221results about How to "Improve inversion accuracy" patented technology

The invention relates to the technical field of satellite positioning navigation enhancement, in particular to a global ionosphere inversion method based on BP neural network fusion multi-source data, and the method comprises the steps: inverting STEC through ground-based GNSS observation data, then employing a projection function to project the STEC to a vertical direction of an ionosphere puncture point, and obtaining VTEC; a BP neural network and a Kriging interpolation method are utilized to fuse foundation observation data into an IRI-2016 ionosphere model to solve the problem that GNSS tracking stations are distributed unevenly in the world, so that the inversion precision of the ionosphere is improved, a spherical harmonic function is adopted to express VTEC space distribution, and then the ocean altimetry satellite observation data is utilized to achieve the purpose of solving the problem that the GNSS tracking stations are distributed unevenly in the world. According to the method, the problem of deficiency of ground-based GNSS observation data in an ocean area is solved by using ground-based GNSS observation data, DORIS observation data and COSMIC occultation data, a global ionosphere observation equation is established, a normal equation is constructed, and finally global ionosphere grid data VTEC is generated through Helmert variance estimation precise weighting calculation, so that the global ionosphere observation precision is improved.

The invention relates to a method for improving satellite data inversion soil salinity by using hyperspectral data of an unmanned aerial vehicle. The method comprises the following steps: collecting a soil sample in a test area, and calculating soil salt content; acquiring and preprocessing satellite remote sensing image data of a test area; storing data, outputting a single-band satellite image, and calculating a spectral index; obtaining and preprocessing hyperspectral image data of the unmanned aerial vehicle in the test area; reconstructing and upscaling unmanned aerial vehicle image data pixels to a spatial resolution of 10m; reconstructing the narrow-band spectral data of the unmanned aerial vehicle into broadband spectral data consistent with a satellite image, storing and outputting a single band, and calculating a spectral index; screening a soil salt content characteristic variable, and constructing a soil salt content inversion model by taking the screened characteristic variable as an independent variable and the actually measured soil salt content as a dependent variable; and fitting a function relationship between the upscaling unmanned aerial vehicle soil salt content inversion model and the satellite inversion model by adopting a numerical regression function method to realize correction of the satellite inversion model.

The invention discloses a rock-fill dam material parameter uncertainty inversion method based on Bayesian inference, and the method specifically comprises the steps: building a rock-fill dam finite element model, and determining the prior distribution of to-be-inverted parameters; constructing an sPCE proxy model based on prior distribution of inversion parameters; a rock-fill dam displacement statistical model is established, and actually-measured displacement water pressure components are separated; constructing a rock-fill dam material likelihood function; carrying out uncertainty inversion on the rock-fill dam material parameters; and updating and inverting the uncertainty of the rock-fill dam material parameters. According to the method, a Bayesian inference method is utilized, and uncertainty distribution characteristics of different partition material parameters of a dam body are obtained through actual measurement displacement data, parameter prior distribution and a likelihood function of the dam. According to the method, the defect that parameter uncertainty cannot be considered in a traditional inversion method is overcome, and obtained parameter values are closer to engineering practice; compared with a traditional inversion method, parameter updating inversion can be carried out through monitoring data in different periods, and continuous tracking of parameter timeliness evolution is achieved.

The invention discloses a soil organic matter remote sensing inversion method based on a stacking model. The method comprises the following steps: step 1, data acquisition and preprocessing; 2, optimizing spectral characteristic variables; 3, constructing a soil organic matter inversion model; 4, performing model combination optimization based on a genetic algorithm; 5, performing remote sensing inversion on the soil organic matters of the stacking model based on the optimal model combination; according to the method, spectral features and topographic features are introduced as covariables of soil organic matter remote sensing inversion, a plurality of machine learning models are stacked and integrated, a genetic algorithm is adopted to optimize a basic learner combination of the stacked and integrated learning models, and a soil organic matter remote sensing inversion model of a stacked model based on an optimal model combination is constructed. The problems that a single machine learning model is prone to overfitting and weak in generalization ability are solved, the generalization ability of the model is enhanced, and the inversion precision of the soil organic matter is improved.

The invention discloses a spaceborne hyperspectral detection laser radar atmospheric particle inversion method. The method comprises the following steps of solving a laser ranging value and an altitude corresponding to a spaceborne hyperspectral resolution laser radar (HSRL) echo signal sampling point; reducing the noise of the echo signal by using a noise removal method; carrying out distance and system constant correction on the de-noised signal; a forecasting model is used for reanalyzing the data set, and S6, a molecular scattering signal transmittance lookup table is established through an atmosphere model and an iodine molecule hyperspectral filter absorption spectrum; calculating back scattering coefficients and scattering ratios of the cloud and the aerosol; performing hierarchical identification by using a scattering ratio and voltage signal-to-noise ratio threshold method; determining an effective inversion region through hierarchical region processing; calculating an initial extinction coefficient to a radar ratio; and performing Gaussian smoothing filtering on the radar ratio to obtain a final radar ratio, and calculating a final extinction coefficient profile. The method is characterized in that cloud and aerosol level identification is carried out on spaceborne HSRL echo signals, a molecular transmittance lookup table of a hyperspectral channel is established, and cloud and aerosol optical parameters are accurately and efficiently solved.

The invention discloses a PPI wind shear extraction method and device based on a laser wind finding radar and a storage medium, and the method comprises the following steps: S1, obtaining the radial data of N light beams in a period based on PPI scanning, and carrying out the screening and marking of the radial data; s2, inverting a horizontal wind field based on a light beam sliding window to obtain a horizontal wind speed and a horizontal wind direction of the light beam; s3, grid division is carried out on the horizontal wind field, and interpolation is carried out based on the detection distance; compared with the prior art, the method takes radial data obtained by PPI scanning as a basis, adopts a sliding window method, and takes the horizontal wind speed and the horizontal wind direction calculated by a previous sliding window as initial values of model parameters to be substituted into the gradient descent solving model of the current light beam, so that the horizontal wind speed and the horizontal wind direction of the current light beam are inverted; the wind field inversion error when the wind direction and the light beam direction tend to be perpendicular is reduced, and the inversion precision of the horizontal wind field is improved.

The invention provides a multi-parameter GNSS-R soil humidity inversion method based on a random forest model. The multi-parameter GNSS-R soil humidity inversion method comprises the following steps: satellite-borne GNSS-R data quality control; satellite-borne GNSS-R data observation quantity extraction and surface reflectance calculation are carried out; eliminating satellite-borne GNSS-R observation data influenced by the open water area; carrying out surface reflectance gridding; preprocessing the SMAP data; constructing an inversion model; and carrying out spaceborne GNSS-R soil humidity inversion and precision evaluation. The method has the beneficial effects that by utilizing the performance advantages of processing a non-linear relationship, good generalization performance, strong robustness and the like of the random forest, after observation data influenced by a water body is accurately eliminated, an inversion model is constructed by combining multi-parameter auxiliary data of a soil humidity active and passive remote sensing satellite; the space-time coverage, the inversion precision and the reliability of soil humidity data are remarkably improved, and powerful data support is provided for the fields of climate forecast, agricultural production, hydrological modeling and the like.

The invention relates to a deep tunnel rockburst prediction method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining the engineering geological conditions of a tunnel site area, carrying out a field ground stress test, and carrying out the ground stress inversion through a multiple regression method; establishing and optimizing a tunnel numerical model in combination with the actual condition of the tunnel; based on field monitoring data, surrounding rock test data and a numerical model, a surrounding rock parameter inversion analysis data set is established, a BP neural network is optimized by adopting a particle swarm optimization algorithm to invert surrounding rock parameters, and the surrounding rock parameter inversion precision and efficiency are improved; numerical simulation and surrounding rock parameter inversion are carried out on the current construction section of the tunnel, the potential rockburst risk of the current construction section is analyzed according to the rockburst criterion, and the rockburst grade and the occurrence probability of the current construction section are judged. The method is closely combined with the actual situation of the tunnel site construction section, and the rock burst prediction accuracy is improved; on the premise that the prediction accuracy is guaranteed, a large amount of engineering cost is saved, and reference is provided for rockburst prediction and site construction.

The invention discloses a forest grass vegetation coverage refined inversion method and system based on multi-scale remote sensing, and the method comprises the steps: firstly providing end member types and end member abundance information through the field observation data of an unmanned plane, inputting the prior knowledge into a linear mixed pixel decomposition model, and carrying out the back calculation to obtain end member spectrum information of the corresponding end member types; by taking unmanned aerial vehicle over-resolution observation data as priori knowledge, the end member spectrum extraction precision is improved, and a technical support is provided for improving the inversion precision of the total coverage of forest and grass vegetation; the textural features of high-resolution satellite observation data are fully utilized, the spectral features of medium resolution and the textural features of high resolution are combined, and the optimal response relation between different remote sensing feature combinations and woody vegetation coverage measured data is constructed by utilizing a random forest model. Therefore, high-precision inversion of the woody vegetation coverage in a large-area range is realized, and data support is provided for further estimation of the herbaceous vegetation coverage.

The invention discloses a multi-model PolInSAR forest height inversion method fused with a deep learning normal form, and the method comprises the steps: building a back scattering model similar to a quadratic function based on an RVoG inversion frame, and defining a volume decoherence expression of back scattering; defining respective backscattering contours by using different parameters; in consideration of model heterogeneity, a plurality of backscattering models are obtained through a linear combination quadratic function model and a GVB model; different models are classified according to application ranges, a model pool is constructed, a CNN classifier is designed to complete classification work, a plurality of models are solved through a nonlinear iterative algorithm, CHM data of LIDAR is used for distributing model labels for pixels, and a data set is obtained; according to the method, a CNN (Convolutional Neural Network) is trained, a data set is balanced through random downsampling, proper hyper-parameters are set, after training is completed, data of a research area is input into the network for classification, an optimal backscattering model is solved by using a nonlinear iterative algorithm according to a classification result, and finally a forest height inversion result is obtained.

The invention discloses an in-situ hyperspectral water quality parameter inversion method based on deep learning. According to the method, a parallel network model is constructed by combining a convolutional neural network (CNN) and a Transformer, local features are extracted by using the CNN, global information is captured by using the Transformer, and comprehensive analysis of complex features and waveband information in in-situ hyperspectral reflectivity data is realized. In the model training process, the initial layer of the network is pre-trained by adopting a self-supervised learning strategy of random shielding and noise injection, and then the weight of the network is finely adjusted by utilizing labeled data, so that the understanding of the model on a data feature mode is enhanced. According to the method, the unlabeled data is fully utilized, and the training precision and generalization ability of the model are improved. Compared with a traditional explicit linear or nonlinear model constructed by depending on prior knowledge, the method has the advantages that the recessive correlation between spectrums can be captured, end-to-end water quality parameter inversion is realized, and the result is remarkably improved in the aspects of precision and generalization ability.

The invention discloses an aerosol parameter inversion method and device based on active and passive cooperative detection data, and relates to the technical field of aerosol measurement. According to the method, aiming at aerosol information of active and passive load cooperative detection, starting from characteristics of different loads, and based on a constraint relation of cooperative detection data on radiation / polarization / space characteristics, a multi-parameter inversion method of aerosol optical thickness, extinction coefficient profile and the like based on active and passive load cooperative detection data is provided. According to the method, the total information amount of aerosol parameter inversion is improved, so that the inversion precision is optimized.

The invention relates to the technical field of general image data processing or generation, in particular to a collaborative detection method for ore concentration area transparency, which is executed by a processor and comprises the following steps: acquiring a gravity data set, a magnetic method data set and an electrical method data set; executing structural consistency joint inversion on the gravity data set, the magnetic method data set and the electrical method data set under the constraint of physical constraint information to form an initial model; forming an optimization model; executing hybrid joint inversion on the optimization model under the constraint of physical constraint information to obtain wave velocity distribution data, resistivity distribution data, density distribution data and magnetic susceptibility distribution data; constructing a three-dimensional geology-geophysical model and performing forward modeling calculation on the model; and an optimal three-dimensional geology-geophysics model is obtained. According to the method and the device, the problems that in the process of realizing the transparency of the ore concentration area, the existing joint inversion cannot meet the requirement of realizing the transparency of the ore concentration area on underground high-precision physical property distribution and the modeling precision is low are solved.

The invention relates to the technical field of meteorology and hydrology, and discloses a short and temporary rainfall forecasting method and system based on a raindrop spectrometer and a meteorological radar, and the method comprises the steps: carrying out the preprocessing of radar echo image data and raindrop spectrometer data; quantitatively estimating a radar reflectivity factor according to the raindrop spectrometer data; extrapolating radar echo image data based on an optical flow method and a UNet network model, and extrapolating a radar reflectivity factor based on an LSTM network model; performing dynamic regression on the extrapolated radar echo image data to obtain the radar echo image data of the required prediction period; carrying out global correction on the radar echo image data of the required predicted period according to the radar reflectivity factor of the required predicted period obtained by extrapolation; and forecasting rainfall according to the radar echo image data after global correction. According to the method, radar echo image data can be closer to near-surface rainfall information, and the accuracy of rainfall forecast and the adaptability to complex and changeable meteorological data are improved.

The invention relates to a transient electromagnetic inversion model construction method based on target data set driving, and the method comprises the steps: constructing an initial training set, training a first convolutional neural network through the initial training set, and obtaining an initial transient electromagnetic inversion network model; inputting to-be-measured electromagnetic response data into the initial transient electromagnetic inversion model for prediction to obtain a resistivity value; performing forward modeling on the resistivity value to obtain electromagnetic response data of the forward modeling, and constructing prediction data based on the electromagnetic response data of the forward modeling and the resistivity value; extracting data similar to the to-be-tested electromagnetic response data in the initial training set, and constructing a target data set based on the similar data and the prediction data; and parameters in the initial transient electromagnetic inversion model are migrated to the second convolutional neural network to carry out iterative learning on the target data set pair, and the transient electromagnetic inversion model is obtained. The invention aims to improve the inversion precision of the to-be-tested data by the network when deep learning inversion is carried out on the to-be-tested data through the iterative training strategy.

The invention discloses a soil water content estimation method fusing multi-source time sequence remote sensing data, and relates to the field of optical detection, and the method comprises the following steps: 1, obtaining an SAR remote sensing image of Sentinel-1 and an optical remote sensing image of Sentinel-2 in a designated area, carrying out the image fusion through a PCA algorithm, carrying out the feature set extraction of the fused images, and carrying out the feature set extraction of the feature set; a minimum analysis unit is generated through image segmentation, geometric shape features of the minimum analysis unit are extracted, a crop planting area recognition model is constructed through a convolutional neural network algorithm, and a crop coverage area is obtained; 2, introducing an MIMICS microwave scattering model, and constructing a crop multi-growth-period microwave scattering model; the SAR image and the optical image are fused through the PCA algorithm, the crop planting area recognition model based on multi-source remote sensing image fusion and the convolutional neural network algorithm is constructed, and the accuracy of crop planting area recognition is improved by utilizing complementary advantages of remote sensing images.

The invention discloses a land cover change and thermal environment influence research method based on multi-temporal data, and the method comprises the following implementation steps: 1), obtaining a multi-temporal remote sensing image of a to-be-monitored region, carrying out the land cover classification mapping research of the remote sensing data of different time phases of the to-be-monitored region, and carrying out the statistics of the area of each land cover type; 2) performing surface temperature inversion on the remote sensing data of different time phases of the to-be-monitored area, grading the temperature, and analyzing the temporal and spatial change conditions of different temperature grades; 3) in the aspect of exploring the influence of the land coverage and the change thereof on the surface temperature, firstly, the temperature response of the surface temperature of different land coverage types in different seasons is analyzed; secondly, semi-quantitatively researching the relationship between the surface temperature and the surface temperature by introducing surface characteristic parameters; then quantitatively analyzing the contribution ratio of the land cover type to the surface temperature after the land cover type is changed; and finally, by introducing a distribution function, quantitatively analyzing heat source area information of different land coverage types in the research area.

The invention relates to a lake remote sensing image algae bloom coverage degree inversion method based on machine learning. Compared with the prior art, the defect that algae bloom coverage degree inversion is difficult to carry out on lake remote sensing images is overcome. The method comprises the following steps: acquiring and preprocessing a remote sensing data source; carrying out algae bloom extraction and processing on the Sentinel-2 multispectral image to obtain a coverage sample set; performing feature selection and combination on the MODIS multispectral image to obtain a feature index data set; constructing a lake algae bloom coverage degree data set; constructing a model and training based on machine learning; and obtaining an inversion result of the algae bloom coverage. According to the method, the algae bloom coverage degree of the MODIS pixel scale can be obtained by utilizing Sentinel-2 data with high spatial resolution, and the quantitative estimation model is constructed by combining MODIS spectral characteristics, so that high-temporal-spatial-precision monitoring and change analysis of the lake algae bloom are realized.

The invention discloses a red soil organic matter spectrum inversion method and device for removing influence of free iron, and the method comprises the steps: obtaining soil sample data, storing organic matter data into a set P, storing free iron content data of typical red soil samples into F, storing spectral reflectivity data of red soil survey samples into H, and obtaining a set P; storing spectral reflectivity data of typical red soil samples into D; according to the set H and the set D, utilizing an MODWT algorithm for decomposition to obtain layered discrete detail coefficient sets CH and CD of the red soil survey sample and the typical red soil sample; removing a noise layer based on CD to obtain an effective layer discrete detail coefficient set C; respectively calculating correlation coefficients between each element in the C and the set F and the set P, and screening out a characteristic wave band set CDG; selecting a detail coefficient CHG located on the characteristic wave band from the layered discrete detail coefficient set CH; a prediction model is constructed, and CHG is adopted for training; and predicting the organic matter content by adopting the trained prediction model. The method can eliminate the influence of free iron oxide, and is low in cost and high in efficiency.

The invention relates to the technical field of remote sensing data processing and earth surface parameter inversion, and discloses an earth surface matrix parameter hyperspectral data inversion method and system based on machine learning. Comprising the following steps: constructing a multi-source heterogeneous hyperspectral data set; performing feature screening on the preprocessed hyperspectral data set based on an adaptive band selection algorithm, constructing a dynamic weight matrix by calculating mutual information entropy and inter-class distance measurement between spectral bands to realize intelligent screening of key feature bands, and combining spectral derivative conversion and spectral index calculation to generate an enhanced feature vector; and a multi-task transfer learning neural network model is constructed, and an output layer realizes multi-parameter collaborative inversion based on a multi-task learning architecture. And performing preprocessing and feature enhancement operation which is the same as that of the training data on the hyperspectral image data of the to-be-inverted region, inputting the trained neural network model, and outputting a surface matrix parameter inversion result.

The invention discloses a method for inverting marine primary productivity based on airborne dual-wavelength laser radar remote sensing data, and the method comprises the steps: building a dual-wavelength high-precision water optical characteristic inversion model based on the airborne dual-wavelength laser radar remote sensing data, and obtaining a high-precision marine phytoplankton absorption coefficient vertical section; and inverting the marine primary productivity vertical section distribution coupled with the high-precision phytoplankton absorption coefficient. According to the invention, one laser wavelength is suitable for detecting an offshore sea area, the other laser wavelength is suitable for detecting an open sea area, and the dual-wavelength ocean laser radar improves the maximum detection depth of near and open seas and the inversion precision of the primary productivity vertical section distribution of a typical sea area. And the ocean carbon cycle space-time evolution precision and the observation capability in the aspect of ocean upper-layer energy conversion are further improved.

The embodiment of the invention provides an ionization layer electron density inversion method, device, system and equipment and a medium. The method specifically comprises the following steps: determining a target area of electron density inversion; obtaining observation data corresponding to the signal link in the target area; the signal links comprise a first signal link from a GNSS satellite to a ground station, a second signal link from the GNSS satellite to a low earth orbit satellite, and a third signal link from the low earth orbit satellite to the ground station; determining the slant total electron content corresponding to the signal link according to the observation data corresponding to the signal link; solving an ionized layer chromatography function according to the oblique total electron content corresponding to the signal link so as to obtain an electron density value of a grid in the target area; the ionized layer chromatography function is used for representing a mapping relation between the oblique total electron content and the electron density value of the grid. According to the embodiment of the invention, the inversion precision of the ionosphere electron density can be improved.

The invention provides a satellite solar short-wave radiation inversion method based on knowledge distillation, which relates to the field of solar short-wave radiation inversion, and comprises the following steps: collecting polar orbit satellite data, stationary satellite data, geographic information data and ground observation data; preprocessing the data to obtain input features and labels for training the teacher model and the student model; sorting the importance of the input features by using a random forest model, selecting key features, and removing non-key features; constructing a deep learning model by using the key features, and generating a teacher model of solar short-wave radiation inversion; and taking an output result of the teacher model as a soft label, introducing ground observation data as a hard label, and carrying out knowledge distillation to obtain a student model of solar short-wave radiation inversion. According to the technical scheme of the invention, the problems of data scarcity and high data acquisition cost caused by the fact that a model is constructed only depending on ground observation data in the prior art are solved, and the problem that the inversion precision completely based on satellite data is limited is solved.

The invention relates to the technical field of magnetic exploration, in particular to a deep learning network structure and method for magnetic anomaly inversion, and solves the problems that a traditional linear inversion method is slow in calculation stable speed and relatively low in precision, and a traditional ground and aeromagnetic data inversion method is insufficient in vertical resolution and relatively low in precision. A traditional linear inversion method is always encountered with the problems of low calculation efficiency and poor stability due to the local minimum value problem, and comprises an inversion network encoder, and a residual block and down-sampling method is adopted to realize the characteristics of magnetic anomaly data to extract an encoding network unit in a two-dimensional form. According to the method, feature extraction of magnetic anomaly data is realized, a residual module is added in an encoder to fuse feature maps from different sources, a three-dimensional magnetic susceptibility structure is reconstructed from two-dimensional observation magnetic field data, and in order to solve the problem that a sample library of deep learning inversion needs enough diversity and complexity, the method has high robustness. The generalization ability of the magnetic method inversion neural network is fully improved, and the inversion precision is further improved.

The invention discloses a logging-while-drilling forepoling stratum parameter inversion method based on a deep neural network. A training stage comprises a logging-while-drilling azimuth electromagnetic wave forepoling model construction step, a data set construction step, a data preprocessing step, a resistivity inversion neural network training step and a stratum boundary inversion neural network training step. In the testing stage, the received logging response is input into the trained resistivity inversion neural network and the boundary inversion neural network, the resistivity inversion neural network outputs the predicted resistivity to the boundary inversion neural network, and the boundary inversion neural network outputs the predicted stratum boundary position; and finally, drawing a stratum image according to the predicted resistivity and the stratum boundary position. According to the invention, a formation parameter inversion system composed of the resistivity inversion neural network and the formation boundary inversion neural network based on the resistivity prior is constructed, so that the formation parameters can be inversed in real time with high precision in the logging-while-drilling forepoling process.

The invention discloses a large-area forest vegetation carbon reserve inversion method based on GEE platform multi-source data fusion. The method comprises the following steps: data acquisition and analysis: obtaining forest carbon storage literature analysis data and an index factor data set; performing correlation analysis, constructing a forest carbon reserve index system, and establishing a corresponding characteristic variable relationship between the independent variable index factor and the forest carbon reserve sample; a seq2seq deep learning model based on LSTM (Long Short Term Memory) is imported; precision analysis is carried out; and inverting a large-area forest vegetation carbon reserve data set by using the deep learning model, and analyzing main determinants of forest carbon reserves. According to the method, a Google Earth engine is introduced, factors such as waveband information, vegetation indexes and long-term change trend characteristics are extracted from the Google Earth engine through a huge remote sensing data file of the Google Earth engine and a strong parallel computing capability of the Google Earth engine, and the forest carbon reserve in the loess plateau region is inverted and main determinants of the forest carbon reserve are analyzed based on a seq2seq deep learning model of LSTM.

The invention discloses a QPS numerical ray tracing and multi-target optimization-based high-frequency sky-ground wave radar ionosphere electron concentration inversion method, which comprises the following steps of: firstly, carrying out FFT (Fast Fourier Transform) on radar echo data, constructing a radar echo distance-Doppler matrix, and according to the obtained radar echo distance-Doppler matrix, calculating the electron concentration of a high-frequency sky-ground wave radar ionosphere according to the obtained radar echo distance-Doppler matrix; determining an antenna direct wave group distance and a transmitting station-receiving station ground distance; then determining an optimal emission elevation angle by using randomly generated ionosphere parameters; and finally, taking the emission elevation angle data corresponding to the n groups of ionosphere parameters as an initial population, and carrying out iterative optimization on the initial population by utilizing a population optimization algorithm to obtain an optimal ionosphere parameter result, thereby realizing ionosphere electron concentration inversion. According to the scheme, the radar echo data and the randomly generated ionosphere parameters are combined, the amount of data needing to be calculated is remarkably reduced, accurate inversion of the ionosphere parameters is achieved through ray tracing and multi-target optimization, the structure is simpler, the method does not depend on a complex model or a large amount of experience data, and the method is suitable for large-scale popularization and application. The optimal ionosphere parameter can be approached more quickly and effectively, and the real-time performance and accuracy of ionosphere electron concentration inversion are improved.

The invention discloses an SAR (synthetic aperture radar) water depth inversion method driven by variable window sliding segmentation, belongs to the technical field of synthetic aperture radars, and is used for water depth inversion including image segmentation and water depth inversion. The image segmentation comprises original image regularization adjustment, information storage, coastline identification and noise processing, pixel extreme value judgment, pixel change function determination and window sliding; after a coastline is extracted and a starting point is determined, pixels are expanded to two sides of the starting point on the coastline and pixels invaded by the coastline are eliminated, so that utilization of a near-shore image is improved; and the more accurate surge wavelength is obtained by dynamically adjusting the size of the pixel. Compared with the prior art, the method has the advantages that wavelength information of different areas is accurately extracted, complex manual judgment is avoided, and segmentation efficiency and inversion precision are improved; for the irregular coastline, the utilization rate of the water depth inversion of the image in the area near the coastline is increased by changing the starting point along the coastline and expanding outwards.

The invention relates to a crack sensitive parameter prediction method and device based on anisotropic gradient modeling, and the method comprises the following steps: S1, obtaining the normal crack weakness and tangential crack weakness of a characterization crack based on a linear sliding model; s2, establishing an initial model of the fracture weakness by constructing a relationship between the fracture anisotropy gradient and the fracture weakness; s3, obtaining an azimuth difference longitudinal wave reflection coefficient equation based on the PP wave linear approximate reflection coefficient; and S4, estimating the tangential crack weakness according to the seismic azimuth amplitude difference, and constraining the data of the tangential crack weakness by using the initial model of the crack weakness. According to the technical scheme, the relationship between the fracture anisotropy gradient and the fracture weakness is constructed, the initial model of the fracture weakness is established, the model is reasonable and reliable, and the data of the tangential fracture weakness can be constrained and inverted, so that the seismic inversion precision of the fracture weakness is improved, and the seismic inversion accuracy of the fracture weakness is improved. The method can be used for reliable prediction of the carbonate reservoir fracture development area.

The invention is suitable for the technical field of oil-gas exploration, and provides a seismic data convolution cross-correlation intelligent FWI method, which integrates a proposed seismic data convolution cross-correlation FWI target equation into a deep learning framework, and combines an automatic encoder technology to significantly reduce the nonlinearity of the traditional FWI, effectively avoid the cycle jump problem, and improve the accuracy of the FWI. Therefore, the speed model inversion precision of the FWI is improved. Finally, the FWI technology which can still carry out stable and accurate high-resolution speed modeling under the conditions that seismic source wavelets are unknown, the initial speed is unknown and low-frequency data are missing is realized, and the method has important significance on seismic exploration.