299 results about "Motion recognition" patented technology

The invention discloses an emotion prediction and disease derivation method and system based on multi-modal fusion, and the system comprises a data collection and preprocessing module, an emotion fusion module, an abnormal condition detection and cloud uploading module, and a disease possibility derivation module. The data acquisition and preprocessing module comprises a video part, a text part and an audio part, and the video part comprises face emotion recognition and prediction and human motion recognition and prediction; the text part comprises text content emotion recognition and prediction; the audio part comprises voice-to-text and voice tone emotion recognition and prediction, the system comprehensively captures an emotion state by fusing multi-mode information such as video, text and voice, and the accuracy and prediction capability of emotion recognition are improved; and by predicting the future emotion trend, the abnormal condition is warned in advance, and the response timeliness is improved.

The invention discloses an intelligent data labeling method based on artificial intelligence, and relates to the technical field of intelligent medical treatment, and the method comprises the following steps: obtaining physiological parameter data, historical health parameter data and motion state data of a user; outputting motion state label data through a pre-constructed motion recognition model according to the motion state data; and performing sliding window analysis on the historical health parameter data, and dynamically establishing and updating long-term change data of the user health parameters. According to the scheme, whether the physiological parameter fluctuation is caused by the movement or not can be distinguished through associated retrieval of time window division and the movement state label, for example, when it is detected that the heart rate fluctuation of the user during running exceeds the threshold value, normal physiological response instead of abnormity can be judged in combination with the movement label; the probability that normal physiological parameter fluctuation caused by movement is mistakenly marked as abnormal can be effectively reduced, and meanwhile, the pertinence of anomaly detection is improved.

The invention discloses a motion recognition method and system based on a redox photoelectric memristor, a terminal and a storage medium, and the method comprises the steps: selecting classical motions based on a human body motion data set, extracting time sequence data, and coding the time sequence data into an optical pulse sequence; constructing a reservoir array composed of a plurality of photoelectric memristors, and expanding an optical pulse sequence signal into a high-dimensional state vector; constructing a supervised training model, solving a weight matrix, and constructing a memristive cross array; and outputting an action classification result through simulation domain operation based on the output current multi-path light current signals in combination with the memristor cross array. According to the method, the motion features can be directly fed into a rear-end classification network for action recognition without depending on a complex digital feature extraction algorithm, so that the transmission and processing overhead of redundant data is fundamentally eliminated; a high-efficiency, low-delay and high-robustness hardware solution is provided for real-time and anti-noise motion recognition in scenes such as intelligent monitoring and man-machine interaction.

The invention relates to the technical field of action recognition, in particular to a pattern rope skipping action recognition system and method, and the system comprises a limb angle collection module, a stability recognition module, an action triggering and screening module, a rope body direction analysis module and a cross mapping judgment module. According to the method, by introducing three-dimensional coordinate extraction and included angle change trend analysis of limb key joints in the jumping period, the dynamic characteristics of limbs in the action process can be accurately reflected, and the action amplitude and stability change can be effectively expressed through an angle data sequence constructed by multi-dimensional indexes such as the maximum value, the minimum value and the average value of the included angle in the period; a stability identification mechanism is constructed to carry out local regression fitting and residual error statistics on a periodic angle trend, accurate labeling of a stable state of an action posture is realized, and then a key frame sequence with an actual jump intention and a continuous posture in the jump action is effectively identified through fusion screening of labeled frames and a jump-empty state.

The invention discloses a fan defect detection method and system based on machine vision, and belongs to the technical field of image data processing, and the method comprises the steps: obtaining a historical shot image of a fan blade, generating a training data set, building a deep learning network architecture based on a convolutional neural network, and building a fan blade motion recognition model. Carrying out calculation processing by using a dark channel prior algorithm, obtaining a dynamic adjustment weight, training an AOD-Net model based on the dynamic adjustment weight, establishing a fan blade fuzzy correction model, carrying out real-time shooting on the fan blade, and carrying out motion identification correction by using a fan blade motion identification model. The fan blade fuzzy correction model is used for dust fog removal, the defect detection model is established, and the defect detection model is used for defect detection, so that the fan blade moving at a high speed can be accurately identified, the outline details of the fan blade are clear, and the state of the fan blade can be timely and accurately analyzed during defect detection. And the working performance of the fan blade is improved.

The invention particularly relates to a multi-task motion recognition method based on a physical constraint guide condition diffusion model, which comprises the following steps: acquiring original observation data of multi-task motion based on an inertial measurement unit and preprocessing to obtain preprocessed original observation data; constructing a conditional diffusion model, and training the conditional diffusion model by using the preprocessed original observation data based on a preset loss function to obtain a trained conditional diffusion model; performing data enhancement on a key motion state category determined based on the motion state category label corresponding to the original observation data according to the trained conditional diffusion model to obtain a key motion state category sample, and constructing a sample balance data set based on the key motion state category sample; and based on the sample balance data set and a preset multi-task learning network loss function, constructing a multi-task learning pedestrian motion recognition model and carrying out multi-task recognition. Therefore, the problems of few key state samples, unreasonable data and the like in motion recognition are solved.

The invention provides an interaction method based on finger pinching action recognition, electronic equipment and a storage medium. The method comprises: acquiring an image in real time; determining the type of finger pinching gestures in a preset number of continuously acquired frame images; when same gesture images in the preset number of frame images meet a target condition, the finger pinching gesture is determined to be an effective finger pinching gesture, the same gesture images are images with finger pinching gestures of the same type, and the target condition includes that the number of the same gesture images is larger than or equal to a first number; determining a man-machine interaction instruction corresponding to the effective finger pinching gesture; and executing the man-machine interaction instruction. According to the method, the pinching gesture is taken as a gesture instruction in the man-machine interaction process, so that the pinching gesture is more difficult to trigger by mistake compared with a common gesture, and the safety of man-machine interaction is enhanced.

The invention discloses an action recognition method based on a virtual inertial measurement signal generation model, and relates to the technical field of action recognition, and the method comprises the steps: collecting a surface electromyogram signal and an inertial measurement signal, and carrying out the preprocessing of the collected signals; constructing and training a generator to convert the surface electromyogram signals into virtual inertial measurement signals; inputting the generated virtual inertial measurement signal and the original surface electromyogram signal into an action recognition model, and training to obtain a classification module; the performance of an action recognition model is evaluated by using a test data set, the accuracy rate, the recall rate and the F1 score index are calculated, the recognition effect of the model is measured, a classification model and a generator are optimized according to the recognition effect, and the advantages of two modal signals can be fully utilized by fusing surface electromyogram signals and virtual inertial measurement signals, so that the accuracy of the action recognition model is improved. And the motion information of the fingers, the wrist, the forearm and other parts is analyzed, so that the complex limb motion is identified more accurately, and the performance of the limb motion identification system in practical application is improved.

The invention relates to the technical field of intelligent security and protection monitoring, and discloses a library intrusion intelligent detection method and system based on deep learning, and the method comprises the steps: collecting library video data, carrying out the preprocessing, extracting spatial-temporal features, and carrying out the recognition of the spatial-temporal features; a deep convolutional neural network and a long-short-term memory network are combined to construct a double-flow fusion model, action recognition and abnormal behavior detection are realized, an intrusion event is determined through a multi-level threshold judgment mechanism, and real-time alarm is realized. According to the method, the accuracy and the real-time performance of warehouse intrusion detection are improved, false alarms and missing alarms are effectively reduced, and meanwhile, the method has autonomous learning and environment adaptability.

An electronic device and a method for anchoring an augmented reality object are provided. The electronic device includes a memory, a display and at least one processor operatively connected with the memory and the display. The at least one processor may be configured to control the display to display at least one augmented reality object on an augmented reality space, obtain anchoring attribute information designated to the at least one augmented reality object, identify an anchoring type of content anchored to the at least one augmented reality object according to a user's motion based on the anchoring attribute information, and control the display to display a visual effect representing the anchoring type to the content.

The invention discloses a sonar target motion identification method based on a low-frame-rate bright spot structure sequence. The sonar target motion identification method comprises the steps of 1, roughly intercepting a bright spot structure; 2, accumulating a bright spot structure, sonar platform information and tracking information; step 3, motion compensation of the sonar platform; 4, accurately intercepting a bright spot structure; and 5, searching the optimal radial speed. According to the method, target radial motion parameters are extracted by means of distance steep motion of a sonar target among multiple frames of echoes, and a speed estimation loss function based on bright spot structure global optimal estimation is innovatively designed for the problems that the time interval between adjacent echoes of a low-frame-rate sonar target is long and the bright spot structure fluctuates greatly. According to the method, the global similarity between input period bright spot structures is fully utilized, the distance steep motion measurement precision is improved, the radial speed of the sonar target can be given at the same time, and the motion identification precision and stability of the sonar target are improved.

The invention belongs to the technical field of sensor measurement and recognition, and particularly relates to a human body motion recognition system based on LSTM (Long Short Term Memory), which comprises a data acquisition module, a data calculation module and a data application module, and utilizes a neural network model and a wearable electronic sensor to detect a human body walking state. A new long short term memory (LSTM) recurrent neural network model, namely an LSTM-STRM model, is established, the model combines multi-task learning, an attention mechanism and spatio-temporal feature fusion to accurately classify the walking state of the human body, and the walking state is compared with the original LSTM model. The result shows that the LSTM-STRM model can be used for classifying the time sequence data collected by the measuring unit, the states of the five targets can be classified with high precision, and the recognition accuracy is higher than that of an original LSTM model. The method improves the recognition precision, is high in adaptability, is suitable for the fields of medical treatment, human-computer interaction, exercise training and the like, and has a wide application prospect.

The invention relates to the field of computer vision, artificial intelligence and intelligent fitness, and particularly discloses a fitness action recognition, counting and quality evaluation method based on machine vision, which comprises the following steps of: extracting video frames and standardizing the video frames, realizing background suppression and human body region enhancement through a semantic segmentation or inter-frame difference method, and outputting a standardized image sequence; detecting key joint points by adopting a pre-training model, and outputting a stable skeleton sequence and candidate action stage data through time sequence consistency filtering; reconstructing a feature tensor, fusing spatio-temporal features through double-branch attention collaboration, and outputting action categories and time sequence compensation parameters in a classified manner; a dynamic threshold method accumulates the number of actions, action qualification is judged in combination with biomechanical constraints, the confidence coefficient is optimized, and a structured result is output; bone rendering, error highlighting, voice generation and personalized training suggestions. According to the method, wearable equipment is not needed, the problems of instable identification, miscounting and the like in a complex scene are solved, and real-time accurate analysis is realized.

The present invention relates to a technology capable of identifying a companion animal by analyzing video, image, and voice data collected through CCTV or a camera, for management of companion animals and tracking in case of loss thereof, wherein, by simultaneously or sequentially using at least one identification method among facial recognition, nose print recognition, voice recognition, and motion recognition by analyzing a video, image, or voice, an effect of greatly improving the reliability of object identification for companion animals can be provided.

The invention provides an action recognition method based on adaptive skeleton grouping and direction sensitive space-time modeling, and relates to the technical field of skeleton action recognition. Comprising the steps of skeleton action sequence input and feature embedding, adaptive skeleton grouping and direction weighting space-time modeling, trunk modeling and feature refinement, adaptive time down-sampling, multi-stream feature fusion and classification output and action recognition loss evaluation. The method comprises the following steps: acquiring an introduction result through skeleton action sequence input and feature embedding, acquiring an adjustment result through adaptive skeleton grouping and direction weighted space-time modeling, acquiring a processing result through trunk modeling and feature refinement, acquiring a reconstruction result through adaptive time downsampling, performing multi-stream feature fusion and classified output, and finally executing action recognition loss evaluation. According to the method, the skeleton action recognition precision is improved, and meanwhile, the complexity of long sequence modeling calculation is reduced, so that the skeleton action recognition real-time performance and deployment efficiency are improved, and the problem that the skeleton action recognition precision is not high in the prior art is solved.

The invention discloses a limited space human body motion recognition method based on multi-scale spatial-temporal feature fusion, and the method comprises the steps: firstly collecting motion data when a worker carries out the operation motion during the operation of the worker in a limited space, and building a motion data set; secondly, all action data sets are preprocessed, and actions are subjected to label numbering; and finally, inputting the data in the preprocessed data set into a multi-scale spatial-temporal feature fusion model, carrying out limited space personnel operation action recognition, and outputting a limited space action type label. According to the method, the aggregation connection relation between IMU sensors can be more completely described, the irregular action relation of the spatial dimension of each part of a human body is fully expressed, and the problems of low identification accuracy and low generalization performance of six-axis IMU operation action identification in a limited space in the prior art are solved.

The invention discloses a rehabilitation training system based on a key point recognition model and a graph neural network, and belongs to the field of deep learning neural networks. According to the system, a YOLOv8-Pose attitude recognition model, STGCN space-time diagram convolution and a double-layer full-convolution twin network are fused. The YOLOv8-Pose posture recognition model is used for accurately detecting the posture of the human body and providing basic data for action analysis of the posture of the human body. Based on STGCN space-time diagram convolution after space-time attention improvement and a double-layer full-convolution twin network, action recognition is carried out according to the features of human body postures, the correctness of rehabilitation actions is evaluated, and therefore the rehabilitation progress of a patient is accurately monitored and evaluated.

The invention relates to the technical field of railway maintenance education, in particular to a high-speed railway four-electricity system intelligent teaching system and method based on AI, and the method comprises the steps: a teaching data obtaining module collects teaching process data, student learning tracks and industry fault data; the virtual simulation modeling module is fused with a BIM model and a technical specification text to construct a four-electric system twinborn body; the AI intelligent analysis engine constructs student ability portraits through a machine learning algorithm, generates a personalized learning path and intelligently answers questions; the virtual-real teaching interaction module builds a digital-intelligent practical training scene, carries out three-dimensional visual error correction and remote guidance through real-time motion recognition, and outputs practical training data. The cross-professional fault simulation module carries out cross-professional fault linkage simulation deduction through an association rule mining algorithm and outputs a deduction result; and the teaching effect evaluation module establishes a binary evaluation system combining process and practical operation skill assessment, and outputs teaching effect data. Therefore, the problems of fixed teaching strategy, low precision and the like in the prior art are solved.

A micro-action recognition method based on space-time structure optimization comprises the steps that firstly, a to-be-trained micro-action video sample is preprocessed; secondly, designing a space-time structure optimization module STM, realizing adaptive enhancement of key features in the spatial dimension, interactively fusing information between adjacent time frames in the time dimension, and enhancing the time modeling capability of the micro-action recognition model; afterwards, micro-action joint embedding losses are designed, including cross entropy losses and semantic embedding losses, when the semantic embedding losses are calculated, video features and micro-action labels are mapped to a shared joint embedding space, and video feature embedding vectors and micro-action label embedding vectors are generated; the distance between the video feature embedding vector and the micro-action label embedding vector is shortened, and semantic alignment of the video features and the micro-action labels is realized, so that micro-action categories which are similar visually but different semantically are distinguished; and finally, under the supervision of the micro-action joint embedding loss, predicting a micro-action label of the video, and realizing micro-action identification and classification. According to the method, the accuracy and robustness of action classification are improved.

The invention discloses a dining table turntable control method based on diner action recognition of a side-mounted camera, and belongs to the technical field of image processing, and the method comprises the steps: obtaining an image region calibrated by a rotating disc, and representing the image region through an approximately elliptical polygon; obtaining a minimum circumcircle shape corresponding to the image area, and adjusting one end, far away from the camera, and the other end, close to the camera, of the approximate elliptical polygon to obtain a corrected image area calibrated by the rotating disc; an image of a dining table during dining is collected through a camera, whether the projection of the key points of the hand falls in the corrected image area calibrated by the rotating disc or not is recognized through a pre-trained RTMO model based on the key points of the human body, and whether the projection of the key points of the hand repeatedly moves or not is recognized, so that a motor is controlled. The intelligent dining table solves the problem that an existing intelligent dining table cannot accurately recognize food clamping, serving and other actions of diners under the condition that a camera is laterally installed, and consequently motor control is not accurate.

The application discloses a non-invasive brain-computer interface rehabilitation robot and relates to the technical field of intelligent rehabilitation equipment, and solves the technical problem that the existing rehabilitation robot mainly adopts a set program to perform repetitive rehabilitation training, and patients are not actively involved, the passive training mode limits the autonomous control and participation of the patients, and the rehabilitation effect is limited, especially in the aspect of neural injury repair, and the effect is poor; the application obtains motion data of the patient; an artificial intelligence model is trained based on historical motion data to obtain a motion recognition model; the motion data is recognized based on the motion recognition model to obtain a recognition result; the patient is subjected to rehabilitation training through the rehabilitation robot based on the recognition result; a rehabilitation effect evaluation coefficient of the patient is calculated based on the electromyographic characteristic data; and the recovery progress of the patient is judged based on the rehabilitation effect evaluation coefficient and a preset rehabilitation effect evaluation threshold value, so that the above technical problem is solved.

Disclosed herein is an article of smart clothing which automatically recognizes body configuration and motion. This smart clothing includes inertial motion units which are distal relative to selected joints and also sets of strain sensors which span those joints. The type of energy transmitted through a strain sensor can be electrical energy or light energy. Combined multivariate analysis of data from both inertial motion units and strain sensors provides more accurate recognition of body configuration and motion than data from either alone.

The invention discloses a hierarchical human body activity recognition method and system based on multiple position sensors. The system comprises a plurality of sensor modules, a data fusion module, a feature vector extraction module, a time sequence feature generation module and a human body activity recognition and judgment module. The plurality of sensor modules are used for acquiring and collecting detection data; the data fusion module performs data layer fusion on the collected detection data; a feature vector extraction module extracts representation vectors matched with human body activities from the fused data; a time sequence feature generation module generates hierarchical time sequence features of the human body activity according to the representation vector matched with the human body activity and the time change; the human body activity recognition and judgment module judges and recognizes human body activities according to the time sequence characteristics of the hierarchical human body activities. According to the method, a series of recognition technologies such as data fusion, feature extraction and action recognition are adopted to quickly recognize human body actions, and the actions are analyzed by adopting a plurality of sections of continuous time axes, so that subsequent action prediction is facilitated.

The invention relates to the technical field of somatosensory interaction game systems, and discloses a Qt-based somatosensory interaction and AI real-time plot fusion game system and a control method.The Qt-based somatosensory interaction and AI real-time plot fusion game system comprises a data acquisition module, a somatosensory interaction module, a voice recognition module, an AI decision engine module and a real-time plot generation module, the data acquisition module integrates Ki project V2 / Ite Rea Sense SDK, and the real-time plot generation module integrates the Ki project V2 / Ite Rea Sense SDK; a depth camera data stream is accessed through a Qt plug-in system, and the somatosensory interaction module comprises a basic motion capture module, a biological feature recognition module, an environment interaction module and a cross-device interaction module. According to the Qt-based somatosensory interaction and A I real-time plot fused game system and the control method, through a somatosensory interaction module, the somatosensory interaction module comprises a basic motion capture module, a biological feature recognition module, an environment interaction module and a cross-device interaction module, motion capture, human body micro-motion recognition, smart home interaction and multi-screen cooperation are achieved through the somatosensory interaction module; therefore, the wolf killer game interactivity is higher, and the scene interactivity is higher.

This application discloses a method, device, storage medium, and program product for rowing motion recognition based on an adaptive multi-scale peak algorithm, belonging to the field of computer technology. The method includes: periodically acquiring first acceleration data in a target direction; for each acquired first acceleration data point, performing peak and trough detection on the first acceleration data set within a preset time window to obtain a peak position set and a trough position set; and performing rowing motion recognition based on the peak and trough position sets to obtain one or more rowing motion information. This application uses only the first acceleration data in the target direction, avoiding interference from acceleration data in other directions, thus improving the accuracy of peak and trough detection. Furthermore, both the peak and trough position sets can reflect the rowing motion pattern; using bidirectional temporal constraints to recognize rowing motions improves the accuracy of rowing motion recognition, thereby improving the accuracy of rowing motion counting.

The invention relates to an action recognition system for online education, and the system comprises a real-time snapshot device which is disposed at a client of remote education and is used for carrying out the real-time snapshot processing of an environment where a remote education user is located, so as to obtain and output a corresponding snapshot processing image; and the contact notification mechanism is used for executing on-site broadcasting of notification information corresponding to small action recognition when the contact between the head and the hand is intelligently recognized based on various image processing data by adopting the multiple trained Hough neural network. According to the invention, the Hough neural network after multiple times of training can be introduced to intelligently identify whether the head and the hand are in contact based on various image processing data, and when the head and the hand are intelligently identified to be in contact, on-site broadcasting of notification information corresponding to small action identification is executed. Therefore, intelligent detection and warning of illegal small actions of remote education customers are realized.

The invention relates to an action recognition method and system based on video decoupling, and the method comprises the steps: constructing a visual feature extraction network, a semantic feature extraction network and a nonlinear similarity measurement learning module, and carrying out the connection and integration to form a zero-sample action recognition model; constructing a category name query library according to the training data, inputting the training data into a model for model training and hyper-parameter adjustment, and fixing model parameters after training; updating a category name query library according to the test data, and inputting the test set data into the model to obtain a corresponding test recognition result; and judging whether the accuracy rate of the test identification result is greater than or equal to a preset value, if so, updating the category name query library again according to the actual demand and inputting the actual data set into the model for identification, otherwise, retraining the model. Compared with the prior art, the method has the advantages that video decoupling and nonlinear similarity measurement learning are used, the discrimination of visual features can be enhanced, and the visual-semantic cross-modal correlativity is captured.

This invention relates to a skeletal action recognition method based on spatiotemporal topology learning, belonging to the field of computer vision and artificial intelligence technology. The method includes: constructing a two-dimensional topology coding module to simultaneously capture the static and dynamic topological relationships of the skeletal structure; employing a multi-level spatial attention aggregation module to adaptively weight and aggregate skeletal features at different levels to enhance the expressive power of key nodes and edges; and designing a temporal dynamic sequence pooling method to extract high-order dynamic information from time-series data based on a dynamic modality decomposition strategy, thereby improving the accuracy of action recognition. The advantages are: this method can fully utilize the spatiotemporal structural information of the skeleton data to achieve accurate recognition of human actions, while improving the modeling ability for complex dynamic behaviors, avoiding the limitations of traditional methods in capturing high-order dynamic information.

An electronic device is provided. The electronic device includes a display including a touch screen, a proximity sensor configured to detect an approaching external object, a motion sensor configured to detect movement of the electronic device, memory storing one or more computer programs, and a processor communicatively coupled to the display, the proximity sensor, the motion sensor, and the memory, wherein the one or more computer programs include computer-executable instructions that, when executed by the processor cause the electronic device to use the proximity sensor to identify whether the electronic device is worn on the arm of a user, in case that the electronic device is worn on the arm of the user, use the motion sensor to identify a movement of the electronic device, identify whether the identified movement of the electronic device belongs to a designated movement, in case that the identified movement belongs to the designated movement, identify whether the angle at which the electronic device is tilted belongs to a designated angle range, determine, at least based on whether the angle at which the electronic device is tilted is within the designated range, that the electronic device is in a touch misrecognition state, and in case that the electronic device is in the touch misrecognition state, refrain from executing an operation corresponding to a touch input detected on the display.

The present disclosure provides a motion recognition method and device and electronic equipment, and relates to the technical field of interaction. The technical problem of low accuracy of user motion recognition in somatosensory interaction is solved. The method comprises: in response to the operation of the somatosensory interaction device, acquiring the motion signal collected by the somatosensory interaction device; separating the motion attribute features in the motion signal and sample attribute features by means of knowledge distillation to obtain the motion attribute features corresponding to the motion signal; wherein the sample attribute features are used to represent the self attribute features of the user; and performing motion recognition by a classification neural network model according to the motion attribute features to obtain the motion recognition result of the user.