406 results about "Raster data" patented technology

The invention provides an urban inland inundation simulation method based on CPU-GPU heterogeneous parallel calculation, and aims to reduce the number of calculation grids by adopting a self-adaptive non-uniform grid, consider pipe network calculation and improve the calculation efficiency of low-fluctuation urban area flood simulation at the same time, and comprises the following steps: performing one-dimensional pipe network runoff CPU calculation, constructing a rainwater pipe network model, and constructing a rainwater pipe network model. Extracting an overflow process of an overflow node of the pipeline; dividing a non-uniform grid based on the original raster data and a gradient threshold; non-uniform grid parallel calculation is carried out; a CPU-GPU heterogeneous parallel one-dimensional and two-dimensional coupling model is constructed; and coupling water depth calculation. The method can flexibly adapt to topographic changes and reduce the number of computational grids, and solves the limitation that precision and efficiency are difficult to consider in waterlogging simulation of a uniform grid in a low-fluctuation area in combination with CPU-GPU heterogeneous parallel calculation.

The invention provides a multi-dimensional visual analysis method for meteorological observation and detection data, and relates to the field of visualization. According to the method, an unmanned aerial vehicle platform carries a meteorological sensor to collect data, and primary processing, abnormal value detection and intelligent data compression are carried out; transmitting the processed data to a ground measurement and control terminal; the ground terminal fuses the multi-source data to generate a three-dimensional meteorological data set; constructing an interactive three-dimensional visual scene by utilizing self-adaptive interaction logic and a dynamic data display technology; a user can check and perform multi-view linkage operation through the interactive operation interface; a severe weather and weather raster data set can be constructed based on the three-dimensional weather data set and historical severe weather records, and simulation and display are carried out; and finally exporting the data, verifying the precision and integrity, and generating a visual analysis report.

The embodiment of the invention provides a landslide disaster negative sample optimization method based on an improved frequency ratio, and belongs to the technical field of data processing, and the method specifically comprises the steps: collecting disaster-pregnant environment data, and carrying out the preprocessing of the disaster-pregnant environment data, and obtaining raster data; defining two parameters of grading precision and grading bandwidth, introducing information entropy and a maximum frequency ratio to adaptively set parameter values of the grading precision and the grading bandwidth, and calculating a comprehensive frequency ratio; based on the statistical distribution of the comprehensive frequency ratio, taking a peak point as a negative sample screening threshold value, dividing a negative sample screening area based on the threshold value, then realizing calculation of a positive and negative sample ratio, and further determining a negative sample label under the constraint of a buffer area; according to the frequency ratio distribution of each raster data, an attribute interval with a frequency ratio greater than 1 is extracted as a significant feature, then the significant features of each raster data are classified, and an optimal feature element is determined through multi-collinearity analysis. Through the scheme of the invention, the accuracy and reliability of negative sample selection are improved.

The present invention relates to the technical field of digital twin, and in particular to a private network park digital twin visualization method, apparatus and device, and a computer storage medium. Disclosed in the present application is a private network park digital twin visualization method, comprising: using a clustering algorithm to classify ground objects in a remote sensing image of a park, dividing grids on the basis of formed pixel clusters, and calculating dilution ratios thereof; mapping grid data and point cloud data into a unified coordinate system, and using the clustering algorithm to cluster the point cloud data in each grid; sampling each point cloud cluster on the basis of the dilution ratio of the corresponding grid, and finding out non-edge points by means of a BoundED algorithm; on the basis of the proportion of the non-edge points in each point cloud cluster, calculating an initial dilution threshold of each point cloud cluster; and, after each point cloud cluster has been diluted, rendering and presenting same by means of a three-dimensional rendering engine and returning a rendering frame number and, if the rendering frame number meets a preset condition, ending the process.

The invention discloses a goaf collapse risk assessment data fusion system based on big data processing, and particularly relates to the technical field of geological disaster assessment, and the system comprises three core modules: a multi-source data adaptive weighted fusion module which establishes a unified space-time coordinate system, converts non-raster data into a continuous field through Kriging interpolation, and performs data fusion on the continuous field; combining the information entropy and the correlation coefficient to dynamically distribute weights, and generating an enhanced feature field through self-supervised pre-training; the physical-space-time neural network dynamic prediction module is integrated with elastic-plastic mechanical constraint loss and multi-task learning, and outputs a future multi-time step risk probability field and a deformation prediction field through a space-time convolution-memory network; and the risk field three-dimensional subdivision and emergency response module is used for clustering three-dimensional voxels in a high-risk area, automatically calculating risk body parameters, generating an emergency scheme in combination with DEM data and an A * algorithm, and improving evaluation accuracy and emergency scheme practical operability through digital twinborn deduction evaluation.

The invention discloses an urban district regulation construction management operation and maintenance method and system based on three-dimensional digital twinning, and the method comprises the steps: collecting multi-source data, including geographic raster data, BIM model data and mobile phone signaling data, of an urban district, and achieving the coordinate alignment of the multi-source data through an LM algorithm; the multi-source data generates feature data composed of coding tables defining urban features through feature engineering, and an octree is constructed based on the feature data; defining leaf nodes of the octree as voxels, and applying multi-scale convolution to each voxel to obtain super voxels with space-time labels; constructing a constraint network based on the super voxels and a construction, management or operation and maintenance specification constraint library; and judging the risk area of the city district by calculating the space-time constraint intensity of the constraint edge, and outputting a corresponding risk area list. According to the invention, the problems of multi-source data integration, real-time dynamic monitoring and risk prediction in the planning, construction, management, operation and maintenance processes of the urban district are solved.

The invention discloses a sea area monitoring method and system based on a satellite remote sensing technology, and relates to the technical field of remote sensing, and the method comprises the steps: firstly, generating multi-stage and multi-band remote sensing raster data through a resource set, and secondly, obtaining joint sample vectors at different times through monitoring external environment data in a target monitoring region, a multivariable kernel density estimation algorithm is adopted to obtain a four-dimensional joint probability density distribution function, and an abnormal coverage area is obtained in combination with multi-stage and multi-band remote sensing raster data; then, based on the abnormal coverage area, identifying an enhanced abnormal unit, and performing dynamic correction operation; and finally, after dynamic correction, obtaining an abnormal connected region, analyzing an abnormal unit with an unstable form, determining an abnormal evolution sensitive region, and realizing a local adaptive optimization means based on the change of the abnormal evolution sensitive region.

The invention relates to the field of aided design of construction roads, and discloses a planning method for a temporary construction road of a power transmission line, which comprises the following steps: acquiring elevation data, topographic parameters, tower position data, existing road information and raster data of a construction area of the power transmission line; constructing an initial grid network based on a grid method, converting the elevation data into grid elevation values, and storing the grid elevation values into the initial grid network; constructing a final grid network with a communication relationship by taking a slope angle between adjacent grids of the initial grid network and the communication cost of the topographic parameters corresponding to the grids as constraint conditions; constructing a moving window to search the final grid network, and constructing a weighted directed graph; and determining a starting point and an end point of the power transmission line, and searching the weighted directed graph based on an improved Dijkstra algorithm to obtain a design scheme of an optimal temporary construction road with the minimum total construction cost between the starting point and the end point. The accuracy and reliability of temporary construction road planning of the power transmission line can be improved.

The invention provides a plaque generation land utilization simulation method, and relates to the technical field of geographic information science, and the method comprises the steps: obtaining initial land utilization data, and obtaining change raster data; random point sampling is carried out on the driving factor data and the change raster data to obtain sampling data; training a random forest model through the preprocessed sampling data; obtaining a development probability and a development potential map through the driving factor data and the random forest model; inputting the development potential map and the initial land utilization data into a CA model, and iteratively scanning each pixel of the initial land utilization data to obtain an overall probability; on the basis of the overall probability, a roulette mechanism and a threshold reduction rule are used for introducing competition of land utilization types, and a land utilization simulation diagram is obtained in combination with multi-type random plaque seeds, so that accurate simulation of changes of multi-type land utilization plaques and landscapes is realized, and meanwhile, the method is applied to future land utilization scenario prediction.

The invention discloses an AI-based illumination control system self-adaptive dimming method, which comprises the following steps of: acquiring a two-dimensional coordinate, a motion track curve and voice communication data of a personnel position, and identifying social interaction scene categories of different time slices; determining a social interaction strength score according to the voice communication data, and predicting the most adaptive spectrum configuration data in the current social interaction scene in combination with the social influence factor matrix and the social interaction scene type; according to a historical dimming log and an energy consumption metering curve, identifying a short-term high-frequency dimming segment, and adjusting a real-time response strategy and a disturbance classification processing mechanism of the spectrum configuration data; and adjusting the illumination power distribution values of the high-frequency use area and the low-frequency use area according to the occupation raster data generated by the millimeter-wave radar and the infrared array in combination with the lamp area mapping relation and the power budget constraint. Through multi-source data fusion and a self-adaptive regulation and control mechanism, unification of real-time performance, pertinence and energy-saving performance of spectrum regulation is realized, and the dynamic response capability and the user perception experience of the intelligent illumination system are improved.

The invention discloses a GIS-based main body function area evaluation index system construction method, and relates to the technical field of land resource management, and the method comprises the steps: obtaining multi-source geographic data of a target main body function area, and carrying out the preprocessing to form a unified geocoding evaluation database; generating an evaluation index system of the main body function area by adopting a composite weighting algorithm according to the evaluation database and the main body function area type division result; inputting the multi-dimensional raster data of the evaluation index system into the spatial decision tree classification model to obtain a classification result; performing overlay analysis on the multi-dimensional raster data to obtain comprehensive evaluation raster data; according to the classification result and the comprehensive evaluation raster data, obtaining a function goodness of fit thermodynamic diagram of the main body function area; and according to the function goodness of fit thermodynamic diagram, performing dynamic updating based on the time sequence, and when the change rate of the monitoring data exceeds a preset threshold, triggering an iterative optimization process. According to the invention, land space planning and main body function area optimization adjustment are realized.

The invention discloses a park scale vegetation carbon sink calculation method based on a visible light vegetation index, and the method comprises the following steps: obtaining a visible light orthographic image of a target park, and calculating a blue-green ratio index GBRI of the pre-processed visible light orthographic image, extracting and converting raster data of a vegetation coverage area in an integral pixel of the visible light orthographic image, and inputting the raster data into a visible light vegetation index calculation model to obtain a visible light vegetation index in the range of the target park; collecting the specific information of the vegetation to obtain the carbon reserve of the vegetation in each sample plot and the carbon sequestration of the vegetation in each sample plot; establishing a park vegetation carbon reserve estimation model by using the carbon reserve of the vegetation in each sample plot and the corresponding different visible light indexes in each sample plot; establishing a park vegetation carbon sequestration amount estimation model by using the carbon reserves of the vegetation in each sample plot and the corresponding different visible light indexes in each sample plot; according to the method, efficient and accurate calculation of the park vegetation carbon sequestration is realized.

The invention provides a flood simulation method driven by physical and data fusion, and belongs to the technical field of flood simulation, the method comprises the following steps: pre-training to obtain a qualified urban flood simulation model, a target loss function comprising a data supervision loss item and a physical constraint loss item, the physical constraint loss item is composed of a mass conservation residual error loss item, a momentum conservation residual error loss item and an energy conservation residual error loss item; inputting ground elevation data and historical meteorological data of a target area in a city into the hydrological model to obtain water depth grid data and flow velocity grid data, and inputting the water depth grid data and the flow velocity grid data into the city flood simulation model for prediction processing to obtain water depth grid data and flow velocity grid data of the target area at different time points in the future; and based on the data at different time points, generating an urban flood ponding dynamic evolution graph of the target area. The invention aims to improve the efficiency and accuracy of urban flood scene prediction.

The invention discloses a soil attribute intelligent prediction method and system based on a multi-dimensional dynamic scale, and relates to the technical field of soil prediction.The method comprises the steps that remote sensing images, terrain factors, meteorological factors and soil actual measurement sample data are collected and stored in a multi-source raster data bin in a unified mode; preprocessing the multi-source data, and constructing raster data in a unified format; combining terrain difference, land utilization and climate fluctuation, dynamically dividing a spatial scale, a time scale, a terrain layering scale and a land utilization pattern scale, and constructing a multi-dimensional dynamic scale feature expression vector; carrying out soil attribute modeling and training based on a multi-task deep neural network model; and applying the model to a non-actually-measured area, and outputting a prediction result and a spatial distribution map layer. The system comprises a multi-source raster data construction and management module, a multi-dimensional dynamic scale feature construction module and a soil attribute modeling and prediction module. The method can be used for forestry management and protection, resource management, ecological monitoring, resource evaluation and other scenes.

The invention relates to the technical field of non-point source pollution decision, and discloses a non-point source pollution simulation method and a non-point source pollution intelligent interaction method based on an agent, and the simulation method comprises the steps: extracting land and river attribute data of a target region, and constructing a nitrogen emission coefficient database; associating the land attribute type with a nitrogen emission coefficient library, calculating the plot emission amount through a reference area correction model nitrogen emission coefficient, converting the plot emission amount into emission amount grid data, and generating a nitrogen emission distribution map; converting the emission raster data into point data, calculating each pollution point and a river reach influence area, and constructing a river directed graph network; calculating a river reach absorption coefficient by adopting an exponential decay model, and calculating a pollution load transmission quantity and a river reach absorption quantity step by step along a river digraph network topology through breadth-first search by combining the river reach absorption coefficient based on the digraph network; the total consumption amount of the influence river reach of each pollution point is accumulated, the net discharge load distribution diagram and the consumption amount distribution diagram are generated, the calculation process is high in real-time performance, and the method is suitable for various application scenes.

The invention discloses a method and a system for generating surface soil humidity based on time-space fusion of remote sensing data, and relates to the technical field of soil humidity processing, and the method comprises the following steps: obtaining a vegetation parameter image, an active microwave remote sensing image and a passive microwave remote sensing image of a required date, and superposing the vegetation parameter image and the active microwave remote sensing image to obtain a vegetation parameter image; obtaining a high-resolution parameter image; resampling the parameter image to low resolution, and performing spatial downscaling on the passive microwave remote sensing image through the low-resolution parameter image; the method comprises the following steps: taking a time sequence of each pixel in a passive microwave remote sensing image as a signal, obtaining phase raster data and high-frequency amplitude raster data of different frequencies of the signal, carrying out spatial interpolation, and fitting a spatial interpolation result and a spatial downscaling result in a time dimension to obtain a daily scale surface soil humidity image. According to the method, fitting of the spatial resolution and the temporal resolution is achieved, the soil humidity image with the fine temporal-spatial resolution is obtained, and the method has high application value.

The invention discloses a hydraulic runoff modeling method capable of reducing dimensions and simplifying, electronic equipment and a storage medium, and belongs to the technical field of water environment engineering. In order to improve the efficiency of urban hydraulic runoff modeling, the method comprises the following steps: acquiring rainfall time sequence data, raster data of an underlying surface digital elevation model (DEM) and underlying surface land utilization data; dividing and constructing a square grid, four vertex positions of each square grid, and a plane coordinate value and an elevation value at the geometric center of each square grid; model simplification topology construction is carried out, simulation inspection well nodes, simulation pipe duct connection and simulation sub catchment area polygons are constructed based on an SWMM model, and all simulation sub catchment area polygons are associated with the simulation inspection well nodes located in the centers of the simulation sub catchment area polygons; the method comprises the following steps: calculating runoff of each grid, calculating runoff of each grid, taking rainfall time sequence data as input of an SWMM model, constructing an SWMM simulation rain gauge, constructing an SWMM rainfall flood model, and calculating to obtain real-time dynamic simulation water quantity and water depth data in each grid. The simulation calculation efficiency is improved.

The invention discloses a method for extracting a steep slope, a steep cliff and a scarp based on an elevation angle extreme value, which comprises the following steps of: 1, acquiring a digital elevation model (DEM) of a research area, and constructing a multi-direction translation model of the DEM based on matrix convolution operation; 2, calculating a three-dimensional maximum elevation angle (3D maxEA) and a three-dimensional minimum elevation angle (3D minEA) of DEM data through a translation model; 3, determining the critical terrain gradient ST of the abrupt slope, the abrupt cliff and the scarp, comparing the relative sizes of the 3D maxEA, the-3D minEA and the ST, and extracting the raster data of the abrupt slope, the abrupt cliff and the scarp; and 4, extracting vector lines of the 3D maxEA and the-3D minEA which are equal, superposing area vectors of the abrupt slope, the abrupt cliff and the scarp, and extracting vector line data of the abrupt slope, the abrupt cliff and the scarp through a cutting method. According to the abrupt slope, abrupt cliff and scarp extraction model provided by the invention, the landform abrupt change is prevented from being smoothed, and the abrupt slope, the abrupt cliff and the scarp which are easily hidden by slope smoothing are excavated.

The invention discloses a photovoltaic power station vector data set construction method based on a deep learning semantic segmentation algorithm. Firstly, open source satellite remote sensing image data and multi-source heterogeneous data are obtained and subjected to geographical calibration processing, and the consistency of the image and the multi-source data is ensured. Secondly, segmenting a research area vector by adopting a fishing net division method, taking multi-source data as priori knowledge, and obtaining a sampling area through a spatial stratified sampling algorithm; a photovoltaic power station in a sampling area is manually marked, and a semantic segmentation training set and a verification set are constructed; and establishing a photovoltaic power station segmentation model based on the DSFA-SwinNet network and the training data. And finally, carrying out segmentation processing on the residual research area by using the model, optimizing a segmentation result through filtering and morphological processing, and converting raster data into a vector format to complete construction of a photovoltaic power station vector data set.

The invention discloses a heterogeneous feature fusion-based multi-task traffic situation cognitive calculation method, and belongs to the field of intelligent traffic. The method specifically comprises the following steps: firstly, preprocessing multi-source traffic observation data to obtain normalized traffic flow, OD raster data and traffic state data; then, for a current frame, carrying out dynamic space-time convolution and diffusion diagram convolution on traffic flow and OD raster data to obtain node-level space-time features; meanwhile, carrying out feature extraction on the traffic state data by utilizing 3D convolution to obtain grid-level spatial-temporal features; and then, inputting the features and the common features into a heterogeneous cross attention fusion module to obtain a unified fusion feature h of the current frame, performing multi-task prediction, and respectively obtaining flow, OD and traffic state prediction results of future K steps. And finally, inputting the prediction result of each task and a true value to carry out multi-task loss calculation, carrying out parameter updating and completing model training. According to the invention, multi-task prediction can be realized by using multiple prediction heads.

The invention discloses a soil organic carbon density spatial non-stationary dominant factor identification and high-resolution mapping method, which uses raster data of characteristic variables as input to train an RF model, so that the trained RF model can correspondingly output and predict spatial distribution of soil organic carbon density, and high-resolution prediction of SOCD in space is realized. Meanwhile, the SHAP value spatial distribution diagrams of all the feature variables are superposed, the SHAP value of each feature variable on each pixel can be obtained, and then the dominant factor of each pixel is obtained, so that the specific influence of the feature variables of different regions on SOCD spatial distribution is obtained, and interpretive analysis of the spatial level is provided.

The invention discloses a target segmentation vector boundary simplification method, electronic equipment and a storage medium, and relates to the technical field of remote sensing. The method comprises the following steps: acquiring a remote sensing image, and performing target segmentation on the remote sensing image to obtain raster data; converting the raster data into vector data; combining adjacent surface element vectors with common boundaries to obtain a plurality of combined surface elements; extracting an intersection point of each common boundary and the boundary of the corresponding combined surface element as an anchor point; according to the anchor points, performing segmentation simplification on the boundary and the common boundary of each combined surface element to obtain a plurality of simplified broken lines of which the head ends and the tail ends are connected; and performing line cutting on the multiple simplified broken lines to obtain a boundary simplification result. According to the target segmentation vector boundary simplification method provided by the embodiment of the invention, the condition that a plurality of broken lines in an intersection relationship exist in a vector result of target segmentation is solved, and the problems that topological errors such as overlapping and gaps occur and the topological consistency is changed are avoided.

The invention discloses a DEM landform analysis method based on a large language model, and relates to the technical field of DEM data analysis, and the method comprises the steps: loading a DEM semantic mapping tool set, and achieving system initialization; after the system receives user input, questions are divided into single-tool questions, multi-tool collaborative questions and non-tool answerable questions through a question classification Agent on the basis of a pre-training model; respectively processing a single tool question, a multi-tool collaborative question and a non-tool answerable question; and performing existing memory retrieval on each problem based on a bidirectional knowledge evolution memory iteration mechanism, and performing multi-dimensional analysis in combination with the large language model to realize DEM landform analysis based on the large language model. The technical problems that in the prior art, LLM is difficult to directly and effectively analyze the spatial topological relation contained in DEM raster data, and iteration updating of geographic knowledge in a complex dynamic analysis task is difficult are solved.

The invention provides a mine land occupation instance segmentation method, and relates to the technical field of image processing, and the method comprises the steps: inputting obtained mine land occupation remote sensing image raster data into a trained mine land occupation instance segmentation model, and outputting a corresponding instance mask, the mine land occupation instance segmentation model comprises a boundary enhancement module and a boundary optimization module; the boundary enhancement module is used for processing the feature map of the raster data of the mine land occupation remote sensing image to obtain an initial mask of the remote sensing image; and the boundary optimization module is used for extracting boundary blocks along the boundary of the initial mask according to a preset sliding window, optimizing the boundary blocks by fusing different scale features through multi-resolution branches to obtain optimized boundary blocks, and splicing the optimized boundary blocks to generate an instance mask. According to the invention, the segmentation precision of the mine land occupation instance can be effectively improved.

The invention belongs to the cross technical field of disaster emergency management, remote sensing information processing and intelligent traffic scheduling, and particularly relates to a snow disaster emergency snow removal path intelligent planning method fusing a multi-target ant colony algorithm, and the method comprises the steps: S1, inputting multi-source heterogeneous data: receiving and integrating the input data, weight parameters such as operation workload, operation difficulty, passing urgency and rescue urgency are set; s2, constructing a road network line segment structure: based on the road vector data, generating basic unit road network line segments by extracting cross points and carrying out topology cutting; s3, calculating the snow removal volume and the operation time consumption: constructing a line segment buffer area according to the snow depth grid data and the road snow removal width, obtaining snow depth slices, calculating the volume, estimating the snow removal time consumption of each line segment in combination with the snow removal capacity of the rescue team, and calculating the snow removal volume and the operation time consumption by adopting the fusion of the road buffer area and the snow depth grid slices; and the accumulated snow volume and the operation time consumption of each road segment are accurately estimated.

The invention belongs to the technical field of meteorology, and particularly relates to a sand storm monitoring method based on aerosol window analysis and spectral indexes. Preprocessing the Himawari-8 data, and generating a raster data set including channel data, latitude and longitude information, elevation angle information and zenith angle information; a lookup table is constructed through a 6S radiation transmission model, and the optical thickness of the aerosol is inverted by using spectral information in the remote sensing image; scanning an image through a sliding window, and screening a sand storm feature region in combination with weighted average, standard deviation and abnormal judgment of aerosol concentration; establishing a sand-dust spectral index by using the sand-dust sensitive wave band; and combining the aerosol concentration with the sand and dust spectral index to extract sand and dust storm information. According to the method, the aerosol concentration and the sand dust spectral index are combined, sand dust storm information can be effectively identified and extracted, and important data support is provided for environmental protection and disaster early warning; the diversity of the sand storm detection method is increased, and the obtained result is accurate and reliable.

The invention discloses a co-seismic landslide susceptibility evaluation method and system based on heterogeneous ensemble learning, relates to the field of co-seismic landslide susceptibility evaluation, and solves the problem in the prior art that a deep learning model is difficult to capture the internal relationship between landslide occurrence and disaster-inducing factors through single-machine learning under complex and hard mountain conditions. According to the method, landslide boundary vector data and disaster-inducing factor raster data are integrated, and invalid disaster-inducing factors are removed according to correlation analysis and importance analysis to obtain a training data set; constructing a CNN-RXStack landslide susceptibility evaluation model to predict co-seismic landslide susceptibility of an earthquake area, converting the predicted landslide occurrence probability into a landslide susceptibility grade to obtain a landslide susceptibility map, and verifying and analyzing a landslide susceptibility result; according to the method, the heterogeneity of the disaster-inducing factor characteristics is improved, and the model can better capture the characteristic difference between the landslide and the non-landslide, so that the accuracy and the reliability of model prediction are improved.

The invention discloses an unmanned aerial vehicle image ground object recognition method based on an improved DeepLabV3 + network, and the method comprises the steps: obtaining a visible light sequence image of an unmanned aerial vehicle, carrying out the preprocessing and dividing, carrying out the data enhancement and pixel-level marking of a training set, and obtaining a marking data set; inputting the annotation data set into the improved DeepLabV3 + network for training, and monitoring a training process and performance indexes by using the verification set; an ASPP module of the improved DeepLabV3 + network adopts an initial voidage combination, and the voidage of the ASPP module is dynamically adjusted based on a verification set performance monitoring index in a training process to obtain a trained semantic segmentation model; inputting the test set image or the to-be-identified unmanned aerial vehicle image into a semantic segmentation model to obtain an initial segmentation result; the method comprises the following steps: firstly, carrying out semantic segmentation on a model, then, carrying out automatic optimization on hyper-parameters of the model based on a Bayesian optimization framework, carrying out retraining or prediction on the model by utilizing the optimized hyper-parameters to obtain an optimized semantic segmentation result, and then, converting a picture format into a raster data format with geographic information to obtain a ground feature recognition graph.

The invention discloses a method and system for quantifying the influence of human activities on surface vegetation, and belongs to the technical field of environmental monitoring, and the method comprises the steps: firstly processing obtained remote sensing vegetation index data to obtain NDVI raster data, and then obtaining KNDVI raster data according to the NDVI raster data; then, a buffer area is constructed for the research area according to the radiation range to obtain an affected area, on this basis, the buffer area continues to be constructed, the original area is removed to obtain a contrast area, raster data of the area except natural vegetation is removed for the contrast area, original KNDVI raster data is used for the affected area, and the original KNDVI raster data is used for the contrast area; and the influence of human activities on the ground vegetation of the research area is quantified. A novel remote sensing vegetation index is adopted, and the defects that a traditional observation period is long, a model is difficult to simulate, and a remote sensing calculation method is large in processing workload and has no universality are overcome.

The invention discloses a Spark-based geographic big data processing method, and relates to the technical field of data processing, and the method comprises the steps: obtaining a preset configuration file, and carrying out the configuration analysis of the configuration file, and obtaining an input parameter; obtaining geographical contour information according to the input parameters, analyzing the geographical contour information into a polygon object, and generating polygon object raster data according to the polygon object in a preset raster system; determining a data structure according to the data volume of the polygon object raster data, and converting the polygon object raster data into polygon data of the data structure; determining a storage mode according to the data volume of the polygon data and a preset memory resource of the computing node, and storing the polygon data according to the storage mode; and operating a data processing command defined in the input parameters to process the polygon data to obtain output data, and outputting the output data. The geographic big data calculation efficiency is improved.