93 results about "Plane segmentation" patented technology

The invention relates to the technical field of medical image analysis, and discloses an HER2 positive breast cancer auxiliary prediction method based on multi-modal data fusion. According to the method, preoperative MRI, clinical pathology data and RNA-seq data meeting quality control standards are obtained through dynamic case screening, traditional image features, multi-scale Hessian matrix blood vessel-structure features and topological persistent coherent curvature features are extracted through multi-plane segmentation, and key features are obtained through three-level screening. A multi-factor logistic model fusing radiomics, clinical features and immune scoring is further constructed, a pCR probability value and a feature contribution heat map are output, and MRI data quality control, real-time analysis and interactive report generation are achieved through a matched computer system. The method breaks through the limitation of single modal analysis, dynamically associates the image features with the molecular mechanism, and provides interpretable decision support for precision medical treatment.

The invention relates to the technical field of unmanned aerial vehicle navigation, and discloses an unmanned aerial vehicle navigation method and system based on computer vision, and the unmanned aerial vehicle navigation method based on computer vision comprises the following steps: collecting a current visual field image of an airborne camera of an unmanned aerial vehicle, and carrying out adaptive image enhancement; performing scene semantic understanding; semantic perception feature extraction is carried out; feature matching and outer point elimination are carried out; floor identification is carried out, and multi-sensor fusion positioning is carried out; plane segmentation, structural constraint extraction and map normalization are carried out; performing visual navigation evaluation, judging a navigation mode according to an evaluation result, and generating an exploration control instruction quantity based on the navigation mode; generating a cross-floor path sequence, optimizing a track in real time, and performing control fusion in combination with a navigation mode identifier and an exploration control vector; according to the method, the navigation problem of the unmanned aerial vehicle in complex scenes such as smoke interference, multi-story buildings and illumination dramatic change in a GPS denial environment is solved, and high-precision autonomous navigation is realized.

The invention provides an inclined real spot cloud building top surface segmentation method, and the method comprises the steps: carrying out the segmentation of an inclined building top surface point cloud through employing a clustering-recognition-detail supplement progressive fine classification method, and obtaining a building top plane point cloud. Firstly, the normal of each point is determined by adopting a point cloud gridding method, the plane and the vertical face of the top of a building are distinguished according to the angle of the normal, and point cloud category marking is achieved. Carrying out rough clustering segmentation on the top surface of the building by adopting a hierarchical clustering method based on height statistics, and reducing the interference between clusters in a plane segmentation process; and finally, edge point cloud supplementation is carried out on the building top plane point cloud to realize supplementation of holes in the plane, and the effect of perfecting the details of the point cloud plane is achieved.

The invention provides a building instantiation segmentation method and device, a terminal and a storage medium, and relates to the technical field of three-dimensional point cloud processing, and the method comprises the steps: carrying out the plane segmentation of a building point cloud of a target building, and extracting a potential vertical plane point cluster from a building plane point cluster obtained through the segmentation; mapping all three-dimensional point clouds in the potential vertical plane point cluster to a two-dimensional feature space to obtain a two-dimensional point set, and processing the two-dimensional point set by using a preset dual-threshold judgment mechanism to extract a building vertical structure point cluster; generating a building distance map based on the building vertical structure point clusters, and generating instance anchor points on the distance map; extracting a multi-level contour in the distance map based on the instance anchor points, constructing a tree-shaped hierarchical relationship corresponding to the multi-level contour, and constructing a mark matrix based on the tree-shaped hierarchical relationship; and building two-dimensional instantiation segmentation is carried out on the distance map based on the mark matrix. Through the unsupervised hierarchical segmentation framework, the dependence on the annotated data can be eliminated, and the segmentation precision is improved.

The invention discloses a deformed workpiece twin modeling method based on measurement point cloud and three-dimensional CAD model matching. The deformed workpiece twin modeling method comprises the steps that an improved GFO-ICP registration algorithm is adopted to solve the initial coarse registration problem of the measurement point cloud and a CAD model in combination with geometric feature constraint and self-adaptive step length optimization; extracting a key geometric curved surface of the point cloud by adopting a clustering-based plane segmentation algorithm, and extracting a semantic feature vector of a segmented region in combination with a Pointnet + + network; a cosine similarity matching mechanism is constructed by integrating mixed descriptors of geometric features (areas and normal vectors) and semantic features, and accurate binding of a point cloud curved surface and a curved surface of a CAD model is achieved. According to the method, the problem of assembly geometric deviation caused by workpiece deformation due to manufacturing errors in compressor and fan rotor assembly is solved, multi-link error accumulation is effectively processed, digital twin modeling of high-precision deformed workpieces is achieved, and the assembly simulation effect and intelligent manufacturing application are improved.

The invention provides an environment perception and path planning method and system based on a vehicle controller, and the method comprises the steps: processing the data of an environment perception sensor, and obtaining a standardized image and a two-dimensional event counting graph; respectively processing the standardized image and the event counting graph to obtain an image multi-scale feature graph set and an event multi-scale feature graph set; carrying out attention mechanism fusion on the image multi-scale feature image set and the event multi-scale feature image set to obtain a fusion feature pyramid; performing feature enhancement processing according to the fused feature pyramid to obtain a line segment descriptor set; performing time sequence matching on the line segment descriptor set to obtain optimized line segment parameters; performing topology analysis and plane segmentation processing according to the optimized line segment parameters to obtain an environment structure model with a confidence score; and according to an output result of the environment structure model and a preset path planning algorithm, outputting a path indication identifier for assisting vehicle driving.

The invention discloses a point cloud welding seam recognition method and device based on dimension transformation. The method comprises the steps that three-dimensional point cloud data of a target welding area are obtained; projecting the three-dimensional point cloud data to a two-dimensional plane to obtain a depth map of the target welding area; inputting the depth map into a preset part segmentation model to obtain at least one welding workpiece mask; for each welding workpiece mask, inputting the welding workpiece mask and the depth map into a preset plane segmentation model to obtain a plurality of plane masks; based on the depth map, back-projecting all the plane masks into the three-dimensional point cloud data to obtain coordinate information of a plurality of planes; according to the coordinate information of the multiple planes, all adjacent planes are determined, and according to the relative position relation between every two adjacent planes, the types and coordinate information of all welding seams are determined. According to the method, the complexity of point cloud data processing can be greatly reduced, and the welding seam area can be accurately identified while the redundancy and noise influence in the point cloud data processing is reduced.

The invention relates to a robot dog dynamic balance and fall self-recovery method and device for a complex terrain, and the method comprises the steps: building a robust local elevation map through fusing multi-sensor data and quantifying the motion uncertainty of a robot; and a safe foot region set defined by the convex polygon is extracted through geometric feature analysis and plane segmentation. Then, collaborative planning of the trunk and the landing point is carried out in combination with a user instruction and the region set. In a control layer, a nonlinear model predictive control algorithm is adopted, a safety area is used as a hard constraint, and incomplete speed constraint is applied to a wheel type foot end, so that an optimized trunk track and a foot end force reference are output. The whole-body controller converts an upper-layer instruction into an accurate joint torque through a multi-task optimization framework with a weight. When instability is detected, the system can autonomously plan and execute a recovery action sequence including curling, controllable rolling and cooperative supporting.

The invention provides a weld joint extraction method based on multi-dimensional information fusion. The method comprises the following steps: S1, collecting a Texture image and depth data of a weld joint; s2, inputting the Texture image into an image recognition model to obtain a mask image of the welding seam; s3, binarization processing is carried out on the mask image, and pixel coordinates of the welding seam are extracted; s4, aligning the pixel coordinates with the depth data to generate an initial welding seam point cloud; s5, the initial welding seam point cloud is processed through plane segmentation and a fitting algorithm, and welding seam feature points are extracted; and S6, obtaining an accurate welding seam point cloud by using a straight line fitting algorithm or a plane intersection fitting algorithm. By combining the two-dimensional image segmentation and the three-dimensional point cloud processing technology, the problem that accurate three-dimensional information extraction and background noise interference resistance of the weld joint are difficult to consider simultaneously in a complex industrial scene by a traditional method is solved, the precision and robustness of weld joint extraction are improved, the processing efficiency is ensured, and the method is suitable for popularization and application. And a high-reliability three-dimensional path planning basis is provided for an automatic welding system.

The invention discloses an abnormal behavior detection method based on millimeter wave radar action recognition. The method comprises the following steps: S1, collecting synchronous millimeter wave radar data; s2, performing radius filtering and other preprocessing operations on the acquired radar data; s3, performing RANSAC (Random Sample Consensus) plane segmentation on the preprocessed data, and distinguishing the ground and other objects on the ground; s4, clustering the point cloud data by using a DBSCAN algorithm, and dividing the point cloud data into a plurality of independent parts to distinguish a human body from a building; s5, analyzing a clustering result, extracting characteristics such as the length-width-height ratio and the point cloud density of each independent part, and analyzing the characteristics to obtain a human body classification result; s6, carrying out dimension conversion from three dimensions to two dimensions, and then carrying out human body key point positioning on the point cloud data by adopting an OpenPose method; and S7, performing abnormal behavior detection by using the space-time diagram convolutional network.

The invention discloses a sugarcane live-action three-dimensional phenotype extraction method based on a 3D Gaussian splashing technology. The method comprises the following steps of data acquisition, wherein a 360-degree omnibearing shooting mode is executed around a single sugarcane plant for shooting or video recording to obtain complete image or video data of the sugarcane plant; preprocessing the data; 3D model construction comprises the following sub-steps: SfM sparse reconstruction: instance segmentation; the method comprises the following steps: carrying out 3D Gaussian Splicing guided by a mask; point cloud data processing includes the following sub-steps: point cloud preprocessing including voxel downsampling, statistical filtering and radius filtering; segmenting and identifying stalks / leaves; the segmentation comprises RANSAC plane segmentation, DBSCAN clustering and region growth; the phenotype analysis comprises grid reconstruction and area / volume / perimeter / plant height calculation. According to the invention, the problems of difficult data acquisition, large workload and incomplete information acquisition in the prior art are solved.

The invention provides a turbine blade wear-resistant layer welding path planning method based on point cloud processing. The method comprises the steps that firstly, a structured light camera and a welding robot tail end coordinate system and a conversion matrix of the welding robot tail end and a world coordinate system are determined through calibration; scanning the turbine blade by using a structured light camera to obtain original three-dimensional point cloud data of the turbine blade, and recording coordinates of a welding gun under a tail end coordinate system of a welding robot; performing point cloud filtering, Euclidean clustering and plane segmentation on the original three-dimensional point cloud data to obtain point cloud data of a to-be-welded surface; the point cloud of the surface to be welded is cut through a principal component analysis algorithm to obtain a preliminary path, a complete path can be obtained through straight line fitting and path logic sorting processing, and then coordinates of a welding gun in a coordinate system at the tail end of the welding robot and the complete path are substituted into a coordinate transformation matrix; coordinates of all the path points in the world coordinate system can be obtained, so that path planning is completed. According to the method, automatic path planning can be realized, and production efficiency is improved.

The invention provides a point cloud plane segmentation method and device based on super-point and deep learning, and belongs to the technical field of three-dimensional point cloud processing, and the method comprises the steps: determining a rough super-point and a preliminary segmentation result based on an input point cloud; optimizing the boundary of the preliminary segmentation result, and determining unfitted points in the input point cloud; respectively refining the rough super points and the unfitted points to obtain refined super points; extracting depth features of each three-dimensional point in the input point cloud, and determining superpoint features of the refined superpoints based on the depth features; and inputting the super-point feature into a mask prediction model to obtain a super-point mask, and inputting the super-point mask into a segmentation model to obtain a target segmentation result. According to the method, the refined super points are used as calculation elements in the mask prediction model, so that the accuracy of a target segmentation result is improved.

The invention discloses a dispensing detection method and system and electronic equipment, and relates to the technical field of dispensing testing, and the dispensing detection method comprises the following steps: converting depth map data into a point cloud data set; and segmenting the point cloud data set into a vertical point cloud and a horizontal point cloud according to the normal vector of the cloud data set. And carrying out multi-plane segmentation on the horizontal point cloud, and taking the maximum plane in multiple planes as a target plane. And fitting the target plane to obtain a plane equation. And visualizing the plane equation, and selecting a reference frame and a frame to be measured. Obtaining a height threshold range according to the reference frame, and obtaining the number of detection points of the to-be-detected frame in accordance with the height threshold range. And when the detection point number accords with the point number threshold range, determining that the to-be-detected frame is successfully dispensed. Based on the scheme, whether dispensing succeeds or not is determined by detecting whether the number of the point clouds meeting the height threshold range exists in the to-be-detected panel or not, so that the dispensing condition of the to-be-detected panel can be more accurately detected, and the production quality of the electronic equipment is improved.

The invention relates to the technical field of point cloud data processing, in particular to a method and device for generating an outdoor large-scale scene laser radar point cloud map, and the method comprises the steps: obtaining a preliminary segmentation result of a target laser radar point cloud map based on a region growth segmentation mode, and carrying out the optimization processing of boundary points, obtaining a target plane segmentation result of the point cloud map; calculating target normal vectors of all segmented planes in the target plane segmentation result, performing plane point cloud refinement processing on the plane segmentation result, performing plane segmentation on a point cloud map subjected to refinement processing, determining a target wall corner structure of a new target plane segmentation result, and removing target miscellaneous points in the wall corner point cloud of the wall corner structure to obtain a new wall corner structure; and a refined outdoor large-scale scene laser radar point cloud map is generated. Therefore, the problems that the point cloud precision cannot be increased from hardware and algorithm levels in a short time, data noise is increased, and the requirement for generating the point cloud map in a complex scene cannot be met in the prior art are solved.

The invention discloses a unit division method and device based on a grid model, equipment, a medium and a product, and the method comprises the steps: obtaining point cloud data of a target region, and carrying out the preprocessing of the point cloud data; generating a mesh grid model of the target area according to the processed point cloud data; performing plane segmentation on the mesh grid model by adopting a segmentation algorithm based on annular neighborhood flatness to obtain a slope division result of the mesh grid model; and carrying out vector conversion on a slope division result to obtain a slope unit of the target area. A mountain grid model can be constructed based on point cloud data, plane segmentation is performed on the grid model which retains geometric mountain ridge and ridge features of a mountain by adopting a segmentation algorithm based on annular neighborhood flatness, automatic slope unit division based on a landslide grid model is realized, the problem of DEM distortion is solved, manual parameter setting is not needed, and the construction efficiency is improved. And the slope unit division precision is improved.

The invention relates to the technical field of vehicle assembly, in particular to a hinge hole assembly reference determination method and device and electronic equipment, and the method comprises the steps: obtaining the point cloud data of the upper and lower end faces of a hinge hole; extracting contour edges of the upper and lower end faces of the hinge hole based on clustering processing and plane segmentation of the point cloud data of the upper and lower end faces of the hinge hole, and performing circle fitting on the point cloud data of the contour edges; and identifying circle center coordinates and axis direction vectors of the upper and lower end faces of the hinge hole in a circle fitting result, and determining an assembly reference of the hinge hole according to the circle center coordinates and the axis direction vectors. Therefore, the problem of how to efficiently and accurately determine the assembly reference of the hinge hole in the vehicle door assembly work is solved.

This invention relates to a method for inverting remote sensing data from overlapping strong and weak absorption lines of greenhouse gases, belonging to the field of satellite remote sensing technology. Specifically, it includes the following steps: Step 1: Constructing an optical system, including a satellite platform, camera 1, and camera 2; camera 1 and camera 2 are mounted at the bottom of the satellite platform; camera 1 and camera 2 are mirrored; the imaging spectral band of camera 1 moves from the weak absorption spectral band to the strong absorption spectral band along the positive push-broom direction to the negative push-broom direction; the imaging spectral band of camera 2 is in the opposite direction; the optical system moves horizontally from left to right to achieve horizontal push-brooming of greenhouse gases on the ground, obtaining a dual-camera fused image; the dual-camera differential absorption spectroscopy algorithm is used to invert the dual-camera fused image to obtain the greenhouse gas column concentration; this invention, based on the spectral characteristics of strong and weak absorption lines of greenhouse gases, adopts a detector focal plane segmentation method, and the dual cameras achieve differential absorption and inversion in the same region, which can guide the design of greenhouse gas detection micro / nano satellite payloads.

The present invention relates to the technical field of three-dimensional plane projection. Specifically disclosed are a plane projection method and plane projection apparatus on a surface of a three-dimensional object. The method comprises: determining a projection distance of a projection apparatus; determining a projectable range on the basis of the projection distance; performing plane segmentation on a three-dimensional object for projection, so as to obtain a plurality of sub-planes; on the basis of the sub-planes and the projectable range, determining an arrangement mode of the projection apparatus; and arranging a plurality of projection apparatuses on the basis of the arrangement mode, and executing a screen projection operation on said three-dimensional object by means of the plurality of projection apparatuses. By improving a conventional screen projection method, a surface of a three-dimensional object is segmented, and projection is separately performed on different sub-planes, thereby ensuring sufficient projection clarity and satisfying actual requirements.

The invention provides a binocular image-based parallax plane segmentation method and device, and the method comprises the steps: obtaining an original image and camera internal parameters, the original image comprising left and right views, a depth image and a normal vector image; generating a plane segmentation image and plane parameters according to the original image, and obtaining a true value of plane parallax according to the camera internal parameters; the plane segmentation image, the plane parameter and the true value of the plane parallax are input into a pre-trained binocular parallax plane segmentation network to obtain a plane segmentation result output by the binocular parallax plane segmentation network, and the plane segmentation result comprises a predicted plane segmentation image, a plane parameter image, a parallax image and a three-dimensional reconstruction result; wherein the binocular parallax plane segmentation network is obtained by training a pre-constructed deep neural network by using an image sample. The position and direction information of the plane is fully utilized, the reconstruction precision of the shielding area is improved, and the three-dimensional reconstruction effect is improved.

The invention relates to the technical field of ship plane segmentation, and discloses a plane segmentation assembly line manufacturing method which comprises the following steps: step 1, main board splicing operation; step 2, SAW welding and fixing; step 3, carrying out segmented turning-over operation; step 4, configuring and scribing a longitudinal bone; 5, longitudinal installation operation is carried out; step 6, longitudinal welding operation; 7, conveying to a jig frame section by section; 8, pulling-in operation of the inner structural material is carried out; ninthly, inner structural material assembling work is conducted; 10, welding operation is conducted on the inner structural material; step 11, checking and repairing operation; step 12, pre-outfitting operation is carried out; and step 13, carrying out subsection transportation operation. The flow from one station to the next station is segmented according to the fixed rhythm, so that the waiting time in the traditional fixed station production is eliminated, and the production process is smooth; the labor investment is effectively reduced; the conveying efficiency is effectively improved through the conveying trolley and the plate chain conveying device, conveying is carried out according to a specific rail, the problem of conveying errors is avoided, and the conveying accuracy is improved.

This invention provides a method for training a curved surface intelligent spraying model, a spraying method, and a system thereof, relating to the field of robot path planning technology. The curved surface intelligent spraying method includes: acquiring point cloud data of an object using a depth camera; performing voxel filtering, statistical filtering, and RANSAC plane segmentation preprocessing on the point cloud data; calculating the centroid and PCA principal component orientation of the preprocessed point cloud; rasterizing the space containing the point cloud according to a preset grid size and number of rows and columns to generate a grid map; using a DQN model for path planning; and having the robot perform spraying operations according to the planned path. This invention also provides a method for training a curved surface intelligent spraying model and a curved surface intelligent spraying system to achieve optimal path coverage of all sprayable areas, avoiding unnecessary movement.

The present invention discloses a method, device, storage medium and electronic device for forklift pallet positioning. The method includes: obtaining a depth image acquired based on an RGB-D sensor; obtaining a planar image acquired by a camera; identifying a forklift pallet using the planar image and determining the area where the forklift pallet is located in the depth image; performing planar segmentation processing on the area where the forklift pallet is located in the depth image to determine a planar set containing the forklift pallet; matching the planar image with a pre-prepared pallet template to obtain a matching target pallet template; using the determined planar set containing the forklift pallet to match with the target pallet template to mark the position of the target pallet template in a second target area; converting the position of the target pallet template in the second target area into coordinates in a three-dimensional space to obtain the position of the forklift pallet. The present invention determines the pallet area through image mapping, solving the problem of large amounts of image recognition data in the prior art.

The invention discloses a modular room size automatic measurement method based on a structure chart and a measurement chain, and belongs to the crossing field of computer vision and artificial intelligence. The method comprises the following steps: firstly, acquiring three-dimensional point cloud data of a concrete module unit to be measured by using a high-precision laser scanner, and obtaining a structure plane set with millimeter-level precision through structure priori guided graph-cut plane segmentation; semantic classification is carried out by extracting multi-scale geometric features and room-level context features, and a plane set with labels is output; constructing a structure chart based on the spatial topological relation and the semantic information; then defining a basic measurement unit, and automatically compiling a high-level measurement task into an ordered measurement chain; and finally, sequentially executing the measurement chain, binding real point cloud coordinates, and outputting a measurement result with deviation judgment. According to the method, full automation from the point cloud to the structure diagram to interpretable measurement is realized, and the method has millimeter-level precision, interpretability and module reusability, and is suitable for factory quality inspection of concrete modular buildings.

The application discloses a point cloud filtering method and system for a laser scanner, and the point cloud filtering method comprises the following steps: acquiring point cloud data to be processed; generating a two-dimensional image according to the spherical coordinates of all data points in the point cloud data, wherein the horizontal coordinate of each pixel of the two-dimensional image corresponds to the azimuth of a data point, the vertical coordinate corresponds to the elevation of the data point, and the pixel value of the pixel corresponds to the depth value of the data point; for each data point, the size of the filtering core of the data point is acquired according to the depth value; for a target data point, the neighbor data points of the target data point are found on the two-dimensional image according to the size of the filtering core; the depth value of the target data point is acquired by using the neighbor data points, and the coordinates of the target data point are updated by using the filtered data. The application greatly shortens the operation time of point cloud processing, obtains point cloud data with less noise, thinness and accuracy in a shorter operation time, and improves the accuracy of point cloud measurement and the quality of plane segmentation.

The application provides a welding seam extraction method based on multi-dimensional information fusion, comprising the following steps: S1, collecting a Texture image and depth data of a welding seam; S2, inputting the Texture image into an image recognition model to obtain a mask image of the welding seam; S3, performing binaryzation processing on the mask image to extract pixel coordinates of the welding seam; S4, aligning the pixel coordinates with the depth data to generate an initial welding seam point cloud; S5, processing the initial welding seam point cloud by using a plane segmentation and fitting algorithm to extract characteristic points of the welding seam; and S6, obtaining an accurate welding seam point cloud by using a straight line fitting algorithm or a plane intersection fitting algorithm. By combining two-dimensional image segmentation and three-dimensional point cloud processing technology, the application solves the problem that a traditional method is difficult to simultaneously consider accurate three-dimensional information extraction and anti-background noise interference of the welding seam in a complex industrial scene, improves the precision and robustness of welding seam extraction, and guarantees the processing efficiency, thereby providing a high-reliability three-dimensional path planning basis for an automatic welding system.

The disclosed embodiments disclose a method and device for extracting scene accessory components based on point cloud data, which comprises acquiring point cloud data collected in a specified scene; performing plane segmentation processing on the point cloud data, and obtaining a point cloud cluster belonging to a wall plane in the specified scene based on the result of the plane segmentation processing; projecting the point cloud cluster of the wall plane onto a corresponding wall plane to obtain an occupancy grid map; detecting occlusion areas in the occupancy grid map, and performing filling processing on the occlusion areas; wherein the occlusion areas are areas of the wall in the specified scene that are occluded by objects; extracting a bounding box of a hole area in the filled occupancy grid map, and determining the position information of the accessory components of the wall in the specified scene based on the bounding box of the hole area. Thus, the position information of the accessory components can be robustly and accurately extracted from the point cloud data, thereby improving data processing efficiency.

The present invention relates to the field of ship monitoring technology, and discloses an integrated safety monitoring method and system for ship berthing and mooring. The point cloud data of the berthing and mooring area to be monitored is acquired by radar, and then the point cloud segmentation calculation is performed by plane segmentation, contour extraction, and point cloud point clustering to identify the position and posture of the ship, so as to obtain the monitoring index of the ship berthing and mooring. The point cloud data of the horizontal plane is determined by comparing the horizontal distance of each point cloud with a first distance threshold, and then the point cloud of the non-water surface object is obtained by separating the horizontal plane point cloud. By comparing the horizontal distance between the point cloud points with the second distance threshold, the ship and other obstacles are accurately separated. Accurate and real-time integrated safety monitoring of ship berthing and mooring is achieved.

The invention discloses a building BIM modeling method based on three-dimensional point clouds, and the method comprises the following steps: S1, carrying out the collection of densely matched building point clouds and laser building point clouds, and carrying out the denoising and downsampling of the building point clouds; s2, registering the cross-source building point cloud data based on a feature point four-point set registration algorithm to obtain complete building point cloud data; s3, plane segmentation is carried out on the building point cloud, a building boundary is extracted, and components are extracted based on nearest neighbor search; and S4, carrying out fusion, regularization and classification on component boundaries, extracting component parameters, carrying out secondary development based on Revit, reading the component parameters, and realizing BIM automatic modeling. According to the method, creation of the BIM model can be accurately completed, manual participation in the process of generating the BIM model through the point cloud is greatly reduced, the efficiency and accuracy of generating the BIM model through the point cloud can be improved, and certain method support is provided for construction of a smart city.

The present invention relates to the technical field of point cloud data processing, and particularly relates to a method and device for generating an outdoor large-scale lidar point cloud map. The method includes: obtaining a preliminary segmentation result of the target lidar point cloud map based on a region-growing segmentation method, and performing boundary point optimization processing to obtain a target plane segmentation result of the point cloud map; calculating the target normal vectors of all the segmented planes in the target plane segmentation result, and then performing refined processing on the plane point cloud of the plane segmentation result to perform plane segmentation on the refined point cloud map, determining the target corner structure of the new target plane segmentation result, and removing the target miscellaneous points in the corner point cloud of the corner structure to generate a refined outdoor large-scale lidar point cloud map. Thus, the problem in the related art that it is impossible to increase the point cloud accuracy from the hardware and algorithm levels in a short time, which increases data noise and cannot meet the requirements for generating a point cloud map in complex scenarios, is solved.