162 results about "Vector map" patented technology

The invention relates to the field of homogenizer intelligence, and discloses a homogenizer intelligent pressure increasing and reducing control system based on servo synchronous transmission, which comprises the following steps: extracting the operation inertia difference of each plunger mechanism under a multi-pump parallel structure, generating a synchronous offset vector map, and calculating the pressure increasing and reducing of the homogenizer based on the synchronous offset vector map. Calling a slip registration algorithm and an inertia field decoupling network, extracting reconstructed plunger stroke and pressure response mapping data, constructing a three-dimensional flow-path-stress tensor model, judging whether fluctuation-wear coupling sensitive points exist or not, activating a frequency conversion regulation and control strategy set of a servo driving layer based on a disturbance window and the coupling sensitive points, and carrying out the control of the servo driving layer. And judging whether structural thermal drift feedback unbalance exists or not, dynamically adjusting servo redundancy and a buffer response strategy in combination with a preset life curve and an abnormal response backtracking channel, and outputting a synchronous regulation and control instruction set. The method has the advantage of improving the long-term operation reliability of the homogenizer.

The invention relates to a power transmission line path planning method and device, computer equipment, a storage medium and a computer program product. Relates to the field of power equipment intelligent planning. The method comprises the following steps: fusing radar point cloud data, satellite remote sensing image data, aerial remote sensing image data, digital elevation model data and vector map data of a to-be-planned area to obtain fused data; based on the fusion data, determining a digital orthoimage and a three-dimensional geographic scene of the to-be-planned area; based on the digital orthoimage and the three-dimensional geographic scene, determining buffer area radiuses corresponding to different areas in the to-be-planned area; determining a target area of the to-be-planned area based on the buffer area radius, and outputting a path topology network corresponding to the target area; and determining the shortest cost path in the path topology network as the power transmission line path of the to-be-planned area. By adopting the method, the power transmission line path planning efficiency can be improved.

The invention relates to the technical field of computer aided design, and discloses a multi-source data fusion CAD drawing method and system based on deep learning. The invention discloses a multi-source data fusion CAD drawing method based on deep learning. Intelligent fusion processing of a building CAD base map, laser point cloud data and a high-definition video stream is realized through multi-source space-time reference alignment, reflection adaptive point cloud enhancement, cross-modal feature fusion, progressive vector generation, engineering constraint optimization and navigation output. Performing point cloud splicing by adopting an ICP (Inductively Coupled Plasma) algorithm, establishing a spatial corresponding relation by using a Zhang Zhengyou calibration method, extracting multi-modal features through a deep learning network, and realizing cross-modal feature fusion by utilizing an attention mechanism; and finally, a high-precision vector map conforming to engineering specifications is generated and is suitable for automatic generation of a large-scale parking lot navigation map. According to the method, the drawing efficiency and precision are improved, the labor cost is reduced, and the method has good engineering application value.

Provided are methods for high definition map fusions for 3D object detection. Some methods described also include obtaining, with at least one processor, raster maps, vector maps, and point cloud data and extracting, with the at least one processor, features from the raster maps, vector maps, and point cloud data to generate respective bird's eye view (BEV) representations. The methods also include fusing, with the at least one processor, the BEV representation of the raster map features, the BEV representation of the vector map features, and the BEV representation of the point cloud features into a fused BEV image. Additionally, the methods include detecting, with the at least one processor, objects in the fused BEV image. Systems and computer program products are also provided.

The invention provides an automatic driving positioning method and device based on a look-around image and a semantic vector map, and the method comprises the steps: obtaining the look-around image formed in an automatic driving process of a vehicle, and a local semantic vector map matched with the look-around image; obtaining map prior features corresponding to the local semantic vector map based on the local semantic vector map; based on the map prior features and the look-around image, obtaining enhanced visual perception features after map prior enhancement; and determining the pose of the vehicle based on the enhanced visual perception feature and the map prior feature, and positioning the vehicle based on the pose of the vehicle. According to the method, the map priori features are learned from the semantic vector map, and the perception effect of the look-around image is enhanced to obtain the enhanced visual perception features after map priori enhancement, so that the online perception quality is improved, and the accuracy of the vehicle pose determined based on the enhanced visual perception features and the map priori features is effectively improved.

The embodiment of the invention provides a method and system for surveying and mapping the area of a peripheral area along a railway based on an unmanned aerial vehicle, and belongs to the technical field of area surveying and mapping. The method comprises the following steps: performing coordinate registration processing on laser point cloud data collected by an unmanned aerial vehicle and a railway red line vector diagram to obtain standard point cloud data and a red line reference boundary under a unified coordinate system; based on the standard point cloud data and the red line reference boundary, an edge point set of an actual occupied area outside the red line is extracted, and a closed boundary contour is obtained through fitting; based on the boundary contour obtained through fitting, unified plane projection and subdivision processing is executed, and a two-dimensional measurement grid used for area determination is generated; and based on the two-dimensional measurement grid, accumulating the area of each unit, and summarizing and outputting the total area and corresponding boundary coordinates of the red line external expansion region. According to the scheme, high-precision and full-process area surveying and mapping of the non-standard occupied area along the railway are realized, and the objectivity of boundary judgment and the stability of an area extraction result are improved.

The application relates to the technical field of computer vision, in particular to a model training method and device, a key point selection method and device, equipment and a medium. The method specifically comprises the following steps: determining a first high-dimensional feature map of training image data and a second high-dimensional feature map of transformed image data through a feature extraction module; constructing a loss function item based on a first predicted key point probability map, a first predicted descriptor vector map, a second predicted key point probability map and a second predicted descriptor vector map output by a key point detection module; and training a key point detection model based on the loss function item. In the application, the training image data containing an RGB image and a depth image and the transformed image data corresponding to the training image data are used as input items of the key point detection model to be trained, the RGB and depth information are fused, multi-scale features are effectively extracted, the key point detection model can effectively utilize the two-dimensional image information and the depth information, and the detection accuracy of the key point detection model can be improved.

The specification provides a vector graph index merging method and device, the method comprising: obtaining a plurality of vector graph indexes to be merged; wherein each node in the vector graph index represents a vector, and each edge represents that the vectors represented by the nodes connected by the edge are similar; for each target vector in each target vector graph index, determining the vectors similar to the target vector from each vector graph index based on an approximate nearest neighbor search algorithm to form a cross-graph candidate neighbor set of the target vector; based on the cross-graph candidate neighbor set of each vector in the plurality of vector graph indexes, constructing a global neighbor graph; wherein each node in the global neighbor graph represents each vector in the plurality of vector graph indexes, and each edge represents that the vectors represented by the nodes connected by the edge are similar; and converting the global neighbor graph into a merged graph index that can be used for approximate nearest neighbor search.

The invention belongs to the field of robustness artificial intelligence, and relates to a robustness semantic vector map construction method and system based on a multi-vehicle cooperation mechanism, and the method comprises the steps: carrying out the restoration processing of all-round view images of a main vehicle and a plurality of proxy vehicles through an image restoration model, and obtaining a restored image; based on the restored image, constructing a semantic vector map by using a semantic vector map construction model, which comprises the following steps: obtaining three-dimensional features of the main vehicle and the plurality of proxy vehicles based on the restored image; performing mask processing on the three-dimensional feature of each proxy vehicle by using the position relationship between the main vehicle and the proxy vehicle; carrying out hierarchical fusion on the three-dimensional features of the main vehicle and the three-dimensional features of the proxy vehicle after mask processing; and constructing a semantic vector map by using the fusion features. According to the invention, efficient perception of traffic environment information can be realized, and the reliability and safety of automatic driving are further improved.

The invention relates to a map displaying and plotting method and system for a full-localization display control system, and relates to the technical field of digital map generation. The map displaying and plotting method comprises the following steps: converting and partitioning original vector map data to obtain a plurality of map data blocks, and calculating a memory occupied by the corresponding blocks; creating a plurality of data receiving threads, receiving and analyzing positioning protocol data transmitted by the Beidou navigation module, and extracting satellite positioning information; converting and filtering the satellite positioning information according to a screen plane coordinate system to obtain screen positioning information; plotting screen positioning information to a map data block according to a double-buffer differentiated drawing mechanism, and rendering and displaying in combination with the occupied memory of the block; smooth map operation and high-frequency target refreshing under finite computing power are realized through map block loading, double-buffer drawing and an incremental updating mechanism; and through the multi-thread architecture design, the resource allocation balance is ensured, and the low delay from data receiving to map display is ensured.

The application discloses a kind of high-precision vector map automatic generation method, mainly according to laser point cloud intensity, geometric structure feature extraction lane marking, through ground filtering, ground, non-ground point cloud is segmented;Adopt block grating to convert three-dimensional point cloud into two-dimensional projection image;Using image filtering, edge extraction operator, dynamic adaptive threshold segmentation, morphological analysis and other processing extraction road marking pixel, based on point cloud image mapping relationship extraction marking point cloud;Using the point cloud contour matching method based on knowledge template library carries out marking classification.Based on deep learning convolutional neural network, road sign detection classification is carried out to road image, and the point cloud region of interest is established by image point cloud fusion;Using clustering screening candidate point cloud, based on principal component analysis carries out local projection, based on knowledge template library point cloud matching extraction traffic sign;Finally, based on GPS data, calibration, parameter fitting, vectorization, data association are carried out, and the construction of high-precision vector map is completed.

The invention discloses a highway meteorological data acquisition method and system based on meteorological bulletin images, and belongs to the technical field of meteorological data processing and geographic information systems. The method comprises the following steps: acquiring a meteorological bulletin image from the China Meteorological Bureau, and extracting effective meteorological data through preprocessing operations such as image boundary determination, scale-pixel conversion and data denoising; converting image pixel coordinates into latitude and longitude coordinates, and superposing the latitude and longitude coordinates with a road vector map to generate a road weather chart; identifying meteorological values based on a color threshold method, and extracting a longitude and latitude sequence of a target road section in combination with a region growing algorithm; and calculating continuous meteorological data by adopting a point-by-point interpolation method, and finally displaying meteorological distribution along the highway through a visualization technology. The problems that an existing meteorological data acquisition mode is high in cost, complex in operation and low in reliability are solved, low-cost and high-precision automatic extraction of the highway meteorological data is achieved, and real-time data support is provided for traffic management and disaster early warning. The system can be widely applied to the fields of highway meteorological monitoring, geological disaster early warning and the like.

The application discloses a vector drawing method and device based on vector tile data and electronic equipment, and the method comprises the following steps: calculating a vector tile number, constructing a resource link, sending a download request link, and acquiring vector tile data according to the display scale and display range of a map; creating a vector tile object by using the vector tile data based on a defined vector tile data format; selecting geometric data for drawing from the vector tile object or after transformation on a display projection by comparing the vector tile data with data in the display projection; rendering the drawing by using the geometric data; and displaying a vector map. In the application, a vector tile file structure with spatial range, projection information, element attribute and rendering style is provided; the vector tile is quickly loaded through index placeholder information, and the file decoding link is removed.

The invention belongs to the technical field of digital urban and rural and built-up environment simulation, and provides a method for constructing and analyzing a regional urban and rural built-up environment by fusing multi-source data, and the method comprises the steps: obtaining a plane grid base map based on a selected region, and vectorizing the plane grid base map to obtain a two-dimensional building vector diagram; calculating and obtaining a building height base map based on DEM and DSM data, and establishing a three-dimensional building model by using the two-dimensional building vector diagram and the building height base map; a three-dimensional terrain model is obtained based on DEM data, and the three-dimensional building model and the three-dimensional terrain model are combined to form a built environment digital twinborn model of the selected area; and carrying out multi-physics field simulation analysis on the built environment digital twinborn model, realizing multi-physics field coupling simulation by combining a turbulence model, a radiation model and the like, analyzing a simulation result, and displaying analysis data in a visual form. According to the method, multi-source data are efficiently fused, the spatial scale of regional large-scale simulation research is expanded, and multi-physics field coupling analysis and quantitative analysis of a built environment are realized.

The invention discloses a multi-condition constrained multi-modal remote sensing intelligent identification method for an arid region reservoir dam, and the method comprises the steps: firstly, making a VIS-NIR multi-modal reservoir dam image data set for algorithm training and feature learning for a serious problem of target spectrum homogenization of VIS data in an arid region; secondly, designing a DDE-YOLO algorithm, adopting an MF early fusion strategy to perform weighted fusion on input of two modes of VIS and NIR in a symmetric mode, and performing feature learning optimization by using MSC3k multi-scale C3k convolution and TCUB transposed convolution up-sampling; and then providing a multi-condition constraint strategy to screen potential candidate areas of the dam, realizing accurate positioning of a target by using an NMS algorithm, and extracting coordinate information of a center point of the detected target to generate a vector map. The method can effectively solve the problems that under the complex arid region background, under VIS single-mode input, the homogenization of a reservoir dam target and the background is serious, the dam space positioning error is large, the detection precision is low, the false alarm rate is high, and the remote sensing intelligent extraction and automatic drawing level is low.

The invention relates to an open-pit mine area online mapping and boundary line fitting method based on sparse perception, belongs to the technical field of open-pit mine area automatic driving, and solves the problems that in the prior art, a step-by-step processing mode is used in an open-pit mine area environment perception method, information transmission loss exists, and environment perception is incomplete. Extracting a multi-scale feature map; s2, decoding the multi-scale feature map to obtain a dynamic instance and a static instance; s3, constructing a local vector map, and dynamically updating the local vector map; s4, key points of road boundaries are extracted from the local vector map, and matching point pairs meeting matching conditions are obtained; s5, carrying out breadth-first search on the matching point pairs to obtain an initial road boundary sequence; s6, screening the obtained initial road boundary sequence to obtain a screened road boundary sequence; and S7, re-projecting the screened road boundary sequence back to the original image to obtain a fitting road boundary.

The application provides a vector map construction method and device, a storage medium and electronic equipment, comprising: performing low-light image enhancement processing on N frames of current images to obtain N frames of enhanced images; performing multi-view BEV conversion processing on the N frames of enhanced images to obtain BEV current features; performing fusion processing on the BEV current features and BEV features in a historical window to obtain BEV fusion features; generating a vector curve based on the BEV fusion features, and completing vector map construction based on the vector curve. In a low-light environment and a dynamic scene, the perception ability of an automatic driving system, the stability of map construction and the calculation efficiency can be effectively improved; through fusion of high-definition enhanced images and space-time information, high-precision and low-delay vector map prediction is realized, the robust application requirements of automatic driving technology in a complex environment are met, and more reliable technical support is provided for the development of a future intelligent transportation system.

This application discloses an autonomous navigation and deception defense system and method for unmanned aerial vehicles (UAVs) in GPS-denied environments. The system includes a multi-source sensor array, and parallel deception detection and isolation threads and a tightly coupled multi-source fusion navigation thread. The deception detection and isolation thread performs multi-dimensional consistency checks on received navigation data. When a dual comparison of geomagnetic fingerprints indicates a poor match between GPS position and geomagnetic readings, but a good match between visual position and geomagnetic readings, the GPS data is isolated from the fusion module. The tightly coupled fusion navigation module employs an error state Kalman filter with an IMU as its core. A geomagnetic deviation term is introduced into the state vector, and the original observation geometric information is directly fed into the filter for tight coupling. The difference between the geomagnetic vector obtained from a geomagnetic vector map query and the measured value from the magnetometer is used for fusion and updating, achieving high-precision continuous navigation in GPS-denied environments.

This disclosure provides a method and apparatus for drawing a vector map of an accident scene. The method includes: performing object recognition on a global top-view of a target accident scene to determine at least one object included in the target accident scene; symbolizing the at least one object in the global top-view to obtain at least one object symbol; determining at least one reference element based on the distribution of the at least one object in the global top-view; determining annotation information for at least one key accident point in the global top-view based on the at least one reference element; and obtaining a symbolized vector map of the scene with annotation information based on the annotation information of the at least one key accident point and the at least one object symbol. This disclosure enables the automatic drawing of a standard vector map of a target accident scene while automatically annotating the details of the accident scene.

The invention relates to the technical field of image detection, and discloses a floor processing quality detection method and system based on machine vision. The method comprises the following steps: collecting multiple frames of images of the surface of a floor through multi-angle structured light illumination, constructing a pixel-level luminosity response vector, and decomposing the pixel-level luminosity response vector into a diffuse reflectance graph, a specular reflectance graph and a normal vector graph based on a physical illumination model; and respectively detecting material flaws, coating unevenness and geometric scratches and pits in each component graph by adopting a targeted algorithm, and finally fusing results to generate a high-precision detection report. The system comprises a high-speed camera and a dome-type multi-light-source module which are synchronously controlled, a data processing unit and a defect decision fusion module. According to the method, interference suppression is realized through physical attribute decoupling, and the accuracy and robustness of defect identification are remarkably improved.

The application discloses an agent action prediction method based on multi-scale space perception, comprising map space structure modeling, historical trajectory feature extraction, feature fusion and multi-modal action prediction, wherein: the map space structure modeling uses a multi-scale graph convolutional neural network to extract map features from two-dimensional vector map data in an application scenario map, to obtain high-dimensional map feature information; the historical trajectory feature extraction uses a convolutional neural network and a feature pyramid network to extract high-dimensional trajectory data features of all agents; the feature fusion models and fuses the correlation of high-dimensional map feature information and high-dimensional trajectory data feature information through a self-attention mechanism, to obtain agent trajectory fusion features with direction information; and the multi-modal trajectory prediction uses the agent trajectory fusion features for prediction regression and confidence scoring, to output multi-modal complete trajectory coordinates and corresponding confidence scores of action prediction, so as to provide reasonable auxiliary decision-making for agent action.

The invention relates to the field of agricultural irrigation monitoring, and provides a crop area water level determination method and device and a storage medium. The determination method comprises the following steps: generating a digital elevation model of a crop region, wherein the crop region comprises a plurality of sub-regions; determining elevation data of each sub-region from the raster data of the digital elevation model based on a predefined field boundary vector diagram to obtain elevation matrix data of each sub-region; performing visualization processing on the elevation matrix data of each sub-region to generate an elevation distribution diagram, and marking pixel points of a preset elevation in the elevation distribution diagram; determining a target mounting position of the water level gauge based on the elevation distribution diagram; acquiring water level data acquired by a water level gauge and elevation data at a target mounting position; and determining a water layer coverage area proportion and / or an average water layer height in the crop area based on the digital elevation model, the water level data and the elevation data at the target installation position. The method can provide accurate irrigation decision support for crops.

The application relates to a ground target detection radar B type display method, determines geographical position information of a radar, and radar detection parameters, loads a vector map file into a computer memory, reads and processes point and line elements in the vector map, draws a radar B type display mode, calculates whether the point and line elements in the vector map are in the B type display range, if in the display range, carries out superposition drawing, and draws a target detected by the radar onto the B type display mode. Technical effects are that the target displayed in the ground target detection radar B type display mode can be directly and visually seen to be relative to geographical reference information (such as villages, rivers, roads and the like), time for a radar debugging personnel to calculate the target relative to the reference information (such as villages, rivers, roads and the like) is saved, and a radar operator is more convenient to judge the accurate position of the target.

The application provides a vector map construction method and system based on prior map and multi-stage diffusion reasoning. A unified vector encoder is used to encode multiple prior maps to generate prior features. Real-time data obtained is converted into bird's-eye view features, and multi-stage diffusion reasoning is performed to map the bird's-eye view features of the degradation domain to diffusion features of the normal domain. The prior features and the diffusion features are fused to obtain fused features. The fused features are input into a constructed map decoder to generate an online vector map. In the scheme, multiple prior maps are introduced, and a unified vector encoder is used to efficiently encode different prior maps, thereby enhancing the robustness and accuracy of online map construction. Through multi-stage diffusion reasoning technology, image details are gradually optimized, which not only uniformly processes multiple image degradation problems, but also provides high-quality image restoration and enhancement effect in complex environments, thereby providing strong support for an automatic driving perception system.

The application provides a terminal device positioning method and related equipment, which can improve the accuracy of the positioning result of the terminal device. The method of the application comprises: obtaining, from a vector map, a first map point matched with a first feature point of a current image frame photographed by a terminal device; obtaining, from the vector map, a second map point matched with a second feature point of another image frame before the current image frame; and adjusting a pose of the terminal device when the current image frame is photographed to obtain a pose of the terminal device when the current image frame is photographed after a current adjustment, as a positioning result of the terminal device.

The application provides a simplification method, system, device and medium for complex linear map elements, comprising: marking the kinks on the complex linear vector map by estimating the maximum noise threshold size according to the zoom scale; removing the line segments and corresponding kinks between the kinks on the complex linear vector map, whose length is less than the maximum noise threshold size, to obtain a preliminary screened map; starting from the kink at the starting end of the preliminary screened map, constructing a sliding circle with each kink as the center in turn, and screening the kinks according to the number of intersection points between the sliding circle and the line segments, to obtain a screened map; and smoothing the screened map based on the Bezier curve to obtain a simplified map. The application solves the problem that the simplification process in the prior art causes the loss of complex bending features of part of the map elements, and the simplification is incomplete due to the omission of part of the sharp bending and dense kinks during identification.

Embodiments of the invention provide an image processing method and apparatus, and an electronic device. The method comprises the steps of obtaining images acquired by an image acquisition device when a lens is located at different optical axis positions in different illumination modes; generating a super depth-of-field depth map based on the obtained image; generating a photometric stereo normal vector diagram based on the obtained image, and generating a photometric stereo depth map based on the photometric stereo normal vector diagram; and correcting depth information in the same depth-of-field region in the super depth-of-field depth map according to a photometric stereo normal vector in the photometric stereo normal vector map, and correcting depth information in a low texture region in the super depth-of-field depth map according to a photometric stereo depth value in the photometric stereo depth map. The detail depth information in the same depth-of-field region can be corrected by the photometric stereo normal vector, and the missing depth information in the low-texture region can be restored by the photometric stereo depth value, so that the super depth-of-field depth map which can well reflect the depth details of the collection view field can be obtained.

The disclosure provides an ellipsoid vector mapping method and system based on cubic polynomial splines, and relates to the technical field of computer graphics, which takes any point outside the ellipsoid as an observation point, maps the observed ellipsoid into a vector graph on a plane, and comprises the following steps: based on the ellipsoid equation to be mapped and the coordinates of the observation point, the plane equation on which the observed ellipsoid contour is located is calculated; according to the ellipsoid equation and the plane equation, the implicit equation of the projection ellipse is constructed; the projection ellipse is segmented and converted into a cubic spline curve through the implicit equation of the projection ellipse; based on the cubic polynomial spline method, the closed area surrounded by the cubic spline curve is filled to obtain the plane projection of the ellipsoid, and the vector graph on the plane is completed. The disclosure can map a three-dimensional ellipsoid into a two-dimensional vector graph, and the related technology can be applied to the development of three-dimensional drawing software with a derived vector graph function, and plays an important role in the production of high-quality illustrations of books, documents and courseware.

The invention discloses a contour vector diagram generation method and apparatus, a storage medium, a device and a product. The method comprises the steps of obtaining a depth map covering a whole scene; constructing an agent model based on the depth map, wherein the agent model is a geometric representation of the scene surface; generating a cutting plane sequence with equal height spacing according to the proxy model; performing intersection calculation according to the cutting plane sequence and the proxy model, and extracting an intersection line as a contour line; and outputting a vector diagram file according to the contour line. According to the contour vector diagram generation method provided by the invention, the depth map is adopted to construct the proxy model and the adaptive cutting calculation is carried out, so that the contour vector diagram is efficiently generated in a large-scale scene, meanwhile, the balance between the calculation precision and the resource consumption is ensured, and unnecessary calculation waste is avoided.