40 results about "Landslide detection" patented technology

The present invention relates to a high-resolution remote sensing image-based multi-index fusion landslide detection method, which comprises the steps of (1) acquiring a high-resolution remote sensing image and a stereoscopic image pair; (2) generating a digital elevation model and calculating terrain feature indexes; (3) preprocessing the remote sensing image; (4) conducting the multi-scale image segmentation for the remote sensing image; (5) selecting appropriate terrain feature indexes for different scale layers and obtaining the values of the terrain feature indexes; (6) conducting the fusion processing for terrain feature indexes and ground object feature indexes, and comparing each feature index with a preset rule set to realize the landslide detection. Compared with the prior art, the noise influence caused by the scattered distribution of landslide regions can be avoided, so that the landslide recognition and extraction problem caused by earthquakes or other natural damages can be solved. Therefore, the method provides support for the post-disaster reconstruction and restoration. The above feature indexes can reflect the features of a ground object within a selected region, and a threshold corresponding to the feature indexes is unique. Based on the combination of features and the determination of the threshold, the landslide region can be successfully extracted and studied.

The invention relates to a remote sensing image landslide automatic detection method based on a three-dimensional space-channel attention mechanism. The method comprises the following steps: manuallylabeling remote sensing images, cutting landslide samples to construct a sample library, training a convolutional neural network based on a three-dimensional space-channel attention mechanism, and learning features of various landslide instances on the remote sensing images; and carrying out landslide detection on the new remote sensing image by using the trained network model to realize automaticand accurate recognition of a landslide target on the remote sensing image. The method has the advantages that the robustness is high, and compared with an original backbone network and other existing attention methods, the method is more suitable for a landslide detection task on a remote sensing image; for the condition containing various interference factors, the method has better interferenceresistance, and higher recognition accuracy can be obtained; the method can be used for preventing landslide disasters, reconstructing after disasters, updating landslide databases and the like.

The invention discloses a landslide detection method based on multi-scale feature fusion, which relates to the technical field of target detection. Firstly, the captured images are sent to VGGNet withfocus module in real time. The depth features are extracted from the convolution layer from bottom to top, and then the depth features are fused from top to bottom according to different layers. Then, a selective search is performed on the original input image to generate several candidate boxes which may contain the target region. According to the size of the original image, the size of the candidate frame and the size of each feature map after feature fusion, the mapping size of each candidate frame on three feature maps is calculated. The dimensions of all the mapped feature maps are unified and passthrough all the connecting layers to adjust the position and dimensions of the candidate frames; Finally, the preliminary detection results are obtained by non-maximum suppression on the original image, and the final detection results are obtained. The invention can detect the debris flow phenomenon with different sizes, and greatly accelerates the detection speed.

The invention discloses a landslide detection method based on unmanned aerial vehicle point cloud. The landslide detection method comprises the following steps: capturing a high-resolution geographicmark image of a landslide through unmanned aerial vehicle equipment; refining parameters estimated in the SFM process through an SFM program and a beam adjusting method; generating dense three-dimensional point clouds for each observation epoch by using an image captured only in the specific area through a semi-global dense matching process; calculating the normal distance between the two point clouds obtained at any different time periods, and detecting the volume surface change by adopting a point-to-point three-dimensional corresponding relation; and finally, through comparison of the pointcloud landslide steep slopes obtained at different time periods, calculating the displacement rate in the horizontal direction, and detecting the point cloud based on low-altitude, unmanned, aviationand vehicle images. According to the landslide detection method, a large amount of images from different data sources can be easily processed without a large amount of ground control point information or further processing. The research result has a certain reference value for landslide research.

The invention discloses a satellite selection method for improving RTK positioning precision and belongs to the field of RTK positioning of a global satellite navigation system. According to method provided by the scheme, after a reference satellite is selected through an elevation angle, a dual-difference residual error threshold is set, phase pseudo range dual-difference residual errors are calculated for the reference satellite and each common satellite, and satellite selection is carried out according to value sizes. According to the scheme, the computational complexity is quite small, the threshold of the dual-difference residual errors can be dynamically adjusted according to different phases in an RTK positioning process, the convergence speed of the RTK positioning is accelerated, the RTK positioning precision is further improved, and the method has great significance on such application occasions with ever improving precision requirements as landslide detection and the like.

The invention discloses a novel simple landslide detection device and solves problems of commonly high investment, complex assembling and relatively high operation cost of a landslide detection device in the prior art. According to the scheme, the novel simple landslide detection device comprises an iron protection casing, a solar cell panel, a mobile power source and a sound-light alarm device, wherein an upper surface of the iron protection casing is provided with the solar cell panel, the mobile power source is connected with the solar cell panel and the sound-light alarm device through leads, the novel simple landslide detection device further comprises an underground pile, the iron protection casing is internally provided with a liquid inclination type switch in lead connection with the mobile power source to realize alarm, the liquid inclination type switch is connected with the underground pile, and a lower surface of the liquid inclination type switch is provided with four support columns. The novel simple landslide detection device has properties of simple structure, simple mounting, convenient use and easy maintenance during practical operation, saves a lot of manpower, can satisfy engineering forecast safety requirements and has strong practicality.

The invention belongs to the technical field of mountain landslide detection, and particularly relates to a physical experiment device for exploring a rainfall induced landslide mechanism. The systemcomprises a water collecting groove, a water collecting groove water level control valve, water collecting groove rolling wheels, a rainfall system support body, a rotary rainfall simulation system, alandslide simulation device, a slope surface simulation plate, a hydraulic slope regulator, a flooding simulation device, a flooding device control door, a water pumping pump, a water conveying pipe,a total control plate, a base, a fully automatic camera, an electric wire, a rotating motor, a rotating shaft, a triangular flow guide plate, a guide pulley and a rope fixing post. The device can simulate two conditions including a rainwater induced landslide condition at different contact surfaces, different slopes and different rainfall quantities and a water soaking induced landslide conditionunder extreme conditions. The rainfall simulation system uses a creative rotary rainfall simulation system, and the uniform rainfall on a soil body is ensured. The whole experiment process can be recorded by the fully automatic camera, and can be transmitted to a remote terminal in a wireless way to be watched and stored.

The invention relates to a river levee side slope landslide detection device, and belongs to the field of geological disaster detection equipment. The river levee side slope landslide detection devicecomprises a ground pile implanted into slope ground along a direction perpendicular to a horizontal plane and a detection box arranged at one end of the ground pile, which faces away from the ground;a datum rope is suspended inside the detection box; a through-beam type photoelectric switch and a power supply are also arranged inside the detection box; a transmitter and a receiver of the through-beam type photoelectric switch are respectively arranged at both sides of the datum rope; when the detection box is not shifted, the datum rope is positioned between the transmitter and the receiverfor blocking off light; the through-beam type photoelectric switch is coupled with an alarm mechanism; the power supply provides required electric energy for the through-beam type photoelectric switchand the alarm mechanism; and when the detection box is inclined, the alarm mechanism sends out an alarm signal. The river levee side slope landslide detection device disclosed by the invention replaces artificial cyclic detection to carry out real-time inspection, so that when labor cost is saved, detection accuracy is improved.

The invention discloses an earthquake-induced secondary landslide post-disaster evaluation and rescue auxiliary system, which comprises a special detection unmanned aerial vehicle building module forbuilding a special post-earthquake landslide disaster detection and recognition unmanned aerial vehicle, and a post-earthquake landslide detection planning preparation module for completing various preparation works before disaster detection and recognition, apost-earthquake environment information acquisition module used for acquiring on-site landslide disaster images and position information coordinates, a Beidou navigation positioning module used for assisting above three modules in the aspects of navigation system construction, cruise route setting and information acquisition and transmission; a landslide disaster analysis decision module used for analyzing related attribute parameters such as post-earthquake secondary landslide disaster influence area, trailing edge height and stability, and a post-earthquake rescue planning evaluation module used for planning a rescue route, making a rescue scheme and evaluating the disaster influence degree. According to the invention, the accuracy, the real-time performance and the high efficiency in intelligent identification function requirements of post-earthquake secondary landslide disasters can be met.

Disclosed herein is a test apparatus for early landslide detection fully-connected with pore water pressure, surface displacement and shear surface. The test apparatus calculates a factor of safety of a slope based on variation in pore water pressure, surface displacement and shear surface of a soil mass, and predicts a change in factor of safety, thus making early landslide detection possible. In the test apparatus, while a container of a slider is moved with a soil mass loaded into the container, shear surface and surface displacement environment is provided, and the shear strength and the shear stress of the soil mass can be calculated based on the pore water pressure and the weight of the soil mass. Thereby, the factor of safety of the soil mass can be calculated, and early landslide detection can be realized by using variation of the factor of safety of the slope.

The invention discloses an SAR and optical image fusion method for landslide detection. The method provided by the invention comprises the following steps: step 1, preprocessing an SAR and an opticalimage; step 2, performing HIS conversion on the optical image to obtain three components I, H and S; step 3, performing stationary wavelet transform and high-frequency component energy maximization fusion on I components of the SAR image and the optical image; step 4, performing saliency detection on the low-frequency and high-frequency components of the SAR image and the gray scale information ofthe image, establishing an SAR salient target detection area guidance function, and partitioning the SAR image; step 5, establishing a salient region fusion rule, and realizing image fusion accordingto a regional fusion strategy; and step 6, identifying and extracting landslide disaster information based on the fused image. The method has good adaptability to SAR and optical image fusion for landslide detection, adopts related processing measures in structure maintenance, noise removal and spectrum reservation, and obtains an excellent effect.

This invention discloses a practical method for landslide detection in large space, which comprises the following steps: image synthesis, ice and snow detection, removal of non-potential landslide area, detection of potential landslide area, feature calculation, landslide detection model construction and precision validation; this invention avoids radiometric correction and outlier by detecting landslide from synthetic image. That guarantees practical applicability of the proposal. Firstly, detecting potential landslides can avoid the imbalanced sample distribution issue between background objects and landslides when training the landslide detection model. The landslide is further detected by building a random forest model based on the spectral features and textural features of potential landslide pixels in different neighboring time domains. It fully considers the changes of objects in different time domains, and lays a foundation for efficient landslide extraction. This model is relatively reliable and practical for automatically detecting landslide from large-scale images.

The invention discloses a resistivity data fusion three-dimensional imaging method and system for landslide detection. According to the technical scheme, the method comprises the steps of: obtaining geophysical prospecting data according to a plurality of types of laid geophysical prospecting lines; preprocessing the geophysical prospecting data, and obtaining two-dimensional resistivity data through an inversion method; extracting resistivity data of multiple geophysical prospecting methods at the same position, and fusing the multiple types of resistivity data; and establishing a three-dimensional coordinate system, converting the fused two-dimensional detection data into three-dimensional detection data, and imaging the three-dimensional detection data to obtain a three-dimensional stereogram. Two-dimensional detection data obtained through inversion by a transient electromagnetic method and a high-density electrical method are fused, three-dimensional detection data is obtained through a three-dimensional coordinate conversion formula, a three-dimensional stereogram is finally obtained after imaging of the three-dimensional detection data, and therefore the distribution range of a sliding surface in space is determined.

The invention provides a device for underwater landslide detection. The device comprises a monitor end, a process end and a connection end, wherein the outside of the monitor end is provided with a plurality of sealed watertight protective layers, the inside of the monitor end is provided with a cavity, the cavity is provided with a MEMS sensor and sonar and damping disc, the MEMS sensor and sonar are fixedly on the cavity through the damping disc, the connection end comprises cable, one end of the cable is connected to the MEMS sensor and sonar in the monitor end, the other end of the cable is connected to the process end to transfer the data obtained from the monitor end to the process end and the cable is also used for supplying power to the monitor end. The process end comprises a data processor and a power module. The device is capable of effectively detecting the underwater landslide condition.

The invention discloses a partitioned landslide detection system and method based on a cascaded deep convolutional neural network. According to the partitioned landslide detection method, an image collection module used for acquiring landslide disaster images is included; the database module is used for constructing and storing landslide disaster image samples with different attributes and different regions; the landslide detection model preparation module is used for preprocessing the image, constructing an image pyramid and determining a corresponding region coefficient of each region; and the landslide detection model generation module is used for generating landslide disaster recognition detection models corresponding to different regions, and comprises a visual DLNet detector generation optimization module and a multi-layer cascaded deep convolutional neural network detection model module. According to the method, an artificial intelligence technology is combined with landslide disaster recognition, and a detection model is generated for different regions by using the regional universality of landslide disasters; and a deep convolutional neural network with cascade characteristics is constructed, so that intelligent interpretation of landslide disasters is realized, and the landslide disaster recognition accuracy is improved.

The invention belongs to the technical filed of radar remote control image processing, and provides an estimation method of a terrain azimuth angle under vegetation cover based on polarimetric synthetic aperture radar (POLSAR). By means of strong penetrability of the POLSAR, a vegetation cover area is observed once, and on the basis of an electromagnetic scattering theory and mathematical modeling, and topographic relief information under the vegetation cover in radar observation data is estimated. Due to the fact that earth surface observation can be carried out all day long in four seasons through the radar, and the topographic relief information under the estimated vegetation cover can provide sufficient guarantees for landslide detection, disaster emergency response, territorial resource annotation and others. Due to the fact that only one POLSAR image is used, economic expenses cost in buying radar data can be greatly reduced, and applicable ranges are greatly increased.

The embodiment of the invention provides a landslide detection method, device and equipment and a storage medium, and relates to the technical field of geological disaster prediction, and the method comprises the steps: obtaining pre-earthquake and post-earthquake remote sensing images of a to-be-detected region from an optical remote sensing image system, carrying out the registration, and carrying out the change detection of the post-earthquake remote sensing image according to the registered pre-earthquake remote sensing image; extracting a predicted deformation area from the post-earthquake non-change area according to the image of the post-earthquake non-change area and a preset deformation rate diagram of the to-be-detected area, and combining the predicted deformation area with the post-earthquake change area to obtain a target area; and calculating feature parameters of each object in the target area according to a gray texture feature map obtained by calculating the preset radar image of the to-be-detected area, the post-earthquake remote sensing image, a preset digital elevation model and a preset deformation rate map, and rejecting areas where non-landslide ground objects are located in the target area based on the parameters to obtain a landslide hidden danger area. According to the invention, the accuracy and comprehensiveness of landslide hidden danger detection can be improved.

The invention discloses a quarrying field safety monitoring system and method. The system comprises a control background and a landslide detection subsystem, the control background is in communicationconnection with the landslide detection subsystem and is used for receiving the detection data and carrying out prediction and early warning on the disaster sign, the landslide detection subsystem comprises a quarrying field body landslide module and a waste dump landslide module, the quarrying field body landslide module includes a plurality of sensor detection units arranged on the side slope,the waste dump landslide module comprises a monitoring unit for detecting the moving amplitude of the waste soil. The system has the advantages that the rock condition of the quarrying field body andthe waste soil condition of the waste dump are detected through the landslide detection subsystem, and the disaster sign analysis is carried out through the control background, so that the possible danger can be determined, and the dangerous case content and the dangerous case suspected coordinates are sent to an alarm module in time, and the alarm module is used for informing an operator in the vicinity of the dangerous case suspected coordinates to timely evacuate or strengthen the side slope monitoring and management, so that problems can be found in time and preventive measures can be taken.

The invention discloses a three-dimensional identification method of a landslide slip surface based on unmanned aerial vehicle images and deep learning, and relates to the technical field of disaster monitoring. The objective of the invention is to solve the problem of inaccurate specific positioning of a slip plane in the existing landslide detection method. According to the method, an unmanned aerial vehicle is used for collecting images of at least five different angles of a detected area as detection images, a semantic segmentation network is used for distinguishing types of all pixel points in the detection images, a semantic segmentation result is obtained, the types of the pixel points comprise background pixel points and landslide slip plane pixel points, and the types of the pixel points comprise background pixel points and landslide slip plane pixel points. Reconstructing the detection image by using a three-dimensional reconstruction algorithm to obtain a three-dimensional point cloud containing a landslide slip plane, calculating a two-dimensional coordinate of any pixel point A in the three-dimensional point cloud in a two-dimensional plane, substituting the two-dimensional coordinate of the pixel point A in the two-dimensional plane into the semantic segmentation result obtained in the step 2 to obtain the type of the pixel point A, and obtaining the type of the pixel point A; and taking the type of the pixel point A and the three-dimensional coordinate of the pixel point A as a three-dimensional identification result of the landslide slip plane.

The invention relates to the technical field of landslide detection, in particular to a comprehensive monitoring system for a landslide earth surface and underground cracks. The comprehensive monitoring system comprises a landslide earth surface displacement meter and landslide underground shearing strain meters, the landslide earth surface displacement meter is horizontally arranged on the to-be-detected earth surface, two cement cubes are arranged on the to-be-detected earth surface according to a certain interval, and the two ends of the landslide earth surface displacement meter are arranged on the two cement cubes respectively; the landslide underground shearing strain meters are embedded in the portions, in soil of the to-be-detected earth surface, of the lower ends of the cement cubes respectively, the upper ends of the landslide underground shearing strain meters are inserted in the cement cubes respectively, and one cement cube is vertically provided with a data line and a GPRS data transmission line. The system has little limitation, the temperatures of all parts have a large corresponding range, and therefore the system is wide in application range.

The invention relates to a landslide displacement detection method and device based on live-action simulation, and the method comprises the steps: receiving landslide body information of a landslide detection region, wherein the landslide body information comprises the temperature value, humidity value, positioning data, inclination angle and multi-segment displacement data of each monitoring point; based on a kinetic model fitting algorithm, calculating and analyzing the temperature value, the humidity value, the positioning data, the inclination angle and the multi-section displacement dataof each monitoring point, and establishing a soil layer state curve of each monitoring point; establishing a two-dimensional displacement scene observation model according to each soil layer state curve; determining a sliding trend of the landslide mass according to the two-dimensional displacement scene observation model; based on the elevation data model, utilizing a three-dimensional curved surface matching algorithm to extract elevation data of each monitoring point; according to the two-dimensional displacement scene observation model, the sliding trend of the landslide body and the elevation data, three-dimensional analog simulation display is carried out on the landslide body in the detection area, so the accuracy of landslide prediction and early warning decision analysis is effectively improved.

The invention discloses a method for accurately identifying a new landslide area based on a multi-source time sequence remote sensing image. The method comprises the following steps: acquiring a slope map of a digital terrain model, a vegetation coverage index NDVI time sequence image and a surface radiation temperature time sequence image; calculating a simplified earth surface drought index TVDI based on an empirical parameter relationship between the earth surface temperature and the vegetation coverage index NDVI; tVDI values of remote sensing image files before and after a landslide are extracted to serve as time phase basic data before and after landslide area extraction, TVDI images before and after the landslide are resampled and subtracted to obtain a TVDI-D value of a change area, a threshold value for segmenting the landslide area and a non-landslide area is set, binaryzation is performed on the TVDI-D images, the binaryzed TVDI-D images are converted into vectors, and the vectors are converted into vectors; and screening the landslide candidate objects by taking the gradient information as a filtering condition so as to determine a landslide area vector. The precision of the landslide area detection method is improved in combination with multi-source remote sensing information, the false alarm rate in landslide detection is greatly reduced, and the credibility of a landslide detection result is also improved.

The invention discloses a satellite selection method for improving RTK positioning precision and belongs to the field of RTK positioning of a global satellite navigation system. According to method provided by the scheme, after a reference satellite is selected through an elevation angle, a dual-difference residual error threshold is set, phase pseudo range dual-difference residual errors are calculated for the reference satellite and each common satellite, and satellite selection is carried out according to value sizes. According to the scheme, the computational complexity is quite small, the threshold of the dual-difference residual errors can be dynamically adjusted according to different phases in an RTK positioning process, the convergence speed of the RTK positioning is accelerated, the RTK positioning precision is further improved, and the method has great significance on such application occasions with ever improving precision requirements as landslide detection and the like.

The invention belongs to the technical field of ground-air transient electromagnetic measurement by using a magnetic source, and particularly relates to a semi-aviation transient electromagnetic data real-time return method suitable for landslide detection, which comprises the following steps of: quickly flying by using an unmanned aerial vehicle platform, automatically combining channels and superposing in the air to obtain measured data of a KB level; and actually measured data are transmitted back to a ground receiving station in real time through wireless communication, so that the data can be conveniently processed in real time. Compared with a conventional method, the method provided by the invention can provide a quicker interpretation result in landslide geological disaster implementation measurement, and provides a basis for disaster control under an emergency condition of landslide geological disasters.

The invention discloses a landslide monitoring system for railway disaster prevention. The landslide monitoring system comprises a landslide detection system, a monitoring host system and a landslide sensing system, wherein the landslide sensing system senses landslide change information and transmits the landslide change information to the landslide detection system; the landslide detection system processes and converts the received landslide change information into a sensing signal and transmits the sensing signal to the monitoring host system; and the monitoring host system stores and processes the landslide change information. The landslide monitoring system can perform unmanned real-time monitoring on the landslide change information continuously, and gives an alarm before the changes of a roadbed exceed the safety range to effectively prevent accidents such as rain derail from happening.

The invention discloses a novel simple landslide detection device and solves problems of commonly high investment, complex assembling and relatively high operation cost of a landslide detection device in the prior art. According to the scheme, the novel simple landslide detection device comprises an iron protection casing, a solar cell panel, a mobile power source and a sound-light alarm device, wherein an upper surface of the iron protection casing is provided with the solar cell panel, the mobile power source is connected with the solar cell panel and the sound-light alarm device through leads, the novel simple landslide detection device further comprises an underground pile, the iron protection casing is internally provided with a liquid inclination type switch in lead connection with the mobile power source to realize alarm, the liquid inclination type switch is connected with the underground pile, and a lower surface of the liquid inclination type switch is provided with four support columns. The novel simple landslide detection device has properties of simple structure, simple mounting, convenient use and easy maintenance during practical operation, saves a lot of manpower, can satisfy engineering forecast safety requirements and has strong practicality.

The invention discloses a three-dimensional imaging method and system for resistivity data fusion for landslide detection. The technical solution is as follows: obtaining geophysical data according to various geophysical detection lines laid out; preprocessing the geophysical data, and passing Obtain two-dimensional resistivity data by inversion method; extract resistivity data of multiple geophysical prospecting methods at the same position, and fuse multiple resistivity data; establish a three-dimensional coordinate system, and convert the fused two-dimensional detection data into three-dimensional detection Data, the three-dimensional detection data is imaged to obtain a three-dimensional stereogram. The invention fuses the two-dimensional detection data obtained by the inversion of the transient electromagnetic method and the high-density electric method, and obtains the three-dimensional detection data through the three-dimensional coordinate conversion formula, and finally obtains the three-dimensional stereogram after the three-dimensional detection data is imaged, thereby determining the sliding surface in space distribution range.

The invention belongs to the technical filed of radar remote control image processing, and provides an estimation method of a terrain azimuth angle under vegetation cover based on polarimetric synthetic aperture radar (POLSAR). By means of strong penetrability of the POLSAR, a vegetation cover area is observed once, and on the basis of an electromagnetic scattering theory and mathematical modeling, and topographic relief information under the vegetation cover in radar observation data is estimated. Due to the fact that earth surface observation can be carried out all day long in four seasons through the radar, and the topographic relief information under the estimated vegetation cover can provide sufficient guarantees for landslide detection, disaster emergency response, territorial resource annotation and others. Due to the fact that only one POLSAR image is used, economic expenses cost in buying radar data can be greatly reduced, and applicable ranges are greatly increased.