246 results about "Geometric reconstruction" patented technology

The invention belongs to the technical field of three-dimensional imaging and measurement, and particularly discloses a laser scanning three-dimensional imaging method and device, a medium and equipment, and the method comprises the steps: obtaining laser scanning data of a target object, and the laser scanning data comprises depth data including reflection intensity, a scanning angle, a timestamp and flight time; preprocessing the laser scanning data, and fusing a time sequence and scanning path information to obtain multi-channel input features adaptive to a neural network; constructing a three-dimensional geometric reconstruction model, and training the model; inputting the multi-channel input features into a trained three-dimensional geometric reconstruction model to obtain a dual-channel output tensor containing a dense depth map of the target object and a corresponding uncertainty map; converting the dual-channel output tensor comprising the dense depth map of the target object and the corresponding uncertainty map into a three-dimensional point cloud under world coordinates; and post-processing the three-dimensional point cloud to obtain a three-dimensional image of the target object.

The invention discloses a BIM model automatic generation method and system based on point cloud data, and belongs to the field of building information modelling, and the transmission method comprises the steps: obtaining original point cloud data, and employing a filtering algorithm based on point cloud density adaptive adjustment to carry out the preprocessing of the point cloud data; constructing a voxel octree structure for the preprocessed point cloud data, and performing semantic classification on the point cloud; geometric modeling is carried out based on the segmented point cloud subsets, and a fitting algorithm is adopted to carry out shape completion on a point cloud area; semantic annotation is carried out on the components subjected to geometric reconstruction, a corresponding relation between component types and spatial attributes is constructed, fusion features based on a point feature histogram and a local curvature are adopted, and classification is carried out; a standard BIM component family is converted, and a three-dimensional BIM model is constructed through the mapping relation. According to the method, the voxel octree data structure and the deep semantic segmentation neural network model are combined, division and semantic recognition are performed on the point cloud data, the intelligent degree of the model is improved, and manual intervention is reduced.

The invention discloses a reflecting object inverse rendering method, system and device based on two-dimensional Gaussian sputtering and multi-mode diffusion prior and a medium. The method comprises the steps that images and camera parameters of a reflecting object under multiple visual angles are acquired; calculating an observation angle according to the camera parameters; performing rasterization rendering on the images of the reflecting object under the multiple visual angles by using the trained two-dimensional Gaussian primitives under the observation visual angle to obtain a rendered image under the new visual angle; according to the method, two-dimensional Gaussian is adopted as a scene representation element, surface characteristics are better fitted, and geometric reconstruction precision can be remarkably improved; the trained two-dimensional Gaussian primitive is closer to a real three-dimensional scene, and experimental tests of multiple data sets show that the method not only can accurately reconstruct geometric and material distribution of a high-reflection area, but also can recover clear environment illumination with high-frequency details, realizes efficient and real reflection inverse rendering, and has a good application prospect. The method can be widely applied to downstream tasks such as three-dimensional reconstruction, material editing and relighting.

The invention discloses a road slope modeling method based on an unmanned aerial vehicle inspection route, and relates to the technical field of unmanned aerial vehicle inspection. The method comprises the following steps: measuring the road slope length, the landform and the geological complexity, and selecting an adaptive unmanned aerial vehicle type and sensor combination according to the road slope length, the landform and the geological complexity; planning unmanned aerial vehicle surface inspection and shallow geological exploration routes, ensuring space matching, and synchronously obtaining high-resolution images and ground penetrating radar data; and processing image data through a photogrammetry technology, generating a digital elevation model and an orthophoto map, and combining to construct a three-dimensional model of the earth surface. Geophysical inversion is carried out on the ground penetrating radar data, reflection features are extracted and projected to the ground surface three-dimensional model, and space correlation between the ground surface and a shallow layer structure is established; the reflection features are digitalized and geometrically reconstructed into three-dimensional surface elements, the three-dimensional surface elements are embedded into an earth surface model to form a three-dimensional geologic body model, time dimensions are introduced, and a dynamic evolution model is constructed. According to the dynamic evolution model, the dynamic analysis of the slope stability is realized.

The invention provides a moving target capturing method and system based on AI multi-view vision, and relates to the technical field of data processing.The method comprises the steps that camera parameters in all vision systems are calibrated, and the camera parameters comprise a camera internal parameter matrix and a camera external parameter matrix; extracting 2D feature points of the moving target through each calibrated visual system; reconstructing each 2D feature point by combining a multi-view geometric reconstruction algorithm and a least square method according to the calibrated camera parameters to obtain 3D feature points; constraints are added to the 3D feature points, and the constraints comprise spatial constraints, direction vector constraints and time sequence constraints; reconstructing the 3D feature points added with the constraints through a multi-view geometric analysis algorithm to obtain three-dimensional motion information of the moving target; and outputting the three-dimensional motion information of the moving target to complete the capture of the moving target. And non-contact high-precision capture of the moving target can be realized.

The invention provides a scene reconstruction and semantic understanding method and system based on a visual large model, and relates to the field of computer vision and three-dimensional reconstruction, and the method comprises the steps: obtaining multi-view image data of a target scene, inputting a pre-training visual large model, and outputting a joint feature representation and attention weight matrix; generating three-dimensional coordinate values and semantic probability distribution of spatial sampling points in a three-dimensional space according to the joint feature representation, and constructing a spatial semantic field; clustering the spatial sampling points by using the attention weight matrix, performing semantic consistency enhancement, and converting the spatial sampling points into deterministic semantic tags; constructing a geometric optimization objective function, and adjusting three-dimensional coordinate values; and extracting continuous space sampling points with the same semantic tag to form an object boundary, constructing a scene topological graph, deducing a scene functional structure, and generating a navigation path. According to the method, the unification of accurate geometric reconstruction and deep semantic understanding of the scene is realized, the three-dimensional reconstruction precision and semantic analysis accuracy are improved, and reliable support is provided for intelligent navigation.

The invention discloses a lightweight design method for a thin-wall structure containing a flow channel based on additive manufacturing, and belongs to the technical field of additive manufacturing, and the lightweight design method comprises the following steps: S100, analyzing functional requirements, and constructing a functional requirement model containing the flow channel; s200, a printing direction and additive manufacturing constraint conditions are set, topological optimization is executed on the function demand model, and a preliminary topological structure is obtained; s300, carrying out geometric reconstruction on a topological optimization result to obtain a three-dimensional model suitable for additive manufacturing; s400, carrying out parameter optimization design; s500, the section shape and size of the runner are optimized based on fluid performance requirements; and S600, additive manufacturing manufacturability inspection is conducted on the structure with the section shape and size optimized. And S700, the model obtained after additive manufacturing manufacturability optimization is analyzed and checked. The technical problems that in the prior art, the design requirement of a thin-wall structure containing a runner cannot be met, and the forming quality and the section shape of the inner surface of the runner are difficult to accurately control can be solved.

The invention relates to the technical field of machine vision, in particular to a precision part size automatic detection method and system based on machine vision, precision part images are collected through a high-precision industrial camera, part positioning is carried out, sub-pixel-level topological feature mapping is carried out on interested area images, and precision part size automatic detection is carried out. Comprising the steps of gray histogram equalization, gray morphological processing, edge detection and edge chain code tracking, construction of an edge point topological feature space, execution of sub-pixel subdivision, obtaining of an edge line through contour analysis of a feature distance and a feature angle, and double-constraint geometric reconstruction based on the edge line. The characteristic distance and the characteristic angle are used for rotation matrix conversion and geometric dimension calculation, a relation model of the geometric dimension and the actual dimension of the part is established, precise part dimension measurement is achieved through dynamic error analysis and compensation, the measurement precision is remarkably improved, the risk caused by unreliability of a single characteristic is effectively reduced, and the measurement accuracy is improved. And the measurement stability is improved.

The invention discloses a stator blade shape-following frequency modulation design method based on a bending vibration mode and application, belongs to the field of vibration design of stator blades of aero-engines and gas turbines, and is suitable for vibration frequency modulation design of non-cantilever type stator structure blades (such as fans and gas compressor stators) with inner rings and outer rings in the aero-engines or the gas turbines. According to the method, firstly, static frequency and dynamic frequency modal analysis is carried out through finite element modeling, and dangerous resonance points and corresponding vibration modes in a working rotating speed range are recognized by combining a Campbell diagram; based on a vibration mode amplitude normalization result, setting a vibration degree parameter and a scale factor, selecting a high response node and carrying out directed disturbance, and updating a structure model to improve a target vibration mode frequency; and through frequency amplification verification and iterative optimization, geometric reconstruction and frequency modulation confirmation are finally completed in combination with aerodynamic performance and processing requirements. The method can realize rapid and accurate frequency modulation of a specific bending vibration mode, avoids introduction of new resonance, and has the advantages of high efficiency, strong adaptability and the like.

The invention relates to the technical field of three-dimensional reconstruction, and discloses a monocular depth guided object-level NeRF reconstruction method, which comprises the following steps: firstly, through monocular video sequence input, generating a frame-by-frame initial depth map by using a depth estimation module, and estimating a relative camera attitude between adjacent frames through a relative attitude estimation module; calculating the absolute attitude of the camera in combination with the initial depth map and the relative camera attitude; then constructing a geometrically enhanced NeRF model, and optimizing scene representation through multi-resolution hash position coding and spherical harmonic direction coding; introducing photometric loss, depth contrast loss and density loss to jointly optimize parameters of the NeRF model, and constraining geometric reconstruction of the object by using depth information; and finally, through a four-stage iterative training strategy, alternately optimizing depth estimation, a camera attitude and a NeRF model, and generating an object-level controllable three-dimensional model. According to the method, depth information is fully utilized to optimize the NeRF training process, so that end-to-end reconstruction from a monocular video to an object-level controllable model is completed.

The invention discloses a VR scene automatic generation method based on a parallel projection design drawing, and the method comprises the following steps: S1, importing a two-dimensional CAD drawing containing building design information, and analyzing a primitive set in the drawing; s2, performing component type identification on the primitives, and establishing a mapping relation between a primitive set and spatial semantics; s3, performing geometric reconstruction based on a mapping relation between the primitive set and spatial semantics, and generating a three-dimensional component model with a unique name identifier; s4, establishing a multi-stage mapping model, and realizing size conversion from a drawing scale to a space coordinate system of the virtual reality system; s5, completing three-dimensional coordinate positioning and automatic assembly according to the multi-stage mapping model, and generating a three-dimensional structure layout of an overall building or park level; and S6, importing the assembled three-dimensional structure layout into a virtual reality rendering engine to generate a virtual reality scene. According to the invention, an automatic, precise and integrated conversion process from a two-dimensional design drawing to a three-dimensional virtual reality scene is realized.

The invention belongs to the technical field of three-dimensional scene reconstruction, and particularly provides a shielding object three-dimensional reconstruction method and system based on millimeter wave radar and vision fusion. The method comprises the following steps: acquiring radar point cloud data and visible light image data of a target object; performing space-time alignment on the radar point cloud data and the visible light image data, and establishing a space mapping relation of multi-modal data; generating a three-dimensional geometric structure of the shielded part through a geometric reconstruction model based on three-dimensional space information penetrating through the shielding object in the radar point cloud data; based on the context features of the visible light image data, generating surface textures of the occluded area through a texture completion model; and fusing the three-dimensional geometric structure and the surface texture, and outputting a three-dimensional reconstruction model containing complete information of the occlusion region. Geometric consistency and texture authenticity of virtual-real fusion are guaranteed through space-time alignment and multi-modal data mapping, and the reliability of three-dimensional modeling in a complex shielding scene is remarkably improved.

The invention relates to the technical field of visual perception, and discloses a mechanical arm visual grabbing method and system. Comprising the following steps: collecting depth images of a target scene at different visual angles in real time, and generating initial point cloud data by using all the depth images; processing the initial point cloud data by using a double-flow neural radiation field to generate a geometric field and a capture field; a preset dynamic updating mechanism is adopted, combined dynamic optimization is conducted on the geometric field and the grabbing field, and optimal grabbing parameters corresponding to all moments in the process that the mechanical arm executes the current grabbing task are obtained; and the mechanical arm is controlled to execute the grabbing action based on the optimal grabbing parameters corresponding to all the moments. According to the method, high-precision three-dimensional scene representation is constructed through depth fusion of multi-view depth images, so that the integrity and robustness of geometric reconstruction of an object in a complex environment can be remarkably improved; based on joint modeling of a double-flow nerve radiation field, collaborative optimization of a three-dimensional structure and grabbing parameters of an object is achieved, and stability and adaptability of grabbing performance in a dynamic environment can be effectively improved.

The invention discloses a method and a system for designing a fluid flow channel of a calcium-based thermochemical heat storage fixed bed reactor based on topological optimization and a storage medium. The method comprises the following steps: S1, constructing a geometric model of a design domain of the reactor; s2, constructing an objective function and establishing a two-dimensional topological optimization mathematical model based on a control equation of conjugate heat transfer and fluid flow; s3, solving the mathematical model to obtain an optimal fluid flow channel two-dimensional topological structure; s4, performing geometric reconstruction on the two-dimensional topological structure obtained in the step S3, converting the two-dimensional topological structure into a Bezier curve, and integrating the Bezier curve into a geometric model corresponding to the original design domain; s5, establishing a finite element model of the reactor; s6, temperature field and pressure field distribution is calculated based on CFD simulation, whether the heat storage performance target is met or not is judged, and if not, corresponding control parameters are adjusted to return to the step S3 until the heat storage performance target is met; the device can optimize the heat transfer and mass transfer performance of the reactor at the same time, and the optimization design efficiency is high.

The invention relates to the field of computer graphics, reverse engineering and additive manufacturing, discloses a multi-modal fusion and physical constraint three-dimensional point cloud geometric modeling system and method, and aims at overcoming the defects that traditional three-dimensional reconstruction entity attributes are missing, key features are prone to being lost, semantic understanding does not exist, and process automation is low. The system comprises a point cloud preprocessing module, a semantic segmentation and recognition module, a mixed geometric reconstruction module, a multi-component assembly and physical constraint solution module and an entity model output module. The method comprises the following steps: denoising and sampling an original point cloud, realizing semantic and instance double segmentation through a Point Net + + network, endowing a geometry-function composite label, completing hybrid reconstruction through improved Poisson reconstruction, parameterized B-Rep fitting and standard component library matching, and outputting a multi-format engineering available model after physical constraint verification and automatic correction of a violation model. According to the method, automatic, high-fidelity and high-practicability conversion from the point cloud to the engineering-level entity model is realized, and the method is adaptive to multiple industrial application scenes.

The invention provides a three-dimensional image reconstruction method based on a large model, and relates to the technical field of artificial intelligence, in particular to the technical fields of computer vision, deep learning, large models, meta universe, virtual reality, augmented reality and the like. According to the specific implementation scheme, reasoning segmentation is carried out on a to-be-reconstructed image based on implicit prompt information, at least one candidate object in the to-be-reconstructed image is determined, a mask is added to the at least one candidate object, and a segmented image is obtained; geometric reconstruction is carried out on the to-be-reconstructed image to obtain a three-dimensional grid model, and the three-dimensional grid model represents geometric information of the at least one candidate object and geometric information of a reconstruction scene of the to-be-reconstructed image; fusing the segmented image and the three-dimensional grid model to obtain a three-dimensional reconstruction result of the at least one candidate object; and inputting the implicit prompt information and the three-dimensional reconstruction result of the at least one candidate object into the first large language model, and outputting the three-dimensional reconstruction result of the target object indicated by the implicit prompt information. The invention further provides a training method and device based on the large model, electronic equipment and a storage medium.

The invention discloses an innovative pit body volume detection system based on a three-dimensional laser scanning technology. Acquiring the volume point cloud information of the pit body by adopting a three-dimensional laser scanning technology; the volume is calculated through double-engine cross validation, two algorithms are used for cross validation, and it is ensured that the volume calculation result relative error meets the requirement; comprising the steps that a Matlab layered integral algorithm is adopted, specifically, discrete layered volume integration is achieved through elevation layered cutting, boundary polygon generation of a polyshape function and AlphaShape geometric reconstruction; and a Geomagic curved surface reconstruction algorithm: generating a watertight grid model based on an NURBS curved surface technology, and completing continuous volume calculation. According to the method, a non-contact measurement system is established, so that the measurement efficiency is greatly improved; a three-dimensional point cloud digital processing platform is created, full-process digitalization of scanning data automatic registration, boundary recognition and volume calculation and real-time visual output of measurement results are achieved, secondary errors caused by manual intervention are reduced, the project acceptance data traceability requirement is met, and the measurement comprehensive error rate is greatly reduced.

The invention provides a rapid calibration method and system for angle measurement precision of a laser seeker. The method comprises the following steps: emitting a laser beam to a calibration rod array through a laser seeker, and collecting the displacement of a laser spot reflected by the surface of each calibration rod through a position sensitive device; inputting the laser spot displacement into a spot displacement feature extraction network, and generating an angle and displacement time sequence containing a timestamp; identifying abnormal sampling points in the angle and displacement time sequence; and performing geometric reconstruction on the abnormal sampling points by adopting a space geometric constraint optimization algorithm and combining a fixed space included angle constraint relation to obtain angle measurement error calibration parameters of the laser seeker. According to the technical scheme provided by the invention, the high-precision dynamic calibration of the angle measurement error of the laser seeker is realized through the collaborative analysis of the space coordinates of the calibration rod and the displacement of the laser spot in combination with the neural network and geometric constraint optimization.

The invention discloses a scene three-dimensional reconstruction and vector information extraction method and device based on vehicle return data, equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing of a time sequence image returned by a vehicle and corresponding pose data, and obtaining a semantic mask and a camera pose of each frame of image in the time sequence image; based on the semantic mask and the camera pose, performing three-dimensional geometric reconstruction on the static scene area in the time sequence image to obtain a static three-dimensional scene model; a dynamic target is separated from the time sequence image according to the semantic mask, three-dimensional motion modeling and independent three-dimensional reconstruction are carried out on the dynamic target, and a dynamic target three-dimensional model is generated; and generating a dynamic three-dimensional scene model and vector labeling information based on the static three-dimensional scene model and the dynamic target three-dimensional model. According to the method, the limitation of automatic driving shadow mode data is overcome, the three-dimensional reconstruction of the scene and the automatic generation of the true value information are realized, the data acquisition cost is further reduced, and the efficiency of automatic driving research and development is improved.

The invention discloses a contact force field-based topological optimization design method for a detachable steel structure beam-column joint reinforcing member, which comprises the following steps of: (a) establishing an integral joint finite element model, defining a frictional contact relationship and carrying out nonlinear static analysis; (b) extracting node counter-force components on all contact surfaces and taking the node counter-force components as equivalent loads acting on a reinforcing component; (c) establishing a reinforcing member independent optimization model, applying an equivalent load and defining an inertia release boundary; (d) setting a design variable, a constraint condition and a target function of topological optimization, and solving; and (e) carrying out geometric reconstruction and performance verification on an optimization result. According to the method, a complex contact nonlinear problem is converted into a determined load boundary condition, an inertia release method is utilized to deal with a constraint problem in optimization ingeniously, an optimal topology result of the detachable reinforcing component is obtained finally, and the problem of optimization result distortion caused by the uncertain boundary condition is effectively avoided.

The invention discloses a three-dimensional point cloud geometric information compression method based on implicit neural representation, and the method comprises the steps: constructing a trunk structure of an implicit neural network through a plurality of sine representation network layers which are connected in series, and introducing a variable-scale position coding mechanism on this basis, the method enables a network to obtain higher geometric reduction precision while keeping a compression ratio, and comprises the following steps: (1) inputting space coordinates of divided voxels into a position coding module with adjustable scale parameters; (2) feeding a coding result into an implicit neural network constructed by a network layer based on sine representation, and outputting the occupancy probability of the voxel through an activation function; (3) in a training stage, the model continuously optimizes parameters, so that the output probability distribution is highly consistent with a real occupied label; (4) after model training is completed, a method of combining an AdaRound second-order quantization optimization strategy and quantization perception training is introduced, network weight is finely adjusted, and quantization errors are reduced; (5) in a reasoning stage, judging whether the voxel is occupied or not according to a preset threshold value, and when the prediction probability exceeds the threshold value, regarding the voxel as occupied; and (6) all voxels judged to be occupied are aggregated, and reconstruction of the geometric structure of the point cloud is completed.

The invention discloses a three-dimensional terrain construction method based on multi-modal semantic constraints, and relates to the technical field of three-dimensional terrain modelling, and the method comprises the steps: obtaining an optical remote sensing image and a digital earth surface model, constructing unified space reference data, and carrying out the joint semantic analysis of ground features under the guidance of the multi-modal semantic constraints; generating a semantic mask set with structural consistency; further performing regional geometric reconstruction on the digital earth surface model on the basis of a mapping relation between semantic categories and earth surface physical attributes to form a three-dimensional terrain entity conforming to ground feature space expression characteristics; meanwhile, by analyzing the project cost document and extracting the project cost data with spatial directivity, under the constraint of spatial consistency and semantic consistency, the project cost information and the three-dimensional topographic entity are subjected to association mapping, and finally a three-dimensional topographic data result fusing topographic geometric information and project economic attributes is formed.

The invention discloses a particle geometric reconstruction and morphological parameter automatic evaluation method based on a three-dimensional image, and the method comprises the steps: firstly obtaining a three-dimensional gray image of a particle sample, carrying out the Gaussian noise reduction, binaryzation and watershed segmentation preprocessing, and achieving the precise segmentation of independent particles based on an established particle segmentation defect repair algorithm; and then, according to the segmented and marked three-dimensional particle image, an initial contour surface is reconstructed by using a marching cube algorithm to generate a particle surface grid, noise is suppressed by using a Taubin smooth optimization algorithm, a topological structure is automatically calibrated, and a digital model conforming to real particle surface features is constructed. And finally, importing each particle model into the created particle morphological parameter automatic evaluation software to realize quantitative analysis of particle morphological characteristics.

The invention provides a liquid cooling plate flow channel design optimization method and device based on topological optimization and a liquid cooling plate, and relates to the technical field of thermal management, and the method comprises the steps: S1, determining the geometric parameters of the liquid cooling plate according to the size of a battery module, constructing a double-layer two-dimensional design domain, and setting material, heat source and boundary conditions; s2, a topological optimization algorithm is adopted, material distribution is represented with a relative density variable, a multi-objective function giving consideration to the maximum heat exchange amount and the minimum pressure drop is constructed, a clear fluid-solid interface is obtained through density filtering and a projection function, and a target double-layer two-dimensional flow channel contour is obtained through iterative solution under volume fraction constraint; and S3, taking the target double-layer two-dimensional flow channel contour as a cross section to carry out spatial stretching bending deformation, and generating a double-layer special-shaped flow channel structure with a diffusion layer and a collection layer. According to the method, global optimization of two-dimensional topological optimization and three-dimensional geometric reconstruction are fused, the contradiction between the heat dissipation performance and the flow resistance is effectively solved, and a high-performance heat dissipation solution is provided for battery thermal management.

The invention relates to an interpretable three-dimensional reconstruction method and device based on decoupling characterization learning and Gaussian splash, and a storage medium, and the method comprises the following steps: obtaining a monocular image, extracting two-dimensional features, obtaining compression features through convolution coding, obtaining a decoupled low-dimensional potential code through full connection processing, and obtaining a three-dimensional image; respectively converting into conditional representations of a geometric reconstruction branch and an appearance reconstruction branch; based on the two-dimensional coordinates on the predefined grid, three-dimensional points in a three-dimensional space are mapped through a plurality of MLP networks, a three-dimensional point cloud is obtained, and standard deviation-mean value modulation is carried out on the intermediate features by using conditional representation of geometric reconstruction branches; based on the three-dimensional point cloud, coding and projecting the three-dimensional point cloud to a plane to serve as initial three-plane features, and based on conditional representation of appearance reconstruction branches, coding the initial three-plane features into final three-plane features through a stylized U-Net network; and obtaining three-dimensional Gaussian scatter points based on the three-plane features and the three-dimensional point cloud to realize three-dimensional reconstruction.

The invention belongs to the technical field of three-dimensional surface reconstruction, and particularly relates to a robust three-dimensional surface reconstruction method based on multi-scale fusion. Acquiring three-dimensional point cloud data, fitting local implicit curved surfaces of neighborhood points by adopting a weighted least square method, and generating a geometric reconstruction grid model; extracting geometric features in different receptive fields through multi-scale convolution, introducing channel attention, space attention and a gating mechanism to perform weighted fusion on the geometric features in different receptive fields, enhancing a global dependency relationship by adopting a feature mixing operation, and generating a deep learning grid model; performing double-branch fusion on the geometric reconstruction grid model and the deep learning grid model, highlighting a geometric sensitive area, and obtaining a fused enhanced feature; inputting the fused enhanced features into an implicit function decoder, and predicting a signed distance function value of each vertex; and based on the predicted signed distance function field, a contour surface extraction algorithm is adopted to generate a triangular mesh model, so that three-dimensional surface reconstruction is completed.

The invention discloses a concrete member reinforcing steel bar identification and dislocation connection collaborative design method and system. According to the method, point cloud data of an exposed reinforcement area are collected through high-precision three-dimensional laser scanning, and precise three-dimensional geometric parameters of existing reinforcements are obtained through registration, denoising and geometric reconstruction; on the basis, a multi-target optimization model which takes bearing capacity maximization as a core target and takes distribution uniformity and symmetry into consideration is constructed, a two-stage optimization strategy is adopted to automatically solve an optimal arrangement scheme of newly-added reinforcing steel bars, and a digital result capable of being directly constructed is output after performance checking. The system comprises a data acquisition module, a point cloud processing and reconstruction module, a collaborative optimization design module and a scheme output and check module. According to the method, full-process automation from reinforcing steel bar recognition to optimization design is achieved, and the design quality and efficiency of reinforcement engineering are remarkably improved.

The invention relates to the technical field of weld joint detection, and provides a visual intelligent weld joint detection device and method, and the detection method comprises the steps: obtaining a 2D image of a weld joint; weld joint 3D shape data are obtained, and weld joint point cloud data are generated; generating a training data set according to the welding seam 2D image and the welding seam point cloud data; constructing a neural network model, and training the neural network model through the training data set; performing intelligent detection on the weld seam image according to the trained neural network model; wherein a loss function of the neural network model comprises a 2D positioning loss item, a classification task loss item and a 3D frame regression loss item. According to the method, by obtaining the loss function including the 2D positioning loss item, the classification task loss item and the 3D frame regression loss item, multi-task collaborative optimization can be realized, classification robustness, 2D positioning precision and 3D geometric reconstruction capability are considered, and the method is suitable for scene requirements of high precision and multi-modal fusion in industrial detection.

The invention relates to a high-fidelity three-dimensional reconstruction method and system based on multi-source data fusion, and the method comprises the steps: obtaining multi-source heterogeneous data of a power grid target region, carrying out the time-space alignment and standardization preprocessing, and generating a standardized data set; performing multi-modal feature extraction and fusion coding on the standardized data set to generate uniform multi-dimensional feature representation; performing entity matching and semantic enhancement on the multi-dimensional feature representation and a pre-constructed power grid semantic knowledge graph to generate a semantic enhanced three-dimensional scene graph; carrying out three-dimensional geometric reconstruction and dynamic detail enhancement on the three-dimensional scene graph based on semantic enhancement, and constructing a high-fidelity power grid model; performing multi-level-of-detail optimization processing and format packaging on the high-fidelity power grid model, and outputting a high-fidelity three-dimensional power grid model suitable for the digital twin platform; according to the scheme, multi-source heterogeneous data can be fully fused, a high-precision and high-fidelity three-dimensional power grid model is constructed, and the requirements of fine management and real-time monitoring of a power grid are met.

The invention discloses an aircraft surface geometric reconstruction method based on deep learning. The method comprises the following steps: acquiring a multi-view two-dimensional image of a target aircraft through a camera, extracting a target area mask, and constructing a feature corresponding relation between view angles to estimate camera parameters; constructing a multi-source loss function of fusion scene prior knowledge of the neural rendering network, and inputting camera parameters and the preprocessed two-dimensional image into the network for training until loss convergence to obtain network implicit output; point cloud reconstruction of the target aircraft is carried out, and an initial reconstruction point cloud is obtained after preprocessing; and inputting the initial reconstruction point cloud into a differentiable parametric curved surface model for optimization, outputting a three-dimensional parametric curved surface model after processing, extracting appearance parameters of the target aircraft to reconstruct the surface of the aircraft, and performing geometric reconstruction. The method effectively improves the precision and efficiency of model reconstruction, has good visualization, interaction and engineering integration capabilities, and is suitable for various application scenes such as industrial design and medical modeling.