33 results about "Land cover land use" patented technology

The invention relates to the field of high-resolution remote sensing image land classification, in particular to an object-oriented unmanned aerial vehicle high-resolution image coal fire region landcover classification method. The method comprises the main steps that on the basis of coal fire region high-resolution color images collected by an unmanned aerial vehicle, homogenous objects are divided through an image multi-resolution segmentation algorithm to construct an expert knowledge base adopting spectral signatures, shape features, textural features and height information as main parts,and high-precision classification of a coal fire region is achieved by means of a rule set and the most adjacent classification method, wherein the classification precision is much higher than that of a traditional pixel-based classification method. The requirement for complex coal ire region land cover precise classification is met.

The invention provides a method and a medium for extracting green degree space of an urban agglomeration. The method comprises the following steps: acquiring high-score remote sensing data of urban agglomeration and extracting land cover data of urban agglomeration; Obtaining the social perception data of urban agglomeration, and extracting the social function attribute data of urban agglomerationland; According to the cover data and social function attribute data, the land of urban agglomeration is classified, and the information about the land and greenness of urban agglomeration is obtained. According to the classification information, the greenness of urban agglomeration is extracted. The present invention solves the problem that only a single remote sensing data can obtain the land cover information in the related art, and the social function attribute of the urban green space cannot be obtained, and realizes the statistics and research of the urban agglomeration green space containing the social function attribute.

The invention relates to the technical field of land coverage classification, and in particular to a land coverage classification system and method. The method comprises the following steps of: acquiring interest point information of an open source map by utilizing a search keyword so as to obtain an interest point data set; adding a corresponding land coverage class for each piece of interest point information in the interest point data set according to a preset land coverage class; converting the interest point information into geographic space vector data; segmenting an original remote sensing image to obtain a plurality of unclassified vector polygons; overlapping the vector polygons and the geographic space vector data through address matching, endowing the land coverage classes in the geographic space vector data corresponding to the vector polygons with the vector polygons so as to complete the land coverage classification of the vector polygons. According to the system and method provided by the invention, the land coverage samples can be rapidly and accurately acquired in a large scale, so that the land coverage classification efficiency is improved.

The invention discloses a recurrent neural network-based remote sensing monitoring method and a device for ground surface abnormal phenomenons. The method comprises the steps: inputting a training front-segment sequence into an encoder constructed based on a bidirectional recurrent neural network, and obtaining a front-segment hidden state sequence for representing the global information of the training front-segment sequence; sequentially inputting the numerical value of the last moment into a decoder based on a single-layer recurrent neural network, so as to sequentially obtain predicted values corresponding to the moments in the training back-end sequence; adopting an encoder-decoder model to predict an observation location at a future moment. By the adoption of the above scheme, an overall change rule of a historical sequence can be mined and observation values at multiple moments in the future can be predicted; near-real-time anomaly monitoring and real-time monitoring and supervising are achieved, meanwhile, required data preprocessing and manually set experience parameters are few, the requirements for experience and professional backgrounds of implementers are not high, themethod is suitable for various geographic areas and land cover types, and the method is high in feasibility, robustness and prediction result accuracy.

The invention discloses a method and apparatus for extracting land cover information in a remote sensing image. With the method and apparatus, a problem of inaccurate land cover information identification in the remote sensing image in the prior art can be solved. The method comprises: obtaining a remote sensing image whose land cover information is going to be extracted; on the basis of a thematic layer, carrying out segmentation on the remote sensing image by using a multi-resolution image segmentation method and a nearest neighboring classification method and extracting information of a river, a lake, an intertidal zone, a low-light-reflection building, a high-light-reflection building, middle-low cover vegetation and / or high cover vegetation.

The invention discloses a land cover mapping method, which comprises the following steps of: extracting effective information of an existing land cover product by establishing a multi-source data space consistency recognition and sample statistical screening method, and establishing a multi-source data space consistency recognition and sample statistical screening method on the basis of a remote sensing cloud calculation platform and mass remote sensing images; a machine learning algorithm is applied to correct inconsistent information among various land cover products, so that the problems of poor consistency degree and low precision of current multi-source land cover data are solved; by establishing a method for quickly extracting a large-range reliable training sample, the difficulty of insufficient training samples of cloud calculation platforms such as a Google Earth engine and the like can be effectively solved; according to the technical scheme, on-site land cover interpretation reference sample points and prior knowledge are not needed, a land cover data fusion product with high spatial resolution in a large-range region can be quickly and accurately generated, high-precision and high-efficiency land cover mapping can be effectively realized, and the method has portability.

The invention relates to a coastline dividing method and system based on a land cover utilization principle. The method comprises the steps of dividing a sandy coast formed by wave motion, dividing a powdered sand and silt coast containing silt, dividing a bedrock coast which is of a rock structure, dividing an estuary coast which is distributed in a bending manner and / or dividing an artificial coast including a wharf or a revetment or a dike or a bulwark or a salt pan, conducting connection treatment on the divided coasts, and forming coastlines. According to the coastline dividing method and system based on the land cover utilization principle, the problems that coasts have various types and purposes and are high in natural variability and artificial categorization arbitrariness, and it is difficult to accurately survey and mark out the coastlines on different types of coasts are effectively solved; dividing bases based on the coastal-zone land cover utilization principle are given out for each coast type, and the advantages of being unified in dividing standard, reliable in dividing basis and reasonable in dividing scheme are achieved. The coastline dividing method and system are simple in principle, high in implementation and capable of being widely applied to the field of coastal-zone ecological environment assessment and the like.

The invention discloses a land cover data fusion correction method. According to the method, uncertain information filtering and fine spatial resolution information extraction from a coarse spatial resolution multi-source data consistent area to a fine spatial resolution multi-source data consistent area is achieved through a generated super-pixel object in combination with a principal component analysis method; and through a method of obtaining a reliable training sample through a local sampling technology, a fine spatial resolution multi-source data inconsistent region is corrected, and fusion correction of land cover data is achieved. Compared with the prior art that only fusion correction of coarse spatial resolution multi-source data can be achieved, the technical scheme of the invention achieves fine improvement from coarse resolution to fine resolution, has portability, solves the difficulty of simultaneously improving resolution and precision in the fusion correction process of a multi-source data product, and effectively solves he problem of improving the operation efficiency of mass data.

The invention discloses an automatic sliding-window-based ecological crisscross belt identification method, which comprises the following steps: 1, making grid data about the area percentages of different land cover types under a unit grid in a certain scale by virtue of land cover type space distribution data; 2, recalculating the grid data of each type of land cover by virtue of a sliding window, assigning a new attribute value of each unit grid in the grid data as an average value of attribute values of all the unit grids in a certain range around the unit grid, and assigning an average value of attribute values of effective unit grids to a boundary region; 3, setting an area proportion threshold value of related land cover types in different types of ecological crisscross belts; 4, performing calculation and judgment on the area proportion data of each type of land cover according to the set area proportion threshold value, and generating an ecological crisscross belt type and a spatial range. According to the method, the type and spatial distribution range of the ecological crisscross belt can be effectively obtained, so that reference is provided for the deep development of ecological crisscross belt researches and ecological environment protection.

The invention belongs to water conservation function influence evaluation, and particularly relates to a method for evaluating the influence of polar ecological retreat on a water conservation function by integrating an InVEST model and a multivariate statistical method. The method comprises the following steps: 1), performing type information division on a wetland remote sensing image, and detecting and analyzing a quantitative relationship between wetland type changes and water conservation; 2), carrying out hydrological and geomorphic classification on the wetland according to the wetland coating remote sensing data; 3), establishing an InVEST model; 4), calculating the water conservation amount before and after the spatial pattern change of a regional landscape to be evaluated by usingthe InVEST model; and 5), according to the calculation result, obtaining the degree of the shadow of the landscape space pattern change on the water conservation function by utilizing the principal component and a clustering analysis method. According to the method, an InVEST model is adopted, and the yield is evaluated by comparing the water yields before and after each change coating project isimplemented. In addition, the land cover change characteristics of the wetland are determined, and the influence of the water conservation function is evaluated by considering the land cover, the climate change and the landscape pattern. Accurate and scientific evaluation is carried out on the influence on the water conservation function in the polar ecological retreat project, valuable referenceis provided for the polar ecological retreatproject, and the result provides guidance for regional ecological restoration and ecosystem function evaluation of the government.

The invention discloses a global land utilization and land cover sequence spatialization method integrating GCAM and CA. The GCAM model and the geographic CA are combined, global 283 subareas are achieved, global land use sequence spatial data products from 2010 years to 2100 years are achieved, the spatial resolution is 1 km, and the time resolution is 5 years. By integrating the initial probability of the neural network, the neighborhood influence, the ground class transformation weight matrix of each sub-region and a planning constraint condition, the comprehensive transformation probability of the ground class of each sub-region is obtained, and the land utilization and the land coverage of each sub-region is spatialized. The method is mainly applied to the global land utilization change and coverage modeling under different development scenes in the future, has the capability of producing high-resolution global LUCC products, and promotes the research on the construction of a high-resolution land utilization model under the global scale.

The invention relates to the technical field of urban planning, in particular to an urban ecological vulnerability spatial prediction method based on GIS and CA simulation, which comprises the following steps: creating neighborhood analysis of current land cover data; creating an orthogonally distributed grid; loading ecological vulnerability space evaluation data; identifying a neighborhood ecological vulnerability space of the land for future development; dividing ecological vulnerability space levels of neighborhoods of land for future development; extracting the highest threshold data of the ecological vulnerability space of the neighborhood of the land for the future development; and building an urban ecological vulnerability spatial distribution prediction initial model based on themodel builder, deleting initial data, and completing the prediction model tool for predicting urban ecological vulnerability spatial distribution. According to the method, the defect of empirical judgment in traditional planning is avoided, the urban ecological vulnerability spatial distribution is quantitatively predicted, and the neighborhood becomes logic of the vulnerability spatial probability in the process of predicting the urban ecological vulnerability spatial distribution.

The invention belongs to the field of agricultural remote sensing, and relates to a remote sensing identification method for land cover and planting structure of a fragmented agricultural area, which adopts a machine learning algorithm to extract land utilization, does not depend on actually measured sample points, and obtains high-precision agricultural area land utilization rasterized data after automatic training. According to the method, an agricultural area and a non-agricultural area are classified through a supervision algorithm support vector machine (SVM), and then crop types are further classified and discriminated by adopting a decision tree algorithm based on obtained agricultural area classification raster data. The method does not depend on actual measurement data, classification time and economic cost are greatly reduced, and negative effects of the quality of a sample training set of a traditional machine algorithm on a classification result are effectively avoided. And meanwhile, the land utilization mapping precision required by agricultural remote sensing application is met on the premise of ensuring high classification efficiency.

The invention provides a storm surge simulation method based on WRF land cover data replacement. The storm surge simulation method comprises the following steps: S1, using ArcGIS for processing land cover data to be replaced; s2, processing land cover data in a preprocessing module in the WRF mode; s3, operating the WRF to obtain simulated wind field data; s4, establishing an SWAN wave model and an ADCIRC hydrodynamic model; s5, respectively providing the simulated wind field data after the replacement of the land cover data for the SWAN wave model and the ADCIRC hydrodynamic model; s6, inputting the wave radiation stress obtained by simulation of the SWAN model into an ADCIRC model; and S7, performing storm surge simulation by using the ADCIRC model. According to the invention, the WRF-SWAN-ADCIRC coupling model is comprehensively used, the method can be used for simulating the storm surge in the bay area of the sea mouth, outdated built-in land cover data in the WRF are replaced, andthe accuracy of storm surge simulation is improved.

The invention relates to a flood disaster loss assessment method and a simulation deduction system, and belongs to the technical field of assessment analysis simulation, and the method comprises the steps: constructing and training a flood deduction neural network model; hydrological characteristic information is collected and monitored in real time, and the precipitation of a prediction area is calculated; flood deduction is carried out; setting loss evaluation parameters; and carrying out loss evaluation calculation. According to the invention, satellite remote sensing and a sensor are comprehensively utilized to monitor data, so that the problems of information integration complexity and insufficient real-time performance of collected information are solved; a neural network deduction model is established, the flood deduction capability is continuously optimized through deep self-learning, network parameters are adjusted and optimized according to land cover data, and the loss evaluation accuracy is improved; and a Markov chain method is adopted in precipitation prediction, so that the redundant data volume is relieved, and the efficiency of deduction calculation is improved.

The invention discloses a land cover mapping method, which extracts effective information of existing land cover products by establishing a multi-source data spatial consistency identification and sample statistical screening method, and applies the method based on a remote sensing cloud computing platform and massive remote sensing images The machine learning algorithm corrects the inconsistent information among various land cover products, which solves the problems of poor consistency and low precision of the current multi-source land cover data; It can effectively solve the difficulty of insufficient training samples on cloud computing platforms such as Google Earth Engine; this technical solution does not require reference sample points and prior knowledge for field land cover interpretation, and can quickly and accurately generate high-resolution images of large-scale areas. Land cover data fusion products can effectively realize high-precision and high-efficiency land cover mapping and are portable.

The invention discloses a method and device for extracting land cover information in remote sensing images, which can solve the problem of inaccurate identification of land cover information in remote sensing images in the prior art. The method includes: acquiring remote sensing images for which land cover information is to be extracted; segmenting the remote sensing images based on a thematic layer by using a multi-resolution image segmentation method and a nearest neighbor classification method, and from the remote sensing image objects formed after segmentation Extract rivers, lakes, tidal flats, low albedo buildings, high albedo buildings, low to medium cover vegetation, and / or high cover vegetation.

The invention provides an earth surface cover change intensity evaluation method and system for a polar region. The method comprises the following steps: acquiring remote sensing image data of a to-be-identified polar region; according to the remote sensing data, adopting a random forest method for classification to obtain bare rock and water body information of the polar region; according to the remote sensing data, extracting the glacier outer edge in the polar region to be identified by adopting a visual interpretation method, and identifying the glacier type by adopting a threshold value method; spatial overlay analysis is carried out on glacier areas, rock mass areas and water body areas obtained according to remote sensing image data obtained in different periods, change areas are obtained, and the earth surface cover change intensity is obtained through calculation. According to the method, the remote sensing image is used as a data source to obtain the land cover change condition of a key area, the land cover of the Antarctic coastal zone area is classified, and the land cover change intensity analysis of the polar coastal zone area is realized by combining and applying a space analysis method and a statistical analysis method.

The invention provides a method and medium for extracting urban agglomeration green space. Among them, the method includes: obtaining high-resolution remote sensing data of urban agglomerations, and extracting land cover data of urban agglomerations; obtaining social perception data of urban agglomerations, and extracting social function attribute data of urban agglomeration land; Functional attribute data, classify the land of the urban agglomeration, and obtain the classification information related to the greenness of the urban agglomeration land; according to the classification information, extract the greenness of the urban agglomeration. The present invention solves the problem in the related art that only relying on a single remote sensing data can only obtain land cover information, but cannot obtain the social function attributes of urban green space, and realizes the statistics of green space of urban agglomerations including social function attributes and research.

The invention discloses a remote sensing image super-resolution land cover mapping method based on a street view image. Firstly, the remote sensing image is preprocessed, and the score image is obtained by soft classification method; at the same time, the convolutional neural network model (Convolutional Neural Network, CNN for short) and spatial positioning technology are used to extract the feature information of the street view image. Secondly, according to the set magnification factor, the ground objects in the street view image are located, and the area ratio of the sub-pixels they occupy is calculated. Then, considering the category and location attributes of the ground objects in the fractional image and the street view image, the spatial dependence index of the sub-pixel is calculated. Finally, according to the spatial dependence index of the sub-pixel and the area ratio of the sub-pixel, the class membership degree of the sub-pixel is obtained, and then the super-resolution land cover mapping result is obtained. The invention constructs a theoretical model of a remote sensing image super-resolution land cover mapping method based on a street view image, which can fuse high-resolution image surface feature information, and the mapping result has high precision.

The invention relates to a characterization method of land utilization / land cover (LULC) change, comprising following steps of: taking a static NDVI as a core, directly utilizing the NDVI as a quantitative calculation index, or deriving a new quantitative calculation index based on the NDVI, and using the NDVI as a core to characterize the land utilization / land cover (LULC) change. According to the method, the change of land utilization / land cover (LULC) can be represented, and inspiration can be provided for many applications related to land utilization such as land utilization planning, land utilization evaluation and land utilization suitability evaluation.

The present invention relates to the field of land classification of high-resolution remote sensing images, and in particular to an object-oriented method for classifying land cover of coal-fired areas in high-resolution images of drones. High-resolution color images are divided into homogeneous objects through image multi-scale segmentation algorithms, and an expert knowledge base based on spectral features, shape features, texture features and height information is constructed. Accuracy classification, the classification accuracy is much higher than the traditional pixel-based classification method, and the invention meets the demand for accurate classification of land cover in complex coal fire areas.

The invention discloses a method for determining land use categories integrated into street view images. Firstly, the fine object category information in the street view image sampling points is extracted by means of deep learning convolutional neural network; at the same time, the remote sensing image is preprocessed, and the land cover map is obtained by using the supervised classification method. Secondly, through the spectrum, texture, shape and geographical distribution information of the pixel where the street view sampling point is located, the category of the adjacent pixel is deduced. Finally, the pixel classification information is fused with the land cover map to obtain fine multi-category land use results. The invention starts from remote sensing based pixel classification, combines fine street scene information, and has high classification result accuracy.

The invention provides a sampling point distribution method and device based on landscape heterogeneity and electronic equipment, and the method comprises the steps of collecting a satellite remote sensing image of a target region, and carrying out the clustering processing, and obtaining a land cover clustering result map of the target region; based on landscape heterogeneity of a target area, dividing the land cover clustering result map into a plurality of grids according to a first preset rule, the plurality of grids being congruent square grids; based on the landscape heterogeneity of the corresponding area in each grid, determining sampling points distributed in each grid according to a second preset rule, wherein the landscape heterogeneity is described by different indexes in the first preset rule and the second preset rule. According to the method, the device and the electronic equipment provided by the invention, the distribution of the sampling points is controlled, the sampling cost is reduced, and the sampling efficiency is improved.

The invention provides a cross-checking method and a cross-checking device for land coverage remote sensing products. The cross-checking method and the cross-checking device relate to the technical field of land coverage remote sensing. The cross-checking method for the land coverage remote sensing products defines two error indexes of a mean square root error and a correlation coefficient based on covariance, and obtains an error of each type of each variety of land coverage remote sensing products relative to real land coverage distribution through comparing three varieties of land coverageremote sensing products obtained independently. The cross-checking method for the land coverage remote sensing products improves the efficiency of checking the land coverage remote sensing products, reduces the checking cost, and can realize full coverage checking of the land coverage remote sensing products.

The invention discloses a method for identifying planting area of ​​overwintering crops based on multi-source full-time satellite images under a cloud computing platform. First, the Landsat 7 / 8 and Sentinel‑2A / B satellite images are fused and preprocessed to calculate NDVI, LSWI, EVI and mNDWI classify the land cover to obtain the cultivated land mask, use the 10-day image synthesis method, linear interpolation and Savitzky‑Golay filter to reconstruct the time series; secondly, iterate on the reconstructed time series data set Algorithm, identify the peaks and troughs in the time series curve, use the threshold method to determine the start date and end date of the growing season of overwintering crops, calculate the growth rate of the maximum growth period and the span of the complete growth cycle; finally, according to the phenology of overwintering crops in different phenological periods Indicator, set the threshold to extract the acreage of overwintering crops on the pixel scale. The invention realizes precise remote sensing and automatic identification of planting area of ​​overwintering crops.