105 results about "Abnormal movements" patented technology

The invention is suitable for the technical field of fall detection and help calling, and provides a fall detection and help calling method for the elderly at home, and the method comprises the steps: collecting the multi-source health monitoring data of the elderly through a monitoring camera and a sensor, and carrying out the preprocessing based on the collected data; respectively extracting walking posture features, getting-up and sitting-down posture features, daily activity posture features, motion features, acceleration features, gyroscope features and heart rate features of the elderly; inputting the extracted features into a trained abnormal action detection model for action detection; when the old people have abnormal behavior actions, performing multi-dimensional analysis on the abnormal behavior actions of the old people by utilizing a clustering algorithm; and feeding back the analysis result of the abnormal behavior action of the old person in real time according to a preset help calling rule. According to the invention, the aims of comprehensively monitoring the health state of the elderly, timely discovering abnormity, deeply analyzing reasons and quickly responding to rescue can be fulfilled.

The invention relates to the technical field of medical or health data information, in particular to an intelligent health old-age care safety monitoring system, which comprises an old-age information acquisition module for acquiring face and basic information; the body health monitoring module predicts health parameters through LSTM, and if the health parameters exceed a threshold value or change abnormally, one is added to the alarm frequency; the psychological health monitoring module uses a StyleGAN to generate an enhanced data set and performs emotion classification through a convolutional neural network, and if the emotion deviates from a baseline, one is added to the alarm frequency; the tumble posture monitoring module judges tumble through posture estimation, and if tumble is detected, one is added to the alarm frequency; the position anomaly monitoring module identifies abnormal movement or residence, and if an anomaly is detected, the number of alarms is increased by one. And guardians and medical staff can obtain alarm information and monitoring reports in time according to the alarm times, so that the efficiency of monitoring the health and safety of the elderly is improved.

The invention discloses a method and system capable of identifying medical waste stealing actions, and the method comprises the steps: firstly obtaining visual information of a loading and unloading region through an omnibearing real-time video, carrying out the preprocessing of video data, and removing interference factors to enhance the definition and accuracy of an image; and then distance estimation is carried out on the preprocessed video, the moving distance of the human body to the transfer trolley is analyzed, and whether an abnormal approaching behavior exists or not is judged. If the system recognizes the abnormal approaching behavior, the arm action of the human body is further analyzed, and whether the abnormal action exists or not is judged according to the angle change of the arm. If the arm angle is detected to be abnormal and the system recognizes that medical waste is possibly held in the hand, the system automatically triggers an early warning mechanism. The method effectively overcomes the defects that traditional manual monitoring is low in efficiency and poor in judgment accuracy, abnormal behaviors can be accurately found in real time, and intelligent support is provided for safety management of medical waste.

The invention discloses an abnormal action detection method and an abnormal action detection model generation method, and belongs to the field of intelligent equipment. According to the method, the action tensor of the preset target is obtained, and the action tensor is obtained based on the image containing the preset target; inputting the action tensor to an abnormal action detection model, wherein the abnormal action detection model is pre-trained by a preset normal action tensor; and obtaining an output result output by the abnormal action detection model based on the action tensor, and judging whether a corresponding abnormal action exists or not based on the output result. The problem that the recognition rate of a current model for intelligently detecting abnormal actions is low due to sample imbalance can be effectively solved.

The invention relates to the technical field of auxiliary walking devices, and particularly discloses an auxiliary walking nursing device for neurosurgical nursing, which is characterized in that an inertial measurement unit (IMU) and a pressure sensor are used for respectively acquiring three-dimensional motion data and standing pressure data of a patient object; a neural network model based on deep learning is introduced to carry out data analysis on the three-dimensional motion data and the standing pressure data of the patient object, so that the time sequence correlation change characteristics of the motion state of the patient object and the dynamic change trend of the standing pressure are mined, and then fusion analysis is carried out on the time sequence correlation change characteristics and the dynamic change trend; therefore, intelligent identification of abnormal actions of a patient object is realized, and an early warning prompt is automatically generated and a built-in safety mechanism is started to lock a position when the abnormal actions occur. In this way, active safety protection of the patient object can be achieved, the safety and intelligent level of the auxiliary walking nursing device are effectively improved, and then the safety and efficiency of rehabilitation training are improved.

The invention discloses an intelligent ankle rehabilitation exercise management system based on multi-modal data fusion, which belongs to the field of medical assistance and comprises an acquisition processing module, a feature compensation module, an action extraction module, a structure modeling module, a judgment and detection module, an error correction guide module, a management scheduling module, a decision execution module and a report generation module. The motion state of each patient can be comprehensively reflected, fine-grained detection of abnormal actions is achieved, the patient is guided to correct the actions through multi-modal feedback, the secondary injury risk caused by wrong action accumulation is reduced, and the training effect and the controllability of rehabilitation progress are improved; the training plan can be continuously adjusted along with the rehabilitation progress of the patient, dynamic evolution is achieved, meanwhile, insufficient training or potential risks can be rapidly found, doctors and the patient are helped to clearly know the rehabilitation progress, the rehabilitation scheme is continuously optimized, and the training process is more efficient.

The embodiment of the invention provides an intelligent alarm method and device for tumble monitoring, and the method achieves the effective protection of the privacy of a user through the innovative design of a low-position visual angle collection mechanism, the setting of the installation height of a camera not exceeding the knee joint, and the starting of collection only during the abnormal motion. A detection mechanism based on a human skeleton model is constructed, and an accurate fall judgment strategy is established by combining multi-part vertical distance analysis and motion state parameters. An intelligent alarm mechanism is introduced, and quick response of an emergency contact person is realized through key video packaging and emergency channel establishment. According to the method, while the privacy of the user is protected, the defects of the traditional technology in the aspects of data acquisition, tumble detection, alarm response and the like are effectively overcome, and the practicability and reliability of tumble monitoring are remarkably improved.

The invention relates to a physical training scheme generation method and system for limb feature recognition, and relates to the field of physical training, and the method comprises the steps: collecting a skeleton topological portrait and a body shape proportion portrait of a target user; retrieving a joint number list and a joint standard time sequence coordinate list of the to-be-trained motion type meeting the portrait; when the target user is trained, the joint training time sequence coordinate list is extracted and compared with the joint standard time sequence coordinate list to obtain a joint time sequence deviation coordinate list; traversing the joint time sequence deviation coordinate list to count a time proportion list of which the coordinate deviation distance is greater than or equal to a deviation distance threshold value; the joint number set with the moment proportion larger than or equal to the moment proportion threshold value is extracted, the joint number set is input into the action matching library to execute abnormal action configuration to obtain an abnormal action set, a physical training scheme is configured, and the technical problems that due to the fact that physical training neglects individual body difference, monitoring precision is insufficient, and the individualization degree of the physical training scheme is poor are solved.

The invention discloses a dynamic risk early warning method and system based on student exercise data. The method comprises the steps that S1, original exercise data of student exercise is collected through wearable sensing equipment; s2, obtaining mapping motion data through an improved near-end depth Q network based on the original motion data and the normal motion data; s3, according to the mapping relation data, obtaining an action risk point through threshold comparison and considering duration time; and S4, performing matching detection on the obtained action risk points through an abnormal action chain type detection mechanism based on motion time-space association to obtain a risk prediction result, thereby avoiding misjudgment of a single tiny deviation as a risk, capturing a damage trend of a continuous abnormal action, scientifically fitting a motion damage occurrence rule, and accurately adapting to various campus motion scenes.

The invention provides an ultrasonic evaluation system and method for heart segmental motion abnormality, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting a continuous multi-frame echocardiogram of a patient; carrying out heart structure segmentation on each echocardiogram to obtain a plurality of heart segments, and respectively tracking and extracting motion features of each heart segment in a heart cycle to form a motion feature time sequence of each heart stage; and calculating a difference value between the motion feature time sequence and a baseline feature of a pre-constructed normal heart motion mode, and when the difference value exceeds a preset threshold value, outputting an evaluation result that the heart segment is in the abnormal motion mode. The method has the advantages that diagnosis efficiency is improved, workload of doctors is relieved, personal errors are reduced, accuracy, consistency and objectivity of diagnosis are improved, final evaluation results can provide effective reference for the doctors, diagnosis ability of primary doctors is improved, and dependence on experience of the doctors is reduced.

The invention relates to a motion detection method and system based on a continuous frame difference method in combination with a CNN, and belongs to the technical field of digital image processing. The method comprises the following steps: acquiring a video frame sequence in real time through a camera, and performing differential operation on video frames of the continuous video frame sequence to extract a potential motion area; inputting the extracted potential motion area into a convolutional neural network for deep feature extraction and recognition; and analyzing the extracted depth features, judging whether the motion belongs to an abnormal motion behavior or not, and if the motion is detected to be the abnormal motion behavior, automatically triggering an alarm by the system and recording related data. According to the method, the CNN processing range is reduced, the processing speed is remarkably improved, the changed part in the video can be quickly captured, the calculation is simple, and the method is suitable for real-time processing.

1. The name of the design product: distribution pole tower positioning monitoring terminal. 2. The use of the design product: for the remote real-time monitoring and management of the position reporting and tilt, abnormal movement and other states of the distribution poles in the distribution network. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

The invention discloses a reservoir operation safety supervision system and method, and relates to the technical field of reservoir safety supervision, the system comprises a biological information monitoring module, a fishing information monitoring module and an ecological system regulation and control module, foreign organisms in a reservoir are monitored so as to prevent the foreign organisms from damaging the ecological system of the reservoir, and the safety of the reservoir is ensured. Meanwhile, the types and the number of original organisms in the reservoir are monitored, so that targeted number regulation and control are performed, the balance and the stability of a reservoir ecological system are guaranteed, meanwhile, the fishing information of each fisherman is monitored in a fishing area, the abnormal movement and the abnormal behavior of each fisherman and the safety condition of used bait are analyzed, and the safety of the fisherman is improved. And alarm prompts can be given to all the fishermen and all the workers in time, so that the damage of illegal behaviors of the fishermen to the ecological system of the reservoir is avoided, and the safety of all the fishermen is also ensured.

A system is provided.SOLUTION: A system comprising: means for collecting performance data of an athlete; means for performing a biomechanical analysis to identify abnormal movements and loads; means for suggesting a personalized training program based on the analysis; and means for providing real-time advice and care to the athlete.SELECTED DRAWING: Figure 1

This invention relates to the field of polishing machines, specifically disclosing a concrete polishing machine with protective functions. It includes a base with a hand handle, a mounting plate on the base, a polishing assembly on the base, a drive assembly connected to the polishing assembly on the base, a push assembly on the base, a brake assembly between the push assembly and the polishing assembly, and an inflation assembly on the mounting plate. This invention provides primary buffering, secondary buffering, braking, and support functions. When the hand is released, it effectively counteracts the polishing machine's continuous spinning and irregular movements, providing effective protection against abnormal movements from multiple angles. This ensures the polishing machine quickly stops working after an malfunction, reducing collision damage and preventing malfunctions, minimizing threats to surrounding workers, improving work efficiency, and accelerating construction progress.

The invention belongs to the technical field of object state monitoring and alarming, and relates to an abnormal action monitoring method for a flame phagocytic experiment of protective equipment. The method comprises the following steps: extracting a protective equipment target contour surrounded by flame in each frame of image in a video stream, and segmenting the protective equipment target contour into a plurality of key body areas; constructing a flame pixel probability map for each key body area in a YUV color space, obtaining anti-interference local box dimensions of the key body areas, and forming fractal feature vectors by the anti-interference local box dimensions of all the key body areas; extracting a time sequence value of each dimensional component in the fractal feature vector in a preset time sequence window to obtain a global descriptor; and calculating a weighted geodesic distance between the global descriptor and a pre-constructed normal action state, and when the weighted geodesic distance is continuously greater than a preset attitude instability threshold in a set continuous frame, judging that an abnormal action is monitored and triggering an alarm. According to the invention, abnormal actions of the dummy or the subject in the experiment can be accurately and reliably monitored.

The present invention relates to the field of vehicle control, and particularly relates to a vehicle control method and a vehicle. When the first door and / or the second door rotates, the method provided by this application first obtains the passive rotation data of the first door and / or the second door. Secondly, the passive rotation data is matched and detected with the preset abnormal rotation data, which improves the accuracy of judgment. If the change in the door position is caused by normal environmental factors, its passive rotation data usually does not match the preset abnormal rotation data, and false judgment will not be triggered. Only when the door has a real abnormal movement and the passive rotation data matches the preset abnormal rotation data, the active rotation function will be disabled, so as to accurately distinguish normal situations from abnormal situations, avoid unnecessary false judgments caused by normal environmental interference, and ensure the stable operation of the door.

The present invention relates to a multi-stage abnormal human motion detection method based on a residual network, the steps of which are: cutting a video clip to be detected into video instances of equal length; using a target detection network model to obtain a human target bounding box, position and size appearing in each monitoring video instance; using an action recognition network model to calculate the category and confidence of the human motion in the bounding box in each monitoring video instance according to the human bounding box information; using an anomaly score learning model to give an anomaly score for each monitoring video instance and perform weighted fusion. The present invention can quickly obtain abnormal motion information in a monitoring video, and designs a target detection network model required for detecting a human bounding box, an action recognition network model for analyzing human motion, and an anomaly score learning model for predicting anomaly scores. The present invention realizes abnormal human motion detection in common monitoring video scenarios, the method is simple, the false alarm rate is low, and it has certain practical value.

The invention provides a method for judging an abnormal action in repeated mechanical actions. The method comprises the following steps: acquiring a first continuous frame image corresponding to at least one action period when equipment normally runs; converting the first continuous frame image into a first feature vector; training a feature prediction model through the plurality of first input variables and the plurality of first output variables; acquiring a second continuous frame image corresponding to the mechanical action of the equipment during operation; converting the second continuous frame image into a second feature vector; inputting a second input variable into the trained feature prediction model to obtain a prediction feature vector; calculating the similarity between the predicted feature vector and the actually measured feature vector; and after the plurality of similarities are obtained, calculating frequency domain characteristics of the plurality of similarities, judging whether the frequency domain characteristics meet preset conditions or not, and if not, judging that the equipment has an abnormal action. According to the method, whether the equipment has the abnormal action or not can be timely and accurately judged through the frequency domain characteristics of the mechanical action.

The invention provides a disposable detachable disinfection operating forceps training and checking system and method for the obstetrics and gynecology department. The method comprises the steps that force sense data of hands of a target student in the operation process of virtual operating forceps are collected; determining a plurality of abnormal actions of the hand of the target student in the process of operating the virtual operating forceps according to the force sense data; determining the fluctuation range of the force sense of the target student in the process of operating the virtual operating forceps through all the abnormal actions of the hand and the difference characteristics of the grip strength, and determining the cooperative deviation between the hand actions of the target student and the instrument conversion state of the virtual operating forceps according to the fluctuation range of the force sense; determining an operation energy efficiency value when the target student operates the virtual operating forceps according to the operation time and the cooperative deviation of each operation step when the target student operates the virtual operating forceps; and examining the operation of the target student on the virtual operating forceps based on the operation energy efficiency value. By adopting the scheme of the invention, the action collaboration can be fused to evaluate the action of operating the operating forceps by the target student.

This invention relates to the field of rehabilitation assessment technology, specifically to an upper limb rehabilitation assessment platform and method based on computer vision and the MYO armband. In this invention, an isolated forest is used to perform sparse scoring on the periodic feature vector composed of the mean angle, electromyographic delay, and trajectory fluctuation difference to identify abnormal segments, improving the accuracy of locating repetitive abnormal movements. The graph structure integrates the electromyographic activation difference and mutation trend between nodes, and uses a graph neural network to process the connection sequence of jumping nodes and inter-frame structural changes to obtain the trajectory of multi-node state changes, enhancing the structural expression ability of temporal mutations and electromyographic activation linkage. The number of angle reversals, peak acceleration, and electromyographic fluctuation amplitude are extracted segmentally and then concatenated to form a fusion feature table. The normalized feature values ​​are mapped to level labels using an interval matching method, establishing a traceable feature index path between the segmented behavioral structure and the level output, improving the stage-specific recognizability of the rehabilitation state and the granularity of the output expression.

The invention discloses a knee joint stress loading MR dynamic imaging support, and relates to the technical field of medical instruments.The knee joint stress loading MR dynamic imaging support comprises a base and a cushion table, supporting frames are arranged on the front side and the rear side of the top of the cushion table respectively, binding assemblies are arranged on the tops of the outer walls of the two supporting frames respectively, a lifting mechanism is arranged in a fixing box, and the lifting mechanism is connected with the cushion table. The middle side, close to the bottom of the fixing box, of the top of the base is fixedly connected with a positioning sleeve, the top of the transverse face of the L-shaped plate is provided with an inward and outward turning stress potential detection mechanism, and the lower side, close to the inward and outward turning stress potential detection mechanism, of the top of the base is fixedly connected with an arc-shaped guide plate. An electric telescopic rod in the ectropion and ectropion stress position detection mechanism carries out telescopic motion to drive the front end of a cushion table to slide along an arc-shaped guide plate, the state that a knee joint is pulled by external force is simulated, a supporting frame on the rear side is controlled by a lifting mechanism to move upwards, and relative displacement of a tibia and a femur under stress is detected; a method of identifying joint semi-dislocation or abnormal activity.

The invention provides an L-shaped probe SPECT device and a motion control method thereof. The L-shaped probe SPECT device comprises a first detector, a second detector and a structure base, the first detector and the second detector are L-shaped probe groups; the structure base can rotate; and the first detector and the second detector are jointly mounted on the structural base. According to the L-shaped probe SPECT device and the motion control method, the whole body scanning efficiency can be remarkably improved, and compared with a traditional double-probe system, the scanning time can be shortened by 50%; through synchronous rotation of the L-shaped probe and ellipse adaptive scanning, the photon collection efficiency is improved, and the motion artifact risk caused by abnormal movement of a patient after long-time scanning is reduced; meanwhile, the transverse moving range of the L-shaped probe is increased, the L-shaped probe can adapt to the contours of patients with different body types, and the problem that after a large probe approaches, a small probe cannot approach the human body due to interference is solved.

The application belongs to the technical field of wearable exoskeleton devices, and provides a wearable exoskeleton device suitable for rehabilitation training of infants with congenital cerebral palsy and an exoskeleton control method. The device comprises a wearable exoskeleton device main body, hip joint drives symmetrically arranged on the left and right sides, knee joint drives, ankle joint drives, a controller and multiple groups of sensors. The controller stores joint target trajectory data corresponding to multiple training modes, retrieves corresponding trajectory data based on the selected training mode and generates control instructions to actively drive the affected lower limbs to move. The multiple groups of sensors comprise electromyography sensors, inertial measurement units and pressure sensors. The controller identifies abnormal movement modes such as combined responses and extensor spasticity based on sensor signals and dynamically reduces the drive control gain to release abnormal stress in a compliant manner. The application realizes multi-mode active guidance training and asymmetric assistance for infants with congenital cerebral palsy, and has the ability of abnormal mode identification and safe compliant response.

The invention discloses a boom stretching safety control method and a related assembly thereof, and relates to the technical field of boom control. The number of abnormal actions of a boom is obtained; if the number of the abnormal actions reaches a preset number threshold value, setting an abnormal action alarm of the cantilever crane; if the abnormal action frequency is smaller than the preset frequency threshold value, the current input operation of the boom is obtained; starting output control corresponding to the current input operation of the arm support; determining an actual output action of the arm support after a preset period of output control corresponding to the current input operation of the arm support is started; and performing corresponding control on the arm support based on the input operation and the actual output action. It can be seen that the abnormal action alarm of the arm support is set when the number of the abnormal actions of the arm support reaches the preset number threshold value, the risk that the arm type aerial work platform is overturned is avoided, and the safety of construction work of the arm type aerial work platform is improved.

The invention discloses a lifting device and a transfer robot. The lifting device comprises a guide rail, a movable part, a sensor and a locking mechanism, the movable part is movably arranged on the guide rail, the sensor is used for detecting abnormal movement of the movable part and sending abnormal movement signals, the locking mechanism comprises a ratchet strip, ratchets and an electric cylinder, the electric cylinder is arranged on the movable part, the ratchet strip is arranged on the guide rail, the ratchets are arranged on the electric cylinder, and the ratchet strip is arranged on the guide rail. The electric cylinder is electrically connected with the sensor, and the electric cylinder is configured to drive the ratchet to be matched with the ratchet strip so as to lock the movable part when the sensor detects a motion abnormal signal. According to the lifting device, the locking mechanism is controlled through the sensor, when the movable part slides and falls due to various reasons (such as chain or belt breakage), the change of the movable part can be sensed, the sensor autonomously senses the sliding and falling action of the movable part, the locking mechanism is controlled in time through the electric cylinder to lock the movable part, and therefore the safety of people and equipment is protected; and automatic response is achieved in the whole process, and higher reliability is achieved.

The application discloses a muscle tension rating and abnormal muscle dynamic positioning system and method, and the muscle tension rating and abnormal muscle dynamic positioning system comprises an information acquisition module, an information processing module and a man-machine interaction module, the information acquisition module is used for collecting personal information and surface electromyogram signals of a detected person, the information processing module is used for muscle tension rating, determination of abnormal movement types and positioning of abnormal muscles, the man-machine interaction module is used for information management, evaluation result display, supervision and tracking and input of diagnosis information, and the like. The application can automatically identify muscle coupling abnormalities of a patient in various movement states, analyze internal muscle factors causing movement symptoms of the patient, evaluate a muscle tension disorder level of the patient and assist in positioning of abnormal muscles.

The system provided by the embodiment of the invention aims to improve the safety and convenience of the daily life of the old people. A system according to an embodiment includes a recognition unit, an analysis unit, a provision unit, a detection unit, and a notification unit. The recognition unit analyzes the movement of the pupil and recognizes an object at which the elderly is watching. The analysis unit analyzes the information recognized by the recognition unit and generates voice guidance. The provision unit provides the voice guidance generated by the analysis unit. The detection unit detects an abnormal motion or state of the elderly using an IoT sensor technology. The notification unit notifies a designated contact or medical institution of the abnormality detected by the detection unit.

The application relates to the field of guitar teaching, and specifically discloses a guitar intelligent teaching method, a system and a storage medium, the method comprising the following steps: acquiring input guitar playing data; performing feature extraction through a preset voice algorithm to acquire audio feature information; comparing the audio feature information with preset standard audio features to acquire a difference time interval set; extracting image sequence frames from video data based on the difference time interval set; performing feature extraction on the image sequence frames through a preset hand motion recognition algorithm to acquire a hand motion feature information set, comparing the hand motion feature information set with preset standard hand motion feature information to acquire an abnormal motion information set, and finally outputting corresponding data analysis information according to the current playing. The application detects and recognizes the guitar playing sound and hand motion from two dimensions of audio and video through the guitar playing data input by the user, finds out the problems in playing, and helps the user to adjust and improve in a targeted manner.