39 results about "Geomorphological mapping" patented technology

The embodiment of the invention is suitable for the technical field of artificial intelligence, and discloses an unsupervised landform classification model training and landform map construction method, comprising the steps: obtaining an unlabeled landform sample training data set; performing adversarial training on the generative adversarial network by using the unlabeled geomorphic sample training data set to obtain target pseudo-geomorphic data output by the generative adversarial network; wherein the target pseudo-geomorphic data is pseudo-geomorphic data output when the training effect ofthe generative adversarial network is optimal; training a pre-trained landform classification model by using the target pseudo-landform data; and after the training is completed, using a landform classification model to construct a landform map. According to the embodiment of the invention, manual labeling is not needed in the landform classification model training process, and the unlabeled sample data is directly used for model training, so that the labor cost of manual annotation is saved, and the landform mapping cost is reduced.

The invention discloses an underwater multi-source acoustic feature extraction method and system. The method comprises the following steps: extracting geomorphic characteristic quantity of a water bottom geomorphic image, topographic characteristic quantity of a water bottom topographic image and stratum characteristic quantity of a water bottom stratum image, wherein the topographic characteristic quantity comprises a water bottom slope, a water bottom slope direction, a water bottom surface curvature and fractal dimension characteristics of a water bottom earth surface, the stratum characteristic quantity comprises an interface line characteristic quantity, an interface layer characteristic quantity and a surface section characteristic quantity, the interface line characteristic quantityrepresents the complexity of a water body substrate interface line, the interface layer characteristic quantity is a gray average value of the water body substrate interface layer, and the surface profile characteristic quantity comprises a gray scale statistical characteristic of the surface profile and a vertical texture characteristic of the surface profile; and determining the geomorphological characteristic quantity, the topographic characteristic quantity and the stratum characteristic quantity as a multi-source acoustic characteristic set. According to the invention, the extraction ofmulti-source acoustic features is realized, complementation of geomorphic, topographic and stratigraphic feature quantities is formed, the feature extraction is more comprehensive, and the classification precision is high when the method is used for subsequent sediment classification.

The invention discloses a method for constructing a 3D flying scenario of a near space aerostat based on STK. The method includes the following steps: converting the format of a MDL file of the near space aerostat, correcting the color of the MDL file, and increasing movements and defining the movements of the near space aerostat, establishing the flying scenario of the near space aerostat in combination with a synthesized landform map of a flying region, obtaining a 3D flying scenario of the near space aerostat by accessing flight track data and external data drive, and realizing dynamic simulation of the flight process of the near space aerostat by real-time updating the content of the 3D flying scenario. According to the invention, the method can better and dynamically display the flying scenario of the near space aerostat of the air above of a specific area, can switch the center and range of a field of view. The method can produce fine models, has high map resolution, and has expandable data interfaces.

The invention relates to the technical field of deep sea oil and gas development, in particular to a deep sea side slope stability analysis method and device. The method comprises the steps that whenvarious rock and soil parameters of a deep sea side slope are not obtained, whether the deep sea side slope is dangerous or not is judged based on related geomorphic attribute information in a deep sea geomorphic map; when various rock and soil parameters of the deep sea side slope are obtained, based on the plurality of rock and soil parameters, a stable value of the deep sea side slope is obtained. Based on the stability value of the deep sea slope, the stability grade of the deep sea slope is determined, on one hand, qualitative analysis can be conducted on the basis of a submarine topography map, on the other hand, quantitative analysis can be conducted on the basis of geotechnical engineering geological data, a reliable slope stability analysis result can be obtained on the basis of limited data, expenditure is saved, and project safety is guaranteed.

The invention provides a well position optimal selection method and device based on ancient and modern geological elements. The well position optimal selection method based on the ancient and modern geological elements comprises the steps that a structural map of an oil-gas-containing stratum is generated; a structural map of the oil-gas-containing stratum in the strong hydrocarbon generation period is generated; an underlying or overlying palaeogeomorphic map of the oil-gas-containing stratum is generated; a reservoir distribution map of the oil-gas-containing stratum is generated according to seismic data of a target block; and the well position of a target work area is optimally selected according to the structural map of the oil-gas-containing stratum, the structural map of the oil-gas-containing stratum in the strong hydrocarbon generation period, the underlying or overlying palaeogeomorphic map of the oil-gas-containing stratum and the reservoir distribution map. According to thewell position optimal selection method, based on ancient and modern geological structure forms, four geological factors of palaeogeomorphology and oil-gas-containing reservoir distribution are combined, the efficient well position is comprehensively and optimally selected, the drilling success rate can be increased in actual oil-gas exploration, then the production cost is saved, and the economicbenefits are increased.

The invention discloses an underwater multi-source acoustic image substrate classification method and system based on decision-making level fusion. The method comprises the following steps: acquiringa geomorphic image, a topographic image and a stratum profile image under a unified space coordinate system; determining a landform segmentation image, a terrain segmentation image and a stratum profile segmentation image; performing feature extraction on each segmented image to obtain landform features, topographic features and stratum features; respectively inputting the geomorphic features andthe topographic features into a plurality of trained machine learning classification models to obtain a plurality of first classification results and a plurality of second classification results, andinputting the stratum features into the trained machine learning classification models to obtain a third classification result; and fusing the plurality of first classification results and the plurality of second classification results based on a D-S evidence theory, or fusing the plurality of first classification results, the plurality of second classification results and the third classificationresult to obtain a substrate classification result of the underwater multi-source acoustic image. According to the invention, automatic and high-precision classification of underwater substrates canbe realized.

The invention discloses a landslide and debris flow monitoring and early warning method, device and system and a medium, and the early warning method comprises the following steps: obtaining real-time monitoring data of a monitoring area, the real-time monitoring data comprising landform data and sound wave data; performing feature extraction on sound wave data in the real-time monitoring data to obtain a real-time feature value group; comparing the real-time characteristic value group with a characteristic value group corresponding to historical sound wave data collected when no debris flow occurs in the monitoring area to obtain sound wave similarity; fusing the landform data in the real-time monitoring data and the landform data collected when the debris flow does not occur in the monitoring area in a simulation 3D landform map, and calculating the landform data difference; when the sound wave similarity is smaller than a first preset threshold value and/or the landform data difference is larger than a second preset threshold value, early warning information is generated. Meanwhile, early warning is realized based on the landform data and the sound wave data, the misjudgment rate can be reduced, and the early warning accuracy is high.

The invention belongs to the field of quantitative characterization of sedimentary facies, particularly relates to a quantitative characterization method and system for sedimentary facies evolution based on geology and geophysics. The invention aims to solve the problem that quantitative characterization of sedimentary facies evolution cannot be systematically and systematically realized by utilizing a geophysics technology under the constraint of a geological model in the prior art. The method comprises the following steps: establishing a high-resolution sequence stratigraphic frame by using a core sample, a logging data body and an earthquake data body, and acquiring a three-dimensional sequence interface on an earthquake data volume; acquiring an approximate ancient landform map of a sedimentary period through combination with a travel time difference representing the terrain height of the sedimentary period; with the high-resolution sequence stratigraphic frame as a model, carrying out logging-seismic combined wave impedance inversion; carrying out stratigraphic slice attribute analysis on a wave resistance body; establishing a sand-to-ground ratio-wave impedance mapping relation; and dynamically and quantitatively representing a sedimentary facies evolution process based on the sand-to-ground ratio-wave impedance mapping relation. According to the method, the dynamic and quantitative characterization of the sedimentary facies evolution process in the high-resolution sequence stratigraphic frame can be realized.

The invention discloses an environment-friendly seabed reef cleaning construction process. The environment-friendly seabed reef cleaning construction process comprises the following steps that an offshore rock drilling platform is driven to a position above a to-be-cleaned reef and then is fixed and located; an industrial test of the crushing effect on a typical area is carried out; according to the test result, rock drilling construction parameter sensitivity analysis is carried out, regression analysis is carried out on the rock drilling effect evaluation index, and rock drilling construction parameters of a reef clearing area are calculated; a three-dimensional geomorphologic map of the reef clearing area is established; rock drilling points are arranged, rock drilling and crushing areconducted at all the rock drilling points through rock drilling rods, and crushed reefs are cleaned through a grab bucket; and the offshore rock drilling platform is driven out of the reef clearing area, the three-dimensional geomorphologic map of the reef clearing area is re-established and analyzed, and construction is stopped until reef clearing is conducted to the designed elevations of all the areas. By means of the construction process, excavation of a seabed foundation trench can be accurately and efficiently achieved, construction errors are effectively reduced, the repeated construction operation condition caused by under-excavation and over-excavation is avoided, the construction efficiency is greatly improved, and the construction cost is reduced.

The invention relates to the technical field of reservoir bottom interface identification, in particular to a weathering crust reservoir bottom interface identification method based on mathematical statistics. The method comprises the following steps: S1, constructing a rock physical model on the basis of well data; S2, determining the relation between granite weathering crust reservoirs with different thicknesses and seismic response; S3, determining the ancient landform of the granite buried hill through an impression method research, selecting different exposure time points on an ancient landform map, counting the relative time difference between the top and bottom interfaces of a weathering crust reservoir stratum corresponding to different exposure time points on a seismic section, and carrying out multi-point fitting analysis by taking forward modeling as a reference to obtain a corresponding quantitative relational expression; and S4, taking the top interface of the whole-region weathering crust reservoir as input to obtain a calibration result of the bottom interface of the whole-region weathering crust reservoir and a quantitative result map of the time thickness of the weathering crust reservoir. The method is excellent in feasibility and reliability, and a good thought and method are provided for delineation work of the weathering crust of the granite buried hill.

The invention discloses a multi-terminal cooperative forest crown geomorphic mapping system and method, the system comprises a satellite navigation system and an unmanned vehicle, and the unmanned vehicle is provided with a three-dimensional laser radar, an inertial navigation system, a satellite navigation module and a processing module. The method comprises the following steps: initializing correction time, wherein the three-dimensional laser radar acquires three-dimensional point cloud data of the surrounding space environment of the unmanned vehicle, the inertial navigation system acquiresinertial navigation positioning data of the unmanned vehicle, and the satellite navigation module acquires satellite navigation positioning data of the unmanned vehicle; fusing the inertial navigation positioning data and the satellite navigation positioning data of the unmanned vehicle to obtain real-time positioning information of the unmanned vehicle; and according to the three-dimensional point cloud data and the real-time positioning information, orderly stacking to generate a three-dimensional landform map. According to the method, the inertial navigation positioning data and the satellite navigation positioning data of the unmanned vehicle are fused, so that the positioning precision of the unmanned vehicle is improved, the precision of a three-dimensional landform map is improved,and direct basis and guarantee are provided for an optimal fire extinguishing search and rescue path and scheme when a fire occurs.

The invention discloses a landform monitoring and management system based on the Internet. The landform monitoring and management system comprises a camera, an identification unit, a database, an analysis unit, a judgment unit, a classified storage unit and a sending unit. The camera user monitors the landform and appearance in real time, obtains geomorphic video information automatically, and transmits the landform video information to an identification unit; geomorphic image information, natural resource image information and position data and shape data corresponding to the geomorphic imageinformation and the natural resource image information are stored in the database; the identification unit acquires geomorphic image information from a database, and the geomorphic image informationand the landform video information are subjected to identification operation together; through the arrangement of the classified storage unit, the types of natural resources on the surface of the earth are stored in a classified mode, and resource quantity sorting is carried out in different types, so that the resources are divided into first-class resource storage and second-class resource storage, the orderliness of storage is improved, workers can conveniently and rapidly search for the resources, time is saved, and the working efficiency is improved.

The invention relates to a sedimentary facies identification method, device and system, and belongs to the technical field of oil-gas field development geology. The identification method comprises the following steps: carrying out ancient landform recovery on a research area to obtain an ancient landform map after deposition of the research area; converting the ancient landform map into a vector structure map; selecting a calculation section on the vector structure map along the object source direction; selecting a plurality of deposition points on the calculation profile, including the highest deposition point; obtaining gradient and horizontal displacement of each deposition point relative to the highest deposition point according to the coordinates of each deposition point; performing staged fitting on the gradient and the horizontal displacement of each deposition point relative to the highest deposition point, finding out inflection points between stages, and taking a gradient value at the inflection points as a critical gradient value; and identifying each sedimentary facies according to the critical gradient value. According to the method, the accuracy of sedimentary facies identification is improved, the boundaries between the sedimentary facies can be quantitatively found out through the critical gradient values, and the sedimentary facies region is accurately divided.

The invention relates to a land survey planning method based on an unmanned aerial vehicle, and the method comprises the steps: S1, obtaining land information, and generating a landform map according to the land information after the land information sent back by the unmanned aerial vehicle is obtained; s2, region division: dividing the generated landform map into a plurality of region grids with the same area according to division information selected by a planning designer, and each region grid contains land information in the region grid; and S3, planning a region, after obtaining planning information selected by a planning designer, screening out region grids meeting planning information requirements according to the corresponding land information in each region grid, and generating prompt information. The method has the advantages that the working efficiency of planning and designing personnel is greatly improved; and compared with manual work, the method has the advantage that the accuracy of data analysis and arrangement is improved.

The invention discloses an electric power engineering project intelligent surveying and mapping method based on a remote unmanned aerial vehicle and sensor technology, and the method comprises the steps: collecting landform images of all pre-construction regions in an electric power engineering project, carrying out the comparative analysis of the type landform corresponding to each pre-construction region, detecting the average wind speed of each pre-construction region in each hour in each day, comparing and counting the duration that the wind speed of each direction in each day of each pre-construction region meets the requirement, screening the average wind speed of each direction in each hour meeting the requirement in each day of each pre-construction region, and calculating the average wind speed of each direction in each hour meeting the requirement in each day of each pre-construction region; meanwhile, obtaining the altitude of each pre-construction region in the electric power engineering project, and calculating the annual power generation efficiency estimation coefficient of each orientation in each pre-construction region, comparing and screening the highest annual power generation efficiency estimation coefficient and the corresponding orientation in each pre-construction region, and displaying the highest annual power generation efficiency estimation coefficient and the corresponding orientation, so that the surveying and mapping level of the wind power engineering project is improved.

The invention discloses a karst ancient landform restoration method with hydrothermal cause siliceous rock as an isochronous datum plane. The karst ancient landform restoration method comprises the following steps: acquiring rock core, slice and logging information; stratigraphic division and comparison are carried out, and the hydrothermal cause siliceous rock bottom boundary is determined to serve as an isochronous reference plane; finely explaining an isochronous reference surface and a target layer erosion surface; calculating the residual stratum thickness between the isochronous reference surface and the target stratum erosion surface; and obtaining a karst ancient landform map according to the residual stratum thickness. According to the method, the principles that the reference surface is isochronous and close to the erosion surface and the earthquake response is clear and easy to track are fully considered, and the recovery precision of the ancient karst landform is greatly improved.

The embodiment of the invention discloses a landform migration method and device, equipment and a medium, and the method comprises the steps: determining a terrain height map region and a ground surface control map region corresponding to a target migration region according to the selected target migration region, terrain height map and ground surface control map; determining an overall migrationamount of landform migration, and sampling a surface height value and a control coefficient before migration corresponding to each data point in the terrain height map area and the surface control maparea based on the overall migration amount; and determining a final height value and a final control coefficient of each data point according to the initial ground surface height value and the initial control coefficient of each data point, the sampling value of the ground surface height value, the sampling value of the control coefficient and the overall migration quantity so as to complete landform migration. According to the method of the embodiment of the invention, the problem of low landform engraving landform manufacturing efficiency required by landform migration is solved; the targetlandform image manufacturing efficiency can be improved.

The invention discloses a karst stage ancient fracture description method and device, and the method comprises the steps: extracting well bypass seismic wavelets for synthetic recording, and achieving the well seismic calibration of a karst reservoir top interface and all overburden velocity control layers including the maximum flooding surface, and the skeleton section interpretation and encryption interpretation; time-depth conversion is carried out on well bypass seismic wavelets and horizon interpretation results; horizon leveling is carried out on the depth domain seismic well bypass seismic wavelets; performing imaging enhancement processing on the leveled and reconstructed seismic data volume; fusing the extracted sensitive seismic attributes on the processed seismic data; fitting the calculated plane extension length of the fracture with the maximum vertical fault displacement on the karst ancient fracture plane prediction map; and classifying and grading the ancient fractures in the karst period based on the fitting result, and projecting the ancient fractures to the ancient landform map of the top surface of the reservoir in the karst period to finish final description. According to the method, rapid identification and grading classification of karst fractures can be realized, and the prediction precision and efficiency are improved.

The invention relates to a landform map construction method convenient for efficient navigation of an autonomous mobile robot. The method comprises the following steps: constructing a landform segmentation network: constructing a high-precision lightweight semantic segmentation network by using a depth separable convolution and pyramid pooling module; according to physical and geometric properties of different landforms, the passable landforms are divided and labeled in more detail, and a landform data set is constructed in a targeted manner for model training; and constructing a landform map: constructing a mapping relationship between the landform image and the occupied grid map according to a geometric transformation relationship between the original image acquired by the sensor and the grid map, and creating the landform grid map. According to the method, the landform in the working environment of the mobile robot is divided more meticulously by using the semantic segmentation network, and the landform grid map containing rich information is created through pixel point scanning, coordinate conversion and grid mapping, so that the navigation efficiency of the mobile robot is improved.