation angle data, compared with the data in the gait database, the processor performs gait cycle identification and motion state identification; the processor according to Motion state recognition, control the damping size of the hydraulic cylinder, and change the action of the prosthesis. The invention overcomes the shortcoming of insufficient moment when the pneumatic cylinder supports the phase, and proposes a control method for the user of the prosthesis in different motion states, so that the user can move more naturally.

The present invention provides a method and device for identifying a customer. The method includes: determining the video image of the same customer based on the collected video; and extracting the first facial feature data corresponding to the video image and the first Gait feature data; when determining that the client corresponding to the video image is an old customer, the first facial feature data is identified based on the face database, and the second facial feature data is determined; the first gait feature data is processed based on the gait database Identifying and determining the second gait feature data; performing cooperative matching based on the second facial feature data and the second gait feature data to determine the identity information of the customer. In this solution, through the above method of customer identification, the identity of the customer can be accurately determined, and then precise services can be provided to the customer.