76 results about "Spatial interaction" patented technology

The invention discloses an intestinal cancer immune response graph neural network prediction method, a medium and equipment, and the method comprises the steps: collecting pathological image information, immunodetection information and basic clinical information, extracting a tissue space distribution characteristic spectrum through a deep convolutional network, and constructing a graph neural network model in combination with an immunomarker expression characteristic matrix; spatial interaction characteristics of a tumor microenvironment are modeled by adopting a graph attention mechanism, finally a treatment response probability, an optimal treatment opportunity and an adverse reaction risk are predicted through a multi-task learning framework, and a clinical decision report containing a prediction response curve, a risk early warning threshold and a treatment time window suggestion is output. According to the method, through multi-modal data fusion and spatial interaction modeling, accurate prediction of intestinal cancer immunotherapy response is realized, and a more comprehensive reference basis is provided for clinical decision making.

The invention provides a multi-feature trajectory prediction and cross-defense early warning method for edge and coast defense control objects, relates to the technical field of trajectory prediction and intelligent control, and solves various limitation problems of poor trajectory continuity, lack of environmental constraints and the like of an existing trajectory prediction technology in edge and coast defense scenes. The method comprises the following steps: firstly, acquiring multi-channel monitoring video data, performing environment adaptation preprocessing, and detecting each target and category; based on a detection result, splicing the track fragments in the single-path monitoring range, and obtaining global track data of the target after preset constraint inspection; extracting features of the global trajectory data from a plurality of preset dimensions, and performing feature fusion to obtain high-dimensional vector representation of the target trajectory; and constructing a multi-dimensional feature constraint trajectory prediction model, respectively modeling through time dependence and space interaction, and outputting a subsequent trajectory of the target under a preset constraint. The method can effectively realize prediction and cross-defense risk identification of future trajectories of key control objects, and is applied to the field of public security.

The invention relates to a multi-view three-dimensional human body posture estimation method and system based on double-flow space-view-time sequence modeling. The method comprises the following steps: inputting a multi-view multi-frame image sequence, and performing human body detection and cutting; performing two-dimensional human body posture estimation on the input image at each view angle and each frame, and extracting image features; constructing an image-attitude bimodal alignment expression; sequentially executing sequential modeling of intra-visual space interaction, cross-visual-angle interaction and intra-visual time interaction on the image flow representation and the attitude flow representation in the same layer to obtain attitude flow fusion features; performing regression on the attitude flow fusion features by using a three-dimensional regression head to obtain three-dimensional skeleton coordinates; training the three-dimensional skeleton by adopting supervised learning; and inputting a to-be-estimated image sequence into the trained model and the regression head to obtain a three-dimensional human body posture estimation result. According to the method, the precision, robustness and deployability of three-dimensional attitude reconstruction can be remarkably improved under the condition of not depending on camera calibration parameters and human body priori.

The invention discloses a remote sensing image segmentation method based on semantic space interaction and edge guidance, which relates to the field of remote sensing image segmentation, and comprises the following steps: carrying out multi-level standard convolution block processing and semantic space interaction module processing on a preprocessed remote sensing image by utilizing a feature extraction network, extracting basic features, and carrying out semantic space information decoupling; and outputting multi-layer enhanced features, performing multi-level standard convolution block and up-sampling processing on the features through a feature fusion network, fusing the features through jump connection and combining a classification head to output a segmentation probability graph. According to the method, feature expression is enhanced through semantic space interaction decoupling, boundary recognition is enhanced through an edge supervision mechanism, and multi-scale fusion and joint loss optimization are carried out, so that the segmentation precision and boundary coherence of the remote sensing image in a complex scene are remarkably improved.

The invention discloses a pedestrian trajectory prediction method based on deep learning. The pedestrian trajectory prediction method is characterized by comprising the following steps of S1, receiving input data and performing preprocessing; s2, extracting spatial features from the input data by using a graph convolutional network GCN, wherein the convolutional neural network GCN simulates spatial interaction between pedestrians through an adjacent matrix; s3, extracting time features from the input data by using an encoder layer of a Transform model, wherein the Transform model captures a dependency relationship of pedestrians in a time dimension through a self-attention mechanism; s4, fusing the spatial features and the time features to obtain spatial and temporal features, and extracting spatial and temporal fusion features by using a Transform encoder; and S5, noise is added to the extracted space-time fusion features, then the features are sent to a full connection layer for pedestrian trajectory prediction, and future position information of pedestrians is output. The method improves the prediction precision and robustness, and is suitable for the fields of intelligent transportation and the like.

The invention relates to a multivariate time sequence anomaly detection method. An anomaly detection model used in the method comprises a U-Net encoder, a U-Net decoder, a time sequence characteristic memory module and a space interaction perception module. Performing multi-scale time sequence feature extraction on the weight-standardized one-dimensional convolutional layer in the U-Net encoder to obtain multi-level encoding features; in a time sequence feature memory module, enhancing long-term dependence features for the multi-level coding features; inputting the memory features into a space interaction perception module, constructing a graph structure between variables through a graph attention network, and outputting space enhancement features; in a U-Net decoder, splicing the corresponding memory features with the up-sampling features of the upper layer, and obtaining reconstruction features through one-dimensional convolution fusion; and calculating a dimension-by-dimension square error sum between the original time sequence fragment and the reconstructed data to obtain an abnormal score. Compared with the prior art, the method has the advantages of high precision, strong robustness, high efficiency and the like.

The present application relates to the technical field of image processing, and more particularly to a lesion segmentation method of adaptive dynamic text prompt, comprising: acquiring a lesion image, and generating an adaptive dynamic text prompt based on the lesion image; constructing a multi-modal enhanced fusion text prompt adapter based on a FiLM global channel recalibration module, a spatial cross-attention interaction module and a zero initialization gate nonlinear integration module cascade; the global channel recalibration module of FiLM utilizes a feature linear modulation mechanism to perform channel-level weighting on a visual feature map; before the feature enters spatial interaction, according to the text semantic enhancement band response, the activation of the background noise channel is inhibited. The present application overcomes the problems of insufficient utilization of multi-modal prompts, dependence on artificial static labeling of text prompts, serious dependence on prior information in the reasoning stage and difficulty in multi-modal feature fusion in a low-contrast environment in the existing medical image segmentation technology.

The invention belongs to the technical field of biological medicine, and particularly relates to a new Spp1 + Carmil1 + and Spp1 + Cav1 + macrophage subgroup, a screening method and application in diagnosis and treatment of rheumatoid arthritis (RA). Aiming at the function limitation of a traditional M1 / M2 classification model, the invention proves the following new subgroup pathological interaction mechanism: osteoclast (ACP5 +) specifically releases TNFSF11 (RANKL), and is co-localized with Spp1 + Carmil1 + and Spp1 + Cav1 + subgroup space; the Spp1 + Carmil1 + subgroup secretes IL-18, and space interaction exists between the Spp1 + Carmil1 + subgroup and Treg (Foxp3 +) and Th17 (Il17a +) cells. The invention provides a novel cell marker and a targeted treatment strategy for accurate diagnosis and treatment of RA.

The invention belongs to the technical field of image segmentation, and discloses an edge prior Mama-based image segmentation detection method, electronic equipment and a storage medium, and the method comprises the steps: preprocessing an input feature through a space interaction enhancement (SIE) module; sS3D is utilized to capture spatial dependence and depth dimension context in the slice, and long-range dependence is calculated through a state space model; meanwhile, an EP component is used for explicitly enhancing local edge and texture features through convolution and background suppression operation; and finally fusing the outputs of the SS3D and the EP. According to the method, the advantage of Mamba efficient modeling long-range dependence is reserved, meanwhile, the detail perception capability of boundaries is remarkably improved, reliable separation of abnormal areas and precise description of local details under complex background interference are achieved, and segmentation robustness and precision in tasks such as industrial defect detection and medical image analysis are effectively improved.

The application discloses a dispatching collaborative optimization method and system based on a segmented agent and a feasible domain evaluation. The method comprises the following steps: obtaining engineering structure and equipment dispatching parameter space and operation conditions, and constructing physical conservation constraints; predicting first safety constraint responses and second safety constraint responses by using a segmented agent model for joint parameter combinations meeting the constraints; dividing a feasible dispatching domain based on flow rate responses, extracting volume indicators of quantitative dispatching scheme proportions, and extracting spatial interaction indicators of quantitative equipment area influence distribution; solving optimal structure parameters by taking minimum outer layer structure engineering quantity as an objective and taking the above two indicators meeting performance thresholds as constraints; and outputting recommended equipment operation strategies by taking minimum inner layer flow rate as an objective in the feasible dispatching domain corresponding to the optimal structure parameters. The application can realize double-layer nested collaboration of physical structure design and equipment operation dispatching, and can improve the cross-condition dispatching adaptability and cross-region response field decoupling capability of an engineering system.

The present application relates to the technical field of data processing, more specifically, the present application relates to a power project demand reserve overall balance method based on intelligent algorithm, comprising: obtaining the declaration text data and electrical scene image data of the power project to be checked; detecting the equipment entity node by using the Grounding DINO model; constructing a spatial interaction matrix based on the spatial distribution characteristics of the equipment entity node; fusing the text semantic features, visual feature vectors and spatial interaction intensity, deducing the electrical topology connection probability and generating the project topology graph; and calculating the topology structure similarity by using the subgraph isomorphism algorithm to determine the duplicate declaration project. The present application realizes the identification of electrical connection relationship by evaluating the spatial relationship and fusing multi-modal information, solves the problem that the traditional method cannot distinguish the same equipment in different topology positions, and improves the accuracy of the power project demand reserve overall balance.

The invention provides a myopathy progress prediction method and device based on a spatio-temporal topology network, and the method comprises the steps: carrying out the image segmentation and quantification of muscle biopsy sections of a patient at a plurality of time points, constructing a static pathological topology network with a single muscle fiber as a node and a spatial adjacency relation as an edge, and enabling the node to comprise the multi-dimensional features of subtypes, forms, textures, cell nucleuses and the like; and forming time-space diagram data according to a time sequence. The data are input into a pre-trained space-time diagram neural network, a space coding layer aggregates neighborhood information through a diagram convolutional network to capture space interaction, a time coding layer learns a time sequence evolution rule through a recurrent neural network, and a decoding layer predicts a network state of a next time point. Pathological network sequences of multiple time points in the future are generated through iterative prediction, then key indexes are extracted to draw a quantitative prediction curve, a visual pathology heat map is generated, crossing from static pathology to dynamic progress is achieved, and an objective and accurate myopathy prognosis evaluation tool is provided for clinic.

The present application relates to the technical field of dangerous rock monitoring, and particularly relates to a dangerous rock monitoring system and a risk assessment method thereof.The present application introduces a graph convolutional neural network to accurately model environmental factors such as wind speed and vegetation, overcomes the limitations of traditional wind force calculation models, corrects the influence of wind force tension, and thus more accurately assesses the stability of the rock mass;By introducing a dynamic adjustment mechanism of the environmental change coefficient and the vibration propagation coefficient, the model can adapt to changes in environmental factors such as temperature, humidity and precipitation in real time, and improve the adaptability and accuracy of the model;In addition, considering the spatial interaction and dynamic effect between rock mass units, an instability propagation equation is established to capture the propagation process of rock mass instability, so as to comprehensively consider the physical properties, spatial interaction and environmental changes of the rock mass, thereby providing more accurate and real-time instability prediction, and providing effective risk warning and emergency decision assistance for mining, slope construction and other engineering.

The application discloses a kind of low air space complex environment under unmanned aerial vehicle cluster target long-time robust tracking method, comprising: by the target detection model of pre-training to the unmanned aerial vehicle target is detected and is combined with environmental state information dynamic optimization;The detection result after optimization is parallelly input light graph convolution network and bidirectional LSTM module, respectively extract the spatial interaction feature and time sequence dynamic feature of target predicted position;Two kinds of features are fused using spatial-time sequence attention mechanism, to obtain more accurate predicted position information;Based on the confidence score of target detection frame, the predicted position of target and the actual detection result of target are two-stage data association.The method effectively improves the tracking accuracy and robustness of unmanned aerial vehicle cluster target in complex dynamic environment, especially in target occlusion, dense formation and background interference scene has strong adaptability.

The invention discloses a dam engineering leakage channel entrance-channel-escape full view identification method and system. The method comprises the following steps: acquiring multi-source monitoring data of a dam engineering area; constructing a background seepage flow field reflecting a macroscopic hydraulic gradient and an anisotropic diffusion tensor field reflecting local permeability characteristics based on multi-source monitoring data; performing bidirectional probability field evolution calculation by using an unsteady convection-diffusion equation set and respectively taking the inlet geometrical characteristics and the escape temperature anomaly as sources to obtain an inlet reachability field and an escape reachability field which are continuously distributed in the dam body space; and determining the three-dimensional full-view probability distribution of the leakage channel based on the space interaction overlapping characteristics of the two fields. According to the method, physical self-consistent reconstruction from discrete monitoring data to a continuous channel entity is realized by constructing a different-field coupling feedback and tensor guiding mechanism, and the topological fuzzy problem of leakage path identification in a complex medium is effectively solved.

This invention discloses an immersive light and shadow art space interaction method and system based on motion recognition. It collects user motion signals and analyzes them into posture features, then uses template comparison to filter valid action inputs. The system performs trajectory analysis on action sequences to extract rhythm distribution features and identifies emotional tendencies based on changes in action amplitude and speed, organizing the queue of actions to be processed according to the urgency of emotional response. For multi-limb synchronous movements, it partitions and decomposes the movements, identifying the temporal coordination relationship between dominant and subordinate limbs and establishing a cascading trigger mechanism. It maps action features to light and shadow control parameters and verifies rendering performance, generating optimized lighting effect execution commands. It monitors user interaction frequency and dwell time distribution to identify deviations from the immersive state, automatically triggering lighting adjustment strategies to restore user engagement. Through preloading and reuse scheduling of rendering tasks, it optimizes system resource utilization and outputs coordinated and unified spatial interaction control commands, achieving a natural fusion of action and light and shadow.

The application discloses a kind of based on graph attention hybrid neural network's multi-aircraft 4D trajectory prediction method, belong to aviation technical field.Utilize Transformer and long short-term memory network to extract different aircraft historical trajectory implicit information in airspace, aircraft in airspace is regarded as node and constructs aircraft connection network graph, aircraft implicit information and aircraft connection network graph are modeled by graph attention network, for capturing the interaction information between aircraft.Graph attention network will integrate the trajectory implicit information of all aircraft in airspace and carry out sharing and updating, finally by two long short-term memory network-based decoder, aircraft trajectory implicit feature information is decoded, the output of the fusion of two decoders is output by fully connected layer, predicted trajectory, compared with the method not considering aircraft spatial interaction information, improve the accuracy of flight trajectory prediction, and accurate trajectory prediction technology provides support for subsequent risk perception technology and conflict resolution technology.

The invention discloses an interaction method, device and system of refrigeration equipment and the refrigeration equipment, and belongs to the technical field of refrigeration equipment. The interaction method of the refrigeration equipment comprises the steps of obtaining interaction data of a user and the refrigeration equipment and space structure information of the refrigeration equipment, wherein the interaction data at least comprises voice interaction data; based on the interaction data and the space structure information, obtaining space interaction features, the space interaction features being used for representing an association relationship between the interaction data and the space structure information; and determining and outputting a target interaction result of the user and the refrigeration equipment based on the space interaction characteristics. The specific article or area pointed by the user can be accurately recognized, and the situation that the interaction accuracy is reduced due to misunderstanding of the user intention is avoided.

The invention is suitable for the technical field of intelligent terminals, and provides a pose estimation method and device of a space interaction device, a terminal device and a storage medium, the method comprises the following steps: obtaining a current frame image of the space interaction device collected by a three-dimensional imaging device, the space interaction device being provided with at least two mark points; identifying visible states of all mark points in the current frame image; acquiring inertial measurement data output by an inertial sensor in the space interaction equipment; according to the visible states of all the mark points, the inertial measurement data and reference space coordinates, a pose estimation result of the space interaction equipment is generated, and the reference space coordinates are determined based on historical frame images. According to the method, the robustness of space interaction pose estimation can be improved, and the interaction continuity and precision of space interaction equipment in a complex scene can be guaranteed.

The invention relates to the technical field of real-time space interaction, and discloses a space interaction method, system and device capable of real-time interaction, and the method comprises the steps: executing a sight line shielding detection strategy, detecting whether moving audiences can shield the sight lines of the audiences, and enabling all the audiences with the shielded sight lines to form a shielded audience set; calculating a key shielding area for each audience in the shielded audience set, and updating the key shielding area in real time; executing a path planning strategy according to the key shielding area, and planning a moving path for each moving audience to avoid the key shielding area; optimizing the film watching experience of each audience through the light field, specifically, calculating a target function of the film watching experience according to a spatial state; executing a reinforcement learning optimization strategy on the target function, and optimizing the film watching experience of each audience; applying and executing a result of the reinforcement learning optimization strategy in the film watching space; and executing an AI adaptive light field optimization strategy, and optimizing the film watching experience of each audience in real time by using space interaction.

The invention discloses an interaction perception comparison coding and multi-mode pocket feature driven molecule generation method and system. The method comprises the following steps: firstly, constructing a pre-training and fine-tuning data set; the method comprises the following steps: acquiring spatial interaction, structure and sequence features of protein pocket-ligand compound data, optimizing the sequence features through cross-modal contrast learning, and fusing the optimized sequence features with the structure features to obtain first multi-modal fusion features; constructing a molecule generation model, pre-training the model by using a pre-training data set, and then finely adjusting the model by using a second multi-modal fusion feature fused with the first multi-modal fusion feature and a corresponding ligand molecule sequence feature; for a specific target protein, the model can generate a molecule adapted to the specific target protein according to the multi-modal fusion feature of the pocket. The invention also discloses a system for realizing the method. The method and the system provided by the invention can effectively generate the candidate molecule which is highly matched with the binding site of the target protein, has high affinity and is excellent in synthesis accessibility.

The invention relates to the technical field of geographic information, in particular to an urban functional area identification method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring multi-source data, and constructing graph structure data based on the multi-source data; and inputting the graph structure data into a preset neural network model, and outputting to obtain an urban functional area identification result, the preset neural network model being obtained by training a graph convolutional neural network model based on a spatial domain in combination with an attention mechanism fusing attribute representation and spatial connection. Therefore, city space interaction is simulated by constructing the graph structure data, and the attention mechanism jointly driven by the attributes and the space connection is utilized to identify the types of the city function areas, so that the problem of inaccurate identification of the city function areas in the prior art is solved, and accurate classification of the city function areas is realized.

The embodiment of the invention discloses a space interaction method, a head-mounted display system, a readable medium and a program product. A specific embodiment of the method comprises the following steps: acquiring an angular velocity information sequence corresponding to the handheld equipment through an inertial sensor of the handheld equipment; generating target angular velocity information according to the angular velocity information sequence; according to the generated at least one piece of target angular velocity information, gesture action information corresponding to the handheld device is recognized, and the at least one piece of target angular velocity information has a time sequence; and in response to determining that the gesture action information meets a preset space interaction triggering condition, controlling a head-mounted display device in communication connection with the handheld device to execute a space operation triggered by the handheld device. According to the embodiment, external equipment and mark points do not need to be arranged, multiple sensors do not need to be used either, the equipment can be lighter, and the mobility and universality of the equipment are improved.

The application discloses a method and system for identifying events in a video. The method comprises: obtaining static features of each frame in the video based on a neural network; obtaining spatial interaction relationship features corresponding to each static feature based on a spatial attention mechanism network; obtaining space-time interaction relationship features corresponding to the set of spatial interaction relationship features based on a time domain attention mechanism network; constructing a space-time interaction relationship graph based on the space-time interaction relationship features, and obtaining features of the space-time interaction relationship graph through a graph convolutional neural network; and identifying events in the video based on the space-time interaction relationship features and the features of the space-time interaction relationship graph. The scheme of the application improves the accuracy and efficiency of identifying events in the video.

The invention provides a method and equipment for determining track confidence and a sensing system. The method comprises the following steps: acquiring a first relationship between a target object and a first environment according to first track information of the target object and the first environment; and based on the first relationship, determining the airspace confidence of the first track information. In the embodiment of the invention, the spatial interaction relationship between the target object and the surrounding environment is introduced to determine the airspace confidence coefficient of the trajectory information of the target object. According to the scheme, on one hand, the possible influence of the surrounding environment on the motion behavior of the target object is considered, and the precision of the track confidence coefficient is improved; and on the other hand, the scheme can be suitable for various traffic driving scenes, such as scenes conforming to the inertia priori hypothesis and scenes not conforming to the inertia priori hypothesis, and the universality and robustness of the track confidence coefficient can be improved.

The invention relates to the technical field of computer graphics and virtual reality, and discloses an unstructured volume data hybrid rendering method based on WebXR and Gaussian splash technologies and a space interaction system. According to the method, a standardized visibility container protocol is constructed, a hysteresis space state machine is combined, and real-time logic cutting, efficient streaming rendering and smooth space interaction of unstructured volume data in any shape are achieved through general rendering entity abstraction, streaming zero-copy rendering pipelines and core algorithm innovation. According to the method, the problems of difficulty in geometric constraint of 3D Gaussian splashing data in the existing WebXR environment, bottleneck of rendering performance of a mobile Web end, single jitter of spatial state perception and the like are solved, and the method is suitable for various scenes such as AR transmission doors, photoscopes, holographic showcases and the like and has wide universality and practicability.

The invention relates to the technical field of data processing, in particular to an electric power project demand reserve overall balancing method based on an intelligent algorithm, which comprises the following steps: acquiring declaration text data and electrical scene image data of an electric power project to be subjected to duplicate checking, and detecting equipment entity nodes by utilizing a Grounding DINO model; constructing a spatial interaction matrix based on the spatial distribution characteristics of the equipment entity nodes; the text semantic features, the visual feature vectors and the space interaction strength are fused, the electrical topology connection probability is deduced, and a project topological graph is generated; and a sub-graph isomorphism algorithm is utilized to calculate topological structure similarity so as to judge the repeated declaration project. According to the method, the electrical connection relationship is identified by evaluating the spatial relationship and fusing the multi-modal information, the problem that the same equipment cannot be distinguished at different topological positions in a traditional method is solved, and the accuracy of overall balance of power project demand reserve is improved.

This invention discloses a multimodal spatial interaction head-mounted system and method for clinical nursing, comprising the following modules: A) a management platform; B) an indoor positioning and wireless communication base station, wherein the indoor positioning and wireless communication base station includes an ultra-wideband (UWB) positioning module and a positioning QR code; C) a nursing workstation, wherein the nursing workstation includes a data terminal, which includes an EHR / HIS integration module, a workflow configuration interface, and a device IoT gateway; and D) a head-mounted glasses terminal. This invention achieves high-precision spatial anchoring through hierarchical fusion positioning of UWB and visual SLAM, introduces multimodal aseptic interaction logic supporting emergency coverage commands to adapt to clinical practice, and innovatively adopts a dynamic privacy rendering and bystander detection mechanism based on the line-of-sight cone, effectively resolving the contradiction between information visualization, aseptic operation standards, and patient privacy protection in clinical nursing.

The invention discloses an assembly clearance detection method based on visual fusion of MLSD and SAR (Synthetic Aperture Radar). The method comprises the steps of calibration of a handheld mobile camera, collection and sampling of images, setting of a threshold value around an assembly gap, linear feature detection based on MLSD, image slice processing, linear feature screening and slope calculation, and linear spacing calculation. According to the method, a Pycharm is adopted based on a pyhton language, and a straight line detection and spacing calculation algorithm is developed by depending on an Opencv image processing library; and developing an SAR display method by adopting Unity 3D based on a C # language. According to the method, accurate and robust detection of linear features is realized by adopting a deep learning-based MLSD algorithm, and a detection result is visually and clearly displayed on a real product assembly site in a virtual-real fusion form by adopting an SAR technology, so that a user can obtain and understand the detection result in natural space interaction. SAR display breaks through the limitation of information display of a traditional two-dimensional screen, a detection result is overlaid to an assembly site, a user interacts with information in a more natural and immersive mode, and perception and cognition of assembly gap detection are enhanced.