67 results about "Computational geometry" patented technology

The invention relates to the technical field of size detection, in particular to a component size detection system based on machine vision, which comprises an interference area identification module, an edge repair module, a structure clustering module, a distortion analysis module and an attitude restoration module. According to the method, the gray level change trend of each pixel and the adjacent points in the image is obtained, the high-reflection suspected interference area is identified, the edge response cancellation processing is executed, and linear interpolation compensation is carried out by combining the effective edge response values of the adjacent areas; after a continuous contour trend is constructed, clustering and sub-grouping division are carried out according to an included angle difference trend between contour geometric vectors, a three-axis direction response linear array variation amplitude is extracted to construct a distortion curvature mapping table, existence of non-equal-interval deformation is judged, reverse angle compensation is applied, and a three-axis intersection point convergence process is completed through a nonlinear optimization algorithm; and according to the three-dimensional coordinates, calculating a geometric dimension and outputting a detection result, thereby realizing self-adaptive detection of different postures and surface interference component dimensions.

The invention discloses a polygonal and quadrilateral mesh automatic subdivision method based on deep reinforcement learning, relates to the field of computational geometry and mesh generation, and constructs an automatic subdivision framework fusing geometric priori knowledge and a data driving strategy. According to the method, nine basic topology filling templates covering triangles to hexagons are predefined as discrete action spaces, and continuous geometric segmentation is converted into a sequence decision problem; learning a mapping relation between the polygon state characteristics and the optimal template selection strategy by using a deep Q network, and balancing exploration and utilization through a dynamic epsilon-greedy mechanism; and in combination with a vertex number priority scheduling strategy and a multi-dimensional quality award function, guiding an intelligent agent to adaptively generate a high-quality quadrilateral grid. According to the method, full-automatic and high-robustness subdivision of the complex polygon area is achieved, the grid orthogonality and the length-width ratio quality are remarkably improved, manual intervention is avoided, and the method is suitable for engineering scenes such as finite element analysis.

The invention discloses a topological network global optimization method based on graph reinforcement learning, and relates to the field of computational geometry and grid optimizing.The method comprises the steps that a topological optimization process is modeled into a Markov decision process, a topological network containing singular points and angular points is converted into heterogeneous graph representation, and a graph state is constructed; selecting a topology editing action by a reinforcement learning agent fused with the graph neural network based on the current graph state; after executing the action, updating a topological structure and a graph state, and calculating an instant reward based on the global grid quality index change; training intelligent agents in batches through experience playback and time sequence differential learning, so that the intelligent agents learn a mapping strategy from a graph state to an optimal action; and applying the trained policy network deployment to automatic optimization of the new topology. According to the method, full-process automation of topological optimization is achieved, non-local negative effects of local modification are effectively avoided, a global optimal optimization strategy is learned, the grid quality and optimization efficiency are remarkably improved, and the method is suitable for simulation grid generation of aircrafts.

The invention discloses a dynamic obstacle recognition and avoidance method based on laser and visual feature level fusion, and relates to the technical field of robot autonomous obstacle avoidance. According to the method, a laser radar and an industrial camera are arranged on the head or the front of a robot, point cloud and image data are collected in real time, geometric features and semantic features are extracted, and the robot autonomous obstacle avoidance is achieved. Generating a fusion feature set F1 through adaptive geometric semantic mapping and graph structure feature fusion; calculating a geometric semantic consistency index CSI based on the geometric offset, the semantic feature gradient and the reprojection residual error, and performing dynamic correction; a modal reliability weight index MRI is calculated through the CWG-Net, and bimodal features are compensated to generate F2; further obtaining the structural stability, the behavior sudden change probability and the visibility of the dynamic obstacle, calculating a dynamic behavior risk index DRI, marking a dangerous obstacle and planning an obstacle avoidance path; and a dynamic obstacle avoidance strategy is generated by using the prediction model and closed-loop control is performed, so that safe autonomous obstacle avoidance and real-time response of the robot in a complex dynamic environment are realized.

The invention relates to the technical field of new energy power generation and control, and discloses a new energy station operation decision support system based on multi-objective optimization, which comprises a multi-source state space reconstruction module, a Riemannian manifold geometry engine module, a self-adaptive inertia Hamiltonian evolution module and a symplectic geometric integral and instruction mapping module. The system collects station data to construct a dimensionless state space, constructs a Riemannian metric tensor according to physical constraints to reconstruct a Riemannian manifold space, and calculates a geometric connection strength factor; the factor is used to adaptively modulate a virtual inertia matrix, and a dissipative Hamiltonian kinetic model is constructed; and finally, solving the steady-state generalized coordinates through a pungent-preserving numerical integration algorithm, and decoding the steady-state generalized coordinates into an equipment control instruction. According to the method, physical constraints are converted into geometric measurements, and a self-adaptive inertia mechanism is introduced, so that the problem of optimization convergence under multivariable strong constraints is solved, and the safety and accuracy of a control instruction are ensured.

The invention provides a method for generating a highway engineering feasibility research report based on local large model training, which is based on a locally deployed large model, integrates multi-source data, and realizes mapping and verification from a fuzzy policy to a structured field through regular extraction of a standard field and construction of a highway engineering knowledge graph. The method adopts TensorRT to accelerate reasoning, injects a specification term table, optimizes term consistency in combination with comparative learning, and generates a specification text through LoRA fine tuning and a reward function. After route parameters are input, geometric features are calculated, a disaster database is associated to recommend a detour scheme, and chapter content is generated. And outputting a formal report based on the template, and ensuring data consistency through cross validation. RPA is adopted to capture standard updating, Kafka and Fiss vector libraries are combined to achieve compliance verification, knowledge graph incremental updating and conflict correction are triggered, and a full-process automatic and intelligent feasibility research report generation system is formed.

The invention provides an adaptive grid generation method and system based on reinforcement learning, and belongs to the field of computer graphics and computational geometry, and the method comprises the steps: S1, according to the input of a known model, constructing a background grid, calculating the curvature characteristics and proximity characteristics of the known model, and generating a corresponding size field; s2, inputting a to-be-predicted unknown model into the information transfer neural network, and predicting to obtain a discrete size field of the unknown model; carrying out interpolation processing on the discrete size field to obtain a continuous size field of the unknown model; and S3, according to the continuous size field and the known model, introducing reinforcement learning, using a leading-edge propulsion method to generate an adaptive grid as a training data set for training the supervised training network, and generating an adaptive grid of an unknown model through the leading-edge propulsion method. On the premise of ensuring the rationality and geometric quality of the grid topological structure, the method remarkably improves the grid precision and overall generalization ability of the key area, and has a wide application prospect.

The invention discloses a convex hull algorithm optimization and applicability analysis method based on rotation contraction. The method comprises the following steps: step 1, processing a point set by a rotation contraction preprocessing strategy; and step 2, convex hull calculation. According to the method, the filtering operation is executed, a large number of non-convex hull points in the point set are deleted, the convex hull is calculated based on the classical convex hull algorithm, the convex hull algorithm optimization and applicability analysis method based on rotation contraction is constructed, the computer memory can be saved when the convex hull is calculated based on the method, meanwhile, the convex hull calculation time can be shortened, and the calculation efficiency is improved. And technical support is provided for the fields of computational geometry, computer graphics, computer vision and the like.

This invention discloses a ground-air collaborative mapping method based on semantic features and ground plane multi-constraint fusion. The method includes: generating sub-map sets from local point cloud data collected by fire trucks and drones; and generating rotation-invariant semantic descriptors using a maximum height value rasterization strategy and rotation group enhancement. By calculating geometric and semantic distances, and combining isomorphic and heterogeneous pattern weighted fusion, candidate sub-map pairs with loop closures are selected and geometrically registered to generate loop closure constraints. Simultaneously, ground point clouds are extracted using the RANSAC algorithm, and the ground plane model is smoothly updated using a sliding window to construct height and pose alignment constraints. The loop closure constraints, ground plane constraints, and pose information are input into the Ceres optimization framework for joint optimization to generate a globally consistent overall map, effectively improving the accuracy and robustness of ground-air collaborative mapping, and is suitable for emergency scenarios such as complex fire rescue.

This invention relates to the field of computer technology, providing a method, apparatus, device, storage medium, and program product for process feature recognition. The method includes: inputting a three-dimensional model into a process feature recognition model to obtain recognition results output by the model; for any given process feature, determining the geometric type of each base surface based on the recognition results; determining the parameter calculation method for each base surface based on its geometric type; and calculating the geometric parameters of the process feature based on the parameter calculation method. This invention can automatically calculate geometric parameters, thereby improving the efficiency and reducing the cost of geometric parameter calculation; and by determining corresponding parameter calculation methods for different base surface geometric types, it can more accurately calculate the geometric parameters of the process feature, thus improving the accuracy of geometric parameter calculation.

The application discloses a UAV path planning method based on computational geometry and deep reinforcement learning, constructs a spatial position model of a UAV and an object, takes a bandwidth utilization index, a photo coverage rate and an energy consumption index as evaluation indexes, adopts a reinforcement learning model for training, and then obtains an optimized UAV shooting action and path. The application can significantly improve the UAV shooting efficiency, the object coverage rate of the UAV shooting and the number of redundant photos of the UAV shooting, thereby reducing the bandwidth consumption.

The present application relates to the technical field of motor design, and particularly relates to a permanent magnet motor iron core size optimization method based on a particle swarm algorithm. The method comprises the following steps: selecting geometric parameters such as stator tooth width to construct particles and initializing a particle swarm; in each round of iteration, an agent model constructed in advance is used to predict the electromagnetic performance index of the particles and to calculate the fitness value; a gradient vector is constructed by perturbing the global optimal solution, and the maximum value of the cosine similarity of the gradient vectors between geometric parameters is calculated as the correlation attribute value; the speed update logic is dynamically corrected in response to the correlation attribute value: an inertia weight inversely proportional to the correlation attribute value is applied to the strongly coupled parameters to suppress oscillation, and an inertia weight proportional to the gradient module length is applied to the weakly coupled parameters to accelerate convergence; until the convergence condition is met, the optimal iron core size is output. The present application effectively solves the oscillation problem caused by strongly coupled parameters, and significantly improves the optimization accuracy and efficiency.

The invention discloses a sparse planting type fruit tree crown adhesion segmentation method based on multi-gradient buffering, and belongs to the technical field of crossing of agricultural remote sensing and geographic information systems, and the method comprises the steps: initializing a working space and parameters, and importing fruit tree crown surface data after semantic segmentation; performing geometric area calculation and screening on the crown surface data, removing small impurities, and distinguishing large adhesion crowns from small independent crowns; performing multi-gradient nested buffering and erasing operation on the large-adhesion tree crowns, and separating adhesion areas; reconstructing a single tree crown surface through line-plane conversion and invalid element elimination; combining single-plant crown surfaces and small independent crowns, and counting the number of fruit trees. The method can efficiently realize single plant separation and accurate plant number statistics of the adhered crowns, provides data support for agricultural insurance claim settlement and yield estimation, is high in operation automation degree, and is suitable for large-scale fruit tree planting areas.

The application discloses an interactive catalytic method and platform based on molecular simulation and artificial intelligence, and realizes intelligent design of a catalyst through data classification, condition optimization, structure modeling and mechanism analysis. After being standardized, the catalytic data uploaded by a user is divided into two categories: data set A is used for recommending optimal reaction conditions by a multi-objective optimization algorithm; and the catalyst structure of data set B is optimized by quantum chemistry software to construct a three-dimensional model, and multiple intermediate, transition state structures and geometric / electronic descriptors are optimized. Through feature screening and dimension reduction technology, a double-model collaborative prediction system of model E and model F is constructed, and finally, model F is used to drive a molecular generation algorithm to screen high-performance catalysts. The application can realize flexible adaptation to specific application scenarios, meet the urgent needs of the industry for efficient and accurate catalyst performance prediction and intelligent design, and effectively help the industry to reduce costs and increase efficiency, and accelerate the catalyst research and development process.

The invention discloses a robot measurement system calibration and uncertainty evaluation method. Firstly, a robot motion error model is established, and an error transmission link is defined; after a theoretical framework is established, N groups of end effector pose data are measured, geometric parameter errors are calculated, and meanwhile, non-geometric parameter errors caused by an approximate external load are experienced. Secondly, estimating discrete probability distribution of related parameters, and obtaining a continuous probability density function and features through approximate fitting; and finally, setting a confidence probability, substituting the confidence probability into the error model to carry out large-scale pseudo-random simulation, obtaining error probability distribution of the end effector, and completing evaluation by combining a measurement mean value and standard uncertainty. According to the method, an uncertainty core source is locked, a parameter influence path is defined, finite sample information is fully utilized, the problems of distribution representation deviation and distortion are avoided, and a precise theoretical basis and reliable data support are provided for uncertainty evaluation of a high-dimensional parameter system.

The application provides a video rain removal method and system based on a space-time difference state space model, and belongs to the technical field of deep learning. The application solves the problem of how to ensure accurate rain removal while retaining the background texture to the greatest extent and breaking the trade-off between "rain removal completeness" and "background fidelity". The method comprises the following steps: S1: extracting shallow space-time features from an input video; S2: constructing anchor points and reference features by using multi-plane scanning, generating a gating map by calculating geometric consistency, and removing rain information in the features by using a difference mechanism; S3: during training, capturing local space-time details by using orthogonal plane convolution, during inference, enhancing the video background texture by using a single-stream convolution after reparameterization, and outputting a rain-removed video. The system comprises a 3D convolution layer module, a consistency-guided difference gating module, and a reparameterized three-plane feedforward network.

The present application relates to a five-axis head RTCP parameter calibration and precision detection method based on a tool setting gauge, and relates to the technical field of machine tool five-axis head precision calibration and detection; including clamping a ball head detection rod, activating the tool setting gauge, executing a measurement and detection program at a set angle, calculating and updating RTCP related parameters, automatically calculating geometric precision and outputting results. The present application executes the prepared calibration algorithm, uses the ball head detection rod clamped on the spindle of the five-axis head and the tool setting gauge installed in the working area of the machine tool for calibration and detection, calculates the five-axis head RTCP parameters that need to be updated through the algorithm, tests the related rotary geometric precision of the five-axis head after the data is updated, and outputs a test report. The items that originally need to be tested manually are optimized for automatic testing, human errors and other disturbances are excluded, the efficiency is improved, the utilization value of the tool setting gauge is improved, and the cost of purchasing the corresponding 3D measurement probe for users with automatic calibration needs is saved.

The application provides a method for generating a feasibility study report of highway engineering based on local large model training, based on the large model deployed locally, integrating multi-source data, extracting standard fields through regularization and constructing a highway engineering knowledge graph, realizing the mapping and verification of fuzzy policies to structured fields. Adopting TensorRT to accelerate inference, injecting a standard terminology table, combining contrast learning to optimize terminology consistency, and generating standard text through LoRA fine-tuning and reward function. After inputting the route parameters, calculate the geometric characteristics, recommend the avoidance scheme by associating the disaster database, and generate chapter content. Based on the template, output the formal report, and ensure data consistency through cross-validation. Adopting RPA to grab the standard updates, combining Kafka and Faiss vector library to realize compliance verification, triggering incremental update and conflict correction of the knowledge graph, forming a full-process automated and intelligent feasibility study report generation system.

The invention belongs to the technical field of laser surveying and mapping, and particularly relates to a basic surveying and mapping geographic entity rapid extraction method and system based on laser three-dimensional scanning, and the method comprises the steps: S100, carrying out the density-adaptive voxelization downsampling of an original laser point cloud, and calculating geometric, reflection and context features from the original point cloud in a multi-scale neighborhood of each representative point; s200, constructing a hypergraph, generating initial hyperpoints through spectral clustering, iteratively performing semantic classification and geometric consistency check, and dynamically merging or splitting the hyperpoints to obtain a refined hyperpoint set with stable semantic tags; s300, subgraphs are constructed, edge weights reflect geometric and boundary compatibility of combination between the super points, and similar super points are clustered into physical entity instances by calculating a maximum spanning tree and cutting at weak joints; and S400, carrying out geometric fitting and regularization on the point cloud of each entity instance, generating a parameterized model, calculating a spatial relationship between entities, and constructing a global semantic topological graph containing complete attributes and relationships.

The invention relates to a dynamic environment vision SLAM (Simultaneous Localization and Mapping) method based on double-source dynamic mask tracking. The method comprises the following steps: inputting a color image into a lightweight instance segmentation model for reasoning, generating a lagged semantic dynamic mask of a prior dynamic object, and asynchronously transmitting the lagged semantic dynamic mask to a front-end tracking thread; in a front-end thread, key points are detected based on a lagged semantic dynamic mask, optical flow tracking is carried out to a current frame, an overlook depth histogram is generated through the current frame, and a semantic dynamic mask of the current frame is obtained through screening and reconstruction. The geometric dynamic mask is calculated after the region is eliminated, and after the geometric dynamic mask which is not detected in the previous frame is supplemented, the geometric dynamic mask and the semantic dynamic mask are combined and collected to obtain a final dynamic mask. And filtering dynamic features by using the final dynamic mask, completing pose estimation and back-end optimization based on static features, updating a TSDF map by only using a static part, removing a changed foreground structure during revisit, and outputting a synchronous positioning and mapping result. By adopting the method, the positioning precision and the mapping quality of the visual SLAM system of the small unmanned platform can be improved.

The present application relates to the technical field of size detection, in particular to a component size detection system based on machine vision, which comprises an interference area identification module, an edge repair module, a structure clustering module, a distortion analysis module and a posture restoration module. In the present application, by obtaining the gray level change trend of each pixel and its adjacent points in the image, the high reflection suspected interference area is identified and the edge response is invalidated, the linear interpolation compensation is performed in combination with the effective edge response value of the adjacent area, the continuous contour trend is constructed, the clustering and sub-grouping division are implemented according to the difference trend of the angle between the contour geometric vectors, the three-axis direction response line array change amplitude is extracted to construct a distortion curvature mapping table and judge the existence of non-equal interval deformation, the reverse angle compensation is applied, and the three-axis intersection point convergence process is completed through a nonlinear optimization algorithm, the geometric size is calculated according to the three-dimensional coordinates, and the detection result is output, so that the adaptive detection of the size of components with different postures and surface interference is realized.

This invention proposes a map data interaction method and system based on urban road networks, comprising: acquiring open-source geographic data; identifying topological relationships between road entities using a computational geometry-based topological relationship reasoning algorithm; constructing an urban road knowledge graph; anchoring semantic nodes through spatial indexing, extracting their neighborhood association subgraphs, and using viewport adaptive computation and force-directed layout algorithms to focus the association subgraphs on the knowledge graph view, establishing a mapping relationship from map to knowledge graph; parsing the geometric attributes of semantic nodes, using a viewport fitting algorithm to drive map positioning and zooming, and establishing a mapping relationship from knowledge graph to map; receiving the user's original query statement, performing intent recognition and structured query generation through a large language model, and combining the knowledge graph retrieval results with the current intent recognition results to generate predictive decision support information; this method improves the semantic understanding depth, human-computer interaction efficiency, and decision-making level of urban traffic.

A method for generating a relief representation model using a plurality of N pairs of stereo images, each of the N pairs of stereo images being associated with a map of photogrammetric disparities which is obtained by correlation calculation based on the pair of stereo images in question, comprises: a calculation of maps of geometric disparities based on a predetermined relief representation model, by projection of the predetermined relief representation model into an epipolar geometry of the pairs of stereo images; a calculation of a cost function, representative of a difference between the geometric and photogrammetric disparities, associated with the predetermined relief representation model; and an updating of the predetermined relief representation model by optimization of the gradient descent type of the cost function, and iteration until a predefined endpoint criterion is reached. Also a computer program and a system.

The invention belongs to the technical field of twinborn simulation, and particularly relates to a bridge jig frame point location intelligent arrangement method based on a digital twinborn model, and the method comprises the steps: obtaining a three-dimensional solid model of a bridge component, carrying out the discretization processing, and calculating the geometric potential energy density based on the plate thickness and principal curvature of an extracted grid infinitesimal central point; calculating a thermal softening index at a grid infinitesimal central point according to the welding seam track set and the welding parameters, superposing the thermal softening index with the geometric potential energy density to obtain the supporting demand density, distributing an initial jig frame supporting point set on the grid, constructing discrete deviation energy, and carrying out iterative optimization on the jig frame supporting point set by adopting a weighted Thiessen polygon algorithm to obtain the supporting demand density. The updated jig frame supporting point coordinates are obtained; and carrying out physical interference detection and edge constraint correction on the updated coordinates of the support points of the jig frame to generate a construction drawing. According to the method, on-demand distribution and global optimization of the supporting point positions are realized, and the manufacturing precision of the component is guaranteed.

The invention belongs to the field of computational geometry, and particularly discloses an arithmetic progression-based three-dimensional discrete dot matrix generation method and application thereof, and the method comprises the following steps: S1, defining an initial set; s2, generating an initial dot matrix; s3, recursively generating a nested dot matrix; s4, constructing a uniform-edge polyhedron; through a three-dimensional discrete dot matrix generation algorithm based on an arithmetic progression and derivative geometric object construction, dot matrix density and distribution are accurately regulated and controlled through dynamic tolerance, cube symmetry is kept so as to adapt to engineering standardization, the method can be expanded to an n-dimensional space, connection lines are selected according to symmetry of an n-dimensional hypercube, and the number of the connection lines is larger than that of the n-dimensional hypercube. And the nested structure is generated through iteration of symmetric connecting line intersection points, all segmentation does not need to be re-calculated, high efficiency is achieved, and self-similarity is kept.

This invention discloses a method for calculating inter-satellite communication windows for remote sensing satellite missions, belonging to the field of satellite communication technology. The technical implementation steps are as follows: Setting the orbital parameters and simulation time period of the satellite constellation; establishing an inter-satellite communication geometric constraint model considering Earth obstruction and maximum distance limitations; based on orbital dynamics principles, analytically deriving the upper bound of the distance change rate and the upper bound of the line-of-sight tangent point altitude change rate, characterizing the relative motion of the constellation; initializing the simulation time; calculating the geometric margin based on the current relative positions of the two satellites; and calculating the adaptive safety step size for the current time in conjunction with the aforementioned upper bounds of the change rates; if a visibility state jump is detected, numerical iterative search is used within the step size interval to lock the window boundary. This invention introduces dynamic boundary parameters to construct an adaptive step size mechanism, significantly reducing the number of invalid calculations while ensuring the accuracy of communication window calculation, thus significantly improving the computational efficiency of inter-satellite link planning.

The application discloses a kind of polygon quadrilateral grid automatic subdivision methods based on deep reinforcement learning, it is related to the field of computational geometry and grid generation, the application constructs a kind of automatic subdivision framework of fusion geometry priori knowledge and data-driven strategy.The method predefines 9 kinds of basic topological filling templates covering triangle to hexagon as discrete action space, converts continuous geometry segmentation into sequence decision problem;The mapping relationship between polygon state characteristics and optimal template selection strategy is learned using deep Q network, and exploration and utilization are balanced through dynamic epsilon-greedy mechanism;Combining vertex number priority scheduling strategy and multidimensional quality reward function, guide agent to adaptively generate high-quality quadrilateral grid.The application realizes the full-automatic, high robustness subdivision of complex polygon region, significantly improves the grid orthogonality and aspect ratio quality, avoids manual intervention, and is suitable for finite element analysis and other engineering scenarios.

This invention discloses a parametric generation method for structured meshes of a three-dimensional pipeline system, comprising: acquiring the geometric structure of the three-dimensional pipeline system; dividing the geometric structure into multiple parts and acquiring the geometric parameters of each part; acquiring the coordinates of the starting section and the coordinates of the center point of the starting section in the coordinate system based on the diameter of the starting section of the geometric structure; calculating the coordinates and normal vector of the center point of the starting section of the geometric structure; performing a spatial matrix transformation on the coordinates of the starting section of the geometric structure based on the geometric parameters and the normal vector of the starting section of the geometric structure to obtain the interface coordinates and the center coordinates; acquiring the coordinates of the interface coordinates; the interface of the geometric structure includes the starting section of the geometric structure; generating a dictionary file of the interface coordinates and corresponding geometric parameters of each part of the geometric structure in the three-dimensional pipeline system according to the blockMeshDict rule using Python; and executing the dictionary file to generate the structured mesh of the three-dimensional pipeline system.