255 results about "Spatial maps" patented technology

The invention discloses a BIM model feature extraction-based unmanned aerial vehicle intelligent inspection path generation method for the interior of a building. The method comprises the steps of extracting a spatial geometric feature set, a spatial topological feature set and a semantic feature set from original BIM model data; constructing a basic space graph representing a space structure and a semantic enhancement graph representing a function relationship based on the feature set, and generating a multi-level semantic graph; task targets and dependency relationships are extracted from the inspection task description and the multi-level semantic graph, and a task weight matrix is generated; generating an optimized path set based on the multi-level semantic graph and the task weight matrix; performing multi-dimensional scoring and real-time optimization on the optimized path set to generate an emergency adjustment strategy; and finally verifying the inspection path. According to the method, through multi-level feature extraction and multi-level semantic graph construction, accurate understanding of the building space is realized, and the intelligent level and reliability of routing inspection path planning are improved.

The invention discloses a contrast learning unsupervised cross-modal hash retrieval algorithm based on a graph attention mechanism, and relates to the technical field of cross-modal retrieval. According to the graph attention mechanism-based contrast learning unsupervised cross-modal Hash retrieval algorithm, feature extraction is performed on image and text data through a pre-trained CLIP model to obtain image modal features and text modal features, and additional positive samples are generated through data enhancement; the similarity is calculated through Euclidean distance and cosine similarity on the basis of a graph attention network, and high weight is given to positive samples and low weight is given to negative samples by comparing adjacent matrixes. The method has the advantages that the features of the image mode and the text mode can be effectively fused and mapped to the common Hash space, a graph attention mechanism can adaptively adjust the connection weight between the features of the different modes, samples with high semantic correlation are highlighted, and therefore the semantic consistency between the image and the text is ensured.

The invention relates to the technical field of ozone concentration prediction, in particular to an ozone concentration prediction method, device, equipment and medium based on a space-time diagram convolutional network model.The method comprises the steps that historical ozone concentration and meteorological data of a target area are obtained, and training and prediction samples are constructed after standardization; a space diagram structure is established, nodes represent monitoring points, and edges represent the monitoring points with correlation exceeding a threshold value; extracting a spatial dependency relationship by adopting GCN and GAT, and generating time sequence data; tCN or RNN is adopted to extract time-dependent features, and spatial-temporal feature comprehensive representation is obtained; training a model by utilizing the training sample and the spatial-temporal characteristics, adopting MSE (mean square error) as a loss function, adjusting parameters by an Adam optimizer, and optimizing hyper-parameters to minimize errors; newest meteorological and historical data are input, future ozone concentration is predicted, GNN extracts spatial features, RNN processes a time sequence, time sequence data can be processed in parallel through convolution operation, and high calculation efficiency is achieved.

The invention discloses a ship target tracking method based on radar imaging in a shielding environment, belongs to the technical field of target tracking, is used for ship target tracking, and comprises the following steps: obtaining multiple frames of echo images of a time sequence, obtaining a potential target on each frame by using a ship image feature detection method, and obtaining a time accumulation space diagram; initial tracks are extracted from the time accumulation space diagram, and a tracking ID is distributed to each initial track; a frame of echo image is read, the position of a potential target is detected, and the position of the potential target is associated and matched with the tracking ID; and updating the position of a potential target which is not normally associated and matched by using a ship target tracking method combining unscented Kalman filtering and LSTM. According to the method, more potential targets are extracted, and a local matching result is optimized; more and more accurate target information is provided for the VTS system, so that the VTS system can have more prior information of early warning, and safe passage of a main channel is ensured.

The invention provides a three-dimensional space collision avoidance system based on multiple tower cranes, and relates to the technical field of port hoisting operation, a three-dimensional voxel grid VoxelGrid space map is constructed, and the three-dimensional space collision avoidance system comprises a first horizontal coverage map, a second descending coverage map, a third fan-shaped area occupation map, a fourth track occupation area map and a fifth interference area map; the system can identify the space overlapping area of the tower crane operation path at the voxel level, breaks through the limitation of the traditional path planning method on the dimension and precision of space modeling, and enhances the fine granularity and accuracy of collision detection. A dynamic acquisition module and a dynamic interference module are introduced, dynamic disturbance caused by factors such as stormy waves and tides on a ship in a port operation environment can be sensed in real time, the deviation trend of the tail end of the lifting appliance is predicted, target correction and track compensation are carried out before track execution, and the situation that the lifting appliance is accidentally touched or falls off due to track drifting is effectively promoted to be reduced.

The invention discloses a hybrid network method for hyperspectral image water quality inversion, and relates to the technical field of hyperspectral image analysis. The model comprises a spectral feature extraction module, an improved Transform encoder, an LSTM (Long Short Term Memory) module, a cross-time attention module and an adaptive feature fusion module. The model extracts spatial and spectral features through a deep convolutional network, enhances feature expression by using multi-scale embedding and an attention mechanism, and optimizes prediction precision in combination with a dynamic weight fusion strategy. Through end-to-end training of the model, water quality parameters (total nitrogen, dissolved oxygen and the like) in a hyperspectral image can be effectively estimated, and a spatial map of water quality distribution is generated. Compared with a traditional method (such as support vector regression, partial least square regression and the like), the hybrid network provided by the invention has the advantage that the accuracy and robustness of water quality parameter inversion are remarkably improved. Experimental results show that the method has excellent performance in a complex water body environment, and an efficient and reliable technical means is provided for accurately monitoring and evaluating water quality pollution.

The invention discloses a database-based space conflict treatment method and device and a medium, and relates to the technical field of electric digital data processing. The method comprises the steps of determining conflicting pattern spots with conflicts in space pattern spots through a preset space analysis function according to a geometrical relationship between the space pattern spots corresponding to a planning land in a space database; analyzing the spatial relationship and the ownership planning relationship corresponding to the conflict pattern spots to determine the conflict intensity corresponding to each conflict pattern spot; based on the spatial relationship network, determining a hotspot conflict area with spatial aggregation according to the conflict intensity; determining treatment priorities corresponding to the conflict pattern spots according to a descending order of the conflict intensity, and adjusting the treatment priorities of the conflict pattern spots according to conflict linkage paths among the conflict pattern spots in the hotspot conflict area; and generating a space governance path in the hotspot conflict area through the adjusted governance priority, so as to realize the space conflict governance of the planned land through the space governance path.

A method includes identifying a zone blocked from line of sight (LOS) of a gNB; receiving a 3D spatial map of a first area including the zone and a second area that surrounds the zone and that is within both a coverage area and the LOS of the gNB; determining whether the second area includes a mountable surface to which a passive RF reflective metasurface is attachable; and determining whether the metasurface, if attached to the mountable surface, generates a reflection path to the zone, based on a determination that the second area includes the mountable surface and based on estimated propagation paths from the gNB to the zone. The method includes determining whether to add a metasurface to the second area based on: a determination result whether the reflection path from the metasurface attached to the mountable surface to the zone includes reflected signals satisfying a threshold bandwidth condition.

The invention provides a geometric and physical fusion method for predicting the assembly deviation of an easy-to-deform structure, and belongs to the technical field of aeronautical engineering manufacturing, and the method comprises the following steps: carrying out the structural decomposition of the easy-to-deform structure, obtaining a space vector point set, and representing and calculating a consistent manufacturing deviation, an assembly positioning deviation and an assembly deformation deviation; constructing an assembly deviation transfer diagram, and mapping the assembly deviation transfer diagram into an assembly state space diagram by using the three multi-source deviations; performing finite element analysis on the assembly precision of the easily-deformed structure by utilizing the assembly state space diagram, and converting the obtained assembly deviation cloud diagram into an assembly deviation contour map; constructing an initial generative adversarial network model, and training and verifying the initial generative adversarial network model by using data in the assembly deviation contour map to obtain a final generative adversarial network model; and inputting the actual deviation image of the easily-deformed structure into the final generative adversarial network model for assembly deviation prediction. The assembly precision and the assembly efficiency of the easy-to-deform structure can be improved.

The invention discloses a three-area three-line database automatic quality inspection evaluation method based on deep learning. The method comprises the following steps of S1, obtaining three-area three-line database data and performing preprocessing; s2, checking a directory structure and a file name of the original data set; s3, extracting spatial layer data from the hierarchical data set; s4, performing segmentation and attribute information extraction on various pattern spots in the layer data after spatial verification; s5, inputting the pattern spot relation characteristic data into an artificial fish swarm algorithm; s6, automatically comparing the pattern spot area and distribution data with a quality inspection standard preset in a three-area three-line database; and S7, performing parallel processing on an automatic comparison result through a multi-thread thread pool. According to the method, an artificial fish swarm algorithm, a relational attention network and a task allocation and thread pool management technology are combined, and three-area three-line database automatic quality inspection evaluation based on deep learning is realized.

The invention relates to a few-sample image classification method based on hyperbolic space image-text local feature alignment, and belongs to the technical field of image recognition and artificial intelligence, and the method comprises the steps: generating word-level attribute description for a support set image through employing a multi-mode large language model; encoding the image and the text by adopting a vision-language model; constructing a hyperbolic local feature alignment module in a hyperbolic space, screening most relevant image local features for text local features by calculating hyperbolic cosine similarity, and fusing by using hyperbolic weighted average; designing a hyperbolic cross attention module, and aggregating key information from the multi-modal local features of the support set to construct a category prototype by taking query image aggregation features as guidance; and finally performing classification based on the hyperbolic geodesic distance. According to the method, the hierarchical modeling capability of the hyperbolic space and the semantic priori knowledge of the large language model are fully utilized, fine-grained multi-modal feature alignment is realized, and the small sample image classification performance is improved.

A map-building method for a robot is provided. The method includes: controlling the robot to travel toward a target point; processing the target point according to the exploration point information; and in response to that, the robot arrives at the target point, collecting map-building data of the region to be explored in order to update an environmental spatial map of the robot.

The invention discloses an intelligent flight route planning method and system for a low-altitude logistics aircraft, and the method comprises the steps: obtaining the three-dimensional space map data of the current position of the aircraft, including the coordinate information of a building and a telegraph pole, and the boundary range of a no-fly zone; acquiring wind speed, wind direction and rainfall data of the current area from a meteorological data source, and calculating meteorological impact factors by combining aircraft performance parameters; dynamically adjusting the candidate flight routes according to the maneuvering characteristic parameters of the aircraft, and calculating the flight time, the energy consumption cost and the safety risk score of each route; meteorological changes in the flight process are monitored in real time, and if the meteorological conditions exceed a preset threshold value, the meteorological data are obtained again, and the meteorological influence factors are updated; and carrying out local optimization adjustment on the current flight route according to the updated meteorological influence factor, generating a new flight route segment and replacing the original segment.

According to one embodiment, a method, computer system, and computer program product for dynamic image reconstruction is provided. The present invention may include training a physics-informed neural network (PINN) using received training data; receiving wind field velocity measurements in a geographical domain from one or more weather information sources and air pollutant concentration measurements of one or more air pollutants from the geographical domain using a sparse sensor network to produce a plurality of input data; processing the plurality of input data through multiple physics-informed masked convolutional layers in the trained PINN to perform a three-dimensional partial convolution process; and performing the dynamic image reconstruction on the processed plurality of input data using the trained PINN to generate a reconstructed continuous spatial map of the geographical domain.

The invention discloses a mangrove forest distribution information extraction method based on deep learning and remote sensing images, and the method comprises the steps: firstly, constructing a spatial diagram fusing ecological priori knowledge such as hydrological connectivity, and extracting high-dimensional features containing a complex space-time dependence relation from an irregular satellite image time sequence through employing a PG-STGT network; a physical information guided antagonism decoupling mechanism is subsequently adopted to explicitly decompose the high-dimensional feature into three subspaces. And finally, according to the method, a high-robustness distribution diagram is generated by applying the decoupled identity feature vectors, and a multi-dimensional mangrove forest health index is generated by using residual error abnormal features. According to the method, the physical mechanism and deep learning are deeply fused, the time consistency of the classification result is improved, and the ability of diagnosing the health condition of the ecological system is provided.

The invention discloses a spatio-temporal feature fusion modeling method for multi-source sensing data, and relates to the technical field of data processing, and the method comprises the steps: obtaining multi-source original spatio-temporal data of a sensor, including densely distributed data and sparse dotted data; for the dense distribution data, extracting statistical distribution characteristics of the process state data, calculating an operation entropy and an operation temperature, and generating regional dynamic characteristic data; for sparse point data, extracting space and time sequence statistical characteristics of environment parameters, calculating environment difference degree and environment stability, and generating regional environment characteristic data; generating target spatio-temporal feature data based on unified spatio-temporal granularity alignment, asynchronous missing detection and dynamic interpolation; modeling is carried out through a time sequence branch, a space map branch and a cross-source fusion branch, and a prediction result is output; the sensing accuracy and the prediction reliability of the industrial operation environment coupling system are improved by introducing the dynamic characteristics and the environment characteristics.

The invention relates to the technical field of image recognition, in particular to a method and system for early warning below a portal crane, and the method comprises the steps: obtaining site environment sensing data, constructing a space map, and recognizing the position and motion state of equipment in real time; predicting an equipment motion track and a load swing range, and generating a predicted dangerous area; image acquisition devices are installed at a plurality of fixed positions, field images are acquired, an image recognition model is used for detecting a target, and the spatial position and the posture state of the target are obtained; tracking a target trajectory based on a motion estimation model and predicting a future position of the target trajectory; dynamically switching the image acquisition devices according to the motion state of the equipment; and comparing the future position of the target with the predicted dangerous area in real time, when the future position and the predicted dangerous area coincide, carrying out potential interference prediction, calculating predicted interference time, and carrying out grading processing in combination with the attitude state of the target. According to the invention, equipment dynamic perception and personnel behavior prediction are realized, and the intelligent protection capability of the operation area is improved.

According to the OPC result correction method and the result verification device thereof provided by the invention, the NILS value is introduced as an index for measuring correction quality on the basis of the existing OPC correction method, so that the absolute value of the EPE value of the target graph after OPC correction is ensured to meet the requirement of the preset EPE standard value, and the NILS value of the target graph can be improved; according to the method, the target graph after OPC correction has better spatial graph contrast, and better imaging quality is obtained in the actual exposure process, so that the problem of disconnection or bridging caused by size deviation of the target graph is effectively avoided, and further, the image quality is improved. The better spatial pattern contrast can bring better photoresist morphology and a larger photoetching process window, so that the generation of pattern defects is reduced, and the improvement of the yield of products is more facilitated.

A method of generating a spatial map includes obtaining a captured image of a target area in a space, obtaining light detection and ranging (LiDAR) scan data by scanning a depth of the target area with respect to a first height, performing object detection on the captured image, and, according to the performed object detection, generating a spatial map of the target area, based on all of the LiDAR scan data and the captured image.

The invention relates to the technical field of AGV intelligent control, and discloses an AGV intelligent control method and system based on end-cloud collaboration, and the method comprises the steps: obtaining the position, obstacle and environment data of an AGV in real time, constructing a dynamic space map, carrying out the grid processing, and when the channel width is detected to be not enough for safe passing, carrying out the grid processing of the dynamic space map; the method comprises the following steps: firstly, adopting an A * algorithm to quickly generate an alternative path, then carrying out multi-AGV cooperative conflict detection through path overlapping comparison, carrying out global optimization on a scheduling sequence and a passing priority of multiple AGVs by applying a genetic algorithm, and finally generating a conflict-free efficient scheduling instruction. According to the method, through multi-algorithm fusion and a real-time decision-making mechanism, the passing efficiency and the space resource utilization rate of multiple AGVs in a dynamic environment are improved.

The invention provides a pedestrian trajectory prediction method based on a multi-scale spatial-temporal feature fusion network, and belongs to the technical field of computer vision. The problem of how to effectively learn complex spatio-temporal characteristics and interrelation thereof in pedestrian trajectory prediction is solved. According to the technical scheme, the method comprises the following steps of S1, acquiring data of a data set; s2, using a self-attention mechanism to learn time dependence characteristics of each pedestrian; s3, converting the pedestrian coordinates into a space diagram; s4, cross-space-time interaction features are extracted through multi-head self-attention; s5, time features, space features and cross-space-time interaction features are acquired; and S6, performing multi-track prediction on the fused spatial-temporal features through a time extrapolation convolutional neural network (MTCN). The method has the beneficial effect that multi-track prediction is realized through the multi-layer time extrapolation convolutional neural network MTCN.

The invention discloses a robot sound source localization method, device and equipment based on graph signal processing and a storage medium, and relates to the technical field of sound source localization, and the method comprises the steps: collecting an original voice signal through a robot microphone array; constructing a space diagram topological structure and a time diagram topological structure, and combining the space diagram topological structure and the time diagram topological structure; mapping the space-time diagram signal of the original voice signal to a diagram frequency domain through joint diagram Fourier transform, and calculating the power spectral density of the diagram frequency domain voice signal in each microphone channel; and sound source localization is carried out based on the MPDR filter of the graph frequency domain voice signal. According to the method, the construction method of the graph signal topological structure of the voice signals in the time dimension is designed, the situation that the voice signals change along with time is fully and effectively analyzed, and the feature information of the multi-channel voice signals is comprehensively analyzed from two dimensions in combination with the space graph topological structure constructed through the microphone steering vector according to the space layout.

The invention discloses a spatial data version processing method and device based on Postgres and a readable medium, and the method comprises the steps: importing vector data of a spatial layer into a Postgres database, and generating a layer data table; creating an extended version management table and a branch extension table corresponding to the extended version management table; creating a current user, creating a corresponding local architecture and a local version management table, updating the local version management table and creating a difference table corresponding to the branch extension table, a local version view and a view trigger if the vector data of the space layer is migrated to the local architecture corresponding to the current user; if the current user edits the vector data of the space layer in the local architecture or the branch extension table is updated, and no conflict exists or the conflict is solved, updating the version number of the corresponding layer in the local version management table, and submitting version record information corresponding to the edited content of the current user to the branch extension table; universal version management functions can be realized.

The invention relates to the technical field of water quality monitoring, and discloses a sewage quality on-line monitoring method and system.The sewage quality on-line monitoring method comprises the steps that time alignment processing and data quality evaluation are conducted on original water quality sequences collected at all monitoring point positions, and aligned data and data quality weights are obtained; constructing a time delay space diagram; performing joint estimation on the global water quality state to obtain global state estimation and uncertainty; performing anomaly detection and identifying an abnormal propagation path and a suspected source to obtain an anomaly score and a source confidence sorting list; formulating a self-adaptive sampling strategy, and determining the sampling frequency of each monitoring point position; on the basis of the time-delay observation equation and weighted optimization, online joint estimation of the global water quality hidden state is achieved, reverse traceability is carried out by combining the residual standardization anomaly score and the time-delay space diagram, the anomaly can be found, the possible source and propagation direction of the anomaly can be located, and the pertinence and interpretability of disposal are improved.

The invention relates to a bridge crane operation attitude control method and a control system, which are used for accurately hoisting a heavy object in a narrow space, and the method comprises the following steps: obtaining operation area space constraint information to generate a space map, and determining a hoisting path and a path width; planning a driving track of the driving mechanism along the central line of the path, marking direction switching nodes, and calculating the walking distance of each straight line segment; determining an allowable maximum swing range according to the width of the lower-section path; a speed control scheme is formulated according to the walking distance and the swing space, and it is ensured that weight swing does not exceed the limit during direction switching. By actively planning the swing space and intelligently adjusting the motion parameters, the hoisting efficiency is remarkably improved while the safety is guaranteed.

The invention relates to a ship route planning method and system integrating meteorological data and historical data, belongs to the technical field of ship route planning, and solves the problem of low feasibility of route planning caused by unreasonable space grid division and no consideration of meteorological data in the prior art. Comprising the steps that after historical route data are collected, a space diagram grid is constructed through downsampling and clustering; according to a to-be-planned route and ship information, space nodes corresponding to all the necessary passing points in the space diagram grids are obtained to serve as all the path points; predicting meteorological data and sea condition data of each space node in the space diagram grid, reconstructing the space diagram grid in combination with ship information, and calculating the weight of each edge in the reconstructed space diagram grid; and on the basis of the reconstructed space diagram grid, obtaining an optimal path passing through each path point according to the sequence of each path point and the weight of each edge, and then replacing each path point in the optimal path with each necessary point to obtain a planned route path. And safe and reliable ship route planning is realized.

A robotic device scans at least a portion of an environment. The robotic device includes one or more LiDAR sensors to scan the environment. A management system receives sensor data generated by the one or more LiDAR sensors to determine a three dimensional (3D) point cloud representing at least the portion of the environment. A layout of the environment is determined, where the layout represents a location of objects within the environment. At least one discrepancy between the 3D spatial map and the layout is determined. Based at least in part on at least one discrepancy, a map of the environment is generated.

In various examples, embodiments are directed to identifying map data (e.g., relevant to a route) using a quad-tree spatial index. In this regard, spatial map data that indicates various map features is represented in a quad-tree spatial index for use in identifying map data. To identify map data, bounding shapes may be generated in association with various segments of a route. An indication of an object-oriented bounding shape may be used to query the quad-tree spatial index to identify map data related to the object-oriented bounding shape. In embodiments, an object-oriented spatial index may be generated that indexes the object-oriented bounding shapes associated with the route. The object-oriented spatial index may be used to query the quad-tree spatial index to identify map data related to the corresponding object-oriented bounding shapes. Alternatively, the quad-tree spatial index may be used to query the object-oriented spatial index to identify map data.

The invention provides a large-space real-time positioning method and system based on SLAM and visual graphics, and relates to the technical field of space positioning and intelligent navigation, and the method comprises the steps: obtaining a space scene image and an environment point cloud, and carrying out the preprocessing; detecting the scene image based on a deep learning target detection model, and identifying a dynamic target and a corresponding category and a bounding box; static space feature points of the environment point cloud are reserved, and an initial space point cloud map is constructed; oRB feature extraction is carried out on the static spatial feature points to obtain feature descriptors, and a spatial visual feature map is constructed; performing joint optimization on the initial spatial point cloud map and the spatial visual feature map to generate a multi-modal global spatial map; the dynamic target is subjected to pose estimation in combination with space inertial measurement information to obtain an initial pose, and the initial pose is optimized to obtain a final pose, so that the positioning precision and robustness in a large-space dynamic environment can be improved, the consistency of a multi-modal map can be optimized, and the positioning stability in the dynamic environment can be enhanced.