Satellite image feedback-based landslide monitoring apparatus and method, terminal, device, and medium

Abstract

A satellite image feedback-based landslide monitoring apparatus and method, a terminal, a device, and a medium, which relate to the technical field of geological disaster data monitoring. The monitori

Description

[0002] The present invention relates to the technical field of geological hazard data monitoring, and in particular, to a landslide monitoring apparatus, method, terminal, device, and medium based on satellite image feedback. BACKGROUND

[0003] Geological hazards in mountainous areas comprise mainly avalanches, landslides, and mudflows. Systematic research has been carried out on the causal mechanisms of different geological hazards and on a variety of measures for their prevention and control, but it should be noted that the early identification of geological hazards plays a pivotal role in prevention. At present, the most common approach adopted is to monitor the hazards already occurred in site, to prevent their further expansion, and the check of potentially unstable hazard points is mainly carried out through rough manual inspection. However, in the manual inspection, due to the complexity of terrain and geomorphology and experience accumulation of survey staff, some hazard points that may result in disasters are ignored. In addition, in some areas, the cost of manual inspection is too high to allow for full coverage.

[0004] Some survey units use satellite photography for inspection. From a macro application perspective, this method is more generally applied and performs manual interpretation based on the satellite images, which greatly saves costs and improve work efficiency. However, satellite image interpretation poses certain requirements on work experience, which greatly reduces its universality. Even experienced staffs are prone to errors in the interpretation process. This is mainly because different geological hazards are located in different environments and these environments are always changing, thereby increasing the complexity of geological hazard identification. In the existing technology, the identification of geological hazards in mountainous areas is carried out using multi-source and multi-temporal high-resolution satellite data, using a combination of unmanned aerial vehicle (UAV) images and digital elevation models (DEMs), or using live 3D modelling of UAV tilt photogrammetry, and the deformations that can give rise to avalanches are measured and observed on site several times and then are identified.

[0005] Based on the above analysis, the problems and deficiencies of existing technology are as follows. In the process of identifying geological hazard through high-definition satellite images in the existing technologies, the hazards in satellite images are interpreted mainly through a combination of site survey and satellite images, and the key problem of this method is that it can only identify geological hazards that have already occurred, and hazards such as potentially unstable slopes and landslides cannot be discerned, that is, landslide hazards that disaster-causing effects have already occurred can be identified. In addition, as some areas are inaccessible to manual survey, some hazards will fail to be found and thus cannot be interpreted on the satellite images, and an artificial intelligence (AI) learning model that is trained based on the corresponding data has limited ability geological interpretation from satellite images.

[0006] The key problem with using UAV images for geological identification is that they can only identify landslide hazards that have already occurred, and cannot identify potential hazards that are about to occur. Also, an UAV can only identify single landslides. For geological hazard investigations in certain areas, UAVs are expensive and time-consuming to operate, and the subsequent workload of image processing is also high. SUMMARY

[0007] To overcome problems in the related art, the embodiments disclosed by the invention provide a landslide monitoring apparatus, method, terminal, device, and medium based on satellite image feedback. The accuracy and reliability of slope monitoring is improved and costs are saved.

[0008] The described technical solution is as follows. A landslide monitoring apparatus based on satellite image feedback includes: a composite monitoring box configured to monitor indicators of groundwater chemical composition, a pore water pressure, a horizontal displacement, a horizontal stress, a vertical stress, and mineral composition change in a monitored wall, to obtain relevant data at different depth positions of a geological body;

[0009] a data acquisition control system, configured to acquire, aggregate, and store on-site data, cache the on-site data acquired for transmission by a wireless transmission device, wherein aggregated data is transmitted to an indoor satellite image interpretation terminal;

[0010] a flow monitoring system, configured to monitor flow of surface water at different locations on a slope, test and analyse a chemical composition of the surface water and Pondus Hydrogenii (pH) data, integrate flow data acquired with meteorological data, and analyse flow indicators and rainfall indicators at the different locations;

[0011] a laser scanning monitoring system, configured to monitor and analyse a scour pattern of the slope in real time, obtain data on the scour pattern at the different locations on the slope, calculate a scour situation by comparing the data on the scour pattern with an original slope pattern, monitor flow and water level at a ditch bottom, and provide early warning of a dangerous situation occurred; and

[0012] a water level monitoring system, configured to monitor the water level and the flow at the ditch bottom in real time, where the water level and flow monitored are transmitted to the data acquisition control system and compared with rainfall data, and compared and analysed with data obtained from a moisture sensor and a pore pressure sensor inside the slope to obtain a rainfall-flow-pore water pressure relationship.

[0013] In some embodiments, the landslide monitoring apparatus further includes: a power supply system, consist of a solar power supply module and a wind power supply module, using solar power and wind power to supplement electric energy for supplying power to the whole monitoring system;

[0014] where the power supply system adopts automatic frequency conversion control, in a case that a change of any parameter detected continuously over a week by a same sensor is less than or equal to 0.1%, one-third of power of a pass-class sensor is automatically cut off to keep it in a state of standby for detection, and in a case that a fluctuation in current is fed back to the power supply system, the power supply system starts to re-power, and the sensor returns to normal.

[0015] In some embodiments, the landslide monitoring apparatus further includes:

[0016] a wireless transmission antenna, configured to transmit data and receive command; and

[0017] a meteorological monitoring station, configured to monitor meteorological indicators comprising local rainfall, temperature, humidity, wind direction, wind speed, and barometric pressure in real-time, and store and transmit data, wherein the data is eventual transmitted to the data acquisition control system.

[0018] Another object of the present invention is to provide a method for monitoring a potentially unstable landslide hazard including:

[0019] step 1, delineating a selected geological hazard investigation area in a satellite image, and deploying the landslide monitoring apparatus in site based on the delineating;

[0020] step 2, monitoring deformation and stress in slope evolution, feeding real-time data to the indoor satellite image interpretation terminal, and iteratively training an established learning model to establish a correlation between transport elements of the slope in site and pixel changes in the satellite image; and

[0021] step 3, performing monitoring and early warning on a landslide hazard in site based on pre-programmed safety thresholds.

[0022] In one embodiment of the invention, in the step 1, a learning model for satellite image interpretation is trained and learned using monitored data of the potentially unstable landslide hazard, and a potentially unstable slope is identified using the learning model.

[0023] In the step 2, in machine learning, the machine is trained iteratively by setting up a dataset of changes in key indicators such as slope gradient, slope direction, elevation, and rainfall, and the occurrence of landslide hazard is predicted and analysed based on a neural network method to carry out identification of the landslide hazard in site; where the learning model is M = F(xl, x2, x3, ...), where xl, x2, and x3 are the slope gradient, the slope direction, and the rainfall, respectively.

[0024] In the step 3, the safety threshold is determined in conjunction with a safety coefficient of the landslide hazard, where the safety coefficient is a ratio of an anti-slip force to a sliding force or a ratio of an anti-slip moment to a sliding moment, the safety threshold is a function of the safety coefficient, and a law of change of the safety threshold is determined according to a change in the safety coefficient.

[0025] A further object of the present invention is to provide an indoor satellite image interpretation terminal for implementing the method for monitoring the potentially unstable landslide hazard as described.

[0026] A further object of the present invention is to provide a storage medium for receiving user input programs, the stored computer programs causing an electronic device to implement the method for monitoring the potentially unstable landslide hazard as described.

[0027] A further object of the present invention to provide a computer device comprising a memory, a processor, and a computer program stored in the memory, where the computer program, when executed by the processor, causes the processor to perform the method of monitoring the potentially unstable landslide hazard as described.

[0028] In combination with all the technical solutions described above, the advantages and positive effects of the present invention are as follows.

[0029] First, in view of the technical problems of the prior art described above and the difficulty of solving the problem, closely combining the technical solution to be protected by the present invention with the results and data during the research and development process, etc., how the present invention solves the technical problems is analyzed in detail and deeply. The specific description is as follows.

[0030] In the present invention, by providing a landslide monitoring apparatus and technical method based on satellite image feedback, the selected geological hazard investigation area in a satellite image is first circled, and monitoring equipment is deployed on site based on the selected geological hazard investigation area. The monitoring technique in the present invention is able to monitor the core elements of engineering geology (formation lithology, meteorology, geological parameters, hydrology, etc.) in real time. On the one hand, the technique monitors the deformation and stress in slope evolution, feeds real-time data to the indoor satellite image interpretation terminal, and repeatedly trains the learning model established to link the transport elements of the slope with the pixel changes in the satellite images, thus enhancing the accuracy of satellite image interpretation. On the other hand, based on pre-programmed safety thresholds, it monitors and warns of landslide hazards on site, thereby providing local disaster prevention and mitigation services, deeply integrating transportation disaster prevention and mitigation with rural construction disaster prevention and mitigation, and reducing the costs of disaster prevention and mitigation.

[0031] The application of this technique can monitor potentially unstable slopes, thereby solves the problem of not being able to identify impending geological hazards in a satellite image. In addition, the technique provided in this invention is based on the application of big data artificial intelligence identification methods and can provide important support for the identification of landslide geological hazards based on satellite images, improve work efficiency, and prevent geological disasters.

[0032] Second, considering the technical solution as a whole or from the perspective of a product, the technical advantages possessed by the solution to be protected by the present invention are as follows.

[0033] The present invention addresses the shortcomings of previous methods and proposes a landslide monitoring apparatus and method based on satellite image feedback, in which a typical disaster point is selected from satellite images, the monitoring is performed at the typical disaster point, and the monitoring information is fed back to an indoor interpretation terminal for interpretation, which improves the accuracy of the satellite image interpretation results.

[0034] On the one hand, in the present invention, the on-site information of the typical disaster point is timely fed back to the indoor satellite image interpretation terminal, facilitating the smooth implementation of interpretation work. On the other hand, the monitoring apparatus in the present invention, in addition to monitoring the general displacement and stress change characteristics of a disaster, systematically monitors the lithology, geological structure, surface scour, meteorology, groundwater, and other information associated with the disaster, and gives the typical characteristics of the monitored disaster site from the engineering geological perspective. These features are then integrated into specific indicators such as strain and stress, and the deformation and stress indicators are linked to the pixel changes in the satellite images, allowing for more accurate interpretation of potential hazards, effectively improving the accuracy and reliability of hazard identification. In addition, based on accurate information from monitoring and artificial intelligence, machine learning is applied to an indoor interpretation method, and later the artificial intelligence method is used to interpret the satellite images of typical disaster points, greatly improving the work efficiency.

[0035] Third, as supporting evidence for the inventiveness of the claims of the present invention, the technical solution of the present invention fills a gap in the industry in China and abroad. The implementation of the present invention brings great convenience to geological disaster investigations, enables large savings in terms of human and material resources, and enables region-wide geological disaster investigation and ranking, providing important support for the safe construction and operation of highway projects and other industrial and civil engineering projects. At the same time, the technological basis of the invention is conducive to its integration into current intelligent traffic information systems, bridging the gap in this area. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] FIG. 1 is a flow diagram of a method for monitoring potentially unstable landslide hazards according to embodiments of the present invention.

[0037] FIG. 2 is a schematic diagram of the method for monitoring potentially unstable landslide hazards according to embodiments of the present invention.

[0038] FIG. 3 is a schematic diagram of a landslide monitoring apparatus based on satellite image feedback according to embodiments of the present invention.

[0039] FIG. 4 is a schematic diagram of the composite monitoring box according to embodiments of the present invention.

[0040] FIG. 5 is a schematic diagram of the flow monitoring system according to embodiments of the present invention.

[0041] Reference numerals: 1, composite monitoring box; 1-1, automatic shear system; 1-2, loading device; 1-3, data collector; 1-4, integrated monitor; 1-5, water chemistry monitoring sensor; 1-6, pore water pressure sensor; 1-7, mineral composition monitoring sensor; 1-8, moisture sensor; 1-9, horizontal displacement monitoring sensor; 1-10, vertical displacement monitoring system; 2, data acquisition control system; 3, power supply system; 4, slope top; 5, ditch bottom; 6, flow monitoring system; 6-1, seepage solute test sensor; 6-2, water temperature monitoring sensor; 6-3, pH monitoring sensor; 6-4, anti-silt flushing system; 6-5, water storage tank inlet; 6-6, water inlet; 6-7, water outlet; 6-8, data memory; 6-9, monitoring water tank; 6-10, tank overflow channel; 6-11, flow monitoring sensor; 6-12, particle composition monitoring system; 7, laser scanning monitoring system; 8, wireless transmission antenna; 9, water level monitoring system; and 10, meteorological monitoring station. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In order to make the above-mentioned objectives, features, and advantages of the invention more obvious and comprehensible, specific embodiments of the invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set out in order to facilitate a full understanding of the invention. However, the invention can be implemented in many other ways than those described herein, and improvements could be made by those skilled in the art without contradicting the content of the invention, and thus the invention is not limited by the specific embodiments disclosed below.

[0043] I. Explanatory description of embodiments

[0044] The method for monitoring a potentially unstable landslide hazard in embodiments of the present invention innovatively provides a machine learning model training method based on monitoring of potentially unstable landslide hazards.

[0045] The invention also innovatively proposes a landslide monitoring apparatus based on the satellite image feedback, which greatly improves the accuracy of landslide hazard discrimination by analysing factors such as changes in lithological and mechanical conditions, groundwater seepage, water chemistry, and mineral composition.

[0046] The method for monitoring the potentially unstable landslide hazards according to embodiments of the invention is based on learning from monitoring data and simultaneously, analysing the deformation of geological bodies and regional meteorological data, etc., to determine the critical conditions for instability of geological bodies from multiple perspectives and to improve the accuracy of the training model in identifying hazardous geological bodies. In the process of identification, geological hazard points that have already occurred on site are used for reversed verification, and the generalisability of the invention is improved through the training of several models.

[0047] Example 1

[0048] As shown in FIG. 1, the method for monitoring the potentially unstable landslide hazards according to embodiments of the present invention comprises steps S101-S103.

[0049] In step S101, a selected geological hazard investigation area in a satellite image is delineated, on the basis of which a landslide monitoring apparatus based on the feedback from the satellite image is deployed on site.

[0050] In step SI02, core engineering geological factors such as formation lithology, meteorology, geological parameters, and hydrology are monitored in real-time, and deformation and stress in slope evolution are monitored, then the real-time data is fed back to the indoor satellite image interpretation terminal, and the established learning model is iteratively trained to establish a link between the transport elements of the slope in the site and the pixel changes in the satellite images, thereby enhancing the accuracy of satellite image interpretation.

[0051] In step S103, based on pre-programmed safety thresholds, monitoring and warning of on-site landslide hazards are conducted, thereby providing local disaster prevention and mitigation services, deeply integrating transportation disaster prevention and mitigation with rural construction disaster prevention and mitigation, and reducing costs of disaster prevention and mitigation.

[0052] In step SI02, in machine learning, the machine is trained iteratively by setting up datasets of changes in key indicators such as slope gradient, slope direction, elevation and rainfall, and the occurrence of landslide hazards is predicted and analysed based on a neural network method to carry out the identification of landslide hazards in site. The learning model isM= F(xl, x2, x3,...) , where xl, x2, and x3 are key indicators such as slope gradient, slope direction, and rainfall.

[0053] The occurrence of landslide hazards is predicted and analysed based on the neural network method to carry out identification of landslide hazards of site, which includes the following.

[0054] Based on changes of the indicators such as slope gradient, slope direction, and rainfall obtained from on-site monitoring, combined with the displacement in the remote sensing satellite image, the deformation state of the slope is determined, and based on the deformation state and deformation rate, the stability of the landslide is analysed, and in turn the potentially unstable slopes can be identified accurately.

[0055] In the prediction process, on the basis of a large number of identifications in the early stage, continuous training is conducted, specifically repeated training is carried out based on the data set; and typical area identification is combined with manual discrimination to improve the accuracy of identification, and latterly, the established methods are used for potential ground hazard identification.

[0056] In step S103, the safety threshold is set in combination with the safety coefficient of the landslide hazard, where the safety coefficient is the ratio of the anti-slip force or anti-slip moment to the sliding force or sliding moment, the safety threshold is a function of the safety coefficient, and the law of change of the safety threshold is determined according to the changes in the safety coefficient:

[0057] F = N / T, where F is the safety coefficient; N is the anti-slip force or anti-slip moment, kN; and Zis the sliding force or sliding moment, kN.

[0058] In the embodiment of the present invention, the satellite image interpretation learning model is trained and learned using monitoring data on potential unstable landslide hazards, which is very different from the training basis of traditional learning models. The learning model can be used to accurately identify potential unstable slopes at a later stage to improve disaster prevention and mitigation, which breaks through the technical bottleneck of the prior art that only already happened hazards are identified.

[0059] In the embodiment of the present invention, the proposed landslide monitoring apparatus based on the satellite image feedback is improved based on the traditional monitoring method. In addition to the deployment of sensors for monitoring displacement, stress, and pore pressure, an additional monitoring system for hydrology, water chemistry, and mineral composition changes is incorporated to reveal the inner mechanism responsible for landslide disasters from the perspective of material composition changes and groundwater seepage field changes. The accuracy and reliability are significantly better than those of existing landslide monitoring techniques and methods. The ultimate aim of this monitoring apparatus and method is to achieve early warning of potential disasters and to aid in the local disaster prevention and mitigation efforts.

[0060] In the embodiment of the present invention, in terms of changes in material composition, the main focus is on the content of clay minerals inside the geological body. An increase in clay minerals is extremely detrimental to the stability of a geological body, which may easily lead to landslide disasters under the influence of triggering factors such as groundwater or earthquakes, and thus has a disaster-causing effect. The formation of clay minerals is closely linked to changes in the composition of certain anions and cations and mineral composition in groundwater, so the underlying mechanisms of geological hazard formation can be revealed from the perspective of material composition and groundwater changes.

[0061] Example 2

[0062] As shown in FIG. 2, the method for monitoring potentially unstable landslide hazards according to embodiments of the present invention comprises the following:

[0063] (1) selecting study area based on satellite image, in which satellite images are used to select the study area where the geological hazard investigation is to be carried out;

[0064] (2) on-site monitoring, in which several typical hazard sites in the study area are selected for the deployment of monitoring apparatus^.

[0065] (3) data acquisition, in which various types of data is obtained through the on-site monitoring, and the data is transmitted to the indoor satellite image interpretation terminal;

[0066] (4) satellite image identification, in which a machine learning is carried out based on the monitored data, and using the learned model and artificial intelligence, accurate identification and batch identification of landslide hazards are carried out in the study area, for disaster prevention and mitigation services.

[0067] In FIG. 2, the potentially unstable geological body is taken as the research object. The landslides or collapses only occurs after a geological body has become unstable. The potentially unstable geological body is monitored repeatedly, in which the traditional monitoring data, such as displacement, stress, and meteorological data, also are key elements for the machine learning in a later stage in addition to data from the on-site monitoring facilities in the present invention. Further, the apparatus proposed in the present invention also can monitor accurately situations such as the flow generated by surface water and the scouring of surfaces of geological bodies.

[0068] Example 3

[0069] As shown in FIG. 3, the landslide monitoring apparatus based on satellite image feedback according to the embodiment of the present invention comprises the following:

[0070] Composite monitoring box 1: The device contains numerous monitoring systems for monitoring indicators such as groundwater chemical composition, pore water pressure, horizontal displacement, horizontal and vertical stresses, and mineral composition changes in a monitored wall. Many sensors are integrated for deployment, which can reduce the number of holes that need to be excavated for the on-site monitoring, greatly improving the installation efficiency due to high level of integration. The monitoring box is arranged vertically to obtain relevant data at different depths. When the parameters of adjacent layers change significantly, in combination with other data, a comprehensive determination is made on whether the geological body will be destabilised.

[0071] Data acquisition control system 2: The system is mainly for the collection and aggregation of the on-site data, have a preliminary data storage function, and can also cache data collected in the site for subsequent transmission through the wireless transmission device. The data are aggregated and transmitted to the indoor satellite image interpretation terminal for being repeatedly learned by the indoor satellite image interpretation terminal, to enable artificial intelligence-based landslide disaster identification, thereby improving the accuracy of disaster identification. In addition, the system has data fusion and processing functions, which can compare and analyse the data collected by underground sensors and data collected by weather stations, to establish correlation therebetween through built-in algorithms, for analysis by monitors.

[0072] Power supply system 3: The system mainly consists of two modules: solar power supply module and wind power supply module, which supplements electric energy by using solar power and wind power and has the function of storing electric energy, for supplying power to a whole monitoring system. The power supply system 3 uses automatic frequency conversion control, when a change of any parameter detected continuously over a week by a same sensor is less than or equal to 0.1%, the power supply system 3 automatically cuts off power of one-third of the sensors for communication to keep it in a state of standby for detection. If any fluctuation in current is fed back to the power supply system 3, the power supply system 3 starts to re-power, so that the sensors returns to normal. This process can reduce power loss during monitoring, and reduces the load on the memory, considering that the repeated collection of through-class data has limited significance for later analysis. Such process set in the present invention has a significant innovation, and is different from the existing technology that turns on all sensors at the same time for monitoring.

[0073] Slope top 4: The slope top 4 is subjected to simple levelling for monitoring.

[0074] Ditch bottom 5: It is positioned where the shady and sunny slopes meet.

[0075] Flow monitoring system 6: The system monitors the flow of surface water at different locations on the slope. The monitoring system is provided with a water storage tank, and is used to test and analyse data such as the chemical composition of the surface water and Pondus Hydrogenii (pH) by combined with the relevant sensors, and to integrate flow data collected with meteorological data to analyse the link among indicators such as flow and rainfall at different locations.

[0076] Laser scanning monitoring system 7: The main function of the system 7 is to real-time monitor and analyse the scour pattern of the slope. Particularly, the system obtains data such as the scour pattern at different locations of the slope, calculates the scour situation by comparing the data with the original slope pattern, and simultaneously, simply monitors the flow and water level at the ditch bottom 5, etc. The system is also equipped with an early warning system for alerting the emergence of dangerous situations, in which an alarm amplifier is set up for alarming, and instructions are issued by the data acquisition control system 2.

[0077] Wireless transmission antenna 8: The antenna is mainly used for data transmission and command reception.

[0078] Water level monitoring system 9: The system is mainly for real-time monitoring of the water level and flow at the ditch bottom 5. These data are transmitted to the data acquisition control system 2, then are compared with rainfall and other data, and are compared and analysed with data from the moisture and pore pressure sensors inside the slope body. Based on the above, a rainfall-flow-pore water pressure relationship is established, thereby laying a foundation for the determination of potential instability of the geological body.

[0079] Meteorological monitoring station 10: The station mainly monitors local meteorological indicators such as rainfall, temperature, humidity, wind direction, wind speed, and air pressure in real time, stores and then transmits the data to the data acquisition control system 2.

[0080] Example 4

[0081] Based on the landslide monitoring apparatus based on satellite image feedback according to embodiment 3 of the present invention, as shown in FIG 4, the composite monitoring box 1 comprises the following:

[0082] Automatic shear system 1-1: The main function of the system 1-1 is to measure the mechanical parameters of the geological body at a certain underground layer. The system is mainly based on the principle of a vane shear meter, and configured for monitoring and analysing the initial shear strength parameters and changes thereof in real time and transmitting then to the data collector 1-3.

[0083] Loading device 1-2: The main function of the loading device 1-2 is to apply a load in the shear test, and control the buried depth of the shear plate. Since a new shear zone is formed for each shearing, in the subsequent shearing after the current shearing, the shear plate needs to be loaded deeper into the un-sheared soil for a second shearing (or multiple shearings) to ensure that accurate test results are obtained.

[0084] Data collector 1-3: The main function of the data collector 1-3 is to aggregate and store data from the entire monitoring box and transmit the data to the data acquisition control system 2.

[0085] Integrated monitor 1-4: The wall material of the box containing the integrated monitor 1-4 has a certain degree of rigidity and is waterproof, which is obtained by custom machining.

[0086] Water chemistry monitoring sensor 1-5: The sensor monitors the water chemistry characteristics of the layer. If the formation is located below the water table, the sensor can be used directly for monitoring, and if the formation in which the sensor is located is above the water table, the sensor monitors the water chemistry composition including parameters such as anions, cations and pH during water migration.

[0087] Pore water pressure sensor 1-6: The sensor monitors the pore water pressure in the geological body of the layer, and provides the core parameters for calculation of effective stress throughout the geological body.

[0088] Mineral composition monitoring sensor 1-7: The sensor monitors and analyses the mineral composition such as clay minerals, quartz, and feldspar of the layer.

[0089] Moisture sensor 1-8: The sensor monitors the moisture content of the layer, to reflect conditions such as moisture migration within the geological body.

[0090] Horizontal displacement monitoring sensor 1-9: When the geological body undergoes horizontal displacement, the monitoring box as a whole is subject to distortion, and the horizontal displacement is monitored by the horizontal displacement monitoring sensor 1-9, to obtain horizontal displacement deformation indicators.

[0091] Vertical displacement monitoring system 1-10: When the geological body undergoes vertical settlement, this sensor can collect vertical displacement in real-time.

[0092] Example 5

[0093] Based on the landslide monitoring apparatus based on satellite image feedback according to embodiment 3 of the present invention, as shown in FIG 5, the flow monitoring system 6 comprises the following:

[0094] Seepage solute test sensor 6-1: The sensor mainly monitors the chemical composition of the surface water, including the type and content of anions and cations in the surface water.

[0095] Water temperature monitoring sensor 6-2: The sensor mainly monitors the temperature of the surface water in real time.

[0096] pH monitoring sensor 6-3: The sensor monitors the pH of the surface water to obtain data on changes in the chemical environment of the surface water.

[0097] Anti-silt flushing system 6-4: The main function of the system 6-4 is to periodically flush the monitoring water tank 6-9, and remove solid material remaining at the bottom by flushing force, to ensure normal operation of the tank.

[0098] Water storage tank inlet 6-5: The main function of the tank inlet 6-5 is to allow water to enter the water storage tank. The tank has a built-in solenoid valve that can automatically close or open as needed.

[0099] Water inlet 6-6: Through the water inlet, water flows into the flow monitoring system 6, and the water inlet allows groundwater to enter a system channel.

[0100] Water outlet 6-7: Through the water outlet, water in the flow monitoring system 6 is discharged.

[0101] Data memory 6-8: The memory mainly collects and stores the data collected by the sensors in the flow monitoring system 6 and finally transmits the data to the data acquisition control system 2 in FIG. 3.

[0102] Monitoring water tank 6-9: The monitoring water tank mainly serves monitoring sensors for the surface water, allows sensors to be fixed and installed thereon, and storing water for testing, to ensure stability of the test results.

[0103] Water tank overflow channel 6-10: Excess water in the monitoring water tank 6-9 can be discharged through the water overflow channel 6-10.

[0104] Flow monitoring sensor 6-11: The sensor mainly monitors flow of the surface water at the monitoring location to provide basic data for later slope scour analysis.

[0105] Particle composition monitoring system 6-12: The main function of the system is to test the soil particles carried in the surface water at the monitoring location to obtain the particle size of substances carried in the water. The system has an automatic flushing function for flushing particles remaining at the bottom away after each test, to ensure a clean flow channel.

[0106] In the above embodiments, the description of each embodiment has its own focus and the parts that are not detailed or documented in one embodiment can be found in the relevant descriptions of other embodiments.

[0107] The information interaction between the above devices / units, the execution process, etc., are based on the same concept as the method embodiment of the invention. Their specific functions and technical effects can be found in the method embodiment section and will not be repeated here.

[0108] It will be clear to those skilled in the field that, for the sake of convenience and brevity of description, the above-mentioned divisions of each functional unit and module are only given as examples. In practice, the above-mentioned functions can be assigned to be performed by different functional units and modules as required, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The various functional units and modules in the embodiment can be integrated in a single processing unit, or be physically exist separately, or two or more units can be integrated in a single unit. The integrated units can be implemented either in the form of hardware or in the form of software functional units. Furthermore, the specific names of the functional units and modules are only for the purpose of distinguishing them from each other and are not intended to limit the scope of protection of the present invention. For the specific working processes of the units and modules in the above system, reference can be made to the corresponding processes in the aforementioned method embodiments, and these will not be repeated here.

[0109] II. Application examples

[0110] Application example 1

[0111] The monitoring apparatus provided by the embodiment of the present invention is used to investigate geohazards in the Bailong River Basin. The Bailong River Basin is selected using satellite images, several typical disaster sites are selected in this basin, the on-site monitoring apparatus is deployed, on-site monitoring data are collected and collated, these data are transmitted to the indoor terminal, and the changes in the slope stress, displacement, and other data collected at the site are used to determine the key indicators of slope, rainfall, etc.. These data are input into the system and the learning model is repeatedly trained to identify landslide hazard as typical hazard points. The key pixel points are extracted from the satellite images during the model training, and the pixel points are linked to the changes in displacement, stress, and other indicators monitored in the site. The trained model is used to identify other similar disaster points within the entire watershed, which can greatly save manpower and resources.

[0112] Application example 2

[0113] The monitoring method provided by this embodiment of the present invention is run on a computer device. The computer device comprises at least one processor, a memory, and a computer program stored in the memory and runnable on the processor (or processors), and the processor executes the computer program to implement the steps in any of the method embodiments described above.

[0114] Application example 3

[0115] The monitoring method provided by this embodiment of the present invention is run on a computer readable storage medium, the computer readable storage medium has a computer program stored on it, and the computer program when executed by the processor implements the steps in each of the method embodiments described above.

[0116] Application example 4

[0117] The monitoring method provided by this embodiment of the invention runs on an information data processing terminal. The information data processing terminal is used to provide a user input interface to perform the steps in each of the above method embodiments when implemented on an electronic device, and the information data processing terminal not is limited to a mobile phone, a computer, or a switch.

[0118] Application example 5

[0119] The monitoring method provided by this embodiment of the invention runs on a server. The server is used to provide a user input interface to perform the steps in the method embodiments as described above when implemented on an electronic device.

[0120] Application example 6

[0121] The monitoring method provided by this embodiment of the present invention runs on a computer program that, when runs on an electronic device, enables the electronic device to perform the steps in each of the method embodiments described above.

[0122] The integrated unit may be stored in a computer readable storage medium if implemented as a software functional unit and sold or used as a standalone product. Based on this understanding, the present invention implements all or part of the processes in the method of the above embodiments, which may be accomplished by means of a computer program to instruct the relevant hardware. The computer program may be stored in a computer readable storage medium that, when executed by a processor, implements the steps in the respective method embodiments described above. Where the computer program comprises computer program code, and the computer program code may be in the form of source code, in the form of object code, in the form of an executable file, or in some intermediate form, etc. The computer readable medium may comprise at least any entity or device capable of carrying the computer program code to a photographic device / terminal device, a recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, mobile hard drives, disks, and CD-ROMs.

[0123] III. Evidence of effects associated with embodiments

[0124] The monitoring apparatus and method provided by the embodiments of the present invention integrates the application of air-sky integration technology in the process of use for the monitoring and identification of landslide hazards. In the monitoring landslide hazards, the satellite is used for precision positioning, the discrimination is performed combined with real-time ground monitoring, and indoor terminals collect aggregated data is used for learning and issuing identification commands to improve the accuracy of the whole identification technology. In the process, remote sensing geology, disaster geology, and sensing technology are effectively integrated, multidisciplinary crossover and mutual verification are carried out, greatly improving the functionality and efficiency of the whole monitoring system.

[0125] By using satellite images to identify a landslide on site, the scope of the landslide is initially determined, and monitoring is carried out on site, which is focused on analysis of changes in groundwater and material composition. The occurrence of a landslide is the result of the formation of a slip zone, which is the macroscopic expression of the reduction of geological indicators (cohesion and friction) at a certain depth within the landslide body.

[0126] The above mentioned is only a specific and preferred embodiment of the present invention, but the scope of protection of the present invention is not limited to it, and any modification, equivalent substitution, and improvement, etc., made by any person skilled in the art within the technical scope disclosed by the present invention and within the spirit and principles of the present invention shall be covered by the scope of protection of the present invention. 09 01 26

Claims

1. A landslide monitoring apparatus based on satellite image feedback, comprising:a composite monitoring box (1), configured to monitor indicators of groundwater chemical composition, a pore water pressure, a horizontal displacement, a horizontal stress, a vertical stress, and mineral composition change in a monitored wall, to obtain relevant data at different depth positions of a geological body;a data acquisition control system (2), configured to acquire aggregate, and store on-site data, cache the on-site data acquired for transmission by a wireless transmission device, wherein aggregated data are transmitted to an indoor satellite image interpretation terminal;a flow monitoring system (6), configured to monitor flow of surface water at different locations on a slope, test and analyse a chemical composition of the surface water and Pondus Hydrogenii (pH) data of the surface water, integrate flow data acquired with meteorological data, and analyse flow indicators and rainfall indicators at the different locations;a laser scanning monitoring system (7), configured to monitor and analyse a scour pattern of the slope in real time, obtain data on the scour pattern at the different locations on the slope, calculate a scour situation by comparing the data on the scour pattern with an original slope pattern, monitor flow and water level at a ditch bottom (5), and provide early warning of a dangerous situation about to occur; anda water level monitoring system (9), configured to monitor the water level and the flow at the ditch bottom (5) in real time, wherein the water level and flow monitored are transmitted to the data acquisition control system (2) and compared with rainfall data, and compared and analysed with data obtained from a moisture sensor (1-8) and a pore pressure sensor inside the slope to obtain a rainfall-flow-pore water pressure relationship.

2. The landslide monitoring apparatus according to claim 1, further comprising:a wireless transmission antenna (8), configured to transmit data and receive command; anda meteorological monitoring station (10), configured to monitor meteorological indicators comprising local rainfall, temperature, humidity, wind direction, wind speed, and barometric pressure in real time, and store and transmit data, wherein the data is09 01 26eventual transmitted to the data acquisition control system (2).

3. The landslide monitoring apparatus according to claim 1, wherein the composite monitoring box (1) comprises:an automatic shear system (1-1), configured to measure mechanical parameters of the geological body at a certain underground layer, wherein initial shear strength parameters and their changes are monitored and analysed in real time, and transmitted to a data collector (1-3);a loading device (1-2), configured to apply a load during a shear test and control a buried depth of a shear plate;the data collector (1-3), configured to aggregate and store data throughout the composite monitoring box and transmit the data to the data acquisition control system (2);an integrated monitor (1-4), configured for water resistance;a water chemistry monitoring sensor (1-5), configured to monitor water chemistry characteristics of a geological formation, wherein in a case that a rock formation is located below a water table, the water chemistry characteristics of the formation is directly monitored; and in a case that the rock formation where a sensor is located is above the water table, a water chemistry composition during water migration is monitored;a pore water pressure sensor (1-6), configured to monitor pore water pressure in the geological body of the geological formation, and provide core parameters for calculation of effective stresses throughout the geological body;a mineral composition monitoring sensor (1-7), configured to monitor and analyse a mineral composition of the geological formation;the moisture sensor (1-8), configured to monitor moisture content of the geological formation to reflect moisture migration within the geological body;a horizontal displacement monitoring sensor (1-9), configured to monitor horizontal displacement when the horizontal displacement is generated in the geological body, to obtain horizontal displacement deformation indicators; anda vertical displacement monitoring system (1-10), configured to acquire vertical displacement in real time when the vertical settlement is generated in the geological body.

4. The landslide monitoring apparatus according to claim 1, wherein the flow09 01 26monitoring system (6) comprises:a seepage solute test sensor (6-1), configured to monitor the chemical composition of the surface water, comprising a type and content of anions and cations in the surface water;a water temperature monitoring sensor (6-2), configured to monitor temperature of the surface water in real time;a pH monitoring sensor (6-3), configured to monitor the pH of the surface water to obtain data on changes in a chemical environment of the surface water;an anti-silt flushing system (6-4), configured to periodically flush a monitoring tank (6-9);a water storage tank inlet (6-5), configured to allow water to enter a water storage tank, wherein the water storage tank inlet (6-5) is equipped with a built-in solenoid valve that is automatically closed or opened as required;a water inlet (6-6), through which water flows into the flow monitoring system (6), and confirming for allowing groundwater to enter a system channel;a water outlet (6-7), through which water is discharged from the flow monitoring system (6);a data memory (6-8), configured to collect and store data acquired by sensors in the flow monitoring system (6) and eventual transmit the data to the data acquisition control system (2);a monitoring water tank (6-9), configured to serve monitoring sensors for the surface water, allow sensors to be fixed and installed thereon, and store water for testing;a tank overflow channel (6-10), configured to discharge excess water from the monitoring water tank (6-9);a flow monitoring sensor (6-11), configured to monitor flow of the surface water at a monitoring location; anda particle composition monitoring system (6-12), configured to test soil particles carried in the surface water at the monitoring location to obtain particle sizes of substances carried in the surface water.

5. A method for monitoring a potentially unstable landslide hazard using the landslide monitoring apparatus based on satellite image feedback according to any one of claims 1 to 4, comprising:09 01 26step 1, delineating a selected geological hazard investigation area in a satellite image, and deploying the landslide monitoring apparatus in site based on the delineating;step 2, monitoring deformation and stress in slope evolution, feeding real-time data to the indoor satellite image interpretation terminal, and iteratively training an established learning model to establish a correlation between transport elements of the slope in site and pixel changes in the satellite image; andstep 3, performing monitoring and early warning on a landslide hazard in site based on pre-programmed safety thresholds.

6. The method according to claim 5, further comprising:in the step 1, training and learning a learning model for satellite image interpretation using monitored data of the potentially unstable landslide hazard, and identifying a potentially unstable slope using the learning model;in the step 2, in machine learning, iteratively training the machine by setting up a dataset of changes in key indicators comprising slope gradient, slope direction, elevation, and rainfall, and predicting and analysing occurrence of the landslide hazard based on a neural network method to carry out identification of the landslide hazard in site;wherein the learning model is M= F(xl, x2, x3, ...), where xl, x2, and x3 are the slope gradient, the slope direction, and the rainfall, respectively; andthe predicting and analysing occurrence of the landslide hazard based on a neural network method to carry out identification of the landslide hazard in site comprises:determining a deformation state of the slope based on changes in the slope gradient, the slope direction, and the rainfall obtained from on-site monitoring, in combination with a displacement in a remote sensing satellite image, analysing stability of a landslide based on the deformation state and a deformation rate, and carrying out identification of the potentially unstable slope; andin the step 3, determining the safety threshold in combination with a safety coefficient of the landslide hazard, wherein the safety coefficient is a ratio of an anti-slip force to a sliding force or a ratio of an anti-slip moment to a sliding moment, the safety threshold is a function of the safety coefficient, and a law of change of the safety threshold is determined according to a change in the safety coefficient:F = N / T,where F is the safety coefficient, N is the anti-slip force or anti-slip moment, kN, and T is the sliding force or the sliding moment, kN.

7. An indoor satellite image interpretation terminal for implementing the method for monitoring the potentially unstable landslide hazard according to claim 5.

8. A storage medium for receiving user input programs, stored computer programs enabling an electronic device to implement the method for monitoring the potentially unstable landslide hazard according to claim 5.

9. A computer device, comprising:a memory,a processor, anda computer program stored in the memory, wherein the computer program, when executed by the processor, causes the processor to perform the method for monitoring the potentially unstable landslide hazard according to claim 5.CM

Images