64 results about "Elevation map" patented technology

The invention discloses an image recognition system for industrial part defect detection, and particularly relates to the field of part defect detection, which comprises a multi-axis controllable light source array module, a high-speed polarization camera module, an edge computing node module, a double-branch semantic segmentation network module and a physical constraint post-processing module, according to the invention, through combination of time-sharing stroboscopic illumination and polarization image sequence acquisition, multi-dimensional perception of surface topography and material differences is realized; generating an elevation map and a normal map by using photometric stereo solution, constructing a differential rendering layer reverse matching CAD model, and extracting flash sensitive features; a double-branch U-Net network is adopted to fuse geometric and polarization characteristics, the characterization capability is enhanced through a trans-attention mechanism, and a pixel-level mask is output; and finally, mapping a two-dimensional result to a three-dimensional coordinate system by means of calibration parameters, carrying out geometric verification in combination with a tolerance zone and a height threshold value, and automatically generating a structured defect report containing position, size, grade and visual information.

The invention provides a mechanical arm real-time stacking model generation method and device based on a 2.5 D elevation map. The method comprises the steps that depth point cloud data of a stacking area is obtained; preprocessing the depth point cloud data to obtain de-noised point cloud data; performing external parameter calibration by adopting Eye-to-hand to obtain an accurate conversion relation between a camera coordinate system and a mechanical arm flange coordinate system, and performing coordinate system conversion to obtain de-noised point cloud data under a unified world coordinate system; performing space clipping on the de-noised point cloud data to obtain effective point cloud data in the stacking area range; a mechanical arm directional bounding box is generated under the unified world coordinate system, and whether the mechanical arm directional bounding box intersects with the stacking area or not is detected through the separation axis theorem; if not, a 2.5 D elevation map is generated according to the effective point cloud data in the stacking area range, and a real-time stacking model of the mechanical arm is obtained according to the 2.5 D elevation map; therefore, quick response and flexible adjustment are realized in a dynamic scene.

The invention discloses a non-flat terrain autonomous navigation method based on path pruning and trajectory smoothing. The method comprises the following steps: constructing an elevation map with three-dimensional map information as a navigation map according to robot navigation space information; based on the navigation map, performing global path search by adopting a path pruning strategy to obtain an initial path from the starting point to the target point; based on the initial path, track optimization is carried out by using a B spline curve, and a smooth track meeting robot dynamics constraints is generated; in response to an obstacle which is detected on the smooth trajectory and cannot be avoided, performing three-dimensional obstacle crossing trajectory planning based on terrain elevation information of the elevation map, and generating an obstacle crossing trajectory capable of crossing the obstacle; wherein the robot performs autonomous navigation based on the smooth track or the obstacle crossing track. According to the invention, efficient global search, smooth trajectory optimization and intelligent obstacle crossing decision of the robot in unknown and non-flat terrains can be realized.

The invention provides an underground pipe network detection positioning method and device, and the method comprises the steps: obtaining the multi-source data of an underground pipe network of a to-be-detected point and the surrounding environment, including ground image data, inspection well internal image data, adjacent inspection well spacing data, ground elevation data, pipe bottom elevation data, and position coordinates; inputting the ground image data into a semantic segmentation network, and outputting a ground semantic segmentation image; respectively extracting contours of the ground semantic segmentation image and the image data in the inspection well to obtain a first edge image and a second edge image; constructing a pavement information measurement model; pipeline diameter information, pipeline gradient and rain and sewage pipeline mixed connection conditions are determined; and integrating the pavement information measurement model, the first edge image, the second edge image, the pipeline diameter information, the pipeline gradient and the like, and drawing a planar graph and an elevation graph of the underground drainage pipe network. According to the method, high-precision detection and positioning of the underground pipe network are remarkably improved, and the method is suitable for city updating and city informatization construction.

The invention discloses a single-frame unmanned aerial vehicle image pixel positioning method and system based on an elevation map, and relates to the technical field of unmanned aerial vehicle image processing and geographic positioning, and the method comprises the steps: obtaining a single-frame original image shot by an unmanned aerial vehicle, unmanned aerial vehicle attitude information during shooting, camera internal and external parameters, and a digital elevation map of a shooting region; according to the internal reference of the camera, converting the pixel coordinate of the target pixel in the original image into a normalized plane coordinate system of the camera, and carrying out coordinate distortion correction to obtain the back projection point coordinate of the target pixel; converting a camera-unmanned aerial vehicle body coordinate system and a body-geographic inertial reference coordinate system according to camera external parameters and unmanned aerial vehicle attitude information to obtain normalized ray direction vectors corresponding to target pixels; an earth surface height constraint is constructed based on a digital elevation map, and an intersection point of a ray and the ground is determined by performing iterative approximation on a space point along a ray direction. According to the invention, rapid and accurate conversion from the single-frame unmanned aerial vehicle image pixel to the geographic coordinate can be realized.

The invention discloses a digital surface model (DSM)-guided hyperspectral and LiDAR combined unmixing method, relates to the field of multi-modal image processing, and aims to solve the problems of end member confusion and insufficient space structure maintenance caused by spectrum similarity in hyperspectral unmixing. According to the method, a hyperspectral image and LiDAR data of the same area are obtained, a LiDAR elevation map is expanded into a multiband profile through an attribute configuration file method, and a digital surface model (DSM) is generated. Constructing a spectrum and space double-branch auto-encoder, respectively extracting spectrum and space features and carrying out fusion mapping, obtaining an abundance matrix by using a normalized exponential function, and reconstructing a hyperspectral image; a double-branch adaptive mixed channel attention mechanism is designed in a spectrum branch, and a space attention mechanism is introduced in a space branch. In the training process, a self-defined loss function is formed by combining DSM-guided structure entropy regularization, spectral angular distance and root-mean-square error, and the space continuity and boundary retention of the abundance graph are improved. The method is suitable for hyperspectral unmixing and multi-modal remote sensing fine identification under complex terrains.

The invention relates to a training method of a perception and control model, and a robot control method and device. The method comprises the following steps: acquiring a plurality of sample image sets; wherein each sample image set comprises a depth image and an elevation image which represent the same terrain; performing multi-stage training on the initial model according to the plurality of sample image sets to obtain a sensing and control model for controlling the action of the robot; wherein the multi-stage training comprises the following steps of: training an elevation image codec by utilizing an elevation image to obtain a trained elevation image codec; training a depth image codec by using the depth image and the trained elevation image codec to obtain a trained depth image codec; and performing reinforcement learning training on the action strategy neural network by using the trained depth image codec, thereby facilitating the improvement of the model training efficiency through the interaction of perception and control, helping the robot to accurately perceive the working environment, and realizing accurate action control.

The application discloses a kind of single frame unmanned plane image pixel positioning method and system based on elevation map, it is related to unmanned plane image processing and geographic positioning technical field, obtain the single frame original image of unmanned plane shooting and the digital elevation map of the shooting area when unmanned plane attitude information, camera internal and external parameters, shooting;According to camera internal parameter, the pixel coordinates of target pixel in original image are converted to camera normalization plane coordinate system, and target pixel back projection point coordinates are obtained after coordinate distortion correction;According to camera external parameter and unmanned plane attitude information, camera-unmanned plane body coordinate system, body-geographic inertial reference coordinate system conversion is carried out, and the normalized ray direction vector corresponding to target pixel is obtained;Based on digital elevation map, construct ground height constraint, by iterative approximation to space point along ray direction, determine the intersection of ray and ground.The application can realize the fast and accurate conversion of single frame unmanned plane image pixel to geographic coordinate.

Embodiments of the present disclosure disclose an airport abnormal area identification method and device, electronic equipment and readable medium. A specific embodiment of the method comprises: selecting reference area data and contrast area data; performing registration correction on the reference area remote sensing image and the contrast area remote sensing image to generate a corrected reference area image and a corrected contrast area image; performing area shadow detection on the corrected reference area image and the corrected contrast area image; performing constraint feature fusion on the corrected reference area image and the corrected contrast area image respectively to generate reference area main features and contrast area main features; performing abnormal area identification on the reference area main features and the contrast area main features; and performing verification grading on the candidate abnormal area identification information to obtain airport target area abnormal information. The embodiment can perform shadow detection on the airport area to identify abnormal shadow areas, and perform verification on the abnormal areas through a grid elevation map, thereby improving the safety of the airport area clearance.

The application discloses a building height detection method and device, a storage medium and an electronic device, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a target remote sensing image corresponding to a target building, wherein the target building is a building to be detected in height; inputting the target remote sensing image into a target detection model for detection processing, and outputting height information of the target building, wherein the target detection model is a model generated by learning and training a target neural network model by using M remote sensing images, M shadow maps and M target elevation maps, the target neural network model is a neural network model with residual connection, the remote sensing image is a remote sensing image corresponding to a building, the shadow map is a shadow map of the building, the target elevation map is used for representing height information of the building, and M is a positive integer. Through the application, the problem that the effect of detecting the height of a building is poor in the related art is solved.

The embodiment of the invention provides a movement control method for a foot type robot. The movement control method is used for controlling the foot type robot to stably and accurately move in diversified terrains. The foot-type robot is loaded with a depth camera and an inertial measurement unit IMU, and the movement control method comprises the steps that in the walking process of the foot-type robot, point cloud data, IMU attitude data, remote control instruction data and motor state data collected by the depth camera are obtained; generating elevation map data based on the point cloud data and the IMU attitude data; inputting the elevation map data, the remote control instruction data, the IMU attitude data and the motor state data into a pre-trained reinforcement learning model to generate a motor control instruction; the reinforcement learning model is obtained by configuring various terrain scenes in a simulation training environment and training by adopting a priority sampling strategy; according to the priority sampling strategy, the sampling frequencies of different learning action sequences are determined according to the difficulty priorities of the different learning action sequences; and the motor control instruction is issued to a motor control system of the foot type robot, and the foot type robot is driven to execute corresponding actions.

The invention relates to the technical field of robot positioning, discloses a four-foot robot three-dimensional positioning method and system, a storage medium and a robot, and aims to solve the technical problems of height drift and foot end error accumulation of a four-foot robot in a visual degradation environment without a GPS (Global Positioning System) by a traditional positioning method. The method comprises the following steps: firstly, constructing an initial state vector containing motion states of a fuselage and four foot ends, and establishing a nonlinear continuous time state equation considering foot end dynamics; relying on an unscented Kalman filtering framework, a prior state is obtained through Sigma point sampling and time updating; a foot end observation value is corrected through foot bottom rolling compensation, a height absolute reference is provided by combining a local invariant elevation map, and multi-leg slippage detection and observation weight self-adaptive adjustment are synchronously completed; and finally, the optimal state is updated and output through observation. According to the invention, the visual sensor is used for topographic mapping instead of direct positioning, the height drift is effectively inhibited, the robustness and precision of positioning in a complex environment are improved, and the method is suitable for unstructured scenes such as rescue exploration and the like.

The application discloses an extraterrestrial celestial body task area search path planning method based on an elevation map and belongs to the technical field of space robots. The method is as follows: the slope data in a task area is solved through the point set data of the elevation map, and the alternative passing position points of the rover are determined; the planning grid size is determined, and whether the grid is passable is determined according to the proportion of the number of passable points in the grid to the total number as a judgment standard; the exploration value of each grid is determined, and a weight value is assigned to the grid. The current grid is taken as a center point, the detectable nodes in a predetermined range around the current grid are marked as detected nodes, and the weight values of the detected nodes are reduced; the value of each grid in the passable range around the current grid is judged with the current grid as a center point, if there is a grid with a value not less than an end threshold, the grid with the maximum value is selected as a next path node, the above steps are repeated, and the extraterrestrial celestial body task area search path planning based on the elevation map is realized.

Methods and systems for improved generation and georeferencing of floor plans are presented. In one embodiment, a method is presented that includes receiving images that depict sheets of a blueprint of a structure. Subsets of the images depicting floor sheets and elevation sheets may be identified. Exterior contours may be extracted from the images depicting floor sheets and elevation contours may be extracted from the images depicting elevation sheets. A corresponding structure within a three-dimensional map may be identified based on the exterior contours and the elevation contours. A three-dimensional contour of the exterior of the structure may be extracted from the three-dimensional map.

The invention relates to the technical field of robot environment perception, in particular to a real-time ground segmentation method and system. The method comprises the following steps: receiving and preprocessing laser radar point cloud data, and dividing the data into a plurality of concentric annular areas; carrying out adaptive ground fitting on each area and processing the multilayer structure through vertical plane fitting to obtain an initial ground point set; correcting an under-segmentation region of the current frame by using a time sequence backspacing mechanism and updating parameters; constructing an elevation map based on the initial point set, and obtaining a fine ground point set through adaptive resolution CSF simulation; and carrying out confidence evaluation on the two types of point sets, and generating a final result through multi-criterion fusion. Accurate segmentation of the complex terrain is realized, and robustness and real-time performance of a dynamic environment are improved.

The invention relates to the technical field of bulk material flow loading monitoring, in particular to a train compartment loading material flow height online monitoring method and device. The method comprises the following steps: continuously collecting point cloud data of carriages and internal material flows passing through a monitoring area, and performing preprocessing; determining a front boundary surface position parameter and a rear boundary surface position parameter of the carriage in the advancing direction based on the preprocessed point cloud data, and establishing a carriage model in combination with the prior parameters of the carriage; performing space judgment on the preprocessed point cloud data based on a carriage model to obtain a material flow point cloud in the carriage; based on the material flow point cloud, constructing a carriage material flow three-dimensional elevation model and a two-dimensional carriage material flow elevation map; and extracting a height statistical sequence from the two-dimensional compartment material flow elevation map, obtaining a height-time curve in combination with train running speed information, and outputting the three-dimensional compartment material flow elevation model and the height-time curve. The problems that online dynamic monitoring requirements are difficult to meet, and local loading abnormity cannot be found in time can be solved.

A method of estimating a radio map from environmental context comprises designating a training area and for each transmitter location in the training area, generating a channel parameter radio map that is centered on the transmitter and is based on a ray-tracing twin model and generating a two-dimensional elevation map, centered on the transmitter, corresponding to the training area. The method may further comprise training a U-Net-based or any other neural network-based model using the channel parameter radio maps and the two-dimensional building elevation maps corresponding to the transmitters in the area, to produce a trained model, and designating an area to be radio-mapped. The method may further comprise providing the trained with a two-dimensional input elevation map derived from a three-dimensional model of the area to be radio mapped, and producing, by the model, an estimate of the radio map based on the two-dimensional elevation map.

This application relates to the field of image processing technology and discloses an online correction method for the installation error of a UAV's downward-looking camera. The method includes: acquiring a first image frame and a second image frame; obtaining a third set of matching points and a fourth set of matching points based on the first image frame, the second image frame, and a satellite map; obtaining the intersection of the third set of matching points and the fourth set of matching points, and performing uniformization processing on the intersection to obtain a fifth set of matching points; generating an elevation map; obtaining the three-dimensional coordinates of each feature point in the fifth set of matching points based on the pixel coordinates matched in the satellite map, and the corresponding satellite map and elevation map; and obtaining the attitude error by solving a least-squares problem, and compensating for the current attitude of the UAV using the attitude error. This application can calculate the installation error of the downward-looking camera in real time, thereby ensuring the positioning accuracy of the UAV during visual navigation.

The invention relates to a risk aggregation method based on improved conditional value-at-risk in an off-road scene, belongs to the technical field of vehicle control, and solves the problem that off-road environment traffic risks are difficult to accurately and reliably assess due to the fact that risk indexes are isolated analysis or simple fusion and uncertainty of data is not fully considered in the prior art. Comprising the following steps: establishing a conversion relation among a planning coordinate system, a vehicle body coordinate system and a sensor coordinate system by taking a world coordinate system as a reference; acquiring current pose information of the vehicle and current point cloud data of the sensor in real time; performing horizontal processing on the current point cloud data to obtain horizontal point cloud data, and constructing a 2.5 D grid elevation map by adopting a sparse Gaussian process regression method; obtaining the cost of each traffic risk index of each grid based on a 2.5 D grid elevation map; and converting the cost of each traffic risk index to a planning coordinate system, and generating a current comprehensive traffic risk map by adopting an improved conditional value-at-risk method.

This invention discloses a two-phase flow identification system and method based on critical elevation maps, comprising the following steps: acquiring relative position data of all equipment, pipes, and valves in the pipeline system to be analyzed, from the starting equipment to the final equipment, as well as operating temperature, operating pressure, and medium composition data of each material flow point; drawing a critical elevation map based on the collected data; and analyzing the impact of pressure and temperature changes at each material flow point on the medium phase state using the critical elevation map, combined with the thermodynamic properties and flow characteristics of the medium, to identify locations prone to gas-liquid two-phase flow. This invention, by establishing critical elevation maps, can identify pipelines prone to two-phase flow in advance, preventing gas-liquid two-phase flow in process equipment pipelines, which can lead to long-term pipeline vibration, resulting in pipeline fatigue, support structure failure, weld cracking, and flange leakage. Furthermore, for pump inlet pipelines, it avoids the presence of two-phase flow that could cause pump cavitation and damage to the pump equipment.

The invention relates to the technical field of visual inspection, and discloses a PCB multi-parameter synchronous detection method, electronic equipment and a storage medium, and the method comprises the steps: generating a PCB elevation map through the fusion of a 2D image and 3D point cloud data, carrying out the synchronous comparison analysis of the height and contour of a component installation position through the combination of the standard height and contour information in a design file, and carrying out the detection of the 3D point cloud data. And integrated detection is realized. According to the method, by introducing the elevation map generated by the three-dimensional point cloud data, the defect that 2D detection lacks depth information is made up; meanwhile, due to space-time alignment fusion of 2D and 3D data, high-resolution details of a two-dimensional image are reserved, height dimension information is increased, and the coverage dimension and precision of detection are remarkably improved; meanwhile, according to the method, through multi-parameter synchronous analysis, matching judgment of the height and the contour is completed at the same time in one detection process, the detection efficiency is improved, and the missed detection risk caused by multiple times of independent detection is effectively reduced.

This invention discloses a hierarchical reinforcement learning motion planning and control method for legged robots in sparse terrain. It extracts semantically rich terrain features from elevation maps using a multi-scale perceptual encoder. These terrain features, along with the robot's real-time state and target commands, serve as input to a conditional denoising network. This network models landing point planning as a parallel, progressive denoising process, generating a sequence of robot landing points. During this process, an inverse kinematics feasibility verification layer evaluates the kinematic reachability of candidate landing points in real time, thus determining the planned landing point sequence. Based on the planned landing point sequence and the robot's real-time state, robust joint motion commands are output using reinforcement learning to control the legged robot's movements. This invention effectively solves the problems of short-sighted planning, low computational efficiency, and poor robustness in traditional methods in discontinuous, high-elevation, sparse terrain, achieving highly dynamic and stable motion of legged robots in complex environments.

The invention relates to an elevation map generation method, device and equipment, and the method comprises the steps: obtaining the point cloud data of a current frame, and mapping the point cloud data to a 3D voxel grid corresponding to a previous frame, so as to obtain an initial 3D voxel grid corresponding to the current frame; carrying out occupancy probability attenuation on voxels except the current voxel corresponding to the point cloud data of the current frame in the initial 3D voxel grid; voxel removal is carried out on the attenuated 3D voxel grids on the basis of the occupancy probability, so that the 3D voxel grids corresponding to the current frame are obtained; and generating an elevation map based on the 3D voxel grid corresponding to the current frame. The occupancy probability of the voxels which can be observed in the 3D voxel grid is not continuously attenuated, and the occupancy probability of the voxels occupied by the obstacles before entering the blind area is continuously attenuated until the voxels are cleared, so that the dynamic obstacles entering the blind area are prevented from being continuously reserved in the elevation map, the dynamic obstacle residue is avoided, and the accuracy of the dynamic obstacle detection is improved. The accuracy of the elevation map is improved.

This invention discloses an automatic frequency-elevation map measurement method and system based on residual networks, enabling rapid and automatic extraction of key ionospheric parameters from frequency-elevation maps acquired by digital altimeters. The method first parses the header information and echo intensity data from the original binary file of the altimeter to generate a frequency-elevation map image. Then, a small number of samples are manually labeled to establish a training set with critical frequencies such as foF2 and h'F2 as ground truth. Based on this, an end-to-end model including a backbone residual network, pooling layers, and a feature extraction head is built. This model is trained to convergence through backpropagation and further fine-tuned using expanded sample data to obtain an engineering-deployable model. Finally, the frequency-elevation map to be processed is input into the model, and the predicted values ​​of the target parameters are output and written to the results file. This method reduces the workload of manual interpretation and verification and improves the measurement accuracy of irregular and disturbed frequency-elevation maps.

This invention discloses a navigation path planning method for multi-story buildings based on elevation maps, applicable to multi-story buildings containing stairs, ramps, and suspended structures. The core lies in using hierarchical cost modeling to uniformly incorporate clearance and height differences into global path search and model predictive control rolling optimization, achieving highly executable path generation and robust control against dynamic obstacles. The system integrates multi-sensor data to construct multi-slice elevation and cost maps, and introduces a time window decay mechanism to suppress the long-term impact of short-term dynamic structures on the cost field. Based on the hierarchical cost map, connecting areas such as stairs and corridors are raster-labeled, and an improved Theta* is used for arbitrary-angle global search to output a reference trajectory with terrain risk annotations; local layers employ model predictive control rolling optimization to achieve path tracking and obstacle avoidance, generating control commands. This invention significantly improves path executability, control stability, and robustness against dynamic obstacles in complex elevation environments.

The embodiment of the invention discloses an airport abnormal area identification method and device, electronic equipment and a readable medium. A specific embodiment of the method comprises the following steps: selecting reference area data and comparison area data; performing registration correction on the reference area remote sensing image and the contrast area remote sensing image to generate a corrected reference area image and a corrected contrast area image; performing area shadow detection on the corrected reference area image and the corrected contrast area image; performing constraint feature fusion on the corrected reference region image and the corrected contrast region image to generate a reference region main feature and a contrast region main feature; performing abnormal region identification on the reference region main features and the comparison region main features; and verifying and grading the candidate abnormal region identification information to obtain the abnormal information of the airport target region. According to the embodiment, shadow detection can be performed on the airport area to identify the abnormal shadow area, and the abnormal area is verified through the grid elevation map, so that the clearance safety of the airport area is improved.

This application discloses a lane line detection method and apparatus. The lane line detection method can be applied to the field of high-precision map production. The method includes: acquiring a point cloud intensity map and a point cloud elevation map corresponding to the area to be detected; obtaining the point cloud intensity map and point cloud elevation map based on ground point cloud data within the area to be detected collected by a lidar sensor on a vehicle; determining a target preset lane line main direction interval from multiple preset lane line main direction intervals based on the vehicle's driving trajectory points; extracting a target point cloud intensity sub-map from the point cloud intensity map that meets preset intensity conditions based on the target preset lane line main direction interval; extracting a target point cloud elevation sub-map from the point cloud elevation map that meets preset elevation conditions; and determining a lane line segmentation map based on the target point cloud intensity sub-map and the target point cloud elevation sub-map, thereby determining the lane line vectorization result. This application can limit the lane line direction and filter out non-lane line areas to obtain a more accurate lane line vectorization result.

The application relates to the technical field of topographic mapping, in particular to an interactive DEM smoothing method; the method comprises the following steps: S1, inputting original DEM data; S2, identifying abnormal values in the inputted original DEM data; S3, marking the identified abnormal values in a highlighted form on a two-dimensional terrain view in real time, superimposing an original contour line layer, obtaining a superimposed elevation map, setting a frame selection range, setting a frame selection range for each abnormal value on the superimposed elevation map, correcting the elevation values of each elevation point in the frame selection range, and obtaining corrected DEM data; the application improves the efficiency of DEM abnormal point correction.

The application provides a three-dimensional road line extraction method and related equipment, comprising: obtaining binocular image data and inertial navigation positioning data of a research area based on a vehicle-mounted platform; performing parallax estimation on the binocular image data according to the inertial navigation positioning data to obtain a visual three-dimensional point cloud; determining a region with the binocular camera as the center and a preset distance as the radius as a target region in the research area, extracting the visual three-dimensional point cloud in the target region, and converting the visual three-dimensional point cloud in the target region to obtain a road surface bird's-eye view and a point cloud elevation map of the target region; inputting the road surface bird's-eye view and the point cloud elevation map into a lane line detection network to extract a road line, and obtaining a three-dimensional road line and a three-dimensional coordinate of the road line; compared with the prior art, the problems of low precision, poor robustness and high detection cost of the current road line recognition method in vehicle-mounted images are overcome, and the efficiency and quality of road line detection in vehicle-mounted images are improved.