49 results about "Hausdorff distance" patented technology

The invention relates to the technical field of injection molding process control, and discloses an injection molding parameter self-learning collaborative regulation and control system which is characterized in that multi-source parameters such as temperature and pressure are acquired in real time through a data acquisition module, and the multi-source parameters are mapped to a low-dimensional manifold space through a dynamic manifold construction module to decouple a nonlinear coupling relationship; the chaos synchronization control module suppresses external disturbance based on master-slave system design, and the distributed optimization module realizes multi-target real-time decision by adopting a split Bregman algorithm; the execution module converts the optimization instruction into a control signal and triggers closed-loop feedback; and the federated learning architecture realizes cross-device strategy migration by extracting continuous homology features of a manifold space and combining a Hausdorff distance and homeomorphic mapping, and drives a global model to update. According to the invention, the technical problems of difficult dynamic regulation and control of nonlinear parameters, insufficient real-time performance of multi-constraint optimization, low cross-scene repeated parameter adjustment efficiency and lack of self-adaptability of long-term process drift in the injection molding process are solved.

The invention discloses a map information updating method and system, and relates to the technical field of map updating, and the method comprises the steps: according to each road sign object, using an editing distance and a cosine similarity to measure a first similarity between symbol elements in an element set of the road sign object and symbol elements in a map element set; measuring a second similarity between the color elements in the element set of the road sign object and the color elements in the map element set by using the Euclidean distance; measuring a third similarity between a central point position element in the element set of the road sign object and a central point position element in a map element set by using a Hausdorff distance; taking the intersection-union ratio of the contour frame elements in the element set of the road sign object and the contour frame elements in the map element set as fourth similarity; determining whether the four similarities are matched or not through the four similarities; if yes, updating the map. The map can be updated in time, and the real-time performance is good.

The invention relates to the technical field of bolt length detection data processing, in particular to a bolt length detection system, which is characterized in that a three-dimensional data acquisition module generates a high-density point cloud and extracts a thread curvature extreme point, and a dynamic model matching module screens an optimal lead angle parameter based on a Hausdorff distance; and constructing an axial expansion model carrying phase offset gradient features. The projection mapping correction module adopts a genetic algorithm to optimize a compensation coefficient matrix, and generates a dynamic weight correction parameter to suppress geometric distortion; and the phase error correction module combines Fourier spectrum decomposition and an exponential decay weight coefficient to construct a frequency domain transfer function to eliminate an axial accumulative error. According to the method, the problems of projection distortion and phase deviation caused by the lead angle are effectively solved through geometric-phase double-domain correction, dynamic parameter iteration and an error collaborative suppression mechanism, and the spatial positioning precision of the feature points at the tail end of the thread is improved.

The invention relates to the technical field of power equipment monitoring, provides a transformer substation portal frame drainage wire safety intelligent monitoring method based on a point cloud semantic segmentation technology, designs a point cloud semantic segmentation model with higher precision, and introduces an attention mechanism and a feature extraction module into a traditional coding-decoding structure. The extraction of different network layer features is enhanced, and the requirement of a substation scene for segmentation precision is better met. By calculating the Hausdorff distance, the system can monitor the position and the state of the drainage wire in time and give an alarm. Meanwhile, Gaussian filtering and RANSAC methods are adopted, the adaptability of the system to environmental noise and complex terrains is improved, the model is optimized through multi-feature extraction and an attention mechanism, efficient segmentation of complex scenes is achieved, and the generalization ability of the model is enhanced.

The invention discloses a multi-objective optimization method for a radiotherapy plan, and the method comprises the steps: S1, building a multi-objective optimization model, and enabling the multi-objective optimization model to be used for the multi-objective intensity optimization of a radiotherapy reverse intensity modulation plan, and generating a Pareto solution; step S2, inputting a Pareto solution currently generated by the multi-objective optimization model into a weight search network model, wherein the weight search network model predicts a sub-objective weight of a next round of multi-objective intensity optimization of the multi-objective optimization model according to the current Pareto solution; s3, optimizing a single-target strength optimization problem with sub-target weights through a multi-target optimization model to update a current Pareto solution; and S4, circularly executing the steps S2 to S3 until a Pareto solution generated by the multi-objective optimization model forms a Hausdorff distance dHaus (Zin, Zout) between an upper bound and a lower bound of the Pareto in space, the distance dHaus (Zin, Zout) is less than threshold, and the threshold is a predefined threshold parameter which is greater than 0. According to the invention, a deep learning method is creatively used to solve the problem of constraint weight search in a radiation therapy reverse plan multi-objective optimization method.

The invention discloses a ship trajectory clustering method, which comprises the following steps of: cleaning AIS data to obtain a plurality of navigation trajectories of a ship in a set time period; the Hausdorff distance between any two navigation tracks is calculated to serve as the calculation basis of a DBSCAN algorithm; a core point prediction model is constructed, high-probability data points in the navigation trajectory are screened out to serve as candidate core points, and neighborhood query and clustering expansion are performed on the candidate core points through a DBSCAN algorithm; a Bayesian optimization algorithm is improved through an adaptive acquisition function, and hyper-parameters of a DBSCAN algorithm are optimized based on the improved Bayesian optimization algorithm. The DBSCAN algorithm has the beneficial effects that the DBSCAN algorithm is improved aiming at the problem of relatively low operation efficiency of the density clustering algorithm under the condition of large data volume, and the operation efficiency of the algorithm is improved while the calculation precision is ensured. The Bayesian optimization algorithm is introduced to automatically optimize hyper-parameters of the DBSCAN algorithm, and the effect and stability of clustering analysis are improved.

The invention relates to the technical field of image processing, and particularly provides a full-reference image quality evaluation method based on wavelet transform and Hausdorff distance, which comprises the following steps of: realizing efficient evaluation on image quality by combining a frequency domain change technology and deep learning; perceptual relevant features are extracted through wavelet transform, multi-scale perceptual information is captured, and how a human visual system analyzes visual information of various scales and directions in space and frequency domains is simulated. Secondly, the distribution similarity measurement module based on the Hausdorff distance can better evaluate the difference between the feature distribution of the reference image and the feature distribution of the distorted image. The method provided by the invention can accurately capture the perception quality difference without training data or subjective quality scores, and a large number of experiments show that the method is excellent in performance and highlights the ability of the method closely related to the perception of a human visual system.

The invention relates to the technical field of multi-target tracking data processing, and discloses a track association algorithm based on spatial-temporal characteristics, which comprises the following steps of: acquiring track data, constructing an Euclidean distance matrix between targets, screening the targets smaller than a coarse threshold, and constructing a target neighborhood topological structure on the basis; the distance measurement between sets is applied to the neighborhood topological structure, and original distance matrix elements are replaced; calculating the corrected average Hausdorff distance of the track history set; and weighting the neighborhood topological structure distance and the Hausdorff distance to obtain a correlation cost matrix, and realizing track correlation through a linear distribution algorithm. According to the method, a multi-target track association task can be effectively completed, and a good association effect and robust performance are achieved.

The invention relates to the technical field of image processing, in particular to a numerical control machining multi-part positioning method and system based on image processing, and the method comprises the steps: obtaining an image containing a to-be-machined part, extracting an initial edge point, calculating the edge gray transition compactness of the initial edge point, and calculating the edge gray transition compactness of the initial edge point; the quantization module is used for quantizing the concentration degree of edge point gray change; candidate points are obtained through screening based on the edge gray transition compactness, and local curve segments are constructed; and calculating the geometric trajectory smoothness of the candidate points. According to the method, two indexes of edge gray transition compactness and geometric trajectory smoothness are constructed to quantify and distinguish a real workpiece contour and noise interference. Furthermore, a self-adaptive weighted Hausdorff distance matching algorithm is constructed based on the indexes, the sensitivity to abnormal points can be dynamically adjusted, and the positioning precision and reliability in a severe industrial environment are remarkably improved.

The present invention discloses a method and system for fault location of oil and gas pipelines based on multi-sensors. The method includes collecting pressure, flow rate, and vibration data; performing data cleaning and feature extraction to obtain low-dimensional features; constructing a Bayesian network model to analyze the conditional probability between features and generate a feature correlation matrix; when the conditional probability is lower than a threshold, weighted updating the feature values; combining with a pipeline geographic information database to match the topological structure and operation years, and using a weighted voting algorithm to preliminarily locate the fault area; using an LSTM network to predict the fault probability distribution; verifying the location result through the Hausdorff distance, and finally determining the fault location. This method realizes accurate location of the pipeline fault location, improves the pipeline operation and maintenance efficiency, and extends the service life of the pipeline.

The invention belongs to the technical field of computer-aided manufacturing, and relates to a variable-step-size high-precision approximation method for a complex spline curve. The method comprises the following steps: selecting an approximation curve as a straight line or a circular arc according to a local geometric characteristic of an approximated curve, namely a curve curvature, measuring an approximation error between the approximated curve and the approximation curve by using a bidirectional Hausdorff distance, controlling an approximation precision fluctuation range and an approximation curve segment length based on a variable step size method, and controlling an approximation iteration process by using a termination discrimination strategy, therefore, on the premise that the approaching progress is guaranteed, the length of the minimum approaching curve section and the number of machining codes are increased as much as possible, and then the machining efficiency and the machining quality are guaranteed.

The invention relates to the technical field of natural language processing, in particular to an event extraction method based on entity representation and side information enhancement. The method comprises the following steps: fusing part-of-speech information, sequence information and an attention mechanism guided by an entity type through an entity representation enhancement module to optimize entity recognition, and solving sample imbalance and label transfer constraints by adopting a weighted cross entropy loss function; constructing a document graph comprising pseudo event nodes, entity nodes and context nodes, modeling a node relationship through three types of directed edges, and initializing edge feature vector explicit coding semantics; aggregating global information by using a graph attention mechanism, and dynamically integrating cross-sentence event elements through pseudo event nodes; the event decoding module adopts a Hausdorff distance loss function to measure the difference between a prediction event set and a real event set, and event type classification and argument role allocation are achieved. According to the method, the F1 value on a financial announcement data set reaches 83.9%, and the problem that document-level event elements are dispersed is effectively solved.

A system (210, figure 2) and a method 400 is used for predicting a collision between components of a vehicle. The method includes collecting, by a controller (208), data associated with at least two c

This invention discloses a fast e-commerce information retrieval method based on deep learning, comprising: acquiring an e-commerce product information database to obtain a set of product semantic fragments; constructing and training an improved Triplet network; collecting e-commerce search query text to generate a set of query semantic fragments; performing vectorization encoding to form a product vector set library; performing vector aggregation processing to obtain query and product representative vectors; performing vector similarity calculation to select a preset number of candidate products; executing the Hausdorff distance algorithm to output a preset number of products as the retrieval results. This invention, by introducing an improved Triplet network and the Hausdorff distance algorithm, achieves fast candidate filtering and high-precision retrieval of query and product information composed of multiple semantic fragments in e-commerce scenarios, while also meeting key attribute consistency constraints.

This invention discloses a triangular mesh coarsening method for broadband electromagnetic scattering analysis, belonging to the field of electromagnetic computation. The method first identifies the characteristic edges, characteristic edge points, and corner points of the mesh, and constructs normal or tangent vectors at the nodes according to the mesh node type. Then, a local Bezier surface is constructed, and local operations are performed on the mesh edges to ensure that the mesh and the local Bezier surface satisfy the Hausdorff distance constraint. Next, a target size field is constructed and smoothed based on curvature estimation and the Hausdorff distance constraint. Finally, guided by the size field, coarsening iterations are performed through edge folding operations, and mesh quality is optimized through edge swapping and vertex movement. This invention reduces the mesh size while quantitatively controlling geometric errors and accurately preserving geometric features, thus improving simulation efficiency and reliability.

The application discloses a kind of ruled surface wire cutting rough machining path generation and sequencing method, it is related to subtractive manufacturing and numerical control machining automatic path planning technical field.The specific includes: based on differential evolution algorithm to find viewpoint, it is evaluated using comprehensive objective function including volume item, uniformity item, barycenter offset item and overhang constraint term;Along the direction of view point, extract projection profile and utilize cubic B-spline curve to approach, according to Hausdorff distance to compensate offset to eliminate collision interference to generate path;Iterative execution until meeting preset termination condition obtains path set;Each path is modeled as three-node model including entry, internal and exit node, is converted into traveling salesman problem and uses safe sphere to establish weight between paths, determines optimal cutting order.The purpose is to solve the problem of uneven distribution of excess amount, poor machining stability and long idle time in wire cutting rough machining, realize efficient, stable and automated processing.

The application discloses a kind of optimal envelope trajectory tracking control methods of space non-cooperative target under multi-target constraint, mainly include the following steps: design the full-drive multi-finger space capture configuration with adaptive configuration adjustment and anti-escape capability, generate effective multi-finger envelope configuration using bidirectional Hausdorff distance based on finger feature and target edge point feature, and construct multi-finger dynamic capture domain according to target motion characteristics and joint feature length;Desired trajectory parameterization is carried out using continuous high-order Bezier curve, and smooth envelope trajectory is planned under minimum base disturbance based on multi-target constraint and multi-modal particle collaborative learning mechanism;Through the nonlinear coupling characteristics of inverse dynamics compensation system of multi-finger, design multi-finger collaborative hierarchical tubular model predictive control strategy based on disturbance observer to realize envelope trajectory tracking control.The method can effectively overcome the interference caused by target shape complexity, motion uncertainty, nonlinear dynamic coupling, realize the accurate tracking control of space non-cooperative target trajectory.

The application relates to the technical field of image intelligent analysis, and specifically discloses a feature extraction method and system based on deep learning and imageomics, which comprises the following steps: through calculating the overlap rate of deep learning and imageomics ROI boundaries and the absolute difference of Z-score standardization of GLCM-entropy, double determination and accurate identification of ROI boundary double inconsistency are realized; based on a spatial attention mechanism, a pixel weight distribution model is constructed in combination with texture abnormality, abnormal weight elimination, smoothing normalization processing and adaptive filtering of pseudo boundaries are carried out, a unified coordinate system with the imageomics boundary center as the origin is re-constructed, image standardization is realized by minimizing and optimizing affine transformation parameters through the Hausdorff distance, finally, a three-order Bezier curve is used to fit the dislocation area and iterative smoothing correction is carried out, high-precision unified ROI boundaries are established, image correction and efficient fusion of heterogeneous features are realized, and the effectiveness of coupled feature representation is improved.

The application provides a layout fuzzy matching method and device based on Hausdorff distance, a medium, a program product and a terminal. The first layout and the second layout to be executed for fuzzy matching are obtained, and a preprocessing operation is performed on the coordinate point sets in the layouts respectively. The Hausdorff distance calculation operation is performed on the first coordinate point set and the second coordinate point set after the preprocessing operation, and the maximum Hausdorff distance is obtained based on the calculation result. Whether the first layout and the second layout are consistent is determined by judging whether the maximum Hausdorff distance is less than or equal to a preset threshold. The application solves the problem that the existing layout fuzzy matching method has long operation time and low efficiency when processing large-scale layouts. The efficiency and accuracy of layout fuzzy matching are significantly improved.

The application discloses a ship trajectory clustering method, comprising the following steps: cleaning AIS data to obtain the sailing trajectories of a plurality of ships in a set time period; calculating the Hausdorff distance between any two sailing trajectories as the basis for calculation of a DBSCAN algorithm; constructing a core point prediction model and screening high-probability data points in the sailing trajectories as candidate core points; performing neighborhood query on the candidate core points and carrying out clustering expansion through the DBSCAN algorithm; improving a Bayesian optimization algorithm through an adaptive acquisition function, and optimizing the hyperparameters of the DBSCAN algorithm based on the improved Bayesian optimization algorithm. The application has the beneficial effects that: in view of the problem that the operation efficiency of the density clustering algorithm is relatively low in the case of a large amount of data, the DBSCAN algorithm is improved, the operation efficiency of the algorithm is improved while the calculation accuracy is ensured, the Bayesian optimization algorithm is introduced to automatically optimize the hyperparameters of the DBSCAN algorithm, and the effect and stability of the clustering analysis are improved.

The invention discloses a ruled surface linear cutting rough machining path generating and sorting method, and relates to the technical field of subtractive manufacturing and numerical control machining automatic path planning. The method specifically comprises the following steps: searching a viewpoint based on a differential evolution algorithm, and evaluating by using a comprehensive objective function comprising a volume item, a uniformity item, a gravity center shift item and a suspension constraint item; a projection contour is extracted in the viewpoint direction, approximation is carried out through a cubic B-spline curve, and offset compensation is carried out according to the Hausdorff distance to eliminate collision interference to generate a path; iteratively executing until a preset termination condition is met to obtain a path set; each path is modeled into a three-node model containing an entrance node, an internal node and an exit node, the three-node model is converted into a traveling salesman problem, a safety spherical surface is used for establishing weights between the paths, and the optimal cutting sequence is determined. The method aims at solving the problems of non-uniform allowance distribution, poor machining stability and long idle feed time in linear cutting rough machining, and efficient and stable automatic machining is achieved.

The application provides a Bloom filter configuration method for a hash connection, comprising: sampling initial data, analyzing the initial data, dynamically analyzing a variety of change trends, fitting a change trend model, constructing a Bloom filter, evaluating a Bloom filter size, and adjusting a Bloom filter configuration. The application combines probability statistics and mathematical modeling to dynamically analyze part of data sets and estimate code value inclination, uses a change slope and a coefficient of variation to judge a stable trend of data, uses Hausdorff Distance to select an optimal model to estimate a Bloom filter, finally proposes different configuration Bloom filter strategies, efficiently configures a Bloom filter size, fully utilizes memory resources, and improves overall query efficiency of the hash connection.

The invention discloses a distribution network differential protection method based on current amplitude-phase characteristics and a distance algorithm, and relates to the technical field of distribution network differential protection, comprising: obtaining real-time power parameters at both ends of a protection installation in a target detection distribution network area, extracting current phasors and calculating a first signal amplitude characteristic ratio and a phase angle difference at both ends of the protection installation to calculate a first differential protection coefficient, and calculating the Hausdorff distance at both ends based on a Hausdorff distance algorithm, thereby obtaining a second differential protection coefficient; judging the fault state twice according to the first differential protection coefficient and the Hausdorff distance, calculating a reliability improvement coefficient and a safety improvement coefficient, and comprehensively analyzing the fault type of the fault distribution network in combination with the current phasors; the method overcomes the problem that traditional current differential protection lacks comprehensive data analysis of multi-terminal quantities, resulting in unreliable fault judgment and poor protection performance, significantly improves the distribution network differential protection performance and fault judgment efficiency, and improves the reliability of the distribution network differential protection.

The invention discloses a method for evaluating urban space boundary recognition accuracy, a medium and a program product. The method comprises the steps of obtaining a coordinate point set created based on a research area range; obtaining a reference set and a prediction set; a preset evaluation step is repeatedly executed, an overall analysis accuracy evaluation result is obtained, and the evaluation step comprises the steps that the intersection-to-union ratio is calculated based on the reference set and the prediction set, and then the matching relation of polygons in the reference set and the prediction set is judged; for the uniquely matched polygon pair, calculating a Hausdorff distance and a turning function distance; the number of the polygon under the non-unique matching condition and the number of the polygon under the unique matching condition with the intersection-to-union ratio smaller than a set threshold value are fed back to the analytical model, so that the analytical model is locally optimized; and respectively marking polygons in the reference set and the prediction set in the coordinate point set to obtain a first list and a second list, and carrying out confusion matrix analysis to obtain an overall analysis accuracy evaluation result. According to the invention, overall and local evaluation can be carried out on urban space boundary analysis accuracy.

This application discloses a method, system, and related devices for flexible load restoration in distribution networks, belonging to the field of power system fault recovery. The method includes: combining a flexible load extreme operation model with distribution network power flow constraints to form the constraints for the distribution network operator; transforming the opportunity constraints in the operator's constraints into deterministic constraints, and then calculating the distribution network's operational safety domain using a vertex search method based on Hausdorff distance; constructing a distribution network load restoration model based on the operational safety domain; solving the distribution network load restoration model based on the operational safety domain; and performing decision optimization at the distribution network level during power restoration based on the solution results. This method performs load restoration while considering the distribution network's operational safety domain, ensuring the safety and robustness of the power grid restoration process, and ultimately helping the distribution network operator make informed decisions.

The present invention discloses a star map recognition method based on the fusion of neural network and Hausdorff distance, belonging to the fields of astronomical navigation, guidance and control. The method includes the following steps: constructing a navigation feature database and a neighboring star index library; building a self-organizing competitive neural network and integrating the calculation method of Hausdorff distance into neurons; training the neural network model; star map recognition, adopting the same feature extraction method for observed stars as that for navigation stars, extracting the feature vectors of observed stars and inputting them into the neural network to obtain the star map recognition result. The present invention combines the respective advantages of neural network and Hausdorff distance in the field of star map recognition. The feature information of navigation stars is complete and sufficient, and the matching process is simple. Under the influence of factors such as parameter deviation, star point position deviation, magnitude deviation, and false star magnitude interference in star sensors, it has strong robustness and a high recognition success rate.

The invention relates to a Hausdorff distance-based user-transformer relation matching method, equipment and a medium. The method comprises the following steps of S1, acquiring an electricity consumption measurement sampling sequence of M transformers and corresponding Y users within a set length time period in a set area range; s2, clustering the Y users into N user groups by adopting a clustering algorithm according to the electricity consumption measurement sampling sequences of the Y users, and determining an electricity consumption data sequence corresponding to each user group; and S3, for each user group, calculating the Hausdorff distance between the power consumption data sequence of the user group and the power consumption measurement sampling sequence of each transformer, and selecting the transformer corresponding to the minimum value of the Hausdorff distance as a matching transformer of all users in the current user group. Compared with the prior art, the method has the advantage of high user change relation matching accuracy.

The present invention discloses a multi-endpoint star map recognition method based on an optimal path. The method comprises the following steps: S1. constructing a navigation star library and, based on the concept of a minimum weight spanning tree, constructing a navigation feature library containing navigation star characteristic patterns; S2. processing an observed star map and, based on the concept of a minimum weight spanning tree, generating characteristic patterns of observed stars; and S3. matching and identifying the navigation star characteristic patterns with the observed star characteristic patterns based on the bidirectional Hausdorff distance. The present invention integrates graph theory knowledge and the optimal path problem with star map recognition methods, utilizing the concept of a minimum weight spanning tree to extract star point characteristic patterns. Due to the unique distribution of star points, the characteristic pattern of each star point is ensured to be unique, thereby increasing the accuracy of star map recognition.

The present invention discloses a method, system and storage medium for identifying and counting cracks in an image. First, a semantic segmentation algorithm is used to identify crack pixel information, and the crack pixel information is clustered to distinguish each discontinuous crack and remove noise pixels. Then, a judgment is made as to whether two cracks are the same crack, and the judgment criteria include whether the Hausdorff distance between the crack endpoints of the two cracks and the angle between the direction vectors related to the two cracks meet the requirements. Two cracks that are considered to be the same crack are merged, and then iterative calculations are performed again until all cracks cannot be merged into a new crack. Finally, accurate crack count statistics are achieved. The present invention uses the crack direction characteristics to analyze and count cracks, and can classify scattered cracks with regular direction as the same crack. It has good practical value for cracks with obvious direction, unclear texture, and partially obscured areas.

The invention discloses a method for longitudinal protection of main transformer of wind farm based on waveform dissimilarity factor, comprising the following steps: collecting current at protection installation places on both sides of transformer and performing phase correction and per-unit processing on the current; calculating Hausdorff distance between phase currents on both sides in data window in real time; judging the interval where Hausdorff distance between phase currents on both sides is located, if it is between reward threshold and penalty threshold, assigning reward and penalty coefficient to 0, if it is less than penalty threshold, assigning reward and penalty coefficient to -1, except for the above situation, assigning reward and penalty coefficient to 1; calculating waveform dissimilarity factor between phase currents on both sides in data window in real time by reward and penalty coefficient, when it is greater than setting threshold, it is determined as fault in zone. The invention is not affected by frequency offset characteristic of wind farm, and has good anti-excitation surge current and current transformer saturation capability, and has obvious superiority compared with traditional transformer current differential protection.