26 results about "Coordinate vector" patented technology

The invention relates to a BIM-based water conservancy project safety early warning method and device, equipment and a medium. The method comprises the following steps: acquiring multi-source data such as an osmotic pressure gradient field based on a predefined hydraulic engineering monitoring point space coordinate list, and generating a three-dimensional osmotic field matrix after noise reduction, standardization and interpolation processing; physical characteristics of the matrix are extracted by using a pre-trained space-time convolution model, and abnormal parameter labels are determined through threshold comparison and deviation calculation; associating the exception to a BIM component node through a spatial topology matching algorithm; and finally, performing visual early warning based on the BIM abnormal coordinate vector, and generating a structured alarm instruction set by combining risk levels and abnormal type matching treatment measures. According to the method, through multi-source data fusion modeling, feature analysis and extraction, BIM space mapping, three-dimensional visualization decision making and the like, the precision, efficiency and intuition of hydraulic engineering safety early warning are improved, and a reliable technical scheme is provided for guaranteeing safe and stable operation of hydraulic engineering facilities.

The invention discloses an abnormal data processing method and system for a virtual game, and relates to the technical field of big data processing. Receiving an interaction data frame sequence uploaded by the terminal equipment, and analyzing to obtain an input equipment sampling point set and a displacement coordinate vector; performing feature extraction on the sampling point set of the input equipment to obtain a high-dimensional feature vector; mapping the high-dimensional feature vector to a preset behavior feature space, calculating a multi-dimensional statistical distance of the high-dimensional feature vector in the behavior feature space, and generating a first abnormal confidence coefficient; calling space mapping data corresponding to the virtual scene; performing discretization ray stepping detection in the space mapping data based on the displacement coordinate vector to generate a second abnormal confidence coefficient; performing fusion calculation on the first abnormal confidence coefficient and the second abnormal confidence coefficient based on a dynamic weighting algorithm to obtain a comprehensive abnormal score; and judging abnormal data through the comprehensive abnormal score, and executing a state rollback operation for the virtual object.

A phase-difference-based high-resolution azimuth estimation algorithm for a small-aperture array with multiple targets comprises the following steps: 1) according to various array structures, the coordinate vectors of the array elements are subtracted and then combined accordingly to obtain new vectors along each coordinate axis; 2) the phase of each array element is obtained by performing a fast Fourier transform on the array element sampling data, the phase difference between the array elements is obtained by subtracting the phase of each array element, and the phase differences are combined according to the combination of the array element coordinate vector differences to extract the component k of the incident direction along each coordinate axis. x , k y and k z 3) After obtaining the various components of the incident direction, the unambiguous incident angle across the entire spatial range is calculated using inverse trigonometric functions. This invention utilizes a combination of array element sampling data to achieve more accurate target direction estimation, enhances resistance to noise and other interference, and broadens the applicable frequency band, enabling flexible application to various arrays, as well as direction estimation using multi-line spectrum signals and broadband signals.

The invention discloses a rapid whole-network coordinate resolving method for a large-scale GNSS (Global Navigation Satellite System) reference station network, which comprises the following steps of: dividing a GNSS observation station into a plurality of sub-networks, and applying loose constraint to each sub-network to resolve and generate a clock error and a track file; independently carrying out PPP floating point calculation on each subnet by using the clock correction and the track file; combining the PPP floating point solution results of all the subnets, calculating the PPP solution of the whole net, and obtaining the coordinates and covariance matrixes of all the stations under the original framework; an independent baseline set is constructed, a relative coordinate difference vector is generated, and the independent baseline set converts the station coordinate vector into a baseline difference vector through a coefficient matrix; and introducing a known coordinate of the reference station under the new reference frame, and reversely deducing coordinates of other stations of the whole network based on the relative coordinate difference vector and the known coordinate of the reference station to complete reference frame conversion. According to the method, the independence and robustness of the conversion process are enhanced, and the coordinate precision and reliability of a large-scale network under a unified reference framework can be ensured.

The application relates to the technical field of radar signal processing, and provides an extended target detection method and system in the case of uncertain interference information. The application avoids the precision loss of a multi-rank interference model by modeling to-be-detected data of rank-one interference as an unknown vector in a binary hypothesis testing model, and adapts to a radar countermeasure environment with uncertain interference directions. Parameters are estimated based on a to-be-detected data matrix and a training sample data vector under a zero hypothesis, and the uncertainty of a steering vector caused by flicker interference is avoided. A test statistic is constructed by combining an interference coordinate vector, an interference amplitude vector and a noise covariance matrix, a complex detection problem with multiple unknown parameters is converted into a statistical detection problem under the condition that the parameters are known, the model uncertainty caused by the unknown parameters is eliminated by using actual measurement estimation results, a detection threshold is determined in combination with a system false alarm probability, the detection potential is fully tapped on the premise that the false alarm is controllable, and the detection capability and reliability are effectively improved.

The invention belongs to the technical field of road engineering, and particularly relates to a night road waterlogging condition inspection method based on a laser beam, which comprises the following steps of: acquiring a road surface image containing the laser beam; extracting a laser line pixel point coordinate vector and a plurality of lane line pixel point coordinate vectors; extracting a reference line coordinate vector corresponding to the laser line pixel point coordinate vector, and calculating intersection point coordinates of the reference line coordinate vector and each lane line; constructing a mapped reference mapping laser line coordinate vector; constructing a ponding coordinate vector; s5, continuously collecting high-definition pavement images, executing the steps S2 to S5, obtaining the driving distance between each frame of image, and drawing a plane distribution diagram of the road ponding according to the ponding coordinate vector and the intersection point coordinate of each lane line; according to the scheme, laser line and lane line information is extracted through an image processing technology, and the ponding area is analyzed and determined through an algorithm, so that rapid and accurate monitoring of road ponding is realized.

The application relates to the technical field of radar signal processing, and provides an extended target detection method and system under uncertain interference direction vectors. N K A matrix is modeled into a rank-one interference and noise, thereby adapting to a multi-channel extended target matrix form echo.The rank-one interference is constructed into a combined form composed of a coordinate vector and an amplitude vector in a known subspace, so that performance loss caused by mismatch of a multi-rank interference model is avoided. Parameter estimation is respectively carried out under a zero hypothesis composed of only the rank-one interference and noise clutter and an alternative hypothesis containing target signals, algorithm performance is optimized under the condition that all parameters are completely unknown and there is no theoretical optimal detector, and the adaptability of a detector to uncertain working conditions of a flickering interference direction vector is improved. A test statistic is constructed by combining parameters estimated under the zero hypothesis and the alternative hypothesis, a detection threshold is determined in combination with a system false alarm probability, and detection capability and reliability are effectively improved.

The invention discloses a universal grid coding method based on multidimensional dimensions, and relates to the technical field of dimension coding. The method comprises the following steps: establishing a dimension scale grid, and discretizing a dimension to be coded through the dimension scale grid to obtain a dimension level width; based on the dimension hierarchy width, calculating to obtain a dimension coordinate vector and a hierarchy maximum bit width; based on the hierarchical maximum bit width, performing binary conversion and zero padding operation on the dimensional coordinate vector to obtain a dimensional binary hierarchical sequence; the method comprises the following steps of: calculating Shannon entropy of a dimension to be coded to obtain a dimension priority weight, and performing dislocation coding on a dimension binary hierarchical sequence by adopting a Hilbert coding method based on recursive structure judgment based on the dimension priority weight to obtain a dimension sequence code; and performing secure confusion and enhanced optimization on the dimension sequence code by adopting a gray code transformation method based on chaotic mapping to obtain a dimension optimization code, thereby realizing grid coding of the dimension to be coded.

The present invention relates to an image recognition method, which comprises the following steps: inputting an image into a detection model to obtain a heat map tensor, a reference depth tensor, a weight tensor, and a sub-target tensor; obtaining K position index values from the heat map tensor; obtaining a fusion tensor based on the weight tensor and the sub-target tensor; obtaining a predicted depth tensor based on the fusion tensor and the reference depth tensor; referring to the K position index values, extracting K vectors from the predicted depth tensor; performing a transformation of a projection matrix on the K vectors to obtain K coordinate vectors in the real space. The method of the present invention can complete two tasks simultaneously through a one-time inference, namely detecting an object and detecting sub-targets included in the object, without building a model based on individual tasks, thereby reducing the latency on the consumer terminal.

The invention discloses a decision analysis method and system for a standard material purchasing system, and relates to the technical field of purchasing decision, and the method comprises the steps: collecting a historical material purchasing data flow, extracting a sum value, a supplier credit rating, and logistics cost, and combining the sum value, the supplier credit rating, and the logistics cost into a basic feature set; carrying out big data analysis attenuation modeling on the aging attribute of the basic feature set, establishing a hyperbolic attenuation model and outputting a dynamic attenuation factor; mapping the basic feature set into a coordinate vector, inputting the coordinate vector and a dynamic attenuation factor into a Lorentz compression engine, and compressing and storing the coordinate vector and the dynamic attenuation factor into a quotation memory bank; and converting the purchasing scheme into an electronic order, distributing the electronic order to a supplier platform, collecting performance time delay data, and calibrating a dynamic attenuation factor to complete a closed loop. According to the method, dynamic modeling is carried out on the aging attribute of the basic feature set through the hyperbolic attenuation model, and the dynamic attenuation factor is enabled to be adaptive to the market environment change by using a nonlinear coupling mechanism of the market fluctuation rate and the logistics cost drift factor.

The invention relates to the technical field of computer graphics and artificial intelligence, and discloses a mobile terminal-oriented lightweight 3D scenic spot model dynamic generation and rendering method, which comprises the following steps of: firstly, constructing a global material primitive library at a server side, and parameterizing each primitive into a material space coordinate vector; then, a neural field function is independently trained for the scene partition, and the function learns to map space coordinates to field density and a material mixed vector; on the mobile terminal, according to a viewpoint dynamic loading function, generating a geometric surface in real time by using field density through a volume rendering principle; and meanwhile, performing proximity search and weighted interpolation in the material primitive library by using the material mixed vector, and synthesizing a high-precision surface material in real time. According to the invention, huge geometric and texture data are implicitly coded into a lightweight network, so that the data storage and transmission pressure is remarkably reduced, and high-fidelity real-time rendering on performance-limited equipment is realized.

This invention discloses a method for constructing a digital twin for a water electrolysis hydrogen production system, relating to the field of digital twin technology. The method includes: collecting multi-source operational data from the water electrolysis hydrogen production system, extracting all sampling points, constructing proximity relationships between all sampling points to form a proximity connection graph; traversing the proximity connection graph, organizing the shortest path distances between all pairs of sampling points to form a high-dimensional spatial distance matrix, using multi-dimensional scaling transformation to map the sampling points to a low-dimensional space, establishing a correspondence between the low-dimensional coordinate vectors and the high-dimensional original data, forming a digital twin state representation model; based on the state representation model, constructing a probabilistic graphical model containing three types of nodes: components, sensors, and state variables; setting initial node confidence and information transmission rules, iterating until the confidence stabilizes; performing corrections based on node state deviations; and iterating in real-time according to the correction cycle to achieve real-time synchronization between the digital twin and the water electrolysis hydrogen production system.

The application belongs to the field of radar target detection, and discloses a radar distributed target detection method and system when a fixed interference direction vector is mismatched. In order to solve the problem that the prior art cannot guarantee reliable detection performance when the fixed interference direction vector is mismatched, the application provides a radar distributed target detection method when the fixed interference direction vector is mismatched. First, the noise covariance matrix is estimated based on auxiliary data. Then, the amplitude vector of the fixed interference and the direction vector coordinate vector, the target coordinate vector, and the noise power mismatch amount are estimated based on the data to be measured. Further, the detection statistic is calculated based on the generalized likelihood ratio criterion, and the detection threshold is determined according to the false alarm probability. Finally, the size of the detection statistic and the detection threshold is compared. If the former is larger, the target exists; if the latter is larger, the target does not exist. The application can guarantee reliable detection performance when the fixed interference direction vector is mismatched.

The invention discloses a dynamic parameter adjustment small sample behavior recognition method based on text feature guidance, and relates to a computer vision technology. Giving a video data set, randomly extracting frames of each video to form a new video frame sequence, and inputting the new video frame sequence into a visual feature extractor; a traditional linear layer reconstructs an extensible basis matrix library, each linear layer is decoupled into multiple groups of basis parameter matrixes, and the basis parameter matrixes jointly form a parameter space substrate. And inputting the labels of the video frame sequence into a text feature extractor through a text template to obtain text features. The text feature input coordinate vector calculation module uses text information as semantic guidance, and linear layer parameters suitable for a specific task are constructed by generating combined coefficients of a plurality of parameter matrixes. And calculating centroid mutual exclusion loss and contrast clustering loss to realize the characteristics of high cohesion and low coupling among the basis parameter matrix groups. And inputting the video features obtained by the feature extractor into a predictor to obtain a prediction result, calculating a loss function and updating the gradient. And classification performance is improved.

The application discloses a vehicle scratch similarity recognition method and system, and belongs to the technical field of computer vision. The vehicle scratch similarity recognition method comprises the following steps: determining the number of vehicle scratches in each of a first image and a second image to be recognized, and the type and position of each scratch; extracting and classifying sub-images in the first image and the second image according to the type and position of the scratches to obtain the image categories of the sub-images; marking feature points of each sub-image according to the image categories and a preset feature recognition model, wherein the preset feature recognition model is used to determine the coordinate vectors of the feature points of the sub-images; and comparing the coordinate vectors of the feature points of each sub-image in the first image with the coordinate vectors of the feature points of each sub-image in the second image to determine the similarity between the vehicle scratches in the first image and the vehicle scratches in the second image. The problems of low artificial inspection efficiency and low picture recognition accuracy in the related art are solved.

The invention discloses a high-stability structure deformation performance inversion method. The method comprises the following steps: dividing an overall structure body into a plurality of finite structure units; obtaining a structural unit end face parameter set; constructing a structural attitude mapping model and obtaining an attitude angle set of grid cells; constructing a modal response prediction model and obtaining a modal coordinate vector prediction value; and carrying out inversion to obtain an overall strain field of the structure. According to the method, full-field deformation inversion can be realized in real time and with high precision through pose reconstruction and modal coordinate inversion technologies in the absence of comprehensive sensor layout, so that the measurement cost is remarkably reduced, and the prediction precision and the interference resistance are improved.

The invention is suitable for the technical field of slip surface inversion, and provides a slip surface geometric inversion method coupling InSAR deformation data and numerical simulation, and the method comprises the steps: obtaining SAR image data, DEM data and drilling data of a landslide region; constructing a two-dimensional geologic model of the landslide area in three-dimensional simulation software based on the DEM data and the drilling data; optimizing the z coordinate vector by using a particle swarm optimization algorithm to obtain an optimal z coordinate vector which has a minimum fitness function and meets a preset spatial constraint; and constructing a sliding surface curve based on the optimal z coordinate vector. According to the invention, the efficiency and objectivity of sliding surface inversion can be improved.

This invention relates to a method and system for calculating the angle and grid coordinates of a large-size chip array. The method includes: obtaining the connection relationship between adjacent lines based on the pixel coordinates of each chip, and performing main direction identification and cyclic angle calculation using a histogram to obtain the cyclic angle and average distance of the chip array; constructing two coordinate vectors of the chip array based on the cyclic angle and average distance, and obtaining the array angle of the chip array based on the proximity of the coordinate vectors to 90 degrees; constructing preliminary grid coordinates for each chip based on the coordinate direction vectors, and performing block processing on the chip array; converting the local grid coordinates of all chips into chip array grid coordinates based on the local grid coordinates of the chips within the chip block, combined with the coordinate offset relationship between the chip block and the reference block. This invention can stably and noise-resistantly calculate the chip array angle, and effectively avoids coordinate drift of large-size wafers through block processing, achieving accurate chip grid coordinate positioning.

The invention provides a measurement optimization configuration method and system considering new energy space distribution difference, and the method comprises the steps: generating an initial configuration scheme based on the latitude and longitude coordinate vectors of a terrain outer contour curve in a to-be-configured region; generating a measurement area matrix based on the initial configuration scheme, and calculating a integrating degree index in combination with measurement data; and establishing a multi-target optimization configuration model by taking the minimum equipment configuration number and the highest data integrating degree as targets, and solving a Pareto solution set by adopting an epsilon-constraint method to obtain a configuration scheme set corresponding to the equipment number and the data integrating degree. The optimal measurement point configuration scheme is generated under the condition that the constraints of cost and precision are considered, the arrangement of redundant measurement points is effectively reduced, the overall cost of the system is reduced, meanwhile, the precision of real-time environment data collection is improved, and a reliable guarantee is provided for the improvement of the accuracy of load prediction and the new energy consumption efficiency.

The present application belongs to but is not limited to the technical field of target detection, and particularly relates to a radar distributed target detection method and system based on data dimension reduction. First, the to-be-detected data and auxiliary data are projected into a signal subspace to realize data dimension reduction. Then, the noise power mismatch, target echo amplitude vector, signal direction vector coordinate vector and noise covariance matrix are estimated based on the dimension-reduced test data and auxiliary data. Further, a detection statistic is calculated, and a detection threshold is determined according to the false alarm probability. Finally, the size of the detection statistic and the detection threshold is compared. When the detection statistic is greater than the detection threshold, the target exists, and vice versa. The present application projects the to-be-detected data and auxiliary data into the signal subspace, effectively reduces the data dimension, simplifies the target detection mathematical model, and realizes reliable detection performance when the amount of auxiliary data is limited.

This invention discloses a method for analyzing the tension characteristics of a wave-shaped riser, comprising: determining the structural parameters of the wave-shaped riser system and the marine environmental parameters; constructing a three-dimensional model of the wave-shaped riser system, discretizing the riser into multiple reduced beam elements using the absolute nodal coordinate method, obtaining the nodal coordinate vectors of the reduced beam elements, and deriving the mass matrix of the reduced beam elements; calculating the element strain energy corresponding to axial strain and bending strain, and obtaining the generalized elastic force of the elements; calculating the distributed load and concentrated force acting on the riser, and converting them into generalized external forces of the system, obtaining the generalized external force vector of the system; assembling the mass matrix, generalized external force vector, and nodal coordinate vector of each element to form the overall system dynamic equation; solving the system dynamic equation to obtain the absolute nodal coordinate vector of the riser at different times, and then calculating the tension along the entire length of the riser. This invention can accurately predict the tension distribution of the riser in complex marine environments.

Provided are systems and methods for determining 3D structure and 3D motion of a protein molecule from 2D or 3D particle observation images. The method includes: initializing pose parameters and unknown model parameters; performing image formation which includes: generating one or more 3D deformation fields by inputting a latent coordinate vector into the one or more flow generators; performing a convection and projection operation; and performing CTF corruption; fitting the unknown model parameters to the experimental images by gradient-based optimization of an objective function; latent variable search for a given experimental image which includes: performing the image formation one or more times to generate simulated images; and selecting one or more latent coordinate vectors based on similarity; and updating the at least one of the unknown model parameters which includes: generating simulated images; and evaluating the objective function; computing the gradient of the objective function.

The invention discloses a window state recognition method based on corner prompt and image feature interaction, and relates to the technical field of demand analysis, and the method comprises the steps: inputting a whole frame of RGB image into a main coding network, carrying out the feature coding of the input image, and outputting a global shared feature map Fglobal; performing angular point prompting and position coding processing, performing normalization processing on the angular point coordinates to form an angular point coordinate vector, and inputting the angular point coordinate vector into a position coding network; according to the invention, a multi-layer perceptron and a corner space representation feature generated by sine and cosine position coding are combined, and deep interaction of a global feature and a corner prompt feature is realized through a lightweight feature transformation module and a trans-attention fusion module, so that the system can complete efficient feature enhancement and discriminant modeling under the condition of low calculation amount; the reasoning capability and the deployment efficiency of the model are remarkably improved, and the method is suitable for embedded terminals and intelligent monitoring scenes.

This application provides a joint calibration method, joint calibration device, and joint calibration system. The method includes: acquiring multiple first coordinate vectors and multiple second coordinate vectors when the robot's motion trajectory is a straight line; acquiring the coordinates of each trajectory point of the camera trajectory and the coordinates of each trajectory point of the first radar trajectory when the robot's motion trajectory is a circular arc; fitting the first coordinate vectors and second coordinate vectors using a least squares algorithm to obtain a target rotation matrix; transforming the coordinates of each trajectory point of the first radar trajectory to the camera coordinate system based on the target rotation matrix to obtain the coordinates of multiple transformed trajectory points, which form the second radar trajectory; and calculating the vector between the center of the circle corresponding to the camera trajectory and the center of the circle corresponding to the second radar trajectory to obtain the target translation vector. This method solves the problem in the prior art where the camera and radar often do not share a common field of view, causing the target-based calibration method to fail.

This application provides a method for structural deformation and load analysis, wherein the load includes resultant force and resultant moment of cross sections. The method includes: performing modal analysis on a finite element model constructed based on a target structure to extract strain modes, displacement modes, nodal force modes, and nodal moment modes; determining the structural deformation of the target structure based on the displacement modes and modal coordinate vectors by superimposing multiple displacement modes; determining the resultant force of the target structure based on the resultant force modes and modal coordinate vectors, wherein the resultant force modes are obtained by summing the nodal force modes corresponding to all nodes on the target cross section; and determining the resultant moment of the target structure based on the resultant moment modes and modal coordinate vectors, wherein the resultant moment modes are obtained based on the moment reference points, nodal coordinates, cross section normal vector directions, nodal moment modes, and nodal force modes corresponding to all nodes on the target cross section.

This invention relates to a method and system for assisting in the identification of knee alignment anomalies considering anatomical constraints. The method includes: acquiring the original aspect ratio of a raw medical image; adaptively filling the raw medical image based on a comparison between the original aspect ratio and a target aspect ratio; using the filled image as input to a depth coordinate regression model to obtain normalized coordinate vectors of key feature points based on global anatomical constraints; converting the normalized coordinate vectors into physical coordinates in the original coordinate system; calculating the angle between the femoral mechanical axis vector and the tibial mechanical axis vector, and calculating the vertical distance from the center point of the knee joint to the total mechanical axis of the lower limb; and generating knee alignment anomaly identification prompts. Compared with existing technologies, this invention has advantages such as accurate geometric measurements, strong robustness, high clinical deployment efficiency, and reliable identification results.