12 results about "Vector angle" patented technology

The invention discloses a steel truss extraction method based on point cloud segmentation and feature analysis, and belongs to the technical field of three-dimensional point cloud processing. The method mainly comprises the following steps: firstly, carrying out multi-scale filtering denoising on an original point cloud and constructing an octree index; local geometric features such as normal vectors, curvatures and first-order plane fitting errors of all the points are calculated; based on the adaptively selected seed points, carrying out region growth by combining a normal vector included angle, a curvature difference and an adaptive edge criterion to realize preliminary segmentation; and finally, a coarse-to-fine multi-layer cascade classifier is adopted to screen the segmented areas, and the steel truss point cloud is accurately extracted through geometric dimension filtering, shape feature screening and multi-dimensional feature matching in sequence. The method does not need pre-training, has the advantages of strong anti-interference capability, high extraction precision and high calculation efficiency, and is suitable for engineering scenes of bridge detection, building surveying and mapping and the like.

The present application relates to a kind of alignment methods of the oblique angle of the oblique end face of pigtail, belong to optical fiber detection technical field, comprising the following steps: step 1) using ANSYS software adopts stereoscopic scanning method scanning and gathers the model data of pigtail and rectangular capillary;Step 2) the oblique angle that the end of pigtail needs to be ground is set in advance, the end of pigtail is simulated to be ground, and the deformation of pigtail grinding is analyzed;Step 3) using vector angle method analyzes the oblique angle of the end of pigtail after grinding, detects the oblique angle that the end of pigtail needs to be ground and the oblique angle of the end of pigtail after grinding;Step 4) the oblique angle of the end of pigtail after grinding is processed;The beneficial effects of the present application: by comparing the oblique angle that the end of pigtail needs to be ground and the oblique angle of the end of pigtail after grinding, the deviation between the oblique angle of the oblique end face of the pigtail after correction or detection grinding and the oblique angle of the oblique end face of the pigtail actually needed is reduced.

This invention provides a training method and an evaluation method for a trajectory prediction model. The training method includes: obtaining the predicted trajectory of a target from the trajectory prediction model and dividing the predicted trajectory into multiple predicted sub-trajectory segments; obtaining the target's true trajectory and dividing the true trajectory into multiple true sub-trajectory segments, the number of true sub-trajectory segments corresponding to the number of predicted sub-trajectory segments; calculating the vector angles between the predicted and true sub-trajectory segments, including lateral and longitudinal vector angles; calculating the L2 distance loss between the predicted and true sub-trajectory segments; calculating the lateral and longitudinal distance losses based on the vector angles and L2 distance losses; and using the lateral and longitudinal distance losses as the loss function of the trajectory prediction model to optimize the parameters of the trajectory prediction model. This application performs a refined decomposition of the loss function, resulting in a more accurate prediction model after training.

The application discloses a steel truss extraction method based on point cloud segmentation and feature analysis, and belongs to the technical field of three-dimensional point cloud processing. The method mainly comprises the following steps: firstly, performing multi-scale filtering denoising on original point cloud and constructing an octree index; then, calculating local geometric features such as normal vectors, curvatures and first-order plane fitting errors of each point; based on adaptively selected seed points, combining the normal vector angle, curvature difference and adaptive edge criterion to realize region growing and preliminary segmentation; finally, adopting a multi-level cascade classifier from coarse to fine to screen the segmented regions, and sequentially passing through geometric size filtering, shape feature screening and multi-dimensional feature matching to accurately extract the steel truss point cloud. The application has the advantages of strong anti-interference ability, high extraction precision and high calculation efficiency without pre-training, and is suitable for engineering scenes such as bridge detection and building surveying.

This invention discloses a method for measuring the vector angle of a thrust vector nozzle, comprising the following steps: S1, acquiring a high-speed schlieren image of the thrust vector nozzle wake to obtain an original grayscale image; S2, obtaining a preprocessed image and a set of coordinate points representing the main boundary of the wake; S3, obtaining the coordinate range of a first selected region and a second selected region; S4, filtering out a first set of edge points located within the coordinate range of the first selected region and a second set of edge points located within the coordinate range of the second selected region; S5, performing linear fitting on the first set of edge points to obtain a first fitted line, and performing linear fitting on the second set of edge points to obtain a second fitted line; S6, calculating the angle bisector of the two lines to obtain the equation of the angle bisector; S7, outputting the thrust vector angle. This invention enables the direct acquisition of the nozzle wake deflection angle based on a high-speed schlieren image in the measurement of the thrust vector nozzle wake deflection angle.

The application discloses a kind of multi-view point cloud matching feature histogram construction method based on structured light phase information, by introducing phase information as auxiliary feature, to enhance the feature distinguishing ability of fast point feature histogram (FPFH), first by structured light technology obtains phase information, combined with traditional fast point feature histogram algorithm, the extended fast point feature histogram feature descriptor including normal vector angle, Euclidean distance and phase value is constructed.Normal vector angle and distance are quantified to N interval respectively, and phase information is introduced as the second 2N+1 feature dimension, the distinguishing degree and robustness of feature description are significantly improved.In the feature matching process, pixel-level matching is carried out using polar correction and sum of absolute differences (SAD) algorithm, and sub-pixel optimization is carried out combined with the linear characteristics of phase information, and the matching accuracy is improved.The application is excellent in the point cloud registration of weak texture area and flat surface, effectively improves the quality and accuracy of three-dimensional reconstruction under complex scene.

The present disclosure provides systems and methods for quantitative material characterization using multi-energy photon and neutron interactions are disclosed. Extending traditional dual-energy techniques, the disclosure utilizes multi-dimensional vector analysis from multiple energy bins to enhance material differentiation. The approach leverages distinct energy-dependent behaviors of photoelectric effect (PE), Compton scattering (CS), pair production (PP), Rayleigh scattering, and neutron interactions. By calculating differences, ratios, slopes, and vector angles and directions across energy channels, and constructing two-dimensional (2D) and three-dimensional (3D) vectors, unique material signatures are obtained. Vector angles and trajectories through quadrants correspond to materials like clock hands indicating time, demonstrating identification precision. The disclosure integrates neural networks trained on simulated data and incorporates real-time feedback loops correcting for dark current, noise, and detector drift before vector analysis. This eliminates Poisson and detector noise, ensuring vectors represent material signals matching simulation vectors.

The invention relates to the technical field of three-dimensional modeling, in particular to an automatic mapping method and system based on a three-dimensional model-engineering drawing, and the method comprises the following steps: sampling a three-dimensional model to obtain a vertex and a normal vector, calculating a curvature change rate through gradient operation, extracting features through principal component analysis to generate a reference set, and classifying detection value difference threshold values to form a region table; the method comprises the steps of generating a normal vector angle difference correlation area to generate a weight table, calculating a control point through vertex weighting and generating a coordinate set through projection, generating a result through Lagrange interpolation view alignment and layout adjustment, and generating a complete file through interpolation of a supplementary point set for checking coverage insufficiency. The interpolation function unifies multi-view-angle coordinates; the supplementary point set ensures continuous mapping; the output result is globally consistent and local integrity is considered; and gap offset is avoided to improve the mapping stability and integrity.

This invention relates to the field of 3D modeling technology, specifically to a bridge appearance defect detection system based on 3D point cloud reconstruction. The system includes: a point cloud spatial partitioning module, which constructs a 3D bounding box based on the bridge point cloud, calculates a local geometric degradation index, and compares the local geometric degradation index with a flatness threshold and a degradation threshold to obtain an initial 3D polygon mesh. In this invention, by filtering the normal vector angles of density feature patch regions and extracting polygon contours, local geometric abrupt change regions are separated from the overall structure. Furthermore, a dual filtering process is performed using the rate of curvature change and surface area values, ensuring that defect regions possess clear characteristic expressions in both spatial scale and morphological changes. This improves the accuracy of defect localization and ensures that bridge appearance defects remain identifiable even in complex environments.

The application provides a camera focal length and posture calibration method based on two straight lines, and belongs to the technical fields of computer vision, SLAM and photogrammetry. The application only needs to know two straight lines in space, measure the camera position and the two straight line positions, shoot straight line feature images, obtain the imaging relationship of the straight lines in the camera, utilize the important geometric phenomenon that the normal vector angles between two planes are equal, establish a method for quickly solving the camera focal length, and finally convert the corresponding relationship of the straight lines into the corresponding relationship of the space points between two coordinate systems with the same origin to solve the camera posture. Compared with the existing straight line-based calibration method, the application reduces the number of straight lines, shortens the calibration time, has a unique solution, and can simultaneously solve the focal length.

A method for calculating an angle of the inductor flux vector, comprising a plurality of iterations, each iteration of the plurality of iterations except the first comprising: a / obtaining (S100) input data including armature voltage values, armature current values, armature resistance values, and armature inductance values; b / estimating (S101) the inductor flux vector of the iteration by adding to an inductor flux vector obtained in the previous iteration a function of said input data; c / a correction comprising i. determining (S102) the center of a circle passing through at least three points defined by the flux vector obtained by said estimation, and at least two flux vectors obtained in two preceding iterations; and ii. obtaining the flow vector of the current iteration in which we correct (S106, S107) the flow vector obtained by said estimation using said center vector.Figure for the abridged version: [Fig. 2].