338 results about "Geometric distortion" patented technology

The invention discloses a real-time single-stage remote sensing image correction target detection method based on YOLOV8, and relates to the technical field of remote sensing image processing. According to the method, a deformable convolution dynamic prediction local geometric distortion parameter is embedded based on a YOLOv8 backbone network, an adaptive deformation field is generated, pixel-level real-time correction is realized, shallow details and high-level semantic features are fused through a bidirectional path aggregation network, and channel attention and a space gating mechanism are combined, so that the real-time correction of the image is realized. The small target detection capability is enhanced, background noise is suppressed, angle prediction is divided into discrete classification and continuous residual error regression tasks through a decoupling type rotation detection head, angle periodic errors are eliminated in combination with a direction sensitive loss function, and the rotation frame positioning precision is improved. And constructing a dynamic multi-task collaborative loss function, introducing gradient distribution consistency constraint to jointly optimize correction and detection tasks, and realizing feature semantic alignment and model self-enhancement through end-to-end closed-loop training. And the rotating target detection precision and the complex scene robustness are obviously improved.

The invention relates to the field of image comparison, and discloses a multi-layer PCB alignment deviation detection system and method based on image comparison, and the method comprises the steps: obtaining high-resolution image data before and after lamination of a multi-layer PCB, carrying out the unified registration preprocessing of an original image through the combination of a multi-mode image fusion algorithm and a geometric distortion correction technology, and obtaining a multi-layer PCB alignment deviation detection result; constructing a standardized image registration input set; key alignment feature extraction is carried out on the image registration input set, and a multilayer structure graph model is constructed based on a graph neural network; in combination with the alignment reference map, dynamically adjusting an image comparison window and a search region by adopting a region attention mechanism and a local adaptive matching algorithm, and constructing an alignment deviation mapping map; constructing a deviation evolution model by using a time sequence behavior recognition network according to the constructed alignment deviation mapping graph and historical process deviation data; and performing comprehensive evaluation on the image registration input set based on the deviation evolution model and the generated early warning information. The method has the advantage of improving the accurate detection level.

The invention relates to the technical field of industrial product surface quality detection, in particular to a flexible display module surface defect image recognition method, which comprises the following steps: acquiring a plurality of surface images of a flexible display module under different light sources and carrying out distortion removal processing on the surface images; reconstructing and generating three-dimensional reference point cloud data representing the current curved surface form of the module; re-projecting the distorted image to the ideal rigid plane according to the relationship, generating a plurality of corrected images, and generating a plurality of corrected images to eliminate geometric and luminosity distortion introduced by flexible deformation; obtaining a defect candidate area binary image; and extracting a multi-dimensional feature vector of the defect candidate region from the binary image of the defect candidate region, and classifying the feature vector by using a pre-trained defect classification model to obtain a defect identification result. Through the three-dimensional reference point cloud reconstruction and image re-projection technology, the problem of geometric distortion caused by surface deformation of the flexible display module is effectively solved, and misjudgment and missed judgment are avoided.

The invention provides a hull surface defect detection system based on machine vision, and relates to the technical field of data processing. The image correction module is used for carrying out illumination equalization processing and geometric distortion correction; the region construction module is used for identifying a defect-free stable region and generating reference region data which comprises a brightness model and a texture model; the candidate generation module is used for detecting a region where texture interruption or abnormal bright spots exist locally to form candidate defect data, and the candidate defect data comprise pixel positions and local contrast parameters; the stability judgment module is used for carrying out projection matching in the multiple frames of images and simultaneously carrying out joint comparison with the brightness model and the texture model of the reference area data to form real defect data and false defect data; the result output module is used for generating a detection result containing defect coordinates, defect contours, image frame numbers and interference sample prompts; the accuracy of hull surface defect detection is improved.

The invention belongs to the field of image generation, particularly relates to a three-dimensional image generation method based on spherical reflection cavity parameter optimization, and aims to solve the problems of low imaging precision and poor adaptability caused by lack of a collaborative optimization mechanism in the prior art. The method comprises the following steps: acquiring optical characteristics of the inner wall of a cavity to generate a reflection path offset comparison table, measuring deformation deviation of the cavity to construct a geometric compensation mapping relation, and monitoring environmental disturbance in real time and converting the environmental disturbance into a noise intensity coefficient; inputting the target three-dimensional image into a pre-compensation processor, synchronously executing optical path compensation, geometric distortion correction and environmental noise suppression, and outputting a pre-corrected image to projection equipment; capturing an actual projection through an image sensor, extracting difference data with the pre-corrected image, and decomposing the difference data into a system error component and an environment interference component; and feeding back the updated parameters to the processor to regenerate a final three-dimensional image. According to the invention, multi-error-source cooperative correction and dynamic optimization are realized, and the imaging precision and adaptability are significantly improved.

The invention discloses an image video de-disturbing method and system based on a time domain attention mechanism. The method comprises the following steps: inputting a continuous multi-frame turbulence degraded image sequence; performing inter-frame registration and alignment to compensate for geometric distortion; extracting spatial features through a multi-scale feature extraction network; enhancing spatial feature representation through the spatial attention module, performing learnable frequency domain filtering through the frequency domain attention module and fusing time sequence features through the time domain attention module in sequence; and finally, integrating the features to reconstruct a clear image after turbulence is removed. According to the method, a time domain attention mechanism is innovatively introduced, the time sequence correlation between video frames is effectively utilized, and the problem of inter-frame inconsistency existing in video restoration of an existing single-frame processing method is solved. Experiments show that the method is obviously superior to the existing mainstream method in objective indexes such as PSNR and SSIM, and the time sequence continuity and stability of the restored video can be effectively enhanced.

The invention relates to the technical field of projectors, in particular to an intelligent picture geometric distortion control method based on machine vision and an image projector, and the method comprises the steps: obtaining image data of a to-be-corrected projection picture, and carrying out the feature analysis of the image data, so as to obtain a distortion feature value; matching a correction optimization model corresponding to the distortion characteristic value, so as to call a correction parameter from a preset correction parameter library through the correction optimization model; simulating a correction process corresponding to the projection picture to be corrected according to the correction parameter; the correction parameters are adjusted according to the correction process, and the projection picture to be corrected is corrected according to the adjusted correction parameters, the corresponding correction optimization model is matched based on the characteristic values, the adaptive correction parameters can be dynamically called from the preset parameter library, the limitation of a traditional fixed algorithm is changed, and the correction efficiency is improved. The correction strategy is highly coupled with the actual distortion scene, and the correction efficiency and pertinence are remarkably improved.

The invention discloses an image digitization method, system and device and a storage medium, and relates to the field of data processing. The method comprises the following steps: acquiring spectral data, three-dimensional shape data and material reflectivity parameters of a target image; based on the three-dimensional shape data, analyzing surface deformation characteristics of the target image, and generating a geometric distortion distribution diagram; generating geometric correction parameters according to the surface deformation features, and reconstructing pixel positions of the geometric distortion distribution diagram to obtain reconstructed image data; and mapping the spectral data to a preset color space, matching a pre-established color database, and adjusting the color component of the reconstructed image data to obtain corrected image data. By implementing the technical scheme provided by the invention, a set of traditional Chinese painting digitization technical system considering high-precision acquisition, intelligent correction and efficient management can be developed.

The invention relates to the technical field of multispectral imaging, in particular to a multispectral imaging fused pattern recognition image acquisition method and device, and the method comprises the following steps: obtaining a multiband substrate response value through a multispectral sensor, generating response difference distribution through moving average filtering, inputting cross-correlation analysis, extracting a space offset vector, and compensating displacement; after the signal intensity of the mark points is captured, Gaussian fitting is carried out to generate registration coordinates, triangular mesh division is carried out, offset is calculated, interpolation mapping is carried out, a registration error distribution diagram is generated, bilinear interpolation remapping is carried out to collect time sequence parameters, and a multispectral imaging data set is output. According to the method, noise is eliminated through moving average filtering, offset vectors are extracted in a cross-correlation mode, array displacement is compensated, mark points are positioned through two-dimensional Gaussian fitting, a registration error graph is constructed through triangular meshes, parameters are collected through bilinear interpolation remapping, geometric distortion is corrected, feature consistency and collection synchronism are enhanced, and the imaging resolution ratio and the signal-to-noise ratio are improved.

The invention belongs to the technical field of engineering information intelligent processing, and relates to a highway bridge drawing multi-modal information extraction and semantic understanding method and system.The method comprises the steps that firstly, self-adaptive judgment is conducted on a vector drawing and a scanning drawing, geometric distortion correction, drawing frame and title bar positioning and layout segmentation are completed, and then the vector drawing and the scanning drawing are obtained; constructing a hierarchical document structure comprising texts, tables, images and two-dimensional drawing objects; then, a front-end target detection network and a rear-end cross-modal document understanding model are fused, and detection and relation reasoning of elements such as view blocks, labels, symbols and tables are achieved; and image-text feature alignment is further performed by using a visual coding network and a text coding network, structural description conforming to engineering semantics is generated by means of a multi-modal language model, an engineering parameter database is established, and parameter query and multi-modal question and answer output are supported. According to the method, the accuracy and efficiency of automatic acquisition and semantic understanding of the key information of the highway bridge drawing are improved.

The invention provides an unmanned aerial vehicle target identification and positioning method and system based on multispectral fusion. The method comprises the following steps: firstly, acquiring multispectral image data, then analyzing the multispectral image data of multiple time phases into pixel-level data, then identifying an early infection area of the plant diseases and insect pests, then constructing a geometric distortion correction model, and then determining the relative position of the early infection area of the plant diseases and insect pests through multi-view space intersection calculation. And finally, fusing the real-time differential global navigation satellite system positioning data of the unmanned aerial vehicle and the relative position of the disease and insect pest early-stage infection area to output absolute geographic coordinates of the disease and insect pest early-stage infection area. According to the technical scheme provided by the invention, accurate conversion from a local coordinate system to a global geographic coordinate system is realized, an exact spatial position basis is provided for precise agricultural operation, and precise positioning of a pest and disease damage area is realized.

The invention relates to the technical field of computer vision and medical image processing, in particular to an intra-intestinal monocular pose estimation method based on diffusion model deformation field prediction, which comprises the following steps of: inputting two frames of enteroscopy images of an intestinal capsule robot with adjacent time sequences, respectively extracting image features of front and back frames of enteroscopy images through a parameter-shared double-path depth map generation network, and generating corresponding depth maps; solving six-degree-of-freedom pose parameters of the intestinal capsule robot by utilizing the pose estimation network; generating a three-dimensional non-rigid deformation field through the non-rigid deformation field prediction network based on the physical regularization diffusion model; reconstructing a composite image aligned with the real previous frame of enteroscope image through an image synthesis module; completing self-supervised training; the problems that in an existing intestinal capsule robot positioning technology, a non-rigid deformation field lacks biophysical constraints, three-dimensional reconstruction geometric distortion is caused by monocular information, and collaborative optimization is lacked in deformation field and pose parameter decoupling are solved.

The invention relates to a multi-modal image matching method and system based on learning features and epipolar geometric constraints. The method comprises the following steps: carrying out edge enhancement processing on an input image through wavelet transform; extracting a multi-scale dense feature map based on the transformed convolutional neural network, and generating a feature descriptor with rotation and scale invariance in combination with principal direction normalization; adopting an FLANN algorithm and dynamic distance constraint to realize preliminary feature matching; and introducing a basic matrix construction and epipolar geometric consistency verification mechanism, and eliminating mismatching point pairs in combination with an RANSAC affine constraint model. According to the method, image enhancement, deep learning and geometric verification strategies are fused, the problems of radiation nonlinearity and geometric distortion caused by imaging mechanism differences among multi-modal images are effectively solved, the matching precision and robustness are improved, and the method is suitable for remote sensing application scenes such as optical-SAR registration, multi-source image fusion and earth surface change detection.

The invention discloses a farmland furrow three-dimensional point cloud data correction method and system based on a laser radar, and relates to the technical field of three-dimensional image processing based on the laser radar, in particular to a three-dimensional image geometric correction technology. The problems that original point cloud data directly obtained under the dynamic operation condition of an existing agricultural machine has geometric distortion, and the shape of a furrow cannot be truly reflected are solved; meanwhile, the problems that furrow quality evaluation seriously depends on low-efficiency manual means, the adaptability of an existing automatic technical scheme to an unstructured field environment is insufficient, and robust, real-time and high-precision perception of a real three-dimensional shape of a furrow cannot be achieved are solved. According to the method, dynamic tilt correction is carried out through real-time acquisition of point cloud and agricultural machine attitude information, boundary feature points are extracted, boundary lines are fitted, and a three-dimensional point cloud set of a single furrow is reconstructed by fusing a self-adaptive clustering algorithm of spatial prior. The method is suitable for the fields of farmland tillage and soil preparation quality detection, precision agricultural management, agricultural machinery intelligent operation and the like under the dynamic operation condition of agricultural machinery.

The invention relates to a multi-modal perception fusion target following method and system. The method comprises the following steps: adopting an improved KCF algorithm to realize a closed-loop process of target tracking, multi-scale space construction, feature fusion, response calculation, optimal scale decision and model updating; designing a four-level shielding processing mechanism fusing motion prediction and depth verification, and realizing tracking recovery in a shielding scene through shielding judgment, motion prediction, fine search and template protection; constructing a target distance mapping function fusing geometric distortion correction and attitude compensation, and realizing high-precision distance estimation based on monocular vision; laser radar point cloud information is integrated, an obstacle threat degree model is constructed, and cooperative path planning of following and obstacle avoidance is realized in combination with an improved TEB algorithm; and designing a linear velocity control law and an angular velocity control law based on the distance deviation and the azimuth angle deviation, and driving the robot to complete target following motion. According to the invention, high-precision and robust following of the robot to the target can be realized.

The invention discloses a self-adaptive text extraction method and system based on artificial intelligence, and the method comprises the steps: carrying out the analysis of the document structure entropy of an example document set, quantifying the noise density, geometric distortion degree and background complexity of the example document set, and carrying out the self-adaptive selection of a preprocessing assembly line intensity grade according to the above; dynamically configuring image preprocessing parameters and AI recognition model parameters, and generating a recognition engine instance to output a preliminary recognition text; after regularized coarse screening extraction is carried out based on key field description, a multi-candidate generation strategy is started for low-confidence-coefficient candidate text fragments, a multi-person cooperative verification process is triggered for lower-confidence-coefficient fragments, finally all the fragments are processed through a text standardization module, and structured text extraction information is output. According to the method, accurate adaptation of processing intensity is achieved through document quality quantitative evaluation, the extraction accuracy and system robustness of complex heterogeneous documents are effectively improved through a multi-level confidence coefficient verification mechanism, and the identification error risk caused by image quality fluctuation or rule solidification is reduced.

The invention discloses a glomerular segmentation method based on multi-scale Mama and dynamic sampling, and relates to the technical field of medical images. According to the method, the input multi-scale features are efficiently represented by using a continuous state transition mechanism, the structural association information extraction capability among different scales is enhanced, in addition, the problem of geometric distortion in dynamic sampling is solved by introducing position coding into the dynamic sampling method and adding position constraint information, and the dynamic sampling efficiency is improved based on a U-Net model. An MS-Mama module is gradually added in the encoder stage, and an existing bilinear interpolation method is replaced by a position sensing dynamic sampling module; and then, serialized representation is performed on multi-scale features based on a selective state space mechanism, the dependence of Mama on specific scale features is reduced, the ability of extracting heterogeneous morphological features of lesion glomerulus is enhanced, the boundary positioning precision of lesion glomerulus is enhanced by adding position constraint information, and the geometric distortion problem in dynamic sampling is solved.

The invention discloses a multi-modal image registration method and system based on deformation adaptation and computer equipment, and the method comprises the steps: collecting a plurality of groups of multi-modal images, carrying out the gray standardization, and constructing a diversified registration data set; building a registration network model comprising a pyramid coding module, a deformation adaptive module, a cross-modal interaction module and a registration parameter estimation module; inputting an image pair into the modules in sequence, respectively extracting basic feature mapping, deformation feature mapping and interaction enhancement feature mapping, and finally outputting an estimation conversion parameter matrix; a training process is supervised through a preset loss function, optimal network parameters are selected, and a trained registration model is obtained; in practical application, an image pair to be registered is input into the trained model, a conversion parameter matrix is obtained, and image registration is completed. The multi-modal image registration performance can be effectively improved, and the method still has good robustness and adaptability especially under the condition that serious geometric distortion and significant modal difference exist.

The invention discloses a high-fidelity three-dimensional Gaussian sputtering (3DGS) lightweight method for resource-constrained equipment. The method aims at solving the problems of high storage and computing resource consumption caused by the fact that a large number of parameters are stored in an existing 3DGS technology, and geometric distortion possibly occurring when details of a scene center are processed is overcome. The core of the method lies in a multi-stage progressive optimization framework, and the framework cooperatively applies four key technologies of Gaussian cutting and opacity regularization, dynamic spherical harmonic function adjustment, entropy constraint vector quantization and coordinate space shrinkage. Wherein in Gaussian clipping, redundant gauss are eliminated by quantifying the contribution degree of a Gaussian unit; the dynamic spherical harmonic function adjustment adaptively adjusts the order of color coding according to the scene complexity; the entropy constraint vector quantization is used for compressing a plurality of Gaussian attributes so as to realize more compact representation; and the coordinate space shrinkage is realized through nonlinear transformation, so that the rendering precision of details of the center of the scene is remarkably improved. According to the method, while the rendering precision and quality are kept, remarkable storage compression is realized, and the method is particularly suitable for deployment of resource-limited platforms such as mobile equipment.

The invention provides a single building three-dimensional reconstruction method based on improved semantic segmentation, and relates to the technical field of three-dimensional reconstruction. The method comprises the following steps: calculating pose parameters of a multi-view inclined image by using a multi-view image processing technology and removing geometric distortion influence; inputting the corrected image into a semantic segmentation module, extracting a building pixel region and labeling attributes, and generating a dichotomy attribute tag graph; performing depth estimation on the corrected image by using a DPT network model to generate a depth map; combining the dichotomy attribute tag map with the depth map to generate an initial point cloud; performing multi-view initial point cloud fusion by using an ICP algorithm to generate a global scene dense point cloud model; and according to the attribute category, automatically separating the building from the dense point cloud model of the global scene to obtain a point cloud model of a single building. According to the method, accurate recognition and three-dimensional reconstruction of the single building are realized, and the problems of mistaken extraction and over-extraction of the single building in an existing method are solved.

The invention relates to the technical field of unmanned aerial vehicle remote sensing detection, and discloses a change detection method and system for an unmanned aerial vehicle remote sensing image. The method comprises the following steps: acquiring a multi-temporal unmanned aerial vehicle remote sensing image data set containing high-resolution image data acquired at different times in a target area; carrying out geometric correction and radiation normalization processing on the data set to generate a standardized remote sensing image data set, and eliminating geometric distortion and illumination difference between images; then extracting multi-scale spatial features in the standardized remote sensing image data set, generating a spatial feature matrix containing texture, spectrum and structure information, and comprehensively capturing image details; inputting the spatial feature matrix into a time sequence feature fusion network, calculating feature differences among different time phase images, and generating a change feature vector; and finally, constructing a dynamic change detection model based on the change feature vector, and generating a target area earth surface change detection result.

The invention discloses a steep cliff three-dimensional modeling method and system based on structural plane constraint, and relates to the technical field of three-dimensional geographic information system.The steep cliff three-dimensional modeling method comprises the steps that steep cliff area geological data including steep cliff multi-view images and steep cliff surface dense point cloud data are collected, and a geological prior knowledge base is constructed; an image recognition technology is used for processing a steep cliff multi-view image, point cloud geometric analysis is combined, and the structural plane spatial orientation including a rock stratum interface and a joint plane is recognized. According to the method, geological priori knowledge is fused, the geometric distortion problem of a traditional digital elevation model in a cliff area is effectively solved, the spatial orientation of a rock stratum interface and a joint surface is extracted by using an image recognition technology, and the spatial orientation is used as a hard constraint condition to construct a control triangulation network framework; the problems of forced smoothing errors of a regular grid DEM and a suspended surface of a triangular irregular network are avoided, and the rock stratum structure, the crack distribution and the dangerous rock body form of the cliff can be precisely restored through the self-adaptive triangulation algorithm.

A multi-feature three-dimensional reconstruction 3D Gaussian method based on laser vision comprises the steps that laser radar point cloud and camera images are aligned through space-time calibration, and a unified coordinate system is established; extracting geometric features by using point cloud data acquired by the Lidar point cloud, and initializing a Gaussian ellipsoid according to the Lidar point cloud; optimizing the brightness, the contrast ratio and the structural similarity of the rendered image and the real image by combining the mean absolute error L1 and the structural similarity SSIM; the curvatures of Gaussian ellipsoids of K-nearest neighbors are forced to be consistent, and long and short axes and line and surface features of the Gaussian ellipsoids are aligned to reduce geometric distortion; the distribution density of 3D Gaussian is dynamically adjusted through line / surface features and visual structure information extracted by Lidar, and balance between geometric detail enhancement and calculation efficiency is achieved. According to the method, the position, the scale and the rotation parameters of Gaussian are uniformly optimized, and the details and the calculation efficiency of the model are balanced while the consistency of the model structure is improved.

The invention relates to the technical field of electron microscopic imaging, in particular to a correction method and device for geometric distortion in scanning electron microscope measurement, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a scanning electron microscope image of a standard sample; calculating an orthogonal included angle parameter and a length-width ratio parameter of the scanning electron microscope image; iteratively adjusting the angle correction parameter and the gain correction parameter under a fixed scanning rotation angle until a preset local calibration constraint condition is met; verifying the orthogonal included angle parameter and the length-width ratio parameter under each scanning rotation angle; and when the orthogonality included angle parameter and the length-width ratio parameter do not meet a preset global stability constraint condition, adjusting a relative gain coefficient, and re-executing calibration verification under each scanning rotation angle until a full-angle correction requirement is met. By adopting the method, high precision and stability of imaging correction can be kept in a full-angle range.

The invention relates to the technical field of bill text recognition, and discloses a bill text recognition system and method based on deep learning. The method comprises the following steps: acquiring target bill original image data containing a multi-channel pixel matrix and spatial resolution information; based on a matching result of the bill edge features and a preset template, geometric distortion correction is carried out on the original image, and a corrected bill image is generated; inputting the corrected image into a pre-training text region detection network to obtain positioning information containing text line boundary coordinates and region confidence; text line image blocks are extracted according to the positioning information, character segmentation preprocessing is executed, and a character-level image sequence is generated; calling a deep character recognition model to classify the sequence character by character, and generating an initial text recognition result; semantic verification and error correction are performed on the initial result based on the bill type knowledge base, final structured text data are generated, and the processing requirements of bills of different types and qualities can be met.

The invention discloses a G-PCC Trisound-oriented point cloud distortion repairing method, which is characterized by comprising the following steps of: acquiring a distorted point cloud to be repaired and preprocessing the distorted point cloud to obtain a voxelized point cloud and a plurality of cubes; constructing a training set and a to-be-trained geometric distortion repair network, and training to obtain a trained geometric distortion repair network; inputting all the cubes into the trained geometric distortion repair network for multi-scale feature extraction and geometric repair to obtain a complete geometric repair point cloud; performing color restoration according to the voxelized point cloud and the complete geometric restoration point cloud to obtain a geometric and color combined restoration point cloud; the method has the advantages that the geometric structure integrity of the compressed distorted point cloud can be remarkably improved, and details and smoothness lost due to quantization and surface approximation are effectively recovered; meanwhile, color information is accurately recovered, and color distortion and blurring are reduced; the joint repair mechanism efficiently solves the common geometric and color distortion problem in compression, and the visual quality and geometric precision of the reconstructed point cloud are remarkably improved.

The invention discloses a CNN-based high-performance lightweight optimization two-dimensional code recognition method, which belongs to the field of intelligent two-dimensional code recognition and comprises four steps of image preprocessing, lightweight CNN model construction and training, multi-scale feature extraction and decoding post-processing. According to the method, noise is suppressed through improved filtering, two-dimensional code edge details are reserved, and geometric distortion is corrected in combination with perspective transformation; the model parameter quantity and the calculation quantity are reduced through depth separable convolution, a channel attention module is embedded to strengthen key features, and a lightweight CNN model adaptive to the resource-constrained equipment is constructed; complete features are extracted through multi-scale feature fusion, and a decoding result is optimized in cooperation with RS code error correction and repeated recognition voting. Therefore, high-precision and real-time identification of the two-dimensional code in complex environments of uneven illumination, noise interference, distortion and the like is realized on a mobile terminal, an embedded device and the like, and the anti-interference capability and scene adaptability of the method are effectively improved.

The invention discloses a structural semantic guided bridge point cloud Gaussian splash modeling and rendering method, and belongs to the technical field of computer vision and bridge engineering digitization. The method comprises the steps of collecting a multi-view image of a bridge, and obtaining a sparse point cloud and a camera pose through a motion recovery structure technology; dense point clouds are generated, deep learning semantic segmentation is carried out, and key components such as piers, bridge floors and cables are identified; mapping the semantic tag back to the sparse point cloud; performing differential Gaussian element initialization on the sparse point cloud based on semantic tags, and endowing different parts with initial shapes conforming to geometric characteristics of the different parts; a geometric constraint loss function is introduced in the optimization process for joint optimization; and finally, outputting a Gaussian splash model of a high-fidelity and accurate geometric structure. According to the method, the problems of model redundancy and geometric distortion when a traditional Gaussian splashing technology is used for processing a bridge scene are solved, and lightweight and high-precision reconstruction and real-time rendering of the bridge model are realized.

The invention discloses a neural implicit vision SLAM (Simultaneous Localization and Mapping) method based on dynamic perception. The method aims at solving the core technical problems that an existing visual SLAM method is insufficient in robustness, poor in global consistency, large in calculation overhead and the like in challenging environments such as dynamic scenes, weak texture areas and violent illumination changes. According to the method, the feature processing capability of deep learning, efficient dynamic object perception, advanced neural implicit mapping and a global optimization mechanism are integrated, so that more accurate camera pose estimation and higher-quality static environment map construction are realized. In the tracking module, a six-step workflow based on mask guidance is adopted, dynamic objects are filtered from the source, frame-level pre-screening is carried out, and the robustness and the calculation efficiency of the system are remarkably improved. In a dynamic local mapping module, a pixel-level fusion method based on transmission probability and inverse variance weight is innovatively adopted, texture blurring and geometric distortion at the boundary of a plurality of sub-maps are effectively inhibited, and the visual quality of a global map is improved. Besides, by introducing a loop candidate frame reordering strategy based on pose uncertainty weighting in loop detection, visual similarity and geometric credibility can be combined, the false detection rate is effectively reduced, and global consistency and long-term precision of the map are ensured.

The invention relates to the technical field of precision shell contour metering, and discloses an online metering method for a precision shell contour, and the method comprises the steps: obtaining an original sampling point cloud and a feature point group of a to-be-measured workpiece in a conveying state; resolving a real-time pose matrix of the to-be-measured workpiece relative to a preset measurement reference by using a spatial topology constraint relationship of the feature point groups; carrying out differential processing on a real-time pose matrix in the sampling sequence, and synthesizing an instantaneous motion vector corresponding to a sampling moment; in combination with a signal response time delay constant of the sampling system, phase compensation is carried out on a motion displacement deviator induced by signal conversion delay, and a corrected sampling point cloud is generated; according to the method, the geometric distortion of the point cloud generated by response delay of the sensor is solved, the problem of instantaneous drift of the measurement reference in an unsteady state conveying state is solved, and the contour reduction precision in a complex working condition is improved.