49 results about "Spatial similarity" patented technology

The invention provides a slope displacement monitoring data processing system based on an unmanned aerial vehicle laser radar, and relates to the technical field of data processing, and the system comprises the steps: carrying out the spatial interpolation processing of a topographic feature data set, and constructing a digital topographic surface model; the displacement field calculation module is used for performing iterative optimization based on a digital terrain surface model through spatial similarity analysis and fusion with a gradient descent algorithm, calculating a slope surface displacement vector field, identifying a potential sliding surface and a deformation abnormal region, and generating a displacement field calculation result; and the evaluation module is used for inputting a displacement field calculation result into a risk evaluation model and carrying out slope stability quantitative evaluation through a multi-source data fusion analysis platform. According to the invention, the practicability and operability of the monitoring result are improved.

The invention discloses a compliance evaluation method, system, medium and equipment for electric power inspection behaviors, and belongs to the technical field of operation and maintenance intelligentization of an electric power system, and the method comprises the steps: obtaining the track data of an inspector, and preset inspection standard data; calculating a time deviation score and a spatial similarity score according to the trajectory data and the inspection standard data; mapping the trajectory data into a three-dimensional grid, generating a space-time cube, extracting space-time features in the space-time cube through a preset convolutional neural network model, and outputting a space-time score; wherein the spatio-temporal characteristics comprise a motion characteristic, a behavior mode characteristic and a path time sequence mode characteristic; and performing weighted fusion on the time deviation score, the spatial similarity score and the space-time score to obtain an inspection behavior compliance evaluation result corresponding to the inspection personnel. Therefore, through implementation of the method, the problems of spatial-temporal feature splitting and insufficient single-dimension scoring precision during power inspection compliance evaluation in the prior art can be solved.

The invention discloses a landslide-debris flow disaster chain assessment method based on a graph neural network, and the method comprises the steps: generating a composite ground feature unit through the watershed segmentation of a composite curvature field and hydrological analysis, abstracting the composite ground feature unit as a geographic node, optimizing a multi-source environment factor through mutual information screening, constructing a directed weighted graph in combination with spatial similarity, and carrying out the reconstruction of a landslide-debris flow disaster chain. An improved GraphSAGE model which introduces neighbor weight, self-loop and residual error information and cancels random sampling is adopted, a sample set adaptive to a less-data area is constructed through a unit splitting-feature matching-terrain fitting mechanism, and model training and risk prediction are completed. The method solves the problems of single disaster assessment, strong sample dependence and high missed judgment rate in the traditional technology, has excellent performance indexes, effectively reduces the missed judgment risk of a high-risk area, and provides accurate and efficient technical support for disaster chain risk assessment of a small watershed in a small-data mountainous area.

The invention discloses a cross-device multi-target tracking method and system. The method comprises the following steps: converting vehicle driving data collected by a camera and a radar into trajectory data under a global coordinate system; dividing all the sensing devices into a plurality of adjacent device pairs; checking the time dimension and the space dimension of the upstream track and the downstream track which are respectively formed by the adjacent equipment pair in sequence; and if the upstream track and the downstream track meet verification, splicing the track sequence according to a timestamp sequence to perform fusion of the global track, otherwise, temporarily storing the track as an isolated track independent of the global track. The system is used for implementing the method. The method has the advantages that feature extraction is not needed, the calculation complexity is low, and the real-time processing requirement of a large-scale sensing network can be met. Spatial similarity calculation methods are respectively designed for devices of the same type and different types, the cooperation requirements of mainstream sensing devices in an intelligent traffic scene are covered, and the compatibility is high.

The invention discloses a main earthquake group intelligent identification method considering a micro-earthquake contour coefficient and spectrogram clustering. The method comprises the following steps: 1, acquiring micro-earthquake monitoring data; 2, acquiring a spatial similarity matrix, an energy similarity matrix and a time similarity matrix, and weighting to form a comprehensive similarity matrix; 3, forming a feature matrix based on the comprehensive similarity matrix, and mapping data points in the feature matrix into two-dimensional space coordinates; 4, clustering data points in the two-dimensional space coordinates by adopting a DBSCAN algorithm, and obtaining a neighborhood radius and a minimum sample number optimization combination based on a micro-seismic contour coefficient; and 5, under the optimization combination of the neighborhood radius and the minimum sample number, clustering data points in the two-dimensional space coordinates by adopting a DBSCAN algorithm to obtain a plurality of micro-seismic event optimization clusters. And 6, screening out a micro-seismic event cluster with the highest comprehensive feature score as a main seismic group. The method is reasonable in design, and the main earthquake group is selected by fusing evaluation indexes of space density, energy characteristics and time characteristics.

The invention provides a surface water body dynamic monitoring method based on multi-source remote sensing and spatial similarity reconstruction, and relates to the technical field of remote sensing detection. The method comprises the following steps: acquiring a to-be-reconstructed water body distribution diagram sequence collected by an optical remote sensor; searching a reference image with the smallest spatial structure difference with the to-be-reconstructed water body distribution map in the reference image set, and taking the reference image as an optimal reference water body map of the to-be-reconstructed water body distribution map; and carrying out pixel replacement on a missing region in each to-be-reconstructed water body distribution diagram by utilizing a pixel category in the optimal reference water body diagram, and finally obtaining a reconstructed water body distribution diagram sequence. According to the method, pixel-level completion is carried out on a to-be-reconstructed water body distribution diagram acquired by an optical remote sensor by using a pre-constructed missing-free reference image set, so that the integrity of a time sequence water body information image acquired by the optical remote sensor within a period of time is ensured, and the continuity of water body information in time sequence is further ensured; the monitoring requirements of large-range and high-dynamic change water bodies can be met.

The application provides a method, device and equipment for identifying a relationship between trajectories. The method of the application connects the starting points of two target trajectories and the ending points of the two target trajectories to form a closed trajectory line containing the two target trajectories. The total area and total perimeter of the closed area surrounded by the closed trajectory line are determined, the similarity distance between the two target trajectories is determined according to the ratio of the total area to the total perimeter, the average time difference between the two target trajectories is determined according to the time stamps of the trajectory points in the trajectory point sequences of the two target trajectories, and whether the two target trajectories have a relationship is determined according to the similarity distance and the average time difference, the spatial similarity and the time similarity. The method can accurately identify whether the two trajectories have a relationship, has low computational complexity, and improves the efficiency of the relationship analysis.

The application discloses a dynamic prototype multi-model medical image classification method based on feature credibility evaluation, and relates to the technical field of medical image processing.The method effectively solves the problems in traditional medical image subtype classification, such as lack of high-quality labeled data, high uncertainty of pseudo-labels, and difficulty of static prototypes in adapting to dynamic changes of lesions, and the like.Through construction of a multi-structure feature extraction network and completion of hierarchical feature fusion, the method combines feature cross learning, local attention modeling and expanded convolution to supplement context information, and strengthens semantic consistency and structural continuity of lesion region features.Meanwhile, the method constructs a spatial similarity graph through cosine similarity, and models and fuses a spatial uncertainty graph with any and cognitive uncertainty based on a Dirichlet distribution, obtains a reliable evidence graph through exponential fusion, and generates a pseudo-label with sample-level confidence, so that effective supervision information is accurately screened from a feature level, cumulative deviation of false pseudo-labels is greatly reduced, and stability of a semi-supervised learning process is improved.

This invention provides a flexible production line control method and system based on a large vertical domain model. The method includes: constructing a pruning index combining temporal stability and spatial similarity for the original large vertical domain model to evaluate the redundancy of the activation tensors in the intermediate layers, and performing structured pruning accordingly to obtain a sparse structure model; performing mixed-precision quantization on the pruned model: allocating quantization bits according to the sensitivity of each network layer weight to task loss, and calculating scaling factors and zeros based on the range of non-zero weights to represent sparse weights; calculating the output deviation between the quantized pruned model and the original model under the same input, and training a feedforward network with the former's output as input to fit and compensate for the output deviation; combining the quantized pruned model and the feedforward network to form a composite control model; after deployment, adding the output command of the quantized pruned model to the compensation vector during inference to generate control commands.

The application discloses a multi-source sensing fusion expressway slope geological disaster real-time early warning method and concretely relates to the technical field of safety early warning monitoring, and is used for solving the problem of false consistency of multi-source data leading to missed reports in the early stage of deep concealed disasters of the existing fusion system; abnormal correlation marks are generated by analyzing energy conversion efficiency differential entropy mutation through collecting time series data of slope multi-source geological parameters; a cooperative evolution matrix is constructed under the mark state; the space-time overlapping area of the zero point of the second derivative of the main deformation parameter and the extreme value point of the frequency domain of the auxiliary parameter in the matrix is searched, and the phase change critical state is compared by comparing the geological history envelope line mark; the hidden variable decoupling is carried out on the critical state matrix, and the concealed disaster risk index is generated based on the spatial similarity of the main feature vector and the disaster mode vector; when the risk index continuously exceeds the dynamic critical value and the frequency domain characteristics show non-steady-state evolution, an early warning signal is generated, the deep disaster critical state is effectively identified, and the missed report caused by the safety illusion is avoided.

The invention provides a multimedia data security measurement method based on multi-dimensional association, and the method comprises the steps: dividing a plurality of features of a single image in multimedia data into a plurality of clustering groups based on spatial similarity through a spectral clustering algorithm for the multimedia data; and judging whether the multimedia data is safe or not by combining the content safety score, the privacy information safety score, the semantic safety score and the formalized safety score, thereby realizing optimization of safety evaluation and calculation efficiency.

The present application relates to remote sensing image processing technical field, specifically a kind of tensor singular spectrum analysis method for hyperspectral image three-dimensional feature extraction, including based on spatial self-similarity adaptive embedding, decomposition and low rank representation based on t-SVD, feature image and classification, compared with prior art, through adaptive embedding and t-SVD process, a new adaptive embedding operation, utilize the spatial similarity feature of HSI, jointly utilize target pixel and non-local similar pixel, combine with re-projection operation, keep target inter-class difference while enhanced intra-class similarity, by designing a trajectory tensor, combined with t-SVD, jointly represent the global low rank feature of HSI, the arrangement of similar pixel in trajectory tensor makes it have low rank feature, and further extract low rank feature by truncated t-SVD, realize the feature extraction of three-dimensional hyperspectral image, and then improve the class separability in hyperspectral image classification.

The embodiment of the invention provides a method and device for automatically generating a Chinese address element knowledge graph of a generative large language model, and the method comprises the steps: constructing a segmentation frame of Chinese address elements based on the semantic features of the address elements, and segmenting a complete address into an address element sequence based on the segmentation frame; the method comprises the following steps: defining address elements based on national standards, and constructing an ontology framework of a Chinese address element knowledge graph with a hierarchical structure and semantic constraints based on the definition of the address elements; based on the ontology framework, performing text classification on the address element sequence by adopting a GPT-based light supervision classification method; according to the ontology framework and the text classification, constructing a Chinese address element knowledge graph based on a Neo4j graph database; and adopting a semantic-spatial similarity calculation method to optimize the synonymous and synonymous entities in the Chinese address element knowledge graph.

Industrial equipment and its progressive distillation unsupervised anomaly detection method, device and medium, relate to the technical field of equipment anomaly detection. The method comprises: acquiring monitoring time series data and preprocessing. The time neighbor set of each sample is calculated in the time dimension. The spatial similarity is measured in the spatial dimension, and the BallTree index structure is constructed through the hierarchical spatial division strategy, and the spatial neighbor set is searched. The current sample is normalized by the spatiotemporal weighted fusion strategy. After normalization, the adaptive threshold unsupervised anomaly detection teacher model is input for training. The student network is trained in stages by using the progressive distillation strategy, the teacher model is used to guide the student model learning, and the distillation student model is obtained. The data to be detected is input into the anomaly detection model, the reconstruction error between the input sample and the reconstructed sample is calculated, and the adaptive threshold judgment mechanism is combined to obtain the anomaly detection result.

The application relates to a power transmission line fault maintenance method and system under severe weather. The method comprises the following steps: performing baseline correction on a traveling wave current signal to identify a wave peak interval corresponding to the traveling wave, determining the symmetry deviation of the wave peak interval according to the difference between the trend item components of two sequences, screening a plurality of suspected fault traveling waves according to the fitting goodness of each wave peak interval, the shape similarity between the two sequences and the symmetry deviation, determining the spatial similarity according to the similarity between each suspected fault traveling wave and a plurality of suspected fault traveling waves at the opposite end, determining the incident waves at both ends of the power transmission line according to the energy intensity of the wave peak interval and the spatial similarity, correcting the initial speed of the incident wave according to the preset corresponding relationship and the main frequency of each incident wave, and positioning the fault point according to the corrected speed and the arrival time of the corresponding incident wave. The method improves the fault positioning accuracy and reliability, and reduces the operation and maintenance difficulty of the power transmission line.

The invention discloses a data linkage analysis method and system for a test operation scene, and belongs to the technical field of data visualization and intelligent monitoring. The field type of each data field is determined by identifying the field type of collected data, extracting the multi-dimensional statistical characteristics of each field and classifying the multi-dimensional statistical characteristics; obtaining a value score of each field type, screening out a plurality of field types meeting a test analysis target value score threshold value requirement, and carrying out priority ranking on the screened field types; constructing a search tree according to the multiple field types subjected to priority ranking and the value scores of the multiple field types, and generating multiple analysis paths by traversing the search tree; and obtaining a statistical feature vector of a multi-field type combination corresponding to each analysis path, and obtaining an optimal recommendation chart type corresponding to each analysis path and a binding analysis scheme thereof by determining a spatial similarity between the statistical feature vector and each chart attribute vector in a chart type library. According to the method, the incidence relation between the key parameters can be captured in time.

The invention relates to the technical field of artificial intelligence and clean room intelligent operation and maintenance, in particular to a clean room intelligent operation and maintenance method and system based on artificial intelligence, and the method comprises the steps: collecting a business data source, an Internet of Things data source and spatial topology configuration data; based on a production scheduling plan, a personnel authority and standard operation program, an equipment ledger and maintenance record and spatial configuration data, constructing an ideal reference space-time tensor organized according to a preset time granularity and a spatial partition; performing parameterized disturbance by using an abnormal event operator to generate an abnormal simulation state set; differentiating the real acquisition data and the abnormal simulation state with the ideal reference to obtain a real residual matrix set and a theoretical residual matrix set; a diagnosis result is generated based on the time sequence and space similarity, and an operation and maintenance work order or a to-be-observed record is generated according to conditions; and collecting execution and re-checking results, updating disturbance parameters and distribution thresholds, and realizing accurate identification and disposal of personnel violation, equipment degradation and logistics retention.

The present application relates to the field of seabed terrain reconstruction, and provides a multi-beam sonar seabed terrain reconstruction method, system, device, product and medium, comprising obtaining an approximation coefficient and an approximation coefficient variance, thereby obtaining a state estimation point, obtaining an initial filtering data point according to the state estimation point, and performing adaptive filtering on the filtering data point to obtain a filtering data point; calculating attention weights and spatial similarity parameters of multi-source data, and then calculating spatial similarity; calculating a message vector according to the spatial similarity, obtaining a final node representation based on the message vector, calculating a data fidelity term, a smoothing constraint term and a terrain regularization term, and obtaining fused terrain data; inputting the fused terrain data into an encoder, decoding through a decoder, performing dense skip connection, and obtaining decoding output; calculating a terrain reconstruction total loss, adjusting model parameters according to the terrain reconstruction total loss, outputting reconstructed terrain data, and completing the reconstruction of the seabed terrain.

This application discloses a target matching method, a target matching device, and a storage medium. The target matching method includes: acquiring a first image and a second image of the same monitored area captured by different cameras; obtaining a first target set from the first image and a second target set from the second image; calculating the spatial similarity distance between each target pair in the first target set and the second target set, wherein each target pair includes one target in the first target set and one target in the second target set; and matching the two targets in corresponding target pairs whose spatial similarity distance is less than a first preset distance. Through the above method, this application can match targets in different images by calculating the spatial similarity distance between two targets and comparing it with a preset distance, thereby achieving the fusion of the same target and improving the accuracy of target fusion.

The invention discloses an urban road waterlogging risk sensing equipment layout optimization method based on a clustering algorithm, and belongs to the field of monitoring point optimization layout, and the method comprises the steps: constructing a road-pipe network double-drainage model; constructing a space similarity index matrix and a time similarity index matrix of road nodes in the road-pipe network double-drainage model; performing primary classification on the road nodes by using the spatial similarity index matrix to obtain an initial clustering result; according to the initial clustering result, carrying out secondary clustering by using the time similarity index matrix to obtain a time-space similarity cluster, and according to the time-space similarity cluster, determining the installation number of the road risk sensing equipment in the target area; and calculating the drainage risk and road node traffic connectivity of each road in the road-pipe network double drainage model, and determining the installation position of the waterlogging risk sensing equipment. The technical problem that the layout of the waterlogging risk sensing equipment is different from the actual waterlogging risk area is solved.

The application provides a workpiece defect detection method based on a self-supervised twin comparison network, first, subgraphs are generated by dividing defect-free workpiece images through a sliding window, and a self-supervised similarity matrix label is calculated based on the spatial overlap of the subgraphs and the original graph; meanwhile, the subgraphs and the original graph are respectively subjected to enhancement processing; the enhanced subgraphs and the original graph are input into a twin feature extraction network, first, iterative optimization is performed through an affine adaptive architecture containing a FastMatch layer and an STNs layer: the FastMatch layer matches the features of the subgraphs and the original graph and outputs the highest similarity feature pair, and the STNs layer generates an affine matrix based on a full connection network to calibrate the perspective of the subgraphs; then, mutual convolution operation is performed on the optimized feature vectors to generate an initial similarity matrix; subsequently, a spatial similarity distribution graph is generated through a convolution network, and network model training is completed; during detection, the image to be detected and the reference image are input into the trained model after being aligned in perspective, and the existence of defects is determined based on the bidirectional similarity matching result and a threshold.

The embodiment of the invention provides a scene positioning method, a scene positioning model training method and a scene positioning model-based information processing method, which are applied to the technical field of deep learning, and the scene positioning method comprises the following steps: obtaining a scene image of a target scene; inputting the scene image into a scene positioning model to obtain target address information of the target scene, the scene positioning model being obtained by training based on a first feature similarity and a second feature similarity, the first feature similarity being a feature similarity between a sample image of the sample scene and the sample address information, and the second feature similarity being a feature similarity between the sample image of the sample scene and the sample address information; the second feature similarity is the geographic space similarity between the sample images of the sample scene. According to the method, the geographic information contained in the sample image is utilized, it is guaranteed that close image features in the geographic space have high similarity, far image features in the geographic space have low similarity, the scene positioning model obtained through training is distributed more evenly in the multi-modal feature space of the image and address information, and the accuracy of scene positioning is improved.

The invention relates to the technical field of data processing, and discloses a grid downscaling random forest model construction method based on a spatial similarity theory, and the method comprises the steps: obtaining original sample grid data of a to-be-trained region and reference sample grid data obtained after downscaling the original sample grid data; determining original characteristic variables in the original sample grid data; performing feature screening on the original feature variables based on the reference sample grid data to obtain target feature variables; and inputting the target characteristic variable into the random forest model, and training the random forest model by taking the reference sample grid data as a label until the training is completed. By screening the original feature variables, it is ensured that the variables of the input model can reflect key factors for driving target spatial variation, the most important information for downscaling is reserved while the number of features is effectively reduced, and not only is the refinement of the grid downscaling ensured, but also the efficiency of the grid downscaling is improved.

The invention discloses an image feature extraction method and device, equipment and a storage medium, and relates to the technical field of computer vision, and the method comprises the steps: extracting an initial image feature of a to-be-detected image; and carrying out multiple times of feature extraction and feature distillation on the initial image features to obtain target features and a background feature set. And based on a channel attention mechanism, performing feature enhancement on the target features and feature points included in the background feature set to determine position information of the feature points. Dividing the initial image feature into a plurality of grid features according to a preset grid; and performing semantic enhancement on each grid feature according to the spatial similarity of the feature point neighborhoods in each grid feature. And superposing the semantic enhancement features and the grid features to obtain local structure features of the feature points. And image distillation is introduced, so that the feature point detection efficiency is improved. And the initial image features are divided by adopting grids, so that the requirement on computing power is reduced. And through semantic enhancement, the detection precision is improved.

A method for UAV non-orthogonal communication and edge scheduling based on semantic symbol matching comprises the following steps: acquiring and partitioning the dataset, constructing an air-to-ground semantic system, matching semantic symbol sequences, correcting semantic errors and recognizing images, constructing a feature fusion strategy network, determining unloading, training the feature fusion strategy network, and testing the feature fusion strategy network. This invention employs a Transformer sequence prediction model, using semantic information to analyze the spatial correlation of the context, and repairs damaged semantic symbols caused by multi-vehicle non-orthogonal collisions or fading, thus improving the success rate. The UAV calculates the direction vector and spatial similarity of the observed signals, and uses the Hungarian algorithm to complete global matching, enabling concurrent vehicle unloading and improving the processing scale of the vehicle-to-everything (V2X) network. A multi-branch feature fusion strategy network is adopted, combining the policy network and the value network to construct a near-end policy optimization framework, achieving adaptive perception of real-time state changes of vehicles and UAVs, and improving system resource utilization.

The invention relates to the technical field of error correction, in particular to an LED packaging machine table error correction method based on data fusion, which comprises the following steps of: acquiring operation data of an LED packaging machine table in a production cycle, constructing an initial operation data set, slicing and dividing the initial operation data set, and meanwhile, dividing the initial operation data set into an initial operation data set; identifying error features by using a hierarchical iteration mechanism; aiming at the identified error operation feedback parameters, correcting the error operation feedback parameters layer by layer by utilizing a double-layer local correction mechanism, and correcting and verifying the corrected operation feedback parameters by utilizing a stability evaluation function; according to the method, a layered iterative error screening mechanism is adopted, so that the effect of step-by-step screening and sorting of error operation feedback parameters is achieved; by adopting an adjacent slice comparison technology based on spatial similarity, an effect of accurately judging error parameters with propagation trends is realized.

Disclosed are various systems and techniques for tracking surgical tools in a surgical video. In one aspect, the system begins by receiving one or more established tracks for one or more previously-detected surgical tools in the surgical video. The system then processes a current frame of the surgical video to detect one or more objects using a first deep-learning model. Next, for each detected object in the one or more detected objects, the system further performs the flowing steps to assign the detected object to a right track: (1) computing a semantic similarity between the detected object and each of the one or more established tracks; (2) computing a spatial similarity between the detected object and the latest predicted location for each of the one or more established tracks; and (3) attempting to assign the detected object to one of the one or more established tracks based on the computed semantic similarity and the spatial similarity metric.

The invention discloses industrial equipment, a progressive distillation unsupervised anomaly detection method and device thereof and a medium, and relates to the technical field of equipment anomaly detection. The method comprises the following steps: acquiring monitoring time sequence data, and preprocessing; and calculating a time neighbor set of each sample in a time dimension. Spatial similarity measurement is carried out on the spatial dimension, a BallTree index structure is constructed through a hierarchical spatial division strategy, and a spatial neighbor set is obtained through searching. And performing spatio-temporal joint normalization processing on the current sample through a spatio-temporal weighted fusion strategy. And after normalization, inputting a self-adaptive threshold unsupervised anomaly detection teacher model for training. And training the student network by stages by adopting a progressive distillation strategy, and guiding the student model to learn by utilizing the teacher model to obtain a distillation student model. And inputting to-be-detected data into the anomaly detection model, calculating a reconstruction error between an input sample and a reconstruction sample, and obtaining an anomaly detection result in combination with a self-adaptive threshold judgment mechanism.

The invention discloses a ship trajectory abnormal behavior detection method based on an interpretable rule model, and relates to the technical field of ship abnormal behavior detection. The method comprises the following steps: preprocessing acquired ship AIS trajectory data; an abnormal behavior rule base is constructed, and track anomalies are divided into spatial position anomalies, local behavior feature anomalies and overall behavior pattern anomalies; based on the spatial similarity of the tracks, completing spatial clustering and identifying a trunk route, and identifying spatial position abnormal behaviors for non-clustered tracks in combination with rules; in each space cluster, a sliding window is adopted to extract dynamic features of the track, behavior feature vectors are constructed and clustered, and a behavior pattern cluster is generated; dynamically generating a self-adaptive anomaly judgment threshold value based on in-cluster feature distribution; and constructing a multi-label random forest classifier, and carrying out joint identification on various types of abnormal behaviors. The method has good interpretability and adaptability, and is suitable for complex sea area ship behavior detection and intelligent early warning tasks.

The application discloses a spherical subarray profiling and full-space coverage multi-beam subarray dynamic cooperative distribution method, and belongs to the technical field of antenna engineering. The method first completes array profiling by adapting the geometric characteristics of a spherical carrier to system operation indexes, realizes subarray profiling of different array requirements, then constructs an integer programming model based on the spatial similarity of array normal and beam pointing, realizes optimal matching of the array and the beam, and finally synthesizes the full polarization domain directional diagram of each beam by using a convex optimization algorithm, so that full utilization of array resources and optimization of beam performance are realized. The application effectively solves the technical problems of low resource utilization, insufficient spatial matching and difficult suppression of beam interference faced by existing conformal phased arrays in a full-space coverage scene, and provides an efficient and robust solution for full-space multi-target cooperative detection and communication in the fields of communication, radar and satellite navigation.