253 results about "Vector map" patented technology

The invention relates to a remote sensing image ground feature recognition method based on deep learning, and the method comprises the steps: carrying out the data collection and preprocessing of a ground surface target region, eliminating the position deviation through geometric correction, processing the illumination difference through combination with radiation equalization, and generating a ground feature registration image; performing multi-dimensional feature fusion processing on the image, performing tensor fusion on vegetation spectral features, earth surface texture features, point cloud data features and linear ground feature features, and constructing a ground feature fusion matrix; a bilateral convolutional neural network is adopted to extract spectral response characteristics and spatial correlation characteristics, and characteristic interaction is realized through an attention mechanism to generate a ground feature probability distribution diagram; and finally, carrying out noise filtering, boundary refining and vectorization conversion processing on a classification result, and outputting a ground feature classification vector diagram. According to the method, three technical bottlenecks of insufficient cooperative utilization of multi-source heterogeneous data, insufficient spectrum-space feature fusion and poor GIS compatibility are solved, and the operation efficiency of territorial investigation, disaster monitoring and other scenes can be remarkably improved.

The embodiment of the invention provides a scoliosis Cobb angle determination method and device based on Mama, equipment and a storage medium. The method is applied to the scoliosis prediction field. The method comprises the following steps: acquiring image information, and preprocessing the acquired image information to obtain an original image; performing multi-level local feature extraction on the original image through an encoder to obtain a first feature map; performing global feature capture on the first feature map through a Mama module to obtain a second feature map; decoding the second feature map through a decoder to obtain a target feature map; generating, by a prediction module, a central heat map, an offset map and a centripetal vector map based on the target feature map; according to the central heat map, the offset map and the centripetal vector map, extracting central point coordinates of the centripetal body, and according to the central point coordinates of the centripetal body, determining four angular point coordinates of the centripetal body; the Cobb angle is determined through a geometric algorithm according to the coordinates of the four angular points of the cone, and the measurement precision of the Cobb angle is improved.

The invention belongs to the technical field of automatic driving, and particularly relates to an automatic driving local vector map construction method based on a standard precise map and fusion perception. The method comprises the following steps: preprocessing standard precise map information, and constructing a fusion sensing network; in the standard and precise map information preprocessing, original standard and precise map data are processed into vectorization expression suitable for model input; the fusion sensing network comprises a feature extraction backbone network, a BEV (aerial view) view conversion network, a multi-mode BEV feature fusion network and a map element decoding module; in the map element decoding module, the processed standard precise map elements are interacted with BEV aerial view features of the sensor; in local map model training, query denoising is introduced as an additional training task, and query sorting selection is introduced to accelerate the model convergence speed. According to the method, higher accuracy, better robustness and higher operation efficiency can be achieved, and the influence of environmental factors is smaller.

The invention provides a method for generating multi-dimensional landform vector data based on remote sensing image semantic segmentation, and belongs to the technical field of remote sensing image ground object recognition and artificial intelligence, and the method comprises the steps: training a landform recognition deep learning model based on a remote sensing image, adjusting a pre-training weight to optimize an image encoder, and generating a classification result; the landform recognition deep learning model comprises an encoder formed by a multi-layer convolutional network and a decoder of an up-sampling and feature fusion architecture; extracting a single-band label corresponding to each landform type from the classification result, and generating a binary raster image with the single-band labels; and converting the binary raster image with the single-band label into a vector diagram, and combining the vector diagram with the elevation information to generate multi-dimensional geographic data.

The invention provides an online vector map construction method fusing a large language model and thermodynamic diagram attention and electronic equipment, and the method comprises the steps: obtaining an image collected by multiple cameras, extracting a feature map of the image, converting the feature map into BEV features, and optimizing the BEV features through pooling operation; dynamic objects and static objects are separated from the optimized BEV features, semantic categories are marked, BEV semantic features are obtained, and semantic enhancement is conducted on the BEV semantic features in combination with a knowledge base of a large language model; generating a thermodynamic diagram by using the BEV semantic features after semantic enhancement, generating query points in the thermodynamic diagram, and constructing a vector map by using the BEV semantic features and the query points; and performing quality evaluation on the vector map and performing error optimization based on an evaluation result. According to the method, for a complex dynamic environment, a high-precision vector map can be constructed by optimizing the feature fusion capability of the multi-view image, accurately distinguishing a static object from a dynamic object and improving the semantic understanding capability of the environment.

The invention relates to the technical field of embedded maps, and provides a vector map display method under a VxWorks system, which comprises the following steps: S1, analyzing a vector map source file through an MITAB library, and extracting geometric space topological data, geographic coordinate system parameters, style attributes and projection parameters; s2, constructing a hierarchical storage structure based on SQLite3, and storing vector map information including a spatial data table, a style configuration table and a metadata table into a geographic information database; s3, creating a map display control MapWidget inherited from a QGraphics View class based on a Qt framework, loading the geographic information database through a dynamic configuration interface, and initializing a coordinate system conversion module, a multi-level view window module and a covering management system; and S4, dynamically rendering and outputting an interactive vector map interface through a layer priority scheduling strategy in a VxWorks real-time operating system. Through the design of the cross-platform data middle layer, the extensible general symbology system and the resource efficiency optimization, the data compatibility problem of the embedded system and the commercial GIS software is solved.

The invention discloses an unmanned aerial vehicle tilt video and vector map real-time projection sleeving method, which comprises an unmanned aerial vehicle body, and is characterized in that a video data acquisition module and a video preprocessing module for preprocessing the tilt video acquired by the video data acquisition module are arranged in the unmanned aerial vehicle body; a map data processing module is further arranged in the unmanned aerial vehicle body, an improved ORB algorithm is adopted in the feature extraction and matching module, and a posture correction and projection transformation module used in cooperation with the video data acquisition module is installed in the unmanned aerial vehicle body. And a real-time display module in the unmanned aerial vehicle body displays the sleeved video and vector map on terminal equipment in real time through a signal module arranged in the unmanned aerial vehicle body. The problems that the real-time performance is poor, the nesting precision is reduced, and long-time high-precision nesting cannot be achieved during working in an existing unmanned aerial vehicle tilt video and vector map fusion technology at present are solved.

The invention discloses a robust online vector map construction method for space-time simplified query, and relates to the technical field of automatic driving perception, and the method comprises the steps: fusing multi-view features into a unified bird's-eye view for representation; matching instance features of the current frame and the historical frame based on time association; updating query features and optimizing query representations; a high-quality vector map construction model is used as a teacher model, a map construction model focused on taking an instance as a center is used as a student model, and features of a teacher network are adaptively distilled to a student network; and designing a loss function to evaluate the student model, performing training optimization, and if the output requirement is not met, further optimizing the student model through a middle distance perception mechanism. According to the method, the robustness, the precision and the efficiency of online vector map construction are remarkably improved, especially the performance in a dynamic environment and a complex scene, and by means of the mode, multi-view, space-time correlation and complex 3D target detection tasks can be better processed.

The invention discloses a method for constructing a computable road network. Road data in the vector map are extracted, line string data or line segments obtained by splitting the line string data are used as connecting edges, geographic feature codes containing Geohash information are given to starting points and ending points of the connecting edges, and therefore a basic road network map is constructed. And extracting a middle vertex set of the basic road network diagram and an edge attribute set of middle vertex connection edges, and carrying out road network vertex optimization based on edge attributes to obtain an optimized computable road network. Geographic feature codes adopted by the method comprise spatial geographic information and have the characteristics of graph vertex coding, and the constructed road network graph can be used for graph calculation and verification and restoration of road connection relations. Furthermore, in combination with a representation point acquisition scheme provided by the invention, a road network map fusing facilities and equipment can be constructed, so that the application range of the computable road network is expanded. The computable road network constructed by the method is high in operation efficiency and suitable for different production environments.

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 semantic vector map robust construction method and device, a storage medium and a program product, which are applied to the technical field of automatic driving perception, and the method comprises the following steps: obtaining a multi-exposure original image; preprocessing the multi-exposure original image to obtain a high dynamic range image; bEV feature extraction and enhancement are carried out on the high dynamic range image to obtain a multi-scale BEV feature and an enhanced BEV feature; processing the enhanced BEV features to obtain a mixed query, and processing the mixed query and the multi-scale BEV features through a multi-layer mask activated instance decoder to obtain an embedded query and a coordinate query; processing the enhanced BEV features through a mask feature construction unit to obtain binary mask features, determining local block features based on coordinate query, and adjusting lane point coordinates based on the local block features, embedded query and a multi-layer perceptron to obtain a semantic vectorization map. The method has the effect of improving the accuracy and robustness of map construction.

The invention discloses an indoor navigation positioning precision improving method and system based on a filtering algorithm, and relates to the technical field of signal processing. The method comprises the following steps: suppressing signal fluctuation and multipath interference by adopting a Bluetooth positioning technology in combination with a segmented power law regression model and Kalman filtering; ultrasonic-assisted positioning is introduced, the distance is calculated through the time difference between pulses and Bluetooth, and the anti-interference capability is improved in cooperation with temperature correction and spectral analysis; multi-mode dynamic fusion is carried out, when shielding is little, Bluetooth positioning is corrected through high-precision ultrasonic waves, when shielding is serious, Bluetooth data is smoothed through moving average, and positioning coordinates are optimized in combination with a least square method; a base station and a reference object are marked through a vector map, the position of a terminal is dynamically displayed, and positioning reliability is visually presented through an error semitransparent circle; a positioning base station array, a mobile terminal and a rear-end processing module are integrated, and low-power-consumption multi-mode fusion positioning is achieved. The method reduces the positioning error in a complex indoor environment, and is suitable for scenes such as personnel / equipment real-time tracking.

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 invention relates to the technical field of vehicle-mounted equipment connection, in particular to a pairing connection method and device between equipment terminals and equipment. The method comprises the following steps: collecting equipment basic information databases of all vehicle-mounted equipment, performing vehicle-mounted equipment trust level evaluation and differentiated communication protocol stack configuration, and constructing a differentiated protocol adaptation strategy; the method comprises the following steps: monitoring signal state data of a vehicle-mounted environment in real time, predicting a signal state quality evolution trend, and constructing a channel quality trend prediction map; carrying out channel quality interval calculation according to the channel quality trend prediction map, and carrying out optimal pairing opportunity calculation and protocol stack pre-distribution according to a differential protocol adaptation strategy and a quality evaluation value to generate a communication protocol pre-distribution strategy; the method comprises the following steps: collecting a vehicle-mounted equipment operation monitoring log, and constructing a global interaction behavior vector graph; according to the invention, efficient and stable vehicle-mounted equipment connection is realized, the equipment connection efficiency is improved, resource conflicts are reduced, and the adaptability to a complex vehicle-mounted environment is enhanced.

The invention relates to the technical field of automatic driving perception, and provides an online vector map construction method and device, a storage medium and a program product. The map construction method comprises the steps of performing reliability analysis on a vehicle laser radar point cloud and a multi-view image, and fusing the laser radar point cloud and the multi-view image into a fused BEV feature based on a reliability analysis result; generating point column data for representing map elements based on the historical high-precision map; performing cross attention mechanism processing on the fused BEV features and the point column data to obtain current map elements; and processing the historical high-precision map based on the current map element to obtain a high-precision vector map. According to the technical scheme, the point cloud and the image can be fused based on the reliability analysis result of the laser radar point cloud and the multi-view image, so that the influence of unreliable data points on the whole data can be automatically reduced when the visual field is limited or the sensor data is abnormal, and the accuracy of vector map construction can be 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 present invention discloses a method for processing submarine sonar images based on semantic segmentation, comprising the following steps: S1. obtaining a preprocessed sonar image block; S2. generating a priori map of submarine landform boundaries corresponding pixel by pixel to the preprocessed sonar image block; S3. constructing an improved U-Net dual-branch semantic segmentation network, comprising a semantic branch and a boundary branch, and configuring a landform mutual feedback gating module at each scale jump layer; S4. obtaining a fused feature map; S5. deploying the map to an autonomous underwater vehicle to form an onboard inference model; and S6. utilizing the onboard inference model to perform online semantic segmentation inference on the preprocessed sonar image block acquired in real time, outputting a semantic category map and a boundary vector map corresponding to the preprocessed sonar image block. The present invention improves the average mIoU metric of the model in scenarios with small targets and multiple categories.

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 present invention discloses an automated multi-threshold discrimination and extraction method and device for burned areas. The method includes: preprocessing the pre-disaster and post-disaster remote sensing images of the burned area to generate preprocessed images; constructing an NDSI index based on the preprocessed images; processing the remote sensing images based on the NDSI index to enhance the spectral differences between cloud, water, shadow interference factors and the burned area, and generating raster images; automatically threshold-classifying the burned area and cloud, water, shadow interference factors based on the OTSU algorithm; performing interference factor elimination processing on the remote sensing images based on threshold segmentation to generate area elimination images; extracting the NDVI-D index for two periods of pre-disaster and post-disaster for the area elimination images, and performing adaptive threshold segmentation on the extraction results to obtain a binary raster map of the burned area, and obtaining a vector map of the burned area as needed. The present invention can achieve the purpose of accurately extracting the scope of the burned area.

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 invention discloses a semantic vector map construction method supporting beyond visual range and dynamic updating, and belongs to the technical field of semantic vector map construction method.The method comprises the steps that a look-around image is input into a network, dynamic map updating is achieved through a query propagation mechanism according to the time sequence fusion features, the consistency of map elements in the time dimension is ensured, and the map quality is improved. According to the method, the perception and decision-making capability of the automatic driving system in a dynamic environment can be effectively improved, and accurate and stable map support is provided for the automatic driving technology.

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.