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 night pedestrian identification method and a system thereof based on thermal infrared gait. The steps of the method comprise detecting and dividing the pedestrian from infrared video images to obtain image sequences of the thermal infrared gait, employing an average gray image to describe videos of the thermal infrared gait to obtain features of the thermal infrared gait and identifying the thermal infrared gait of the pedestrian in the thermal infrared video images, based on a nearest neighbor classifier and taking the average gray image as the description of identity. The system comprises an acquisition module of the thermal infrared video images and an analysis processing module for the infrared images. Experiments on a CASIA night infrared gait data bank and an NLPR day visible light modal gait data bank prove that the invention has effectiveness for identification of the night pedestrian and strong identification ability of gait representation mode. The invention is applied to security monitoring of special occasions.

The invention provides a lower limb power-assisted exoskeleton robot gait pattern identification method and system. The method comprises the steps that a pressure sensing device embedded into a shoe sole of a lower limb power-assisted exoskeleton robot is used for detecting shoe sole pressure information of a wearer, the obtained shoe sole pressure information is compared and matched with walking modes in a human body gait database, a walking mode is obtained through judgment, movement phases in a gait period are identified, and a movement phase is obtained through judgment; data are processed and analyzed, results obtained after identification and matching are transmitted to a lower limb power-assisted exoskeleton robot control system to be used as a control basis. The walking mode and the detailed movement phase in each gait period of the lower limb power-assisted exoskeleton wearer provide the abundant decision basis for control over the lower limb power-assisted exoskeleton. The lower limb power-assisted exoskeleton robot gait pattern identification method and system not only can provide movement information of the wearer for the lower limb power-assisted exoskeleton, but also can be widely applied to human body gait identification.

The invention discloses a hemiplegic patient self-recovery meter based on a body mirror image theory, and belongs to the field of medical instruments. The hemiplegic patient self-recovery meter comprises a contralateral limb posture signal acquisition device, a signal feature extraction and combination device and an electrical stimulation device, wherein the contralateral limb posture signal acquisition device is used for acquiring the posture information of a contralateral limb of a patient and transmitting the acquired signal to the signal feature extraction and combination device; the signal feature extraction and combination device is used for mapping the posture information of the contralateral limb of the patient from a normal walking gait database according to the received posture information of the contralateral limb and the personal information of the patient, generating a corresponding motion control command, and sending the control command to the electrical stimulation device; and the electrical stimulation device is used for receiving the control command, generating an electrical stimulation signal for stimulating the limb of the patient and making the limb of the patient generate corresponding motion. The recovery meter controls the contralateral limbs of the patient to coordinately move through mirror image copying by taking the posture information of the contralateral limb of the patient, so that effects of balance walking of a hemiplegic patient and reconstruction of a moving function are achieved.

The invention relates to the technical field of human body gait analysis, and provides a human body gait database and an establishment method thereof. The method for establishing the human body gait analysis comprises the following steps: collecting human body gait data, wherein the human body gait data comprises movement data, human body EMG data and plantar pressure data; storing the human body gait data through figure identification; designing a human body gait database interface, linking the database interface to the gait data, and extracting various types of gait data according to the operation requirements of users on the database interface. The human body gait database and the establishment method can meet not only the requirements of researchers on large data volume, but also the requirements of different researchers on various types of gait data.

The invention discloses a gait recognition method based on a 3D dense convolutional neural network. A network in the method employs 3D convolution to extract the transformation features of a gait at atime dimension, and has the feature retaining capability of a DenseNet structure. According to the invention, the method achieves the training of a classification and recognition model with the excellent performances under the conditions that a network depth is smaller and the number of training samples is smaller, wherein the classification and recognition model can recognize the identity according to the gait in a video. The testing on Dataset A of a CASIA gait database indicates that the method can achieve the training of the gait recognition model under the condition that the training samples are not sufficient. Moreover, the method is high in training speed, is small in model parameters, is high in recognition efficiency, and has the considerable recognition capability under the single-view-angle or cross-angle conditions.

Embodiments of the present invention disclose a gait identification method and an electronic device. The method comprises: acquiring a gait parameter of a target object; extracting gait feature information of the gait parameter according to the gait parameter; matching the gait feature information with an abnormal gait model stored in the gait database to obtain a matching ratio; and if the matching ratio reaches a preset threshold, identifying that the current gait of the target object is the abnormal gait. According to the embodiments of the present invention, gait identification accuracy can be improved in a facilitated manner, and the probability of misjudgment can be reduced.

The invention discloses a wearable lower limb gait intelligent detecting walking device. The device comprises a seat mechanism, thigh mechanisms, knee joint mechanisms, crus mechanisms and shoe cover mechanisms; the thigh mechanisms are movably connected with the seat mechanism, and the shoe cover mechanisms are connected with the crus mechanisms; each knee joint mechanism comprises a knee joint outer plate and a knee joint inner plate, an outer plate shaft is fixedly installed on each knee joint outer plate, an inner plate shaft is fixedly installed on each knee joint inner plate, each knee joint outer plate is provided with a straight line slot extending in the vertical direction, each inner plate shaft is movably clamped in the corresponding straight line slot, each knee joint inner plate is provided with a curve slot, each outer plate shaft is movably clamped in the corresponding curve slot, the upper portion of each knee joint inner plate is fixedly connected with the corresponding thigh mechanism, and the lower portion of each knee joint outer plate is fixedly connected with the corresponding crus mechanism. According to the wearable lower limb gait intelligent detecting walking device, a personal gait database is formed and used for doctor diagnosis and rehabilitation, walking and sitting are more convenient, and operation is easy and convenient.

The invention aims at developing a gait recognition and classification system, so as to help a security department to recognize suspicious persons appearing already from a lot of surveillance videos, carry out trajectory tracking on a target pedestrian, then determine the action range of the pedestrian and provide a clue to the security department for detecting the case. The method comprises the following steps: a gait database is built, a skeleton tracking function of an SDK is opened, a WPF program to obtain gait features is written, feature vector data of the gait features are acquired, the data are outputted to a txt text document, and a color image marked through skeleton tracking at the same time and a color video stream at the same time are outputted; a gait recognition and classification system is developed, the recognition and classification system is developed through C++, and an improved nearest neighbor classification algorithm is used for realizing system classification; and trajectory tracking is carried out on the target pedestrian, which is realized by finding out the video section of the target pedestrian at the same time according to the recognized target gait data at the same time.

The invention discloses a knee joint disease prediction apparatus and system. The system comprises an acquisition unit used for acquiring gait data and establishing a gait database; a preprocessing unit used for preprocessing the gait data in the gait database and obtaining standard gait data; a model training unit used for constructing a multilayer neural network model, training the multilayer neural network model according to the standard gait data, and obtaining a trained neural network model; and a prediction unit used for acquiring the gait data of users in need of detection, inputting the data to the trained neural network model, and obtaining a prediction result. According to the apparatus and system, multiple multilayer neural network models are established and trained through thegait data, knee joint diseases can be further predicted, various knee joint diseases can be rapidly and accurately predicted without radiation, and the prediction efficiency is greatly improved. The apparatus and system can be widely applied to the medical field.

The invention provides a customer identification method and device, and the method comprises the steps: determining a video image of the same customer according to a collected video; based on the video images of the same client, extracting first facial feature data and first gait feature data corresponding to the video images; when it is determined that a client corresponding to the video image is an old client, identifying the first facial feature data based on a face database, and determining second facial feature data; recognizing the first gait feature data based on a gait database, and determining second gait feature data; performing collaborative matching based on the second facial feature data and the second gait feature data, and determining identity information of the client. According to the scheme, client identification is carried out through the mode, the identity of the client can be accurately determined, and then accurate service is provided for the client.

The invention relates to a gait database construction method and a gait database construction system under a multi-camera environment. The method comprises the following steps of determining the number of cameras and the coverage angle of each camera to the target to ensure that the target is completely covered by a plurality of cameras; using each camera to extract the gait information of the target and compensating each other for the missing information; after each camera is compensated, adjusting the weight of each camera to obtain the compensated general eigenvalue, and extracting the gaitinformation of the target by using the compensated general eigenvalue to form the gait database. Through the method and the system of the invention, a more accurate gait database can be established,and the application scene of the gait database can be enlarged.

The invention discloses an artificial limb knee joint control method, which is characterized in that a processor builds an artificial limb simplified model, the ground bearing reaction data is collected according to a pressure sensor installed on an artificial foot, and the knee joint corner data is obtained according to an angle sensor installed on an artificial leg knee joint; the data is compared in a gait database according to the bearing reaction data and the knee joint corner data, and the gait period identification and motion state identification are carried out by the processor; the processor is used for identifying and controlling the damping of a hydraulic cylinder according to the motion state and changing the action of the artificial limb. The defect that the moment of force isinsufficient when a pneumatic cylinder supports a phase is overcome, and a control method of an artificial limb user in different motion states is provided, so that the user can move more naturally.

The invention provides a lower limb exoskeleton robot gait stabilization algorithm oriented to medical rehabilitation and assisted walking. A human body gait acquisition device is established by usingan acceleration sensor and an image processing mode to acquire walking gaits of various people so as to establish a primary gait database; the primary gaits are classified according to ages and genders according to a dynamic time warping algorithm, and polynomial curve fitting of data is performed in respective categories to obtain a primary gait model; a virtual robot in the same proportion is constructed to carry out repeated simulation tests, actual tests are performed according to a robot kinetic model, and gait model defects are adjusted according to feedback information such as moment in the test process. The input gait model can meet the hardware condition of the lower limb exoskeleton robot and meet the lower limb rehabilitation requirement of a patient at the same time, and finally the standard gait model meeting the requirements of all ages and genders is obtained.

The invention discloses an intelligent physical examination machine with gait detection and a gait detection method thereof. The intelligent physical examination machine comprises a physical examination machine body with a computer, a walking simulation device, a video acquisition device, and a gait recognition and processing program running on the computer of the physical examination machine body. The walking simulation device is used for collecting gait samples and establishing a gait database. The video acquisition device obtains gait video information of a tester and sends the gait video information to a computer. The computer loads the gait recognition and processing program to recognize and process the obtained gait video information, obtains the corresponding gait characteristics, compares the matching situation in the gait database; and when the gait characteristics match the gait data in the gait database, and gives the prompt information. The gait recognition method of the invention is applied to an intelligent physical examination machine, and aims at finding and extracting the variation features between individuals from the same walking behavior to realize automatic identification.

The invention discloses a pedestrian identity recognition method, device and equipment and a storage medium, which belong to the technical field of computers. According to the invention, cross recognition of footprint features and gait features is realized, the gait characteristics of the pedestrian and the footprint characteristics of the pedestrian can be mutually converted, and the gait characteristics can be converted into the footprint characteristics through the gait footprint conversion model in a scene that the pedestrian does not leave footprints on a crime scene, so that retrieval can be performed in the footprint database under the limitation that a footprint image is not obtained. Besides, in a scene in which a video of the pedestrian is not shot, footprint features can be converted into gait features through the footprint gait conversion model, so that retrieval can be performed in the gait database under the condition that the video is not obtained, the effective information amount of pedestrian identity recognition is increased, and case solving clues are increased.

The invention provides an intelligent terminal unlocking method based on gait recognition. The method comprises the following steps that 1, a gait of a user is recorded and stored in a gait database, the user starts a gait collecting module to record sample gait data of the user, the data is stored in the gait database of an intelligent terminal, and gait recognition is set as an unlocking password; 2, a gait collecting module is started, and when the user wants to conduct the unlocking action, the gait collecting module is started to collect the gait of the user; 3, a user gait recognition module is started, after gait collection is completed, the user gait recognition module is started, and the gait, obtained in the step 2, of the user is recognized and converted into gait data. The method has the advantages that by means of uniqueness of the gait of each person, the gait of the user is used as the unlocking password of intelligent terminal equipment, the security level of unlocking is improved, the pleasure and entertainment of unlocking are increased, and the user can enjoy the pleasure of intelligence in the unlocking process.

The invention discloses a gait recognition platform and method based on a trusted execution environment. The gait recognition platform comprises a key management module, a gait collection module, an encryption module, a preprocessing module, a feature extraction module and a classification recognition module. The monitoring camera equipment acquires a public key which is stored in the trusted execution environment and used for an asymmetric encryption algorithm, encrypts a gait video sequence acquired by the monitoring camera equipment by using the public key, sends the encrypted gait video sequence to the computer, transmits the encrypted gait video sequence to the trusted execution environment, preprocesses the encrypted gait video sequence in the trusted execution environment, sends thepreprocessed image sequence to a feature extraction application program in a trusted execution environment, analyzes and extracts gait features in the feature extraction application program, comparesand identifies the extracted gait features with the gait features in a gait database, and returns a comparison result to a common environment. The gait recognition platform and method can effectivelyprevent malicious programs from embezzling gait information, and defend against time sparse adversarial attacks for gait recognition.

The embodiment of the invention relates to the field of data analysis, and discloses a leg joint gait detection method, which specifically comprises the following steps of: analyzing gait images of a crowd to construct a gait database; combining normalization processing to acquire gait parameters and gait tracks of joint points of a patient in the walking process; constructing a fitness function for the gait track and samples in the gait database, solving the fitness function, and optimizing the gait parameters; and analyzing and evaluating the gait of the patient according to the optimized gait parameters. The gait analysis of the leg joint points of the patient is converted into parameter optimization analysis, so that the analysis of the health degree of the leg joints is quantified, the result is evaluated, the detection result is obtained quickly, and the detection process is short in time. And dynamic images of the patient are analyzed. Compared with traditional static image analysis, the method has the advantages that the overall movement effect of the leg joints of the patient can be analyzed more easily, and the obtained detection result is more accurate and comprehensive.

The invention relates to the technical field of posture correction, and discloses an attitude correction method which comprises the following steps: 1), collecting the current gait of a person with nonstandard gait; 2), comparing the data with a standard gait database; 3), finding out gait data most matched with the gait data, and taking the group of gait data as a reference recovery track for gait correction of the patient; 4), analyzing the motion trail of the patient with poor gait deformation, and finding out the problems of the current patient through analysis; and 5), correcting and recovering the gait of the patient in cooperation with a lower limb exoskeleton rehabilitation system. According to the attitude correction method, gaits after rehabilitation are collected and compared with the standard gait database, problems existing in a current patient are obtained after comparison, the recovery state of each current muscle group of the patient is derived in combination with the inverse kinematics principle, and correlation analysis and rehabilitation training planning are conducted on the recovery states.