117 results about "Human body structure" patented technology

The invention provides a rescue robot autonomous identification and positioning system and method based on industrial vision, and relates to the technical field of robots, and the system comprises a cooperative scanning module which is used for constructing a robot coordinate map; the data fusion module is used for establishing a data fusion frame with a unified time sequence to obtain multi-modal image data; the human body recognition module is used for mapping the thermal imaging image to a visible light image and executing human body local recognition to obtain human body structure features; the target marking module is used for determining a rescue target and marking the rescue target in the robot coordinate map; and the path planning module is used for performing path planning, constructing an updated path and performing rescue through the rescue robot. According to the method and the device, the technical problem of relatively low rescue efficiency caused by difficulty in accurately positioning the rescue target due to shielding and interference in a complex environment in the prior art can be solved, and the precision of positioning the rescue target is improved by combining industrial vision, so that the rescue efficiency is improved.

The invention relates to the technical field of computer vision and neural network processing, in particular to a human body posture recognition method based on a multi-modal neural network framework. Comprising the following steps: collecting event data, and representing the event data by using a five-channel event representation method; dividing the preprocessed five-channel event data into three channels, namely a space channel, an event characteristic channel and a point cloud channel; after channel splitting is completed, feature extraction is carried out respectively, and features of different modals are optimized and integrated through dynamic feature enhancement and cross-modal fusion; a skeleton constraint module and a space-time joint processing module are adopted to enhance understanding of the network on a human body structure and modeling of a space-time dependency relationship; a prediction head module is adopted to output final human body key point coordinates; and carrying out training optimization by adopting a multi-task learning framework to obtain a posture recognition result. The method has the advantages that the sparse asynchronous characteristic of event data is effectively processed, the blind spot problem of static limbs is solved, low power consumption is kept, and high-precision human body posture estimation is achieved.

The invention relates to an electromagnetic radiation dose calculation method and system based on polynomial fitting, and the method comprises the steps: carrying out the variable human body modeling based on a CT image, and obtaining a variable human body model; selecting physique parameters, and sampling according to the physique parameters to obtain a physique parameter sample; establishing a basic human body model based on the variable human body model and the physical parameter sample, and performing electromagnetic exposure simulation on the basic human body model to obtain an electromagnetic radiation sample; fitting, evaluating and screening a plurality of multi-order PCE models according to the electromagnetic radiation sample to obtain an electromagnetic radiation metering model; and performing human body electromagnetic radiation dose evaluation based on the electromagnetic radiation metering model to obtain a metering result. According to the invention, the individual difference of the human body structure can be reflected more accurately, so that the accuracy of electromagnetic radiation evaluation is improved.

The invention belongs to the field of computer vision, and relates to a time-space topology modeling-oriented real-time three-dimensional human body posture estimation method and device. The method comprises the following steps: carrying out label-free pre-training by utilizing a comparative learning normal form, and extracting prior information of a human body topological structure; a lightweight space-time human body topology extraction network is used as an encoder, human body topology structure prior information is utilized, and a double-flow structure is adopted to extract a human body kinematics topological relation in a time domain and a human body geometry topological relation in a space domain in parallel; and carrying out accumulation operation on the human body kinematics topological relation in the time domain and the human body geometry topological relation in the space domain, obtaining the position of the human body three-dimensional posture through linear regression, and outputting corresponding three-dimensional skeleton coordinates. According to the method, the human body structure and kinematics topological characteristics can be accurately captured, and efficient and accurate three-dimensional human body posture estimation can be realized.

Methods and systems for determining changes in muscle activity patterns based on the optical response of subcutaneous human body structures are presented, along with methods and systems for controlling assistive robotic systems based on the measured muscle activity patterns. Elements of a muscle activity measurement system are mechanically coupled to a wearable structure that fits closely to a portion of the body of a human user. The muscle activity measurement system includes multiple emitters and detectors at different spacing along the skin surface. In some embodiments, the illumination intensity of each measurement channel, the programmable gain of each measurement channel, or both, are calibrated to maximize measurement sensitivity. In some embodiments, optical measurement data is employed to more accurately locate the muscle activity measurement system with respect to the human body. In some embodiments, a muscle activity measurement system tracks changes in muscle structure over time.

The invention provides an intelligent motion generation method and system of a human body, and electronic equipment, and the method comprises the steps: obtaining motion capture data based on video data, obtaining target motion information and an initial human body skeleton key point sequence based on the motion capture data, obtaining source skeleton motion data based on a human body structure prior key point sequence of the human body skeleton, and obtaining the target motion information and the initial human body skeleton key point sequence; obtaining a virtual human image based on the target motion information, obtaining a virtual human skeleton structure based on the virtual human image, obtaining redirected skeleton data, obtaining an optical flow graph and a mask graph based on the virtual human image and the redirected skeleton data, and obtaining an initial virtual human motion image sequence based on the optical flow graph and the mask graph; and acquiring a virtual human motion image sequence based on the initial virtual human motion image sequence, and acquiring motion output of the living agent based on the virtual human motion image sequence. Therefore, coherent and natural action generation is realized.

The invention discloses a cross-modal retrieval method fusing human body attribute features and vector embedding. The method comprises the following steps: regularly extracting pictures from a monitoring video stream, and coding and caching; extracting a face feature vector, a human body feature vector, a text vector and a human body structured attribute feature of the picture; and associating the face feature vector with the human body feature vector to realize face fusion, and storing the face fusion in a vector library. And associating the human body structured attribute features with the vector data, and storing the human body structured attribute features and the vector data in a search engine. According to the method, feature extraction and vectorization are carried out on an input image or character, vector data are matched with data in a vector library, then filtering is carried out through a search engine according to human body structured attribute features, and a search result is obtained. According to the method, the human face and human body fusion technology is introduced, the human body attribute features are structured, and the deep learning model is combined, so that the personnel information in the video can be quickly and accurately retrieved under multiple cameras, and the video in the corresponding time period can be efficiently and stably generated for playback.

The invention discloses an adjustable 3D printing temporal bone surgery training model and a training method thereof.The 3D printing technology is adopted, printing materials similar to all structures and textures of a human body are selected to manufacture a simulated temporal bone model, a clinical temporal bone surgery conventional approach is integrated in the model, and a clinical operation scene is simulated; preoperative training and planning of ear neurosurgery are assisted; meanwhile, digital technologies such as CT scanning and 3D scanning are utilized, and personalized customization of the simulated temporal bone model for special cases is achieved through the 3D printing technology.

The invention discloses a human motion classification method, device and equipment based on RFID and a medium, and the method comprises the steps: collecting human skeleton phase data, and carrying out the preprocessing of the data, and forming RFID skeleton activity data; constructing a human body action diagram, and connecting hand joint nodes and foot joint nodes in the human body action diagram to form an initial human body skeleton structure diagram; and performing graph convolution processing on the initial human body skeleton structure graph to generate a human body action prediction classification label. By adding an improved graph structure formed by connection between the hands and the feet in the human body structure graph, the topological distance between the nodes is reduced, the nodes can be aggregated to more neighbor node information, the aggregation capability of joint features is improved, and the joint feature aggregation efficiency is improved. And during action recognition of cooperation of both hands and hands and feet, richer action feature information can be captured through aggregation features among non-physically connected joints, and the recognition precision of the model on complex postures is improved.

The embodiment of the invention provides a low-illumination human body posture estimation method and device based on knowledge distillation. The low-illumination human body posture estimation method based on knowledge distillation comprises the following steps: acquiring a low-illumination picture of a human body posture to be determined; the low-illumination picture is imported into a trained human body posture evaluation model, a target posture estimation result is obtained, and the human body posture evaluation model is obtained through knowledge distillation training based on a double-flow network architecture. According to the embodiment of the invention, through a double-flow network architecture and knowledge distillation training, more effective brightness optimization can be carried out on the low-light image. High-precision extraction of human body structure features can be realized, and the problems of low precision and poor robustness of human body posture estimation under poor illumination conditions such as low illumination or night and the like are solved.

The present invention relates to a method, device, electronic device and storage medium for reconstructing an occluded gait. The method for reconstructing an occluded gait of the present invention comprises: obtaining an occluded gait sequence to be reconstructed; inputting the occluded gait sequence into a generator of a trained generative adversarial network to obtain a repaired gait sequence; wherein the training method of the generative adversarial network comprises the following steps: selecting part of the samples from the unoccluded gait data set as the original sample set, and adding occlusion features to the samples in the original sample set as a training set; putting the samples of the training set into the generator to obtain a repaired sample set; putting the repaired sample set and the original sample set into the discriminator, and using the loss function to calculate the difference between the repaired sample set and the original sample set; adversarially training the generator and the discriminator until the loss function converges, and saving the trained generative adversarial network. The method for reconstructing an occluded gait of the present invention reconstructs a more complete human body structure.

The utility model relates to technical field of sports socks, concretely is a kind of wide-toe moisture-absorbing quick-drying split-toe sports socks, including sock main body, the sock main body includes toe, by thumb toe part and split-toe part, the thumb toe part and split-toe part are connected with widened part, the sock body includes the sock face of wrapping instep and the sock bottom of wrapping foot bottom, the sock bottom of the sock body is equipped with space bin, and the space bin is located metatarsal bone and phalanx connecting portion of sole of foot.This kind of wide-toe moisture-absorbing quick-drying split-toe sports socks, by adding widened part between thumb toe part and split-toe part, increase the distance between thumb toe and two toes, reduce the pressure and friction to thumb toe inner side when wearing, in addition, space bin is additionally arranged in metatarsal bone and phalanx connecting portion of sole of foot, so that the sports socks can better fit foot, and the wrapping property is stronger, so that the sports socks are more in line with human body structure, and wearing is more comfortable.

The embodiment of the invention relates to the technical field of humanoid robots, and discloses a humanoid robot body design method and device, a medium and a product. For the appearance of the humanoid robot, the internal structure layout is deduced reversely with the human body appearance curved surface as the boundary condition, and skin-skeleton-muscle layered coupling design is adopted; for humanoid robot skeletons, heterogeneous mixed joints are configured according to 23 main joint types of a human body, and a spine adopts a four-degree-of-freedom series link mechanism design; the humanoid robot is driven, driving assemblies are configured in a graded mode based on the gravity center migration model, and shape memory alloy bionic tendons are introduced. Through an appearance constraint design method, the appearance of the robot is highly matched with the human body structure, the appearance is more vivid and natural, and the affinity of man-machine interaction is improved. The technical problem that the humanoid robot is stiff and stiff can be at least solved.

The invention is suitable for the technical field of pedestrian re-identification, and provides a pedestrian re-identification method and related equipment. The method comprises the following steps: acquiring a plurality of to-be-cut areas of the same human body in a target image, wherein each to-be-cut area comprises at least two target key points; respectively carrying out rotary cutting on the plurality of to-be-cut areas according to the human body posture and the image view angle by utilizing at least two target key points respectively included in the plurality of to-be-cut areas to obtain a cut image of the human body structure; and comparing the cut image of the human body structure with a reference structure image in a target database to obtain a pedestrian re-identification result. According to the method, the recognition precision of pedestrian re-recognition can be improved.

The application discloses a method for registering a skeleton model to a world coordinate system of a body surface point cloud, which comprises the following steps: S1. scanning a human structure by using a CT scanning device to obtain a CT skeleton three-dimensional model and a CT human structure body surface point cloud; S2. scanning the human structure by using a structured light 3D camera to obtain a 3D structured light body surface point cloud of a 3D structured light human structure; S3. selecting an antegrade registration or a retrograde registration mode to register the body surface and the skeleton of the human structure according to the scanned human structure; the method for registering the skeleton model to the world coordinate system of the body surface point cloud: registering the torso and the pelvic bone of the human body by using the antegrade registration mode; registering the limbs of the human body by using the retrograde registration mode; constructing a virtual puncture channel, so that even if the spatial position of a puncture instrument is adjusted at will in a process of orthopedic puncture, the relative position relationship between the puncture instrument and the human structure can be acquired in real time, and the navigation puncture based on dynamic three-dimensional visualization is realized.

The application provides a visual human-imitating ankle intelligent teaching aid and a preparation method thereof, and relates to the technical field of teaching aids.The visual human-imitating ankle intelligent teaching aid comprises a human-imitating ankle body, human-imitating muscles, a position sensor, a force sensor and a controller.The human-imitating ankle body comprises human-imitating phalanges, human-imitating metatarsal bones, human-imitating cuboid bones, human-imitating wedge bones, human-imitating navicular bones, human-imitating fibula bones, human-imitating anterior tibiofibular ligaments, human-imitating calcaneus bones, human-imitating talus bones, human-imitating calcaneofibular ligaments and human-imitating tibia bones arranged according to human body structure.The position sensor and the force sensor are connected with the controller.The human-imitating muscles wrap the human-imitating ankle body, the position sensor and the force sensor.The acceleration sensor and the angular velocity sensor are arranged on the human-imitating talus near the human-imitating anterior tibiofibular ligaments.The force sensor is arranged on the human-imitating talus near the human-imitating calcaneus bone.The application can quantize the data of bone-setting methods of experts and trainers, so that the deviation of the bone-setting methods of the trainers can be obtained, and the trainers can learn better.

The invention discloses a gait recognition method based on local appearance and micro-motion perception, and belongs to the technical field of computer vision and biological feature recognition. The method comprises the following steps: firstly, preprocessing and aligning an input gait contour sequence; extracting spatial features by using a convolutional neural network, and performing non-uniform part division according to human body structure priori knowledge so as to focus on leg equal-height discriminant regions; further, an independent micro-motion sensing module is designed for each component, and unique short-distance space-time dynamic characteristics of each component are captured through a double-branch time sequence attention mechanism; and finally fusing the enhanced features of all the components to complete identity recognition. By emphasizing the local appearance difference and the micro-motion mode, the robustness and stability of the gait recognition model in a real complex scene are remarkably improved, and the method has important security monitoring and identity authentication application value.

The invention discloses a pedestrian re-identification method and system based on attitude guidance and triple cooperative loss, and belongs to the technical field of computer vision. According to the method, in an attitude guided depth supervision framework, a triple collaborative loss function fusing temperature scaling cross entropy loss, triple loss and circular ring loss is designed, and feature learning is collaboratively optimized through weight configuration; in the training stage, the knowledge distillation of the posture encoder is combined to enhance the understanding of the model on the human body structure, and in the reasoning stage, only the optimized human body encoder is used for recognition. Experiments show that the mAP of the method is improved by 3.1% on a VC-Clothes clothes change data set, the robustness of shielding and clothes change scenes is remarkably improved, and additional calculation overhead is not needed in the reasoning stage.

The invention discloses a human body posture estimation method based on multi-scale data self-adaption, and belongs to the technical field of computer vision. The invention aims to solve the problem that the precision is low because the same neural network model is adopted to carry out attitude estimation on original pedestrian image data with different scales. Comprising the following steps: preprocessing an original image to obtain an input image with a set scale, and obtaining a scale scaling factor; a low-level feature map is obtained from an input image through a first convolution module, and a scale normalized low-level feature map is obtained through a first scale correlation domain bridging module; a human body structure feature map is obtained through a second convolution module, and a scale normalization human body structure feature map is obtained through a second scale correlation domain bridging module; a human body joint point feature map is obtained through a third convolution module, and a thermodynamic diagram predicted value of human body joint points is obtained from the human body joint point feature map through a network head; and realizing human body posture estimation based on the thermodynamic diagram predicted value. The method is used for human body posture estimation.

The present invention relates to the technical field of assessing the degree of response of human organs under vibration impact, and specifically discloses a method for rapidly assessing injuries caused by vibration impact to typical organs of a reclining human body, comprising the following steps: first, obtaining dynamic response data of typical organs of a reclining real person under vibration impact; then, establishing a reclining human body-structure integrated multi-rigid body theoretical model and optimizing model parameters; determining the force damage threshold of the reclining human body under vibration impact; then, verifying the reclining human body-structure integrated multi-rigid body theoretical model in combination with impact experimental data of a reclining real person; calculating the acceleration response data of various parts of the reclining human body, comparing it with the force damage thresholds of different human organs given above, and then rapidly assessing the degree of response of the reclining human body. The present invention adopts the above-mentioned method for rapidly assessing injuries caused by vibration impact to typical organs of a reclining human body, providing core technical support for rapid assessment of injuries to reclining human bodies and anti-impact protection design.

The utility model discloses a one-piece shapewear which comprises a chest binding area, a waist binding area, a crotch area and a back area, the chest binding area, the waist binding area, the crotch area and the back area are connected together according to the structural characteristics of a human body, switch belts are arranged on the two outer sides of the back area, secondary tightening threads are arranged on the inner sides of the switch belts, and the secondary tightening threads are connected with the switch belts. A main compact line is arranged at the central position of the back area; the back area is further provided with a hip lifting area, and a hip lifting line is arranged on the outer edge of the hip lifting area. The body-shaping underwear is good in body-shaping effect, high in air permeability, convenient to put on and take off and capable of better meeting the requirements of consumers.

The invention discloses a movable doll module, a production and manufacturing method of the movable doll module and a striking technique training system. The movable doll module comprises a framework, at least one movable joint structure, at least one counter weight and a coating material. The framework is provided with a head part and a trunk part. The at least one movable joint structure is correspondingly arranged on the head part and the trunk part according to a human body structure so as to form a neck part, an upper limb part and a lower limb part. The at least one counter weight is correspondingly arranged on the head, the trunk and the lower limbs according to the human body structure. The covering material is used for covering the framework, the at least one movable joint structure and the at least one balance weight.

The present application relates to the technical field of process monitoring, in particular to a personnel intrusion detection method based on an artificial intelligence loading system, comprising the following steps: acquiring a target continuous motion image sequence, calculating a node included angle and a direction deflection to generate positioning information, comparing a work boundary to determine an intrusion and generate an identification state, combining a direction vector to determine a device response instruction, loading a trajectory in a three-dimensional scene and marking a risk area, and obtaining a personnel path intrusion calibration record. In the present application, through continuous extraction of a human body structure boundary box in an image frame and dynamic displacement analysis of a trajectory node, a target path with continuous gait characteristics is identified and static and device interference areas are eliminated, whether the target has crossed the boundary is determined in combination with a real-time loading work boundary, a minimum distance between a turning point and a boundary path is calculated to complete intrusion behavior identification, real-time discrimination of a behavior trajectory and efficient linkage of alarm direction control are realized, and the accurate identification and real-time reaction capability of personnel intrusion events are enhanced.

This invention provides a toy skeleton structure, wherein the torso adopts a segmented structure. Ball heads and spherical grooves are respectively provided on the opposite end faces of the first and second torsos. The first torso can be rotatably fixed to the second torso through the rotational engagement of the ball heads and spherical grooves. Simultaneously, the ball heads are connected to the end face of the first torso through a waist-shaped transition section, allowing the torso to rotate and perform small-amplitude swaying movements, thereby increasing the torso's range of motion and fluidity. This invention also provides an assembly mecha toy including the aforementioned toy skeleton structure. The toy skeleton structure provided by this invention has the advantages of simple and compact structure, fewer component parts, simple assembly operation, and easy management. It also has the advantages of multiple and smooth range of motion, ensuring high playability and interactivity for both the skeleton structure and the assembly mecha toy. Furthermore, the structural design of the toy skeleton structure allows its shape to more closely resemble a simulated human body structure.

The utility model relates to the technical field of medical instruments, and discloses a full-porous cervical vertebra fusion cage which comprises a machining area and a porous area. The machining area is provided with a threaded hole for leading in a surgical instrument; the porous area comprises an upper surface, a lower surface, a front surface, a rear surface and two side surfaces; and the upper surface and the lower surface are provided with anti-skid surfaces. The threaded hole in the machining area is used for being matched with a surgical instrument matched with the threaded hole, and surgical implantation is facilitated. The porous area provides more growth space for cells, adhesion, migration, proliferation and differentiation of the cells are facilitated, and finally the bone ingrowth effect is achieved. The full-porous cervical vertebra fusion cage has the mechanical property matched with the human body structure and good biocompatibility, can form a tight bonding interface with the human body tissue, improves the combination stability of the fusion cage and the human body tissue, has the mechanical property matched with the human body tissue, and can be applied to clinical treatment of cervical degenerative diseases.

The invention discloses a highly-simulated transvaginal endoscopic surgery training model, which comprises a model body, the model body is a model from the waist and abdomen of a human body to the thigh root, a simulated vaginal orifice is arranged outside the model body, a simulated pelvic cavity is arranged in the model body, a cover plate is detachably arranged on the model body, and the cover plate is used for opening the model body. A vagina model and a uterus model are arranged in the simulated pelvic abdominal cavity and are mounted according to a human body structure, the uterus model comprises a uterine orifice section, the vagina model comprises a vagina section and a dome section which are detachably connected with each other, and a ferrule is arranged at one end, far away from the vagina section, of the dome section; the sleeve ring is used for sleeving the uterine orifice section and moving in the length direction of the uterine orifice section, a training module is arranged in the simulated pelvic abdominal cavity, the requirements of the transvaginal laparoscopic surgery are met through the arrangement of the cover plate and the dome section, the defect that traditional training equipment is too large in loss is overcome, and meanwhile targeted training of the surgery is achieved through the arrangement of the sleeve ring.

The present application provides a preparation method for biosynthesis of a human body structural material. A polypeptide has the amino acid sequence of SEQ ID NO. 4, 5 or 6. Recombinant type-VII humanized collagen prepared in the present application has high activity of promoting cell proliferation and does not produce an immune response when applied to a human body, and the preparation method therefor is novel and can obtain the recombinant type-VII humanized collagen on a large scale, and is widely applied in the preparation of human body structural materials.

The invention relates to a method and a system for combining B-ultrasonic records and a human body structure model, and aims to provide visual and accurate diagnosis support for doctors through fusion of a high-precision human body structure model and real-time B-ultrasonic data. The method comprises the steps of human body structure model acquisition and preparation, B ultrasonic data acquisition and preliminary matching, real-time matching and data fusion, step-by-step filling and model updating, model checking and dynamic data analysis, result storage and subsequent application and the like. The system comprises corresponding modules, such as a human body structure model acquisition and preparation module, a B ultrasonic data acquisition and preliminary matching module and the like, and the functions of automatic operation of the mechanical arm, abnormal data identification and the like are supported. According to the invention, a doctor can observe the B ultrasonic image information of the target part in real time, dynamically analyze and evaluate the change trend of lesions, so that the clinical decision is optimized, and the diagnosis and treatment efficiency and accuracy are improved.

The invention discloses a clothes changing pedestrian re-identification method based on graph attention human body part dynamic correlation modeling. The method comprises the following steps: acquiring a video sequence of a pedestrian to be identified, and performing part segmentation through a human body analysis model; a part perception multi-scale feature extraction module is used to extract the features of each part under different receptive fields and suppress the background; then, the combination of each frame and the part is constructed into a graph node, and a heterogeneous space-time part graph containing a space edge, a time edge and a non-local edge is constructed; performing message passing on the three types of edges through a graph attention part association modeling module, and fusing to obtain enhanced node features; then, the enhanced node features are mapped back to original spatial feature dimensions; carrying out adaptive weighted integration on different granularity features; and finally, an identity representation vector is generated, similarity between the identity representation vector and image library features is calculated, and a matching result is output. According to the method, by capturing time sequence dynamic association between human body parts, identity representation generated by the model is more focused on a stable human body structure instead of a variable clothing appearance.

The application relates to a graph convolution human skeleton action recognition method based on a structural attention mechanism, and comprises the following steps: S1, constructing a structural attention module, including two submodules of a body part relationship-based attention module and a body symmetry node trajectory-based attention module; S2, adding the body part relationship-based attention module to an adaptive adjacency matrix of an intermediate result of graph convolution to obtain an adaptive adjacency matrix of reinforced body part structure; S3, constructing a related attention module by using the relationship of body symmetry node trajectories to reinforce the attention of a human skeleton action recognition model to body symmetry; and S4, feeding the features reinforced by the structural attention into a classifier to obtain scores corresponding to each action, so that a final recognition result is obtained. Compared with the prior art, the application considers the human structural behavior mechanism in action recognition, and has the advantages of high recognition precision.