21 results about "Inverse projection" patented technology

This invention relates to the field of point cloud computing technology, specifically a hatch positioning method based on multimodal lidar point clouds. First, a multi-dimensional point cloud containing three-dimensional coordinates and reflection intensity is acquired. The coordinate system of the multi-source data is unified and fused through radar extrinsic parameter calibration or point cloud fusion algorithms. Then, the deck area point cloud is segmented based on normal vector filtering and DBSCAN clustering, and converted into a two-dimensional image through orthogonal projection. Next, edges are extracted using an improved Canny operator, and the hatch planar position is determined by least-squares rectangle fitting. Finally, the three-dimensional coordinates of the hatch are restored through inverse projection mapping. This invention overcomes the detection limitations of a single radar, effectively improving the accuracy and stability of hatch positioning under complex conditions. It can provide reliable positional information for the automated control of bulk carrier unloading operations, and has significant industrial application value.

The application relates to a contact analysis method and application of an arc tooth bevel gear based on tooth surface topology modification, and belongs to the technical field of gear transmission analysis. The method is as follows: based on basic geometric parameters and machine tool machining parameters of the arc tooth bevel gear, a meshing equation and a pinion conjugate tooth surface equation are constructed, and then a topology modification model is obtained; based on the meshing equation and the pinion conjugate tooth surface equation, an instantaneous contact line Gamma0 on the pinion conjugate tooth surface is established, and then through projection and inverse projection, an actual pinion tooth surface instantaneous contact line Gamma1 corresponding to the Gamma0 is obtained; the contact area boundary of the tooth surface is demarcated through the topology modification model, and the Gamma0 and the Gamma1 are contact-aligned through an angular positioning angle; distance judgment and contact analysis are carried out on the contact-aligned instantaneous contact line, and corresponding contact characteristics, namely a contact trace, a contact mark and a transmission error, are obtained. The method is helpful to improve the calculation stability of the contact analysis method, reduces the calculation time, and can improve the design efficiency and quality of the arc tooth bevel gear.

The application discloses a mesh texture simplification algorithm suitable for three-dimensional reconstruction, and comprises the following steps: S1, reference three-dimensional model scene construction; using the recovered scene structure in three-dimensional reconstruction and the calibrated image, performing texture mapping on the original fine three-dimensional mesh to complete the reconstruction of the reference three-dimensional model scene; S2, reference three-dimensional model scene image acquisition; according to the three-dimensional space relative attitude relationship between the reference three-dimensional model scene and the view internal and external parameters, using the inverse projection principle to perform three-dimensional mesh to two-dimensional image rasterization calculation pixel by pixel and view by view, and completing the acquisition of the reference image set; S3, mesh and texture simplification; using the QEM algorithm to simplify the mesh, using the reference image set as a data source and using the texture reconstruction algorithm to perform texture remapping and simplification. The application has the advantages that the algorithm can support mesh and texture simplification together, under different texture simplification parameters, the texture almost has no distortion and can significantly reduce the texture data amount.

This invention belongs to the field of industrial automation control technology, specifically relating to an intelligent optimization system for process parameters in automotive steel wire production. The system includes: a data acquisition module, a deviation calculation module, a prediction module, and a control module. The data acquisition module constructs a Lagrange coordinate system with material particles as the core, collects time-series process data throughout the entire process, and maps it to the material particles. The deviation calculation module calculates the historical deviation modulus of the material particles based on the arc-length integral of the state deviation trajectory. The prediction module constructs a dissipative damping model based on the historical deviation modulus and calculates the performance loss of the material particles. The control module generates process parameter compensation quantities through inverse projection based on a preset historical regression gradient vector and performance loss, and implements feedforward control of the equipment. This invention achieves closed-loop compensation for historical accumulated damage, improving the consistency of automotive steel wire product performance.

The invention relates to a method for automatic segmentation of a dental arch that comprises acquiring a three-dimensional surface of the dental arch, in order to obtain a three-dimensional representation comprising a set of vertices, generating virtual views from the three-dimensional representation, projecting the three-dimensional representation onto each two-dimensional virtual view, in order to obtain an image representing each vertex on the virtual view, processing each image by means of a deep learning network, carrying out inverse projection of each image in order to assign, to each vertex of the three-dimensional representation, one or more pixels of the images in which the vertex appears and to which it corresponds, and assigning one or more probability vectors to each vertex, determining the class of dental tissue to which each vertex most probably belongs based on the probability vector or vectors.

Disclosed are a method, device, and system for X-ray CT contrast imaging using magnetic nanoparticles. The method includes: rotating a target object containing magnetic nanoparticles as contrast agent for one circle relative to X-rays, recording multiple consecutive transmission projection images of the target object at all angles; for multiple transmission projection images at each angle, after conducting image post-processing operations, performing inverse projection transform on magnetic nanoparticle imaging projection images at all angles to obtain a tomographic imaging result. The image post-processing operation includes: time-frequency transform: converting the third dimension of the three-dimensional time domain matrix to frequency domain to obtain a three-dimensional frequency domain matrix; feature extraction: selecting a two-dimensional matrix with frequency of kf along the third dimension of the three-dimensional frequency domain image matrix; k is a positive integer; f is the frequency of the excitation magnetic field.

The application discloses a kind of laser radar point cloud prediction method and system based on Transformer, it is related to three-dimensional environment perception technical field.The method first converts distance image by spherical coordinate projection conversion for historical point cloud sequence with semantic label, and utilizes semantic embedding module to map discrete semantic label into high-dimensional continuous feature;Subsequently, the geometric feature and semantic feature are spliced in channel dimension, to form the integrated composite feature tensor;The tensor is input into the improved Transformer network for training, the network adopts encoder-double-branch attention-decoder architecture, and is supervised using the composite loss function combining geometric loss and semantic loss;Finally, the network synchronously outputs future distance image and semantic probability distribution, and generates future three-dimensional point cloud with semantic attributes through inverse projection and color mapping.The application significantly improves the modeling ability of the motion law of different categories of objects and the prediction accuracy in complex dynamic scenarios by explicitly integrating semantic information.

This invention discloses a multimodal intelligent safety early warning system for oil and gas field stations, relating to the field of industrial information security technology. It includes the following modules: a multimodal temporal alignment module for outputting initial multimodal data frame pairs; an acoustic spectral fingerprint extraction module for outputting acoustic spectral fingerprint vectors; a visual micro-perturbation perception module for outputting visual perturbation tensors; a physical constraint cross-modal coupling module for introducing a strain energy gradient-driven crack tip field energy diffusion constraint mechanism to improve the ViLBERT model; a temporal correlation dynamic verification module for outputting verification confidence levels; an early warning decision module for outputting safety early warning signals; and a three-dimensional inverse projection tracing instruction module for generating safety early warning instructions. This invention overcomes the limitations of traditional methods, such as difficulty in cross-modal alignment and susceptibility to background noise interference, providing an efficient solution for intelligent safety early warning in oil and gas field stations.

This invention relates to the field of marine geological 3D visualization technology, specifically disclosing a marine geological VR 3D visualization simulation system that integrates panoramic real-world footage. The system includes a visualization simulation platform, which is communicatively connected to the following modules: a coordinate mapping and alignment module, used to precisely align the spherical UV coordinates of the panoramic real-world footage with the terrain world coordinate system in Unity3D through inverse projection transformation using a unified virtual-real coordinate mapping algorithm, and to smooth the parallax jumps during head rotation using quaternion interpolation. This invention precisely aligns the spherical UV coordinates of the panoramic real-world image with the 3D terrain world coordinate system using a unified virtual-real coordinate mapping algorithm, and uses quaternion spherical linear interpolation to smoothly transition the parallax changes during head rotation, eliminating the visual disconnect between the panoramic background and the 3D terrain, and ensuring the continuity and stability of spatial perception when the user's perspective changes.

This invention belongs to the field of artificial intelligence technology and discloses a method and system for semantic assignment of two-dimensional LiDAR point clouds based on a camera. The method includes: obtaining the projection transformation matrix from the camera image plane to the laser scanning plane; obtaining the pixel set within the target bounding box of the target object image based on the target object image captured by the camera; extracting image features from the target object image to obtain the contour geometric line equation of the target object; inversely projecting the contour geometric line equation of the target object onto the LiDAR coordinate system using the projection transformation matrix to obtain the inverse projection curve; determining the range of laser points hitting the target object based on the inverse projection curve, and assigning semantic information to the laser points within the range. This invention abandons complex clustering algorithms, data fusion algorithms, and a large amount of laser point reprojection calculations. It only requires inverse projection of the contour to filter out the range of laser points hitting the target object, improving the accuracy of point cloud search and the speed of semantic assignment.

This invention relates to a method for 3D reconstruction and parameter identification of parachutes. It generates a 3D physical skeleton of the parachute through fluid-structure interaction numerical simulation, constructs a virtual optical imaging link including a camera projection model based on physically based rendering technology, and generates a synthetic image and a depth ground truth map spatiotemporally aligned with the physical skeleton, forming a high-fidelity synthetic dataset. Bidirectional domain adaptation processing is performed on both the synthetic image and the real airdrop image to eliminate differences in their feature distributions. A depth-sensing network is trained using the processed synthetic image and depth ground truth map, and depth maps are obtained through inference on the processed real image. A relative 3D point cloud is reconstructed based on the inverse projection relationship of the camera projection model. The time-varying scale factor is solved using known geometric invariants as physical constraints, mapping the relative point cloud to absolute physical space and calculating aerodynamic parameters. This invention solves the problems of data scarcity and domain discrepancies in parachute measurement, achieving high-precision 3D reconstruction and parameter identification under monocular video.

The application discloses a color structured light color crosstalk correction method based on unsupervised deep learning, generates a color phase shift fringe image by using a computer, collects a deformed color fringe image modulated by the height of a measured object, extracts three gray scale images from the image, inputs the three gray scale images into three deep neural network modules respectively for color crosstalk correction, outputs three predicted corrected gray scale images, solves a predicted phase by using a phase shift method, performs computer inverse projection simulation by using the predicted phase, and can obtain a deformed color fringe image corresponding to the phase result; loss values of the actually collected deformed color fringe image and the inverse projection simulated deformed color fringe image are calculated, network parameters are optimized through multiple iterations until the loss value is the minimum, and an ideal correction result is obtained. The application does not need to make a large amount of training data and corresponding labels, does not depend on a specific data set, and significantly improves the operation efficiency and generalization performance of deep learning.

The application discloses a dam leakage digital twin diagnosis and early warning method based on thermal infrared data fusion. The method obtains thermal infrared images collected by a multi-source platform on a target dam, platform pose parameters, and a pre-constructed three-dimensional dam surface unstructured grid; based on the platform pose parameters, the thermal infrared images are inversely projected onto the three-dimensional dam surface unstructured grid, and based on an observation quality index, fusion is carried out to obtain a three-dimensional dam surface fusion temperature field; based on the area distribution of the surface element, the three-dimensional dam surface fusion temperature field is subjected to area weighted spatial mean removal to obtain an online observation relative temperature deviation field; the online observation relative temperature deviation field is mapped to a reduced modal space composed of a truncated modal basis to obtain online observation modal coefficients; in the reduced modal space, the online observation modal coefficients are subjected to Bayesian probability matching with scene modal coefficients of each candidate leakage scene in a reduced thermal fingerprint library to infer and output a diagnosis result.

The application provides a compliance control method, system, device and storage medium. The method comprises the following steps: a mapping relationship between an operation space and a null space of a main body of a robot arm is established, and a main task and a secondary task are decoupled through a Jacobian matrix and a pseudo-inverse projection matrix; a composite error is calculated based on joint position error and joint speed error; an error boundary of an adaptive barrier function is dynamically adjusted according to real-time interactive force of the end of the robot arm; stiffness parameters and damping parameters of a null space impedance model are dynamically adjusted based on the real-time interactive force of the end; operation space control torque and null space control torque are respectively calculated, the operation space control torque and the null space control torque are superimposed through the pseudo-inverse projection matrix, and total control torque is obtained; and motion instructions are output by substituting the total control torque into a dynamics equation of the main body of the robot arm. Under the premise of ensuring the trajectory tracking accuracy of the operation space, the application realizes adaptive adjustment of the null space compliance, and ensures that the main task and the secondary task do not interfere with each other.

A virtual reality rendering system includes a processor, memory storing a stereo panoramic image with color and depth data captured using above-horizon and below-horizon projections representing static virtual environment components, and a graphics processing unit. The GPU renders the stereo panoramic image as a full-screen quad using equirectangular projection coordinates, extracts depth values and reconstructs world-space positions by applying inverse projection and view matrices corresponding to capture parameters, seeds a runtime depth buffer with reconstructed positions for depth-based occlusion, and renders dynamic three-dimensional objects using standard rasterization techniques. The dynamic objects interact with the stereo panoramic image through the seeded depth buffer to provide correct occlusion relationships.

This invention discloses a method for measuring the distance of transmission lines based on the mapping of 3D point clouds and 2D images. It acquires 3D point cloud data of the transmission line corridor and 2D images from a monitoring device. The 3D point cloud data is preprocessed to construct a 3D spatial coordinate system. The camera pose parameters of the monitoring device are optimized automatically / semi-automatically. Based on the optimized camera pose, the 3D point cloud is accurately projected onto the 2D plane, achieving alignment and fusion with the 2D image. On this basis, the 3D point cloud data is used to accurately locate the transmission line position. For potential hazards identified in the 2D image, a reverse projection algorithm based on the ground intersection principle is used to locate them in the 3D spatial coordinate system. The shortest 3D distance between the hazard and the transmission line is then accurately calculated, and protection zones and risk levels are delineated according to preset safety distances. This invention significantly improves the automation and accuracy of measuring the distance of safety hazards in transmission line corridors.

This application discloses an engineering machinery and its bird's-eye view processing method, apparatus, and medium, relating to the field of human-computer interaction technology. The method includes: acquiring a distorted original camera image; performing an end-to-end inverse projection from the bird's-eye view template to the original camera image based on a pixel inverse mapping relationship from a predefined bird's-eye view template to the original camera image, to obtain a distortion-free camera-corrected image, wherein the pixel inverse mapping relationship is constructed in relation to camera distortion; and generating a corresponding camera's bird's-eye view based on the camera-corrected image. This application generates a bird's-eye view through end-to-end inverse projection, completing distortion removal and perspective projection in one step, which not only avoids image cropping and field-of-view loss that may result from intermediate results but also saves computational resources and memory usage.

The application discloses a kind of coordinate conversion method and system based on laser radar and UTM projection, belong to indoor V2X simulation test technical field, including: data preprocessing;Local coordinate acquisition;Benchmark establishment and parameter calibration;Coordinate mapping;Inverse projection processing.The application is significantly superior to traditional linear conversion method in positioning accuracy, stability and robustness to test distance;The effectiveness of the method of the application is derived from its strict UTM projection inverse solution, which accurately compensates the geometric deformation introduced by the earth curvature in long-distance coordinate mapping, thereby providing a practical technical solution for carrying out high-precision, high-reliability V2X test in indoor environment.

This invention relates to the field of industrial vision technology, specifically a method for monitoring the quality of high-temperature resistant industrial wire harnesses based on visual analysis. The specific implementation process includes: acquiring a sequence of wire harness images; constructing a response image tensor using optical flow frames; decoupling the Lambertian volume and specular reflection into a reflectivity component map and a surface normal vector map; constructing a cylindrical plane inverse projection transformation matrix based on the wire harness central axis equation; inputting the corrected data into a texture geometry segmentation network to generate a non-defect semantic mask and outputting a wire harness defect feature map; generating topological features of the defect region based on the Hessian matrix for defect monitoring of the wire harness. This invention utilizes photometric stereo decoupling and semantic mask suppression to effectively eliminate high-brightness interference and normal texture noise from the surface of highly reflective wire harnesses, improving the difficulty of defect extraction in low-contrast environments. Through diffusion enhancement and topological descriptors, the sensitivity of wire harness defect recognition is improved, and the false alarm rate of defect detection is reduced.

The present application provides a kind of human body Gauss point cloud adaptive training system based on reference posture geometric constraint, first, based on the geometric constraint mechanism of reference space inverse projection, by the Gauss point inverse mapping back static reference posture space in training process, and with the distance check of reconstructed surface grid, to inhibit the suspended noise generated by false gradient;Second, based on the smoothing pruning strategy of life cycle cumulative score, use the sum of rendering contribution degree and geometric offset degree of multiple iterations as the discriminant standard, not single frame gradient, improve the time sequence stability of dynamic rendering.