107 results about "Point cloud processing" patented technology

This invention provides a method, system, and device for workpiece segmentation and pose estimation in stacked scenarios, relating to the field of 3D point cloud processing technology, aiming to solve the problems of inaccurate segmentation and large registration errors in traditional algorithms. The method includes: downsampling a standard model STL file constructed by CAD software to obtain a point cloud file; acquiring 3D point cloud data of stacked workpieces using a 3D acquisition device; segmenting the point cloud data using an improved Euclidean clustering algorithm after preprocessing to obtain individual workpiece target point clouds; extracting PPF features from the preset model point cloud and the target point cloud, and generating an initial pose through coarse PPF registration; generating multiple sets of perturbed poses based on the initial pose, and verifying and selecting the optimal pose through voxel voting; using the optimal pose as the initial value, obtaining the final pose through fine ICP registration. Ultimately, this improves segmentation accuracy and registration precision, accelerates processing speed, and is suitable for industrial automated sorting and positioning scenarios, demonstrating strong practicality.

This invention discloses a method and system for feature extraction of multi-layer stacked cardboard boxes based on 3D vision. Addressing the problem that existing technologies cannot accurately quantify surface anomalies and multi-dimensional stacking features of cardboard boxes in complex stacking scenarios, this invention acquires the original point cloud and RGB images of the stacking scene; after preprocessing, a height distribution histogram is constructed and clustered to obtain initial point cloud clusters; warped anomalies are eliminated through curvature analysis and normal vector consistency to obtain accurate upper surface point clouds; principal component analysis is used to obtain the length, width, and normal vector direction of the cardboard box; basic geometric features, damage features, occlusion relationships, overhang / overlap features, tilt features, and multi-layer interlacing features are extracted by combining RGB images; finally, the feature vector of each cardboard box is output in a structured manner. This invention achieves refined point cloud processing, effectively eliminates warped points, systematically quantifies complex stacking features, and integrates multimodal data, significantly improving the ability to identify and extract features from cardboard boxes in complex logistics scenarios.

The disclosure provides a point cloud monitoring method, device and equipment, relates to the technical field of image processing, specifically to the technical field of point cloud processing, and can be applied to an automatic driving scene. A specific embodiment of the method comprises the following steps: acquiring point cloud data collected by a laser radar to a field programmable logic gate array (FPGA); acquiring configuration parameters of a first monitoring item configured by an advanced reduced instruction set machine (ARM) to the FPGA; and processing the point cloud data using the configuration parameters on the FPGA to obtain a monitoring result of the first monitoring item. The embodiment is suitable for monitoring more point cloud data and can obtain a good monitoring result.

The present application relates to the field of three-dimensional point cloud processing and machine learning, especially to a welding seam point cloud recognition method based on improved gradient boosting tree, which comprises the following steps: collecting three-dimensional point cloud data of welding seams with different shapes, pre-processing and extracting ROI region, calculating geometric features under a preset neighborhood scale to construct training set and test set, taking LightGBM as a base classifier, optimizing key hyperparameters by using improved glide snake optimization algorithm, obtaining LFMGSO-LightGBM model and verifying it, and realizing accurate recognition and classification of input welding seam point cloud. The welding seam point cloud recognition method based on improved gradient boosting tree provided by the present application improves the population initialization, search step and guide mechanism of the improved glide snake optimization algorithm, improves the initial solution space distribution quality, realizes the dynamic balance of global exploration and local development, and enhances the stability and anti-local optimal ability of the optimization process.

The present application relates to the technical field of laser radar point cloud processing and human posture reconstruction, in particular to a human three-dimensional structure modeling method based on single-station line-scan laser radar, in which, according to the determined effective layering meeting the human characteristics in the height direction and the offset of the center of gravity in the horizontal direction, continuity is judged, the effective layering meeting the continuity is clustered to obtain a human candidate region, the point cloud in the height range corresponding to the human candidate region is intercepted to form a human layered point cloud set, and the boundary range of the human in the height direction is determined. The present application does not need multiple sensors to cooperate or a complex learning model, and is suitable for human three-dimensional perception and structure analysis in complex environments.

This application provides a method and related device for concurrent construction and resource management of 3D point cloud data. In this method, the real-time 3D point cloud processing system first pre-allocates multiple sets of computing resources with fixed physical memory addresses and establishes atomic pointer mappings. After the current frame processing ends, the real-time 3D point cloud processing system switches the current frame pointer to a cleaned set of idle computing resources through atomic pointer mapping switching, achieving instantaneous transition from the current frame to the next frame. Next, the real-time 3D point cloud processing system performs lightweight initialization on the next frame's computing resource set, enabling the processing thread to immediately enter the processing of the next frame's point cloud data. Finally, the real-time 3D point cloud processing system performs asynchronous traversal cleanup on the uncleaned computing resource set, separating the time-consuming depth traversal operation from the main processing path. This method alleviates the technical problem of time-consuming inter-frame switching and improves the real-time processing frame rate under high-load scenarios.

The present application relates to the technical fields of image processing and point cloud processing, and discloses a line laser scanning three-dimensional reconstruction shear deformation correction method, system, computer readable storage medium and computer program product, the method comprises the following steps: according to the ratio of the height measurement value and the height nominal value of the height block, the first direction deflection angle of the line laser scanning camera is calculated;And the included angle between the straight line L and the predetermined horizontal line is measured, and the second direction deflection angle of the line laser scanning camera is obtained;The third direction deflection angle of the line laser scanning camera is obtained by using the constraint relationship that the adjacent edges of the checkerboard grid are perpendicular to each other;The photographed image is sequentially deflected and corrected according to the order of the three direction deflection angles, and then outputted.The present application does not need camera internal and external parameters, greatly reduces the system calibration difficulty, improves the angle solving precision through the joint estimation of the height block and the checkerboard grid, has strong algorithm robustness, can effectively remove the three-dimensional shear deformation caused by splicing, and has high algorithm reconstruction precision.

This invention relates to the field of computer-aided design technology, specifically to a dynamic management system for bridge construction progress based on BIM technology. The system includes a progress point cloud processing module, which generates a downsampled point cloud based on construction point cloud data and the associated computer-aided design boundary volume of the main beam. It then extracts the distribution feature vectors of coordinate points within the downsampled point cloud and the theoretical point cloud, and calculates the bounding box translation and rotation parameters using these feature vectors to generate initial alignment data. In this invention, the feasibility assessment of pouring is more specific, and the adjustment information is more targeted. A continuous correlation is established between progress verification, deviation warning, layout correction, and on-site decision-making during bridge construction, thereby improving the depth of information expression during the construction phase in a computer-aided design scenario, the pertinence of construction organization, and the effectiveness of transforming the design model into construction management results.

The application provides a sorting method and a sorting system for a target object, which comprises the following steps: acquiring a natural language describing the target object and an environment image of a work area where the target object is located; generating a target three-dimensional point cloud including the target object based on the environment image and the natural language; performing spatial clustering on the target three-dimensional point cloud to separate a target three-dimensional point cloud cluster corresponding to the target object; determining a geometric feature of the target object according to the target three-dimensional point cloud cluster, determining a target grasping pose according to a type of a mechanical arm of a robot corresponding to the work area and a geometric constraint; and controlling the robot to sort the target object based on the target grasping pose. The application realizes accurate positioning for unknown objects. Meanwhile, an advanced image segmentation large model and a three-dimensional point cloud processing technology are deeply integrated to construct an accurate mapping and geometric topology analysis mechanism from two-dimensional pixels to three-dimensional physical space, which can be combined with different industrial robots and physical kinematic constraints of end effectors.

The application belongs to the technical field of point cloud processing, and discloses a multi-modal semantic mapping method and system based on a master-slave architecture. The method comprises the following steps: S1, extracting feature points in point cloud data and calculating the covariance matrix of the neighborhood point cloud of each feature point; S2, solving the optimal pose corresponding to the minimum of the target function; S3, converting the obtained point cloud data into a world coordinate system using the optimal pose, fusing all frame point cloud data in the world coordinate system to form a three-dimensional point cloud map, projecting the three-dimensional point cloud map to a camera coordinate system, and inputting the three-dimensional point cloud map projected to the camera coordinate system into a semantic segmentation neural network to obtain the semantic label corresponding to each point cloud in the three-dimensional point cloud map, thereby achieving semantic mapping. Through the application, the problems of weak dynamic adaptability, insufficient model generalization ability and long-term running precision decay of the traditional three-dimensional semantic mapping method in a dynamic complex scene are solved.

This disclosure provides a point cloud processing method, apparatus, computing device, and storage medium. The point cloud processing method includes: acquiring object states in an initial point cloud frame; extracting object point clouds from the initial point cloud frame based on the object states; calculating a shape code of the initial point cloud frame based on a shape model and the object point clouds in the initial point cloud frame, wherein the shape code is used to characterize the surface shape of the object; and calculating the object states and / or shape codes of corresponding objects in other point cloud frames of the point cloud sequence based on the shape code and object states of the initial point cloud frame. Embodiments of this disclosure combine object tracking and shape reconstruction based on a shape model, improving the accuracy of object tracking and the fidelity of shape reconstruction.

The application discloses a point cloud outlier removal method, a point cloud processing method, a device and related equipment. The method comprises the following steps: dividing a point cloud to be processed into multiple layers along a reference coordinate axis; determining whether to remove each point in each layer of the point cloud based on the bounding box of the point, and obtaining a second point cloud; obtaining the average nearest neighbor Euclidean distance of the second point cloud, that is, the nearest neighbor Euclidean distance of each point; and determining whether to remove each point in the second point cloud based on the difference between the nearest neighbor Euclidean distance of the point and the average nearest neighbor Euclidean distance, and the standard deviation of the nearest neighbor Euclidean distance of each point in the second point cloud and the average nearest neighbor Euclidean distance. Through the above steps, the application can remove the outliers in the point cloud more intuitively and conveniently while retaining the detailed features.

This invention belongs to the field of laser processing modification technology, and relates to a teaching-free laser shock strengthening method and system based on a 3D model and laser point cloud. The system includes a 3D modeling module, a point cloud acquisition module, a point cloud processing module, a path planning module, and a control module. It identifies the area to be strengthened using a 3D information model and plans the initial processing path. High-precision point clouds are acquired using a line laser sensor and monocular vision fusion method, and the deviation matrix of the workpiece to be processed is calculated through registration. Based on the deviation matrix, the initial processing path is transformed and corrected point by point, ultimately generating a robot execution program to control the laser processing device to complete the automated laser shock strengthening operation. This invention achieves fully automated operation from 3D information model to robot operation, effectively solving the problems of traditional methods such as reliance on manual teaching, low positioning accuracy, and difficulty in assessing accessibility.

This invention disclosed a full-space intelligent detection method and system for underground drainage networks, as well as storage media, including the following steps: image acquisition, intelligent image denoising, internal pipe defect segmentation, concealed defect detection around the pipe, 3D reconstruction with volume quantification, and pipeline life prediction. This invention introduced a bionic four-wheel-drive, all-terrain detection robot that can effectively navigate through mud and flowing water-challenges that hinder traditional detection devices. By leveraging deep learning algorithms as well as various techniques of computing vision, 3D reconstruction, and point cloud processing, the system thoroughly analyzed collected data to determine defect types and precise locations. Utilizing this analysis, precise location of different defect type and quantitative measurement of their dimensions can be realized. Based on the data analysis results, a deep-learning driven model was developed for predicting pipeline longevity to support maintenance staff with timely information on pipe defects and operational lifespan.

The application discloses a kind of filtering and remaining area spray trajectory replanning method and system across station has been sprayed area, belong to industrial robot vision and automatic spraying technical field.This system is composed of mobile chassis, six-axis robot arm, terminal RGB-D camera, spray gun, station visual label and supporting function module;Robot generates standard spraying trajectory by visual segmentation, point cloud processing in first station and records sprayed area data, after moving across station, relies on visual label to complete multi-frame fusion relocation, calculates inter-station coordinate transformation matrix, maps historical sprayed area to current coordinate system and eliminates from real-time point cloud, re-plans serpentine spraying trajectory for remaining to be sprayed area, and accurately controls spray gun on-off in combination with trajectory segmentation logic.The application solves the problem that traditional mobile spraying robot is prone to repeated spraying, uneven paint film and repeated teaching when working across station, realizes multi-station continuous automatic spraying, and the trajectory runs smoothly with high spraying consistency.

The present application relates to the technical field of point cloud processing, and provides a track structure point cloud weak supervision segmentation method and system based on SAM and PointNet++. The method comprises the following steps: collecting track structure point cloud data to obtain an unlabeled point cloud data set with RGB color; constructing a point cloud processing module based on SAM and a weak supervision point cloud segmentation module based on PointNet++; preprocessing the unlabeled point cloud data set to obtain point cloud training data with pseudo-labels and confidence weights; inputting the point cloud training data into the weak supervision point cloud segmentation module based on PointNet++ for training to obtain a trained point cloud semantic segmentation model, and performing semantic segmentation on the unlabeled point cloud by using the trained point cloud semantic segmentation model. The method realizes a training effect close to that of full supervision point cloud data under weak supervision, reduces the dependence of track structure point cloud segmentation on high-precision labeled point cloud data, and improves the segmentation efficiency and accuracy.

This invention discloses an adaptive compensation method for hydraulic support assembly errors based on laser tracking and positioning, relating to the field of hydraulic support assembly technology. The method includes: deploying laser radars at the four corners of the hydraulic support assembly platform to scan and collect original point cloud data of components and the platform; embedding an adaptive frequency-spatial domain joint filtering algorithm to filter out vibration and dust coupling noise, and collecting effective point cloud data. This invention constructs an accurate three-dimensional digital model of the hydraulic support and assembly platform through multi-laser radar collaborative scanning and high-precision point cloud processing. Combined with real-time coordinate transformation and a closed-loop compensation mechanism, it effectively eliminates subjective errors and random deviations in manual assembly. During dynamic assembly, it can continuously monitor the difference between the actual pose and the virtual model and perform adaptive fine-tuning, significantly improving the positional accuracy and attitude consistency of component assembly, providing a reliable guarantee for the overall structural stability and service life of the hydraulic support.

A point cloud processing method includes: obtaining decoding indication information of a point cloud bit-stream, the decoding indication information including at least one of a first indication parameter or a second indication parameter, the first indication parameter being set in sequence-level parameter information of a point cloud sequence and providing sequence-level decoding indication, the second indication parameter being set in frame-level or slice-level parameter information of the point cloud sequence and providing frame-level or slice-level decoding indication; and decoding the point cloud bit-stream according to the first indication parameter or the second indication parameter.

Embodiments of the present disclosure disclose a point cloud processing method and device, electronic equipment and storage medium, the method comprising: obtaining a reconstructed point cloud and an ideal point cloud, the reconstructed point cloud being obtained by reconstructing an image taken for a target screen, and the ideal point cloud being generated based on design data of the target screen, the target screen being provided with at least one visual marker for carrying corresponding identification information; matching the reconstructed point cloud with the ideal point cloud based on the identification information, determining an initial point set in the reconstructed point cloud that matches successfully, and determining a target point set in the ideal point cloud that matches successfully; determining a transformation parameter according to the initial point set and the target point set, correcting the reconstructed point cloud based on the transformation parameter; scaling the ideal point cloud based on the transformation parameter, completing the corrected reconstructed point cloud based on the scaled ideal point cloud, and obtaining a target point cloud; the embodiments of the present disclosure can improve the point cloud processing efficiency.

The application discloses a kind of based on three-dimensional point cloud image's Pinus massoniana partial phenotypic parameter automatic measurement method. Three-dimensional point cloud image of Pinus massoniana plant under different viewing angles is collected;Pretreatment obtains the three-dimensional point cloud of Pinus massoniana plant;Registration is carried out to obtain fused point cloud;Extract the minimum spanning tree and the skeleton point of main stem, and then obtain the main stem point cloud and leaf point cloud;According to the minimum spanning tree, main stem point cloud and leaf point cloud processing obtain main stem length, Pinus massoniana ground diameter, average leaf length and leaf number.The method of the application can automatically identify Pinus massoniana plant in three-dimensional point cloud image, and complete the strategy of main diameter length, ground diameter, average leaf length, leaf length, improve the detection efficiency, realize high-precision, high-efficiency crop phenotypic data acquisition, achieve fast, low-cost crop phenotypic parameter accurate and nondestructive testing.

This invention discloses a point cloud processing method and system based on spatial locality-preserving serialization and adaptive mesh reshaping. The method includes: first, acquiring raw 3D point cloud data and quantizing and sorting it into a one-dimensional locality-preserving sequence using spatial filling curves such as Morton codes; second, introducing an adaptive mesh reshaping strategy, dynamically configuring a 2D target canvas according to the number of input point clouds, using lossless rectangular mapping for small-scale point clouds, and employing a spatial continuous truncation mechanism to extract dense topological sub-blocks for large-scale scene point clouds; finally, inputting the generated 2D pseudo-image into a 2D attention network to extract features and outputting the processing result. This invention completely eliminates zero-filling artifacts in cross-modal mapping without relying on expensive 3D domain query operators, significantly improving the computational efficiency and feature representation accuracy of 3D point cloud data processing.

The small sample point cloud target classification method based on LSHW-Mamba network relates to the technical field of 3D point cloud processing and deep learning, and solves the problems that the existing technology focuses on feature mining in the Euclidean space domain, and under the condition of insufficient target samples, the feature representation ability of the target geometric structure is insufficient, and it is difficult to capture the key information for distinguishing similar objects. The present application explicitly analyzes the spectral response of the local geometric structure of the point cloud through the spherical harmonic function, combines the local spherical harmonic wavelet transform to realize the geometric feature extraction of the point cloud, simultaneously introduces the geometric gating mechanism to optimize the feature propagation efficiency of the Mamba model, models the target features in parallel with the Transformer attention branch, considers the relevance of the local and global features of the target, and improves the discrimination ability of the network model for the target under the small sample scene. The present application effectively solves the core problems of insufficient feature discrimination and decreased classification accuracy under the small sample condition.

The application discloses a high-precision digital maintenance method for GIS equipment of a transformer substation based on point cloud data, belongs to the technical field of digital maintenance of a power system, and comprises point cloud collection, preprocessing, modeling, checking and packaging. The preprocessing adopts NRGA denoising and VCHS simplification algorithms, introduces point neighborhood radius analysis and voxel convex hull feature calculation, and improves data quality; the modeling is based on an equipment structure segmentation and parameterization method, and ensures high precision; and the packaging adds an interactive function. The method solves the problems of low efficiency and poor precision of traditional technologies, improves the point cloud processing efficiency by 50%, the model error is less than 5%, supports intelligent operation and maintenance, and improves the safety of a power grid.

The present application relates to the field of intelligent filling planning, and more particularly to a goaf intelligent filling planning method based on machine vision and autonomous navigation. The content includes: collecting point cloud data of the goaf environment and registering to obtain point cloud data under the global coordinate system; based on the point cloud data under the global coordinate system, introducing a local spectral weighted index algorithm to generate a structure weighted geometric reconstruction index; based on the structure weighted geometric reconstruction index, identifying the filling points and constructing the filling point set; based on the filling point set, combining the structure weighted geometric reconstruction index, introducing a progressive filling planning algorithm to select the next filling point and generate an ordered filling path. The problems of traditional point cloud processing methods that are difficult to automatically identify structural abnormalities or high-risk areas, and filling path planning that fails to make comprehensive decisions combining structural risk and spatial distance, and is difficult to achieve continuous, controllable and risk-prior intelligent filling planning are solved.

To acquire high-precision point cloud data or 3D data. [Solution] A measuring jig 400 is provided, which includes a first sphere 410, a second sphere 420, a connecting rod 430 as a sphere position maintaining means for maintaining the relative positions of the first sphere 410 and the second sphere 420, and a rod body 440 provided so as to be parallel to the central axis of a virtual line connecting the center of the first sphere 410 and the center of the second sphere 420. The 3D scanner 100 is used to perform measurements with the measuring jig 400 inserted into a hole 6 formed in the object to be measured, with the axial direction of the rod body 440 coinciding with the depth direction of the hole 6. The point cloud processing device 300' calculates the virtual line connecting the center of the first sphere 410 and the center of the second sphere 420 as the depth direction of the hole 6.

The present application provides a point cloud processing method and device, electronic equipment and storage medium, and relates to the field of oral scanning data processing. The method comprises the following steps: acquiring an original oral point cloud, and determining an implant rod target point cloud from the original oral point cloud according to the spatial connectivity, color information and normal information of each sampling point in the original oral point cloud; performing directional bounding box normal segmentation on the implant rod target point cloud to obtain an implant rod pattern surface point cloud; and obtaining an implant rod pure pattern feature point cloud based on the implant rod pattern surface point cloud, thereby improving the feature stability, noise resistance and convergence speed of the subsequent registration algorithm, greatly reducing the calculation cost while ensuring the clinical precision, and meeting the rigid requirements of the oral implant navigation system for real-time performance, robustness and no manual intervention.

The application discloses a kind of based on dangerous chemical warehouse storage safety management system and method, it is related to dangerous chemical warehouse safety management technical field, including acquisition dangerous chemical warehouse three-dimensional point cloud data and carries out space denoising, generates the space point cloud set after purification;Identify high-density aggregation area in point cloud and extract warehouse internal entity partition contour, track profile trend change capture turnaround position and mark as space topological conversion node, connect adjacent node to form closed space segmentation polygon, define as independent storage management domain;Traversal each management domain generates netted inspection line with cross feature, sensor detection site is arranged in the geometric center of netted line, and the global safety management path scheme is output.The application realizes storage domain automatic division by three-dimensional point cloud processing and space topological segmentation, combines netted inspection line with center sensor arrangement optimization management path, adapts dangerous chemical warehouse space safety management demand.

The application discloses a line structure light type gear rotation scanning system and a rotation axis calibration method thereof, three coordinate systems of a camera coordinate system, a base coordinate system and a workpiece coordinate system and conversion relations are established in the scanning system; initial calibration of a hand-eye matrix is completed by using a joint arm measuring machine and point cloud processing software according to changes of incident structure light in the workpiece and the camera coordinate system; comprehensive optimization of whole rotation axis parameters is completed by using an initial calibration output matrix and an ICP algorithm; the method has the advantages of fast calibration speed, high efficiency, simple operation and good optimization effect.

The present application relates to a kind of point cloud processing and path planning method for automatic marking of workpiece, belong to workpiece automation processing technical field, to solve the problem of low efficiency, poor precision and existing system insufficient adaptability of artificial marking.The technical scheme includes: the three-dimensional model of workpiece is converted into point cloud, and the main plane is extracted by random sampling consistency algorithm;Adopt spatial index clustering optimization boundary;Based on principal component analysis, construct local coordinate system;Generate the minimum circumscribed rectangle boundary;Calculate the rotation quaternion posture and sample trajectory point;Define workpiece coordinate system to realize space transformation;Output standardized trajectory file.The method significantly improves the adaptability of complex surface, realizes high-precision trajectory automatic generation, reduces the dependence on artificial experience, and the standardized interface is convenient for integration robot system, promotes the intelligent upgrading of industrial marking process.