55 results about "Abnormal posture" patented technology

The invention relates to a method for detecting fatigue driving and driving in an abnormal posture under multiple postures. The method comprises the following steps: step 1, image acquisition: acquiring driver image information; step 2, image recognition: carrying out real-time face detection on the acquired images; step 3, face feature point calibration: calibrating feature points of eyes and a mouth of the detected face; step 4, judgment of the states of the eyes and the mouth: for the feature points, analyzing the closing states of the eyes and the mouth by adopting an eye state detection method on the basis of face width information; step 5, detecting of three-dimensional posture of the head: detecting the three-dimensional posture of the head through geometrical characteristics; step 6, judging of the states of fatigue driving and driving in the abnormal posture: judging whether the driver is in fatigue or not according to the closing states of the eyes and the mouth, and judging whether the driver is in the abnormal driving state or not according to the three-dimensional posture of the head.

The invention discloses an auxiliary training tool for comprehensive rehabilitation of cerebral palsy. The auxiliary training tool comprises a chassis, a lower limb exoskeleton device and a suspension device, wherein the chassis comprises universal casters with the locking function, a lower chassis, an upper chassis and an exoskeleton device inclination adjusting mechanism; the lower limb exoskeleton device is fixedly mounted on the chassis, and can realize rehabilitation training such as flexion / extension, adduction / abduction and pronation / supination of hip joints, flexion / extension of the knee joint, and dorsi-flexion / plantar-flexion, strephenopodia / strephexopodia and pronation ans supination of feet; the suspension device is fixedly mounted on the chassis, and plays a role in slinging, weight reduction and upper body posture correction of paitents. The auxiliary training tool can help patients with cerebral palsy to correct the abnormal posture of lower limbs, carry out the rehabilitation training on all joints of the lower limbs, promotes the development of all the joints of the lower limbs, and gradually establishes the normal standing and walking functions.

The invention is suitable for the field of screen display, and provides a screen display adjusting method and device. The method comprises the steps that angle information of the screen orientation is acquired; when the angle information of the screen orientation is in accordance with a preset angle range in which a screen faces downwards, the distance between a user face and the screen is detected; when the distance between the user face and the screen is smaller than a preset distance, the current environment light intensity is detected; the screen brightness is correspondingly adjusted according to the current environment light intensity. According to the screen display adjusting method and device, the damage of the screen to the user vision can be reduced when a user operates the screen in abnormal posture.

The invention discloses a driver abnormal posture monitoring method based on reinforcement learning. The method comprises the steps that after a driver driving video is acquired, driver face key points and skeleton key points which are far smaller than image pixel points in number are extracted through an OpenPose system, meanwhile, the face key points are subjected to spline interpolation for three times to construct face characteristics, and driver identity recognition is carried out. According to the key frame extraction method, based on a reinforcement learning mode, a corresponding rewardis obtained through an attitude detection model, and an action value function is updated according to each action of the reward and an attitude until a stable key frame extraction strategy suitable for each driver is obtained. Based on the driver dynamic behavior with time and space changes, a machine learning algorithm is used for training to obtain a posture detection model. And a safety earlywarning mechanism with a guiding property is established in combination with a protection motive theory. According to the invention, the real-time performance and accuracy of detection are improved, and the reliability of safety early warning is enhanced.

The invention relates to a pipe diameter adjusting mechanism. The pipe diameter adjusting mechanism comprises two radial adjusting devices which are basically the same in structure and are symmetrically arranged on the two sides of a detecting device. The radial adjusting devices both adopt screw rod nut pairs to drive three parallel-connecting connecting rod mechanisms to achieve diameter changing, wherein the spaces of the three parallel-connecting connecting rod mechanisms are spaced by 120 degrees. Screw rods in the two radial adjusting devices are the same in screw pitch and rotation direction and are oppositely mounted, and the purpose of synchronous rotation is achieved through mutual cooperation of bosses and grooves in the connecting end faces. The pipe diameter adjusting mechanism can adapt to the pipe diameter change within a certain range, and all supporting wheels can independently cross obstacles without affecting the stability of the whole structure, so that the adaptivecapacity of the operation environment of the pipe diameter adjusting mechanism is greatly enhanced; stable support can be provided for the rotating-type pipeline detecting device, and thus the problems that in the prior art, rotating of the detecting device is obstructed, and the abnormal posture of the detecting device is prone to occurring are effectively solved; the structure is simple and convenient to adjust; and the pipe diameter adjusting mechanism can also be used in other pipeline rotating operation robots needing diameter changing.

A pedestrian re-identification method based on a skeleton posture comprises the following steps: acquiring a pedestrian image, and constructing a target pedestrian template; collecting image frames and sequentially segmenting all pedestrian images under various postures in the image through skeleton detection; performing size standardization on all segmented pedestrian images and skeleton information according to the target pedestrian skeleton template; according to the pedestrian skeleton information, sequentially carrying out local feature region segmentation on all the pedestrian images, and carrying out tilt correction on the local feature regions of all the pedestrians to obtain a local feature image set consistent with the target pedestrian template attitude; and establishing a target pedestrian local feature fusion recognition model, and comprehensively calculating the similarity between the target pedestrian template and all pedestrians detected in real time to realize accuraterecognition of the target pedestrian. According to the invention, effective pedestrian identification under the influence of abnormal postures such as bowing, limb inclination and the like in a multi-person scene in a complex environment can be realized, and the pedestrian identification accuracy is improved.

Methods for diagnosis and treatment are provided for a patient having processing difficulties, integration problems, imbalances or abnormal postures. In one embodiment, a patient is identified having retinal signal problems creating at least one negative medical condition. Then, at least a portion of the patient's lacrimal drainage system is blocked or altered. Further, as a result of the diagnosis and treatment, the patient achieves improved retinal signal processing and at least one of the patient's negative medical conditions is improved. Other embodiments include using in combination the Z-Bell Test and / or light blocking or altering devices for diagnosis and treatment.

A traction method and apparatus for treating irregularities in the lateral curvature and global posture of the thoraco-lumbar, lumbar, lumbo-pelvic and / or pelvic region of the spine involving the application of an intermittent transverse traction load to a region of the patient's body. The patient is placed in a supine or semi-supine position, and a traction sling is positioned about a region of the body of the patient, where the region of the body can be the thoracic region, lumbar region and pelvic region of the body. A transverse traction load having an intermittently varying magnitude is exerted on the sling to induce an extension posture in at least a portion of the spine of the patient and / or a flexion posture of the pelvic portion of the spine of the patient. The flexion and extension postures serve to at least partially restore normal lateral curvature of the thoraco-lumbo-pelvic region of the spine, thereby reducing lower back pain, correcting abnormal posture, and easing other symptoms that are related to thoraco-lumbo-pelvic spine curvature irregularities.

The invention discloses a human abnormal posture detection system and method based on three-dimensional vision, and belongs to the technical field of computer vision. The technical problems to be solved by the invention is how to enlarge the fall detection range, effectively avoiding the blocking of the blocking object in the detection process, at that same time, how to detect the fall accurately.The adopted technical scheme is characterized in that a Kinect somatosensor is mounted on a mobile platform using a Mecanum wheel, so that the mobile platform equipped with the Kinect somatosensor can move with the human body; the kinect somatosensor is used to obtain the distance between the mobile platform and the human skeleton point, and the obtained distance change is converted into an electric signal to be sent to the mobile platform, so that the mobile platform can move with the moving direction and the moving speed of the human body, and the set distance between the mobile platform and the human body is ensured; at the same time, according to the image captured by the camera on the Kinect somatosensor, the human skeleton point information is obtained to establish the fall model and detect the fall of the human body.

The invention belongs to the technical field of image processing, and specifically provides a posture recognition method and system based on an infrared sensor. According to the scheme, an indoor scene is imaged based on the infrared sensor, human body target detection and posture recognition are carried out through the rear-end information processing module. Once an abnormal posture is detected,for example, falling down, an abnormal alarm signal is immediately output through a network, and a recognized abnormal image picture is output to the alarm terminal. The infrared sensor captures humanbody temperature information, so that the problem of human body privacy leakage does not exist, and the system can work all day long. On one hand, a room can be monitored by adopting a single wide-angle infrared sensor, the imaging distance is increased, and the practicability is enhanced. On the other hand, the target detection network and the classification network are adopted, the human body only needs to be within the image range, any position can be automatically detected, then posture classification is conducted on the human body, and practicability is greatly improved.

The invention discloses a head posture detecting method for fatigue driving detection. An image acquiring portion, a face detecting portion, a head posture fatigue feature extracting portion and a fatigue alarm portion are included. A vehicle-mounted camera is mounted in front of a main driving position, and the camera is connected to a vehicle-mounted computer through a USB port and is debugged so that the driver's driving position can be correctly captured. A trained face detection model and a running program thereof are copied to a specified position of the vehicle-mounted computer and tested. A trained head posture estimation model and a running program thereof are copied to a designated position of the vehicle-mounted computer, and the head posture estimation model is debugged in combination with the debugged face detection model. Fatigue features of the head posture are extracted, and whether fatigue is determined according to the ratio of nodding frequency and the abnormal posture. The method can stably detect the head posture, and has certain feasibility and applicability.

Methods for diagnosis and treatment are provided for a patient having processing difficulties, integration problems, imbalances or abnormal postures. In one embodiment, a patient is identified having retinal signal problems creating at least one negative medical condition. Then, at least a portion of the patient's lacrimal drainage system is blocked or altered. Further, as a result of the diagnosis and treatment, the patient achieves improved retinal signal processing and at least one of the patient's negative medical conditions is improved. Other embodiments include using in combination the Z-Bell Test and / or light blocking or altering devices for diagnosis and treatment.

The invention relates to an abnormal posture detection method and device, relates to the field of intelligent posture detection and comprises the steps that a target image of a target user is acquired, position coordinates of key points of the target user in the target image are acquired, the key points comprise preset human body parts of the target user, and whether the posture of the target useris an abnormal posture or not is determined according to the position coordinates. The method can accurately determine whether the posture of the user is the abnormal posture or not when the user hasno obvious displacement change, is high in detection efficiency, does not need to additionally carry a detection device by the user, is convenient to use, and improves the user experience.

The invention relates to an improved passenger sensing system which comprises passenger seats and a sensor part, wherein the passenger seats are arranged on a vehicle; and the sensor part senses a plurality of load values according to passenger positions on the passenger seats; the passenger weight is calculated out on the basis of the load value and then compared with a first setting value, and an air bag is expanded with high pressure when the passenger weight is more than the first setting value; a difference value is calculated out by mutually comparing the load values when the passenger weight is less than the first setting value, the difference value is compared with a second setting value, and the air bag is expanded with low pressure when the difference value is less than the second setting value; and whether a passenger sits at the front part of the passenger seat or not is judged when the difference value is more than or equal to the second setting value, the air bag is expanded with low pressure if the passenger sits at the front part of the passenger seat, and a control part of the air bag is expanded with high pressure if the passenger does not sit at the front part of the passenger seat. The improved passenger sensing system ensures that an adult can not be considered as a child passenger by mistake when sitting on the passenger seat in an abnormal posture, and can identify the passengers correctly and effectively protect the passengers on the seats.

The invention discloses a trunk-supported walking aid lower limb power orthosis. The orthosis comprises a shoulder supporting assembly, a walking aid assembly a waist supporting assembly and a lower limb power orthosis assembly, and the lower limb power orthosis assembly is used for supporting the trunk of a human body to be upright and adjusting abnormal postures in the walking process; according to the shoulder supporting assembly, the width symmetric adjustment design of the shoulder and back portion is achieved through a connecting rod structure; and according to the walking aid assembly and the rear-mounted walking aid, support can be provided when a patient wears the device to move, other supporting tools such as a walking stick are not needed, the trunk can be made to be more upright, and constraint of other supporting tools is reduced. According to the invention, three-degree-of-freedom adjustment of the hip joint, one-degree-of-freedom adjustment of the knee joint and passive assistance of the ankle joint in the walking process are realized; according to the invention, change angles of joints of lower limbs of normal gaits of a human body can be adapted; the size can be adjusted to adapt to people with different figures, and the comfort level of the equipment is improved; and the modularized design can be detached and replaced at any time, different requirements are met, and cost is saved.

The invention provides a posture detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a figure image collected by a photographing device; carrying out posture detection based on the obtained figure image, and determining posture detection information of a target figure included in the figure image; and in response to the posture detection information indicating that the posture of the target person belongs to an abnormal posture, generating posture correction prompt information. The posture correction prompt information for posture correction prompt can be automatically generated under the condition that the posture of the student is abnormal, so that the student is helped to develop a good learning habit.

The invention discloses a tumble detection system, which comprises a wearing device used for wearing of a user, and a processor connected with the wearing device. The wearing device is provided with an inertia sensor used for detecting the exercise data of the user. The processor is connected with the inertia sensor of the wearing type device, and when the tumble state of the user occurs according to the user exercise data, the inertia sensor is used to acquire the exercise data of the user during the time spent in standing up again, and according to comparison between the normal posture condition in a database and / or the abnormal posture condition and the exercise data of the user, the severity of the damage on the user is determined.

The invention discloses a novel spherical robot and a control method thereof, and relates to the technical field of the spherical robot. A driving shaft and a rotating shaft are perpendicular to each other and both are located in the horizontal plane. Actions such as going forward and back, no-radius any angle turning and any abnormal posture adjusting can be realized by virtue of a power source driving motor and a steering motor which are used for controlling the two shafts. The novel spherical robot has the characteristics of omnidirectional rolling and no movement dead angle, and also has the advantage of simple and reliable control; in addition, the unique design of the structure of the novel spherical robot is also capable of guaranteeing excellent stability of a carrying platform; devices such as a camera and a sensor can be built in the novel spherical robot so as to realize movement in an unfamiliar environment, planet exploration and the like.

A fall detection system includes a wearing apparatus for wearing by a user and a processor connected to the wearing apparatus. The wearing apparatus is set with an inertial sensor for detecting user motion data. The processor is connected with the inertial sensor of the wearing apparatus. When the user's fall state is recognized, further obtaining the motion data of the user at the time of the stand by the inertial sensor and comparing the motion data according to a normal posture condition and / or an abnormal posture condition in a database to determine damage severity of the user.

Methods for diagnosis and treatment are provided for a patient having processing difficulties, integration problems, imbalances or abnormal postures. In one embodiment, a patient is identified having retinal signal problems creating at least one negative medical condition. Then the amount of a natural hormone in the eye is increased such that, as a result of the diagnosis and treatment, the patient achieves improved retinal signal processing and at least one of the patient's negative medical conditions is improved. Other embodiments include using the treatment in combination with the Z-Bell Test and / or light blocking or altering devices for diagnosis and treatment.

The invention relates to a method for climbing stairs of a routing inspection robot. The method comprises the following steps that when the posture information is abnormal, a telescopic device is driven to extend a climbing hook and the climbing hook is fixed to the top of the stairs; and the telescopic device is driven to drive routing inspection robot to climb along the stairs. The posture stateof the routing inspection robot can be acquired quickly through a detection device so that the abnormal posture of the routing inspection robot can be found in time; and according to the abnormal posture of the routing inspection robot, the telescopic device is driven so that the climbing hook can be fixed on the stairs stably, and thus the stairs with a large slope are convenient for the routinginspection robot to climb.

The invention relates to the technical field of artificial intelligence, and discloses a three-dimensional human body posture estimation method and device, equipment and a storage medium, and the method comprises: obtaining a to-be-estimated image; inputting the to-be-estimated image into a target three-dimensional human body posture estimation model for three-dimensional human body posture estimation, wherein the target three-dimensional human body posture estimation model is a model obtained by adversarial training based on a to-be-trained generative network, a to-be-trained discrimination network and a human body posture strong geometric constraint model; and obtaining target three-dimensional human body posture data corresponding to the to-be-estimated image according to the three-dimensional human body posture data output by the target three-dimensional human body posture estimation model. The problem of few training samples is effectively solved through adversarial training, the prediction performance of the model obtained through training on interference shielding is improved by introducing the human body posture strong geometric constraint model, and the probability of occurrence of abnormal posture prediction is reduced.

The embodiment of the invention relates to the technical field of image processing and discloses a video generation method, a video generation device, an electronic device and a storage medium. The method comprises the steps: obtaining an input image and a target action type, generating a first posture sequence according to the input image and the target action type and then generating dynamic video according to the input image and the first posture sequence. Thus, the video generation method disclosed by the embodiment of the invention can avoid noise and abnormal postures and can prevent generated video from including mutant video frames, so that video generation controllability is improved.

The invention discloses an intelligent control method and a control system for cosmetic boxes, and a storage medium. The intelligent control method for the cosmetic boxes comprises the following stepsof: acquiring the monitoring video of a video monitoring camera in real time; carrying out face recognition and posture information recognition on the acquired monitoring video, constructing a face recognition model and an abnormal posture recognition model based on a neural network algorithm, and establishing the corresponding relation between the face recognition model and the posture recognition model; judging whether the posture of a user is abnormal or not; sending reminder notification to the user by an artificial intelligence unit talking to the user when the posture of the user is judged to be abnormal; and storing the face data and the posture information of the user. According to the intelligent control method and the control system for cosmetic boxes, and the storage medium, the abnormal behaviors and the environment unfriendly behaviors to the cosmetic boxes in the sharing mode are monitored, and the problem of monitoring the user to the cosmetic boxes in the sharing modeis solved.

The invention discloses a human body abnormal posture detection method. The method comprises the following steps: S11, acquiring a target image; S12, detecting the regional position of a human body in the target image and the posture of the human body; S13, judging whether the human body posture is an abnormal posture or not; S15, if the abnormal posture exists, judging whether the abnormal posture is the abnormal posture of the same human body appearing for the first time or not; if the abnormal posture is the abnormal posture of the same human body appearing for the first time, entering the step S16, otherwise, entering the step S17; S16, storing an abnormal posture category corresponding to the abnormal posture of the human body, an abnormal frequency, a regional position of the human body in the target image and storage time as abnormal target data; S17, finding out the area with the maximum intersection ratio with the area position corresponding to the abnormal posture, and comparing the similarity of the areas; S18, judging whether the similarity exceeds a first preset threshold value or not; and S19, if the first preset threshold value is exceeded, generating abnormal posture early warning.

The invention discloses an examinee posture segmentation and smoothing method, and the method mainly comprises the steps: obtaining posture data, obtaining data of all human body postures in a frame of image, and carrying out initialization; determining the centroid of each posture, and segmenting postures belonging to the same person in the video into a sequence by combining the position information of the human body, so as to achieve the purpose of human body tracking; and detecting the segmented attitude, the change characteristics of the attitude and the rationality of the attitude, eliminating the abnormal attitude and the deviated attitude in the attitude, and smoothing the corrected attitude sequence by adopting a sliding window smoothing method. According to the method, the human body posture in the video can be tracked, the abnormal posture data is screened out, the posture precision in the video is improved, and the posture jitter between the front frame and the rear frame isreduced, so that the influence of abnormal posture change on subsequent use is avoided.

The invention relates to a traditional Chinese medicine lotion for treating abnormal postures of children with spastic cerebral palsy. The traditional Chinese medicine lotion can effectively solve the problem of treatment medication on abnormal postures of the children with spastic cerebral palsy. The solved technical scheme is as follows: the traditional Chinese medicine lotion is prepared from the following crude drugs of: 20-50g of lycopodium clavatum, 20-50g of garden balsam stem, 15-35g of pawpaw, 15-30g of folium artemisiae argyi, 15-30g of flowers carthami, 20-50g of lignum millettiae, 10-30g of radices sileris, 20-50g of prepared rehmannia root, 10-30g of hematoxylon, and 20-50g of ramulus mori, wherein the drugs are soaked for 30 minutes by 6-12fold water at room temperature, decocted for 30-45 minutes and filtered to prepare the lotion. The traditional Chinese medicine lotion is scientific and reasonable in composition, convenient to use, low in cost, good in effect, convenient and safe to use and low in cost, and can be easily accepted by a sufferer, and has practical importance on improvement of the self-care capability of the sufferer, improvement of living quality, improvement of the quality of the population, and reduction of the social and family burdens.

The invention discloses an automatic-ladder-matched safety device. The automatic-ladder-matched safety device specifically comprises photoelectric sensor arrays respectively used for detecting body postures of passengers (whether an abnormal posture such as falling appears or not), pressure sensor arrays for detecting gap pressure between steps and steps and gap pressure between steps and skirt guard panels, magnetic sensors for detecting the running speed and the direction of an elevator, first judging controllers for judgment according to pressure sensor array outputting, second judging controllers according to photoelectric sensor array outputting and magnetic sensor outputting, warning device arrays for sounding warnings to the passengers and scramming devices for elevator scramming. The automatic-ladder-matched safety device has the functions of detecting insecurity and accident behaviors of the passengers and the abnormal running conditions of the elevator, is thoughtful, and can effectively prevent security accidents from occurring.

The invention discloses an abnormal posture correcting, evaluating and training system. The system comprises an abnormal posture correcting device, a signal conditioning module, an analog-to-digital conversion module, a signal capturing device, a data transmission module, a data processing module, a terminal and a power supply device, wherein the abnormal posture correcting device comprises a sensor. When the wearable abnormal posture correcting device is in a motion state, sensing signals of the sensor are conditioned through the signal conditioning module to obtain an analog quantity, then the analog-to-digital conversion module converts the analog quantity into a digital quantity, finally, the digital quantity is analyzed and evaluated through the terminal, and analysis and evaluation results are displayed. In the motion state, the signal capturing device captures a motion sensing signal of the sensor, and after the motion sensing signal is sent to the data processing module throughthe data transmission module for processing, a three-dimensional model is established through the terminal and used for comparison before and after use. According to the system, quantitative indexesare increased while abnormal postures are corrected, and a safe and effective exercise treatment tool for patients suffering from brain injury and limb abnormality.

The invention provides an alarm method, a scooter, a storage medium and an electronic device. The method comprises the steps that the posture of the scooter is detected according to the movement stateof the scooter and movement parameters of the scooter; alarming is carried out on the scooter under the condition that it is detected that the posture of the scooter is an abnormal posture. By meansof the method, the problem is solved that in the prior art, situations which possibly occur to the scooter cannot be predicted and accordingly certain safety hidden hazards exist in the aspects of application and maintenance of the scooter.