Multi-source data processing method and system for intelligent monitoring of geological disasters

Abstract

The invention relates to the technical field of data processing, in particular to a multi-source data processing method and system for intelligent monitoring of geological disasters. Taking debris flow geological disasters as an example, firstly, real-time monitoring data of a to-be-monitored area is acquired through a data acquisition assembly, and the real-time monitoring data comprises a soil humidity value, a subterranean acoustic signal, rainfall and an aerial image; then determining a debris flow risk level of the to-be-monitored area based on the real-time monitoring data, when the debris flow risk level is high risk or medium risk, indicating that the possibility of occurrence of debris flow disasters in the to-be-monitored area is high, further determining whether mutually contradictory real-time monitoring data exists at present, and if the mutually contradictory real-time monitoring data exists, determining that the possibility of occurrence of debris flow disasters in the to-be-monitored area is high; if yes, the debris flow risk level of the to-be-monitored area is determined again based on the contradictory real-time monitoring data and the priority value of each real-time monitoring data, and the re-determined debris flow risk level is more accurate.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a multi-source data processing method and system for intelligent monitoring of geological disasters. Background Art

[0002] Climate change is leading to more frequent extreme rainfall events, exacerbating the risk of geological disasters such as debris flows and landslides. Furthermore, traditional monitoring and early warning methods are ineffective in addressing the increased invisibility, chain reaction, and remote impacts of geological hazards caused by extreme rainfall. However, extreme rainfall-induced clustered geological disasters, flash floods, and debris flows, among other chain reactions, continue to occur, causing significant casualties and property losses. Effective and accurate monitoring and warning of these geological hazards, thereby minimizing losses, is a pressing research area.

[0003] In existing disaster monitoring and early warning programs, it is necessary to use multiple acquisition devices to collect data from multiple different sources. These data are then summarized and analyzed to determine the risk of geological disasters. However, due to the large number of data sources, there may be differences or even contradictions between the data from different sources, resulting in inaccurate or erroneous geological disaster risks. How to process these contradictory and collected data from different sources to obtain accurate geological disaster prediction results is a technical solution that is currently urgently needed. Summary of the Invention

[0004] The main purpose of the present invention is to provide a multi-source data processing method and system for intelligent monitoring of geological disasters, aiming to solve the problem of how to process contradictory collected data from different sources to obtain accurate geological disaster prediction results.

[0005] The technical solution proposed by the present invention is: A multi-source data processing method for intelligent monitoring of geological disasters is applied to a multi-source data processing system for intelligent monitoring of geological disasters; the system includes a central server and a data acquisition component communicatively connected to the central server; the number of the data acquisition components is multiple; the data acquisition components are arranged in the area to be monitored; the method includes: The central server obtains real-time monitoring data collected by the data collection component in the monitored area, wherein the real-time monitoring data includes soil moisture value, underground acoustic signal, rainfall and aerial image; The central server determines the debris flow risk level of the monitored area based on the real-time monitoring data, where the debris flow risk level is any one of high risk, medium risk and low risk; When the debris flow risk level is high or medium, the central server determines whether there are conflicting real-time monitoring data; If so, the central server obtains the priority values of each real-time monitoring data, and re-determines the debris flow risk level of the monitored area based on the conflicting real-time monitoring data and the priority values of each real-time monitoring data, and marks it as the final risk level; If not, the central server will mark the debris flow risk level of the monitored area determined based on the real-time monitoring data as the final risk level.

[0006] Preferably, the data acquisition component includes: a drone equipped with a camera for collecting aerial images, a geoacoustic sensor for collecting underground acoustic signals, a humidity sensor for collecting soil moisture values, and a meteorological sensor for collecting rainfall; the priority values of the monitoring data are, from largest to smallest, aerial images, underground acoustic signals, soil moisture values, and rainfall; the central server determines the debris flow risk level of the monitored area based on the real-time monitoring data, including: The central server obtains the soil type of the area to be monitored, determines the warning humidity value of the area to be monitored based on the soil type of the area to be monitored, and determines the warning level corresponding to the soil humidity value based on the comparison between the warning humidity value and the real-time monitoring data; The central server obtains the warning frequency range and warning sound intensity, and determines the warning level corresponding to the underground sound signal based on the comparison of the warning frequency range and warning sound intensity with the real-time monitoring data; The central server obtains a terrain type of the area to be monitored, determines a first rainfall threshold and a second rainfall threshold for the area to be monitored based on the terrain type of the area to be monitored, and determines a warning level corresponding to the rainfall based on a comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data, wherein the second rainfall threshold is less than the first rainfall threshold; The central server determines the warning level corresponding to the aerial image; The central server determines the debris flow risk level of the monitored area as the highest risk level among the warning levels corresponding to the soil moisture value, underground acoustic signal, rainfall and aerial images, where the warning level is any one of high risk, medium risk and low risk.

[0007] Preferably, determining the warning level corresponding to the soil moisture value based on the comparison between the warning humidity value and the real-time monitoring data includes: When the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period is greater than the first preset value, the central server determines that the warning level corresponding to the soil moisture value is high risk; when the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period is greater than the second preset value, the central server determines that the warning level corresponding to the soil moisture value is medium risk, wherein the second preset value is less than the first preset value; otherwise, the central server determines that the warning level corresponding to the soil moisture value is low risk; The step of determining the warning level corresponding to the underground noise signal based on the comparison of the warning frequency range and the warning sound intensity with the real-time monitoring data includes: The central server obtains the duration corresponding to when the frequency value of the next acoustic signal falls into the warning frequency range within the second preset time period in the past, and marks it as the first target duration; The central server obtains the time duration corresponding to when the intensity value of the next acoustic signal is greater than the intensity of the warning sound within the second preset time duration, and marks it as the second target time duration; When the first target duration and the second target duration are both greater than or equal to the third preset duration, the central server determines that the warning level corresponding to the underground subaerial signal is high risk; When the first target duration or the second target duration is greater than or equal to the third preset duration, the central server determines that the warning level corresponding to the underground subaerial signal is medium risk; When the first target duration and the second target duration are both less than the third preset duration, the central server determines that the warning level corresponding to the underground subacoustic signal is low risk.

[0008] Preferably, determining the warning level corresponding to the rainfall based on comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data includes: When the rainfall in the past first preset time period is greater than the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is high risk; when the rainfall in the past first preset time period is greater than the second rainfall threshold and less than or equal to the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is medium risk; otherwise, the central server determines that the warning level corresponding to the rainfall is low risk.

[0009] Preferably, when the debris flow risk level is high risk or medium risk, the central server determines whether there are conflicting real-time monitoring data, including: When the debris flow risk level is high or medium, the central server determines whether any of the warning levels corresponding to the soil moisture value, the underground acoustic signal, the rainfall, and the aerial image is low risk. If so, the central server determines that there are conflicting real-time monitoring data; The central server obtains the priority value of each real-time monitoring data, and re-determines the debris flow risk level of the monitored area based on the conflicting real-time monitoring data and the priority value of each real-time monitoring data, and marks it as the final risk level, including: The central server marks the real-time monitoring data corresponding to the warning level of soil moisture value, underground acoustic signal, rainfall, and aerial image as target data. The central server marks the real-time monitoring data with the highest priority value among the target data as priority monitoring data; The central server will prioritize the warning level of the monitoring data as the final risk level.

[0010] Preferably, the drone further includes a controller for controlling the flight of the drone body and activating the camera; the data acquisition component further includes an indicator, the indicator including a connecting rod and an indicator light provided on the top of the connecting rod; the connecting rod is used to be vertically embedded in the ground of the area to be monitored; the indicator light is above the ground; the indicator lights of the indicators of the indicators in the same sub-area are different in color; the area to be monitored is divided into a plurality of square sub-areas of uniform area; a plurality of indicator lights are provided in each sub-area and distributed in a matrix, and the number of indicator lights in each sub-area is uniform; the controller can control the start and stop of each indicator light through wireless communication; the central server determines the warning level corresponding to the aerial image, including: The central server sends the aerial location points of each sub-area to the drone; The controller controls the drone to fly to the aerial photography position point of the sub-area every fourth preset time period to perform a photography task, including: When the drone flies to the aerial photography location of a sub-area and hovers, the controller turns on the indicator lights of the indicator components in the sub-area and turns off the indicator lights of all other indicator components in the sub-areas. The drone then starts the camera to take a bird's-eye view of the sub-area to obtain an aerial image, which is then sent to the central server. The central server marks the most recently received aerial image as the current image, and marks the aerial image acquired in the most recent shooting mission of the same sub-area as the reference image; The central server determines the warning level corresponding to the aerial image based on the current image and the comparison image.

[0011] Preferably, the central server determines the warning level corresponding to the aerial image based on the current image and the reference image, including: The central server performs image recognition on the current image to obtain the position points of each indicator light in the current image, and performs image recognition on the reference image to obtain the position points of each indicator light in the reference image; The central server determines the position of each indicator light in the current image based on the color of each indicator light and marks it as a first position point; The central server sorts the first position points in the current image and determines a unique serial number, wherein the serial number of the first position point corresponds to the color of the corresponding indicator light; The central server determines the position point of each indicator light in the reference image based on the color of each indicator light, and marks it as a second position point; The central server sorts the second position points in the reference image and determines a unique serial number, wherein the sorting method of the second position points is consistent with the sorting method of the first position points; The central server obtains the distance between each first position point in the current image and the first base point, and the distance between each second position point in the reference image and the second base point, wherein the first base point is the upper left corner position point in the current image and the second base point is the upper left corner position point in the reference image; The central server calculates the average relative displacement of the sub-regions: , Where, is the average relative displacement of the sub-region; is the distance between the i-th first position point and the first base point in the current image; is the distance between the i-th second position point and the second base point in the reference image, , N is the total number of first position points, which is also the total number of second position points; When the average relative displacement of the sub-region is greater than or equal to a third preset value, the central server determines that the warning level corresponding to the aerial image is high risk; When the average relative displacement of the sub-region is greater than the fourth preset value and less than the third preset value, the central server determines that the warning level corresponding to the aerial image is medium risk; When the average relative displacement of the sub-areas is less than a fourth preset value, the central server determines that the warning level corresponding to the aerial image is low risk.

[0012] Preferably, the central server stores the altitude of the position where each indicator is embedded in each sub-area; the central server obtains the distance between each first position point and the first base point in the current image, and the distance between each second position point and the second base point in the reference image, and then further includes: The central server calculates the distance difference corresponding to each indicator light: , Where, is the distance difference corresponding to the i-th indicator light, which represents the difference between the distance between the i-th first position point and the first base point in the current image and the distance between the i-th second position point and the second base point in the reference image; The central server marks the indicator light corresponding to the distance difference greater than the fifth preset value as the target indicator light; The central server marks other indicator lights whose altitudes are lower than the altitude of the location where the target indicator light is embedded as indicator lights to be analyzed; The central server obtains the distance difference corresponding to the indicator light to be analyzed; When the distance difference corresponding to the indicator light to be analyzed is greater than or equal to a fifth preset value, the central server determines that the warning level corresponding to the aerial image is high risk.

[0013] The present invention also proposes a multi-source data processing system for intelligent monitoring of geological disasters, which applies a multi-source data processing method for intelligent monitoring of geological disasters; the system includes a central server and a data acquisition component communicatively connected to the central server; the number of the data acquisition components is multiple; and the data acquisition components are arranged in the area to be monitored.

[0014] The above technical solution can achieve the following beneficial effects: The multi-source data processing method for intelligent monitoring of geological disasters proposed in the present invention can process contradictory collected data from different sources to obtain accurate geological disaster prediction results; first, real-time monitoring data of the monitored area is obtained through a data acquisition component, and the real-time monitoring data includes soil moisture value, underground acoustic signal, rainfall and aerial image; then, the debris flow risk level of the monitored area is determined based on the real-time monitoring data. When the debris flow risk level obtained by analysis is high risk or medium risk, it indicates that the possibility of a debris flow disaster in the monitored area is high, and it is necessary to further determine whether there are contradictory real-time monitoring data at present, so as to eliminate the occurrence of false alarms of disaster risks. If there are contradictory real-time monitoring data, the debris flow risk level of the monitored area is re-determined based on the contradictory real-time monitoring data and the priority values of each real-time monitoring data. Because real-time monitoring data with higher priority values can more accurately reflect the geological disaster risk situation of the monitored area, the debris flow risk level of the monitored area re-determined based on the contradictory real-time monitoring data and the priority values of each real-time monitoring data is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0016] Figure 1 This is a flowchart of the first embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention. DETAILED DESCRIPTION

[0017] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] The present invention proposes a multi-source data processing method and system for intelligent monitoring of geological disasters.

[0019] As attached Figure 1 As shown, in a first embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention, the multi-source data processing method for intelligent monitoring of geological disasters is applied to a multi-source data processing system for intelligent monitoring of geological disasters; the system includes a central server (such as a cloud server) and a data acquisition component communicatively connected to the central server; the number of the data acquisition components is multiple; the data acquisition components are set in the area to be monitored; this embodiment includes the following steps: Step S110: The central server obtains real-time monitoring data collected by the data collection component in the area to be monitored, wherein the real-time monitoring data includes soil moisture value, underground acoustic signal, rainfall and aerial image.

[0020] Specifically, the above-mentioned soil moisture values, underground acoustic signals, rainfall and aerial images are typical multi-source collected data.

[0021] Step S120: The central server determines the debris flow risk level of the monitored area based on the real-time monitoring data, wherein the debris flow risk level is any one of high risk, medium risk and low risk.

[0022] Specifically, the above-mentioned soil moisture values, underground acoustic signals, rainfall and aerial images can all reflect the possibility of debris flow in the monitored area, so the debris flow risk level of the monitored area can be determined based on real-time monitoring data.

[0023] Step S130: When the debris flow risk level is high risk or medium risk, the central server determines whether there are conflicting real-time monitoring data.

[0024] Specifically, when the debris flow risk level obtained by analysis is high risk or medium risk, it means that there is a possibility of debris flow disaster in the monitored area. In order to protect the safety of life and property, it is necessary to further determine whether there are contradictory real-time monitoring data at present, so as to eliminate the possibility of false alarm of disaster risk.

[0025] If so, execute step S140: the central server obtains the priority value of each real-time monitoring data, and re-determines the debris flow risk level of the monitored area based on the conflicting real-time monitoring data and the priority value of each real-time monitoring data, and marks it as the final risk level.

[0026] Specifically, if there are currently contradictory real-time monitoring data, it is necessary to find the real-time monitoring data with a higher priority value among the real-time monitoring data. The real-time monitoring data with a higher priority value can more accurately reflect the geological disaster risk situation of the area to be monitored. Therefore, it is necessary to re-determine the debris flow risk level of the area to be monitored based on the real-time monitoring data with a higher priority value.

[0027] In addition, when the collected data from different sources are aggregated on the central server and conflicts arise between them, the processing priority needs to be determined based on the priority value of each real-time monitoring data, that is, a higher degree of trust is given to the real-time monitoring data with a higher priority value.

[0028] If not, step S150 is executed: the central server marks the debris flow risk level of the monitored area determined based on the real-time monitoring data as the final risk level.

[0029] Specifically, if there is no contradictory real-time monitoring data at present, it means that the currently obtained debris flow risk level is relatively accurate, and the debris flow risk level of the monitored area determined based on the real-time monitoring data (i.e., the debris flow risk level determined in step S120) is directly used as the final risk level; in addition, when the debris flow risk level determined in step S120 is low risk, since the overall risk is low, there is no need to determine whether there are contradictory real-time monitoring data.

[0030] The multi-source data processing method for intelligent monitoring of geological disasters proposed in the present invention can process contradictory collected data from different sources to obtain accurate geological disaster prediction results; first, real-time monitoring data of the monitored area is obtained through a data acquisition component, and the real-time monitoring data includes soil moisture value, underground acoustic signal, rainfall and aerial image; then, the debris flow risk level of the monitored area is determined based on the real-time monitoring data. When the debris flow risk level obtained by analysis is high risk or medium risk, it indicates that the possibility of a debris flow disaster in the monitored area is high, and it is necessary to further determine whether there are contradictory real-time monitoring data at present, so as to eliminate the occurrence of false alarms of disaster risks. If there are contradictory real-time monitoring data, the debris flow risk level of the monitored area is re-determined based on the contradictory real-time monitoring data and the priority values of each real-time monitoring data. Because real-time monitoring data with higher priority values can more accurately reflect the geological disaster risk situation of the monitored area, the debris flow risk level of the monitored area re-determined based on the contradictory real-time monitoring data and the priority values of each real-time monitoring data is more accurate.

[0031] In a second embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention, based on the first embodiment, the data acquisition component includes: a drone equipped with a camera for collecting aerial images, a geoacoustic sensor for collecting underground infrasound signals (such as a capacitive infrasound sensor, which is buried 2-5 meters underground in the area to be monitored to reduce the impact of surface noise), a humidity sensor for collecting soil moisture values, and a meteorological sensor for collecting rainfall; the priority values of each monitoring data are, from large to small, aerial images, underground infrasound signals, soil moisture values, and rainfall; the aerial images can intuitively and accurately reflect whether landslides or landslides occur in the area to be monitored. The aerial image has the highest credibility because of the displacement. Therefore, the aerial image has the highest priority value. When friction displacement or other phenomena that cause debris flow occur underground, underground acoustics will be generated. Therefore, the underground acoustic signals in the monitored area can be used to more accurately warn of debris flow disasters. Therefore, the priority value of the underground acoustic signals is set to be only less than the aerial image. Heavy rainfall can cause a sudden rapid increase in the moisture content of the mountain, that is, a rapid increase in underground humidity, which may cause debris flow. Therefore, the priority value of the soil moisture value is set to be less than the underground acoustic signal. The effect of rainfall on mountain debris flow has a hysteresis, but it is also an indicator worthy of attention. Therefore, the priority value of rainfall is set to be less than the soil moisture value. Step S120 includes the following steps: Step S210: The central server obtains the soil type of the area to be monitored, determines the warning humidity value of the area to be monitored based on the soil type of the area to be monitored, and determines the warning level corresponding to the soil humidity value by comparing the warning humidity value with the real-time monitoring data.

[0032] Specifically, the higher the soil moisture value, the greater the risk of mudslide disasters, and the warning humidity values corresponding to different soils are different. For example, because clay accumulates moisture slowly and has a strong hysteresis for mudslides, its warning humidity value can be set relatively large, such as 25%; while gravel soil has large pores and rainwater can penetrate quickly, so the warning humidity value is smaller than clay and can be set to 20%; sandy soil has high permeability and is more prone to instability when the humidity suddenly rises, so the warning humidity value should be smaller and can be set to 15%.

[0033] Step S220: The central server obtains the warning frequency range (e.g., 1-10 Hz) and the warning sound intensity, and determines the warning level corresponding to the underground sound signal based on the comparison of the warning frequency range and the warning sound intensity with the real-time monitoring data.

[0034] Specifically, before a mudslide occurs in a mountainous area, the infrasound waves generated by internal movement are usually within a specific frequency range. Within this specific frequency range, the greater the sound intensity value of the generated infrasound waves, the greater the energy of the internal movement of the mountain, and the greater the risk of a mudslide. Therefore, the warning level corresponding to the underground infrasound signal can be estimated by the warning frequency range and warning sound intensity.

[0035] Step S230: The central server obtains the terrain type of the area to be monitored, determines the first rainfall threshold and the second rainfall threshold of the area to be monitored based on the terrain type of the area to be monitored, and determines the warning level corresponding to the rainfall based on the comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data, wherein the second rainfall threshold is less than the first rainfall threshold.

[0036] Specifically, when the moisture content of the soil in the monitored area increases due to rainfall, a mudslide disaster may occur; when the rainfall in the past certain period of time reaches the first rainfall threshold, the corresponding warning level is high risk; when the rainfall in the past certain period of time reaches the second rainfall threshold and is lower than the first rainfall threshold, the corresponding warning level is medium risk.

[0037] In this embodiment, the rainfall thresholds for issuing warnings are different for different terrain types. For example, in a granite mountain area that is relatively compact, the first rainfall threshold can be set to 150 mm (24 hours) and the second rainfall threshold can be set to 120 mm (24 hours). In a loose accumulation area, the first rainfall threshold can be set to 100 mm (24 hours) and the second rainfall threshold can be set to 80 mm (24 hours). In a loess gully area, the first rainfall threshold can be set to 80 mm (24 hours) and the second rainfall threshold can be set to 50 mm (24 hours).

[0038] Specifically, the greater the rainfall within a certain period of time, the higher the risk of mudslides. Specifically, the risk can be compared with the first rainfall threshold and the second rainfall threshold to obtain the warning level corresponding to the rainfall.

[0039] Step S240: The central server determines the warning level corresponding to the aerial image.

[0040] Specifically, by performing image recognition on aerial images, it is possible to determine whether cracks or faults appear in the photographed area, thereby determining the corresponding warning level.

[0041] Step S250: The central server determines the debris flow risk level of the monitored area as the highest risk level among the warning levels corresponding to the soil moisture value, underground acoustic signal, rainfall and aerial image, where the warning level is any one of high risk, medium risk and low risk.

[0042] Specifically, if only one high risk level is detected, the overall debris flow risk level for the monitored area will be determined as high risk. If at least one medium risk level is detected and the other warning levels are low risk, the overall debris flow risk level for the monitored area will be determined as medium risk. If all warning levels are low risk, the overall debris flow risk level for the monitored area will be determined as low risk.

[0043] In a third embodiment of a multi-source data processing method for intelligent monitoring of geological hazards proposed by the present invention, based on the second embodiment, determining the warning level corresponding to the soil moisture value based on the comparison between the warning moisture value and the real-time monitoring data in step S210 includes the following steps: Step S310: When the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period (for example, 24 hours) is greater than the first preset value (for example, 5%), the central server determines that the warning level corresponding to the soil moisture value is high risk; when the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period is greater than the second preset value (for example, 3%), the central server determines that the warning level corresponding to the soil moisture value is medium risk, wherein the second preset value is less than the first preset value; otherwise, the central server determines that the warning level corresponding to the soil moisture value is low risk.

[0044] The step S220 of determining the warning level corresponding to the underground noise signal based on the comparison of the warning frequency range and the warning sound intensity with the real-time monitoring data includes the following steps: Step S320: The central server obtains the duration corresponding to when the frequency value of the next acoustic signal falls into the warning frequency range within the second preset duration (eg, 2 hours) in the past, and marks it as the first target duration.

[0045] Step S330: The central server obtains the duration corresponding to when the intensity value of the next acoustic signal in the past second preset time period is greater than the warning sound intensity (for example, twice the average intensity value of the background noise in the past second preset time period), and marks it as the second target duration.

[0046] Step S340: When the first target duration and the second target duration are both greater than or equal to a third preset duration (eg, 10 minutes), the central server determines that the warning level corresponding to the underground acoustic signal is high risk.

[0047] Step S350: When the first target duration or the second target duration is greater than or equal to the third preset duration, the central server determines that the warning level corresponding to the underground next sound signal is medium risk.

[0048] Step S360: When the first target duration and the second target duration are both less than the third preset duration, the central server determines that the warning level corresponding to the underground acoustic signal is low risk.

[0049] Specifically, the above steps provide a specific solution for determining the warning level corresponding to the underground noise signal based on the warning frequency range and the warning sound intensity.

[0050] In a fourth embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention, based on the second embodiment, determining the warning level corresponding to the rainfall based on the comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data in step S230 includes the following steps: Step S410: When the rainfall in the past first preset time period is greater than the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is high risk; when the rainfall in the past first preset time period is greater than the second rainfall threshold and less than or equal to the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is medium risk; otherwise, the central server determines that the warning level corresponding to the rainfall is low risk.

[0051] Specifically, this embodiment provides a specific technical solution for determining the warning level corresponding to the rainfall based on the first rainfall threshold and the second rainfall threshold.

[0052] In a fifth embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention, based on the second embodiment, step S130 includes the following steps: Step S510: When the debris flow risk level is high risk or medium risk, the central server determines whether any of the warning levels corresponding to the soil moisture value, the warning level corresponding to the underground acoustic signal, the warning level corresponding to the rainfall, and the warning level corresponding to the aerial image is low risk.

[0053] If so, step S520 is executed: the central server determines that there are conflicting real-time monitoring data.

[0054] Specifically, when the debris flow risk level is high or medium, if any of the warning levels corresponding to the soil moisture value, the warning level corresponding to the underground acoustic signal, the warning level corresponding to the rainfall, and the warning level corresponding to the aerial image is still low risk, it can be confirmed that a certain real-time monitoring data has a fault anomaly, resulting in a contradiction between the real-time monitoring data.

[0055] Step S140 includes the following steps: Step S530: The central server marks the real-time monitoring data corresponding to the warning level of medium risk or high risk among the warning levels corresponding to the soil moisture value, the underground acoustic signal, the rainfall, and the aerial image as target data.

[0056] Specifically, the target data is real-time monitoring data that indicates whether the area to be monitored has high or medium risk.

[0057] Step S540: The central server marks the real-time monitoring data with the highest priority value in the target data as priority monitoring data.

[0058] Step S550: The central server uses the warning level of the priority monitoring data as the final risk level.

[0059] Specifically, because the current warning levels already include medium-risk or high-risk warning levels, the risks are relatively high, and even misjudgments need to be taken seriously. Therefore, only the real-time monitoring data with the highest priority value (priority monitoring data) will be selected from the target data in the future, that is, the real-time monitoring data corresponding to the low-risk warning level will no longer be considered (even if the real-time monitoring data corresponding to the low-risk warning level has the highest priority value). In other words, the priority monitoring data is used to determine the specific debris flow risk level of the area to be monitored, so as to avoid the situation where the final risk level is low risk.

[0060] In a sixth embodiment of a multi-source data processing method for intelligent monitoring of geological hazards proposed by the present invention, based on the second embodiment, the drone further comprises a controller (e.g., a single-chip microcomputer) for controlling the flight of the drone and activating the camera; the data acquisition component further comprises an indicator, the indicator comprising a connecting rod and an indicator light (e.g., an LED light; in this embodiment, the indicator lights of different connecting rods can display different colors when activated to facilitate image recognition and distinguish between the indicator lights); the connecting rod is configured to be vertically embedded underground in the area to be monitored; the indicator light is above the ground; the indicator lights of the indicator lights of the indicator lights of the indicator lights of the same sub-area are different in color; the area to be monitored is divided into a plurality of square sub-areas of uniform area; a plurality of indicator lights are provided in each sub-area in a matrix arrangement (in this embodiment, each sub-area is provided with four indicator lights, arranged in two rows and two columns, and the four indicator lights can emit four different colors), and the number of indicator lights in each sub-area is the same. In actual application, the area to be monitored is manually divided into the plurality of sub-areas in advance, and the indicator lights are inserted into the ground of the sub-areas in advance; the controller can control the start and stop of each indicator light via wireless communication; step S270 comprises the following steps: Step S610: The central server sends the aerial photography position point of each sub-area to the drone, wherein the aerial photography position point of the sub-area is located at the center of the sub-area, and the height of the aerial photography position point is a preset height (for example, 100m). The GPS positioning system or the Beidou positioning system can determine the geographical coordinates (including longitude and latitude) of the four corners of each sub-area, and then determine the geographical coordinates of the center position of the sub-area. Combined with the determined height (preset height), the aerial photography position point can be accurately located, thereby ensuring that the position of the subsequent drone is more accurate each time it takes a bird's-eye view of the sub-area.

[0061] Step S620: The controller controls the drone to fly to the aerial photography position point of the sub-area every fourth preset time period (e.g., 1 hour) to perform a photography task, including the following steps: Step S621: When the drone flies to the aerial photography position point of the sub-area, it hovers. The controller controls the indicator light of the indicator in the sub-area to turn on, and the indicator lights of all the indicator lights in other sub-areas are off. Then, the camera is started to take a bird's-eye view of the sub-area to obtain an aerial image, and the aerial image is sent to the central server.

[0062] Specifically, when the drone is hovering, the controller controls the indicator lights of the indicator parts in the sub-area to turn on, and the indicator lights of all indicator parts in other sub-areas to turn off, so as to prevent the indicator lights in other adjacent sub-areas from being captured in the aerial image and interfering with the results of subsequent image recognition.

[0063] Step S630: The central server marks the most recently received aerial image as the current image, and marks the aerial image acquired in the most recent shooting mission of the same sub-region as the reference image.

[0064] Specifically, the current image and the reference image have the same size and resolution to facilitate comparison.

[0065] Step S640: The central server determines the warning level corresponding to the aerial image based on the current image and the comparison image.

[0066] Specifically, by comparing the current image with the reference image, the change of the sub-region within the fourth preset time period can be known, and the corresponding debris flow warning level can be determined.

[0067] In a seventh embodiment of a multi-source data processing method for intelligent monitoring of geological disasters proposed by the present invention, based on the sixth embodiment, step S640 includes the following steps: Step S710: The central server performs image recognition on the current image to obtain the position points of each indicator light in the current image, and performs image recognition on the reference image to obtain the position points of each indicator light in the reference image.

[0068] Step S720: The central server determines the position point of each indicator light in the current image based on the color of each indicator light, and marks it as the first position point.

[0069] Step S730: The central server sorts the first position points in the current image and determines a unique serial number, wherein the serial number of the first position point corresponds to the color of the corresponding indicator light.

[0070] For example, in this embodiment, the colors of the four indicator lights in the sub-area are red, green, purple and blue respectively; they can be set as follows: the first first position point is the position point of the red indicator light in the current image, the second first position point is the position point of the green indicator light in the current image, the third first position point is the position point of the purple indicator light in the current image, and the fourth first position point is the position point of the blue indicator light in the current image.

[0071] Step S740: The central server determines the position point of each indicator light in the reference image based on the color of each indicator light, and marks it as a second position point.

[0072] Step S750: The central server sorts the second position points in the reference image and determines a unique serial number, wherein the sorting method of the second position points is consistent with the sorting method of the first position points.

[0073] Specifically, referring to the above steps, it can be set as: the first second position point is the position point of the red indicator light in the reference image, the second second position point is the position point of the green indicator light in the reference image, the third second position point is the position point of the purple indicator light in the reference image, and the fourth second position point is the position point of the blue indicator light in the reference image.

[0074] Step S760: The central server obtains the distance between each first position point and the first base point in the current image, and the distance between each second position point and the second base point in the reference image, wherein the first base point is the upper left corner position point in the current image (i.e., the leftmost and uppermost position point in the image), and the second base point is the upper left corner position point in the reference image.

[0075] Specifically, because the drone's shooting position remains unchanged when taking aerial photos in each sub-area (all at the aerial photography position), the first base point and the second base point are actually the same position point. Therefore, the current image and the reference image taken at different times can be compared with each other to obtain the position changes of each indicator light.

[0076] Step S770: The central server calculates the average relative displacement of the sub-areas: , Where, is the average relative displacement of the sub-region; is the distance between the i-th first position point and the first base point in the current image; is the distance between the i-th second position point and the second base point in the reference image, , N is the total number of first position points (ie 4), which is also the total number of second position points.

[0077] Specifically, the above average relative displacement can reflect the displacement of each indicator light in the sub-area within the fourth preset time period. The larger the average relative displacement, the greater the displacement, and the greater the risk of debris flow.

[0078] Step S780: When the average relative displacement of the sub-region is greater than or equal to a third preset value (eg, 3% of the length of the current image), the central server determines that the warning level corresponding to the aerial image is high risk.

[0079] Step S790: When the average relative displacement of the sub-area is greater than a fourth preset value and less than a third preset value (for example, 1% of the length of the current image), the central server determines that the warning level corresponding to the aerial image is medium risk, wherein the fourth preset value is less than the third preset value.

[0080] Step S791: When the average relative displacement of the sub-region is less than a fourth preset value, the central server determines that the warning level corresponding to the aerial image is low risk.

[0081] Specifically, this embodiment provides a specific calculation scheme for determining the warning level corresponding to the aerial image based on the current image and the comparison image.

[0082] In an eighth embodiment of a multi-source data processing method for intelligent monitoring of geological hazards proposed by the present invention, based on the seventh embodiment, the central server stores the altitude of the location where each indicator is embedded in each sub-area; step S760, followed by the following steps: Step S810: The central server calculates the distance difference corresponding to each indicator light: , Where, is the distance difference corresponding to the i-th indicator light, which represents the difference between the distance between the i-th first position point and the first base point in the current image and the distance between the i-th second position point and the second base point in the reference image. , N is the total number of indicator lights, which is also the total number of first position points.

[0083] Specifically, the distance difference here represents the displacement of each indicator light itself within the past fourth preset time period.

[0084] Step S820: The central server marks the indicator lights corresponding to the distance differences greater than a fifth preset value (eg, 2% of the length of the current image) as target indicator lights.

[0085] Specifically, when the distance difference is greater than the fifth preset value, it means that the corresponding indicator light has undergone a relatively obvious displacement within the past fourth preset time period, and the risk of mudslide is relatively high. However, in order to ensure the accuracy of the disaster warning, further analysis can still be carried out, so the indicator light corresponding to the distance difference greater than the fifth preset value is marked as the target indicator light.

[0086] Step S830: The central server marks other indicator lights whose altitudes are lower than the altitude of the location where the target indicator light is embedded as indicator lights to be analyzed.

[0087] Specifically, the height of the indicator light to be analyzed is lower than that of the target indicator light, that is, the indicator light to be analyzed is below the target indicator light.

[0088] Step S840: The central server obtains the distance difference corresponding to the indicator light to be analyzed.

[0089] Step S850: When the distance difference corresponding to the indicator light to be analyzed is greater than or equal to the fifth preset value, the central server determines that the warning level corresponding to the aerial image is high risk.

[0090] Specifically, when the distance difference corresponding to the indicator light to be analyzed is greater than or equal to the fifth preset value, it means that all the indicator lights below the target indicator light have also experienced displacement caused by landslide or fault, and not just the target indicator light itself has experienced displacement (for example, when the indicator element is blown tilted by strong wind). In this way, the warning level corresponding to the aerial image can be more accurately determined to be high risk.

[0091] The present invention also proposes a multi-source data processing system for intelligent monitoring of geological disasters, which applies a multi-source data processing method for intelligent monitoring of geological disasters; the system includes a central server and a data acquisition component communicatively connected to the central server; the number of the data acquisition components is multiple; and the data acquisition components are arranged in the area to be monitored.

[0092] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0093] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A multi-source data processing method for intelligent monitoring of geological disasters, characterized in that: A multi-source data processing system for intelligent monitoring of geological hazards; the system includes a central server and a data acquisition component communicatively connected to the central server; There are multiple data acquisition components; the data acquisition components are arranged in the area to be monitored; the method includes: The central server obtains real-time monitoring data collected by the data collection component in the monitored area, wherein the real-time monitoring data includes soil moisture value, underground acoustic signal, rainfall and aerial image; The central server determines the debris flow risk level of the monitored area based on the real-time monitoring data, where the debris flow risk level is any one of high risk, medium risk and low risk; When the debris flow risk level is high or medium, the central server determines whether there are conflicting real-time monitoring data; If so, the central server obtains the priority values of each real-time monitoring data, and re-determines the debris flow risk level of the monitored area based on the conflicting real-time monitoring data and the priority values of each real-time monitoring data, and marks it as the final risk level; If not, the central server will mark the debris flow risk level of the monitored area determined based on the real-time monitoring data as the final risk level.

2. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 1, characterized in that: The data collection component includes: a drone equipped with a camera for collecting aerial images, a geoacoustic sensor for collecting underground acoustic signals, a humidity sensor for collecting soil moisture values, and a meteorological sensor for collecting rainfall. The priority values of the monitoring data are, from highest to lowest, aerial images, underground acoustic signals, soil moisture values, and rainfall. The central server determines the debris flow risk level of the monitored area based on the real-time monitoring data, including: The central server obtains the soil type of the area to be monitored, determines the warning humidity value of the area to be monitored based on the soil type of the area to be monitored, and determines the warning level corresponding to the soil humidity value based on the comparison between the warning humidity value and the real-time monitoring data; The central server obtains the warning frequency range and warning sound intensity, and determines the warning level corresponding to the underground sound signal based on the comparison of the warning frequency range and warning sound intensity with the real-time monitoring data; The central server obtains a terrain type of the area to be monitored, determines a first rainfall threshold and a second rainfall threshold for the area to be monitored based on the terrain type of the area to be monitored, and determines a warning level corresponding to the rainfall based on a comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data, wherein the second rainfall threshold is less than the first rainfall threshold; The central server determines the warning level corresponding to the aerial image; The central server determines the debris flow risk level of the monitored area as the highest risk level among the warning levels corresponding to the soil moisture value, underground acoustic signal, rainfall and aerial images, where the warning level is any one of high risk, medium risk and low risk.

3. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 2, characterized in that: Determining the warning level corresponding to the soil moisture value based on the comparison between the warning humidity value and the real-time monitoring data includes: When the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period is greater than the first preset value, the central server determines that the warning level corresponding to the soil moisture value is high risk; when the latest soil moisture value is greater than the warning moisture value, and the average hourly increase in the soil moisture value in the past first preset time period is greater than the second preset value, the central server determines that the warning level corresponding to the soil moisture value is medium risk, wherein the second preset value is less than the first preset value; otherwise, the central server determines that the warning level corresponding to the soil moisture value is low risk; The step of determining the warning level corresponding to the underground noise signal based on the comparison of the warning frequency range and the warning sound intensity with the real-time monitoring data includes: The central server obtains the duration corresponding to when the frequency value of the next acoustic signal falls into the warning frequency range within the second preset time period in the past, and marks it as the first target duration; The central server obtains the time duration corresponding to when the intensity value of the next acoustic signal is greater than the intensity of the warning sound within the second preset time duration, and marks it as the second target time duration; When the first target duration and the second target duration are both greater than or equal to the third preset duration, the central server determines that the warning level corresponding to the underground subaerial signal is high risk; When the first target duration or the second target duration is greater than or equal to the third preset duration, the central server determines that the warning level corresponding to the underground subaerial signal is medium risk; When the first target duration and the second target duration are both less than the third preset duration, the central server determines that the warning level corresponding to the underground subacoustic signal is low risk.

4. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 2, characterized in that: The step of determining the warning level corresponding to the rainfall based on the comparison of the first rainfall threshold and the second rainfall threshold with the real-time monitoring data includes: When the rainfall in the past first preset time period is greater than the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is high risk; when the rainfall in the past first preset time period is greater than the second rainfall threshold and less than or equal to the first rainfall threshold, the central server determines that the warning level corresponding to the rainfall is medium risk; otherwise, the central server determines that the warning level corresponding to the rainfall is low risk.

5. The multi-source data processing method for intelligent monitoring of geological disasters according to claim 2 is characterized in that: When the debris flow risk level is high or medium, the central server determines whether there are conflicting real-time monitoring data, including: When the debris flow risk level is high or medium, the central server determines whether any of the warning levels corresponding to the soil moisture value, the underground acoustic signal, the rainfall, and the aerial image is low risk. If so, the central server determines that there are conflicting real-time monitoring data; The central server obtains the priority value of each real-time monitoring data, and re-determines the debris flow risk level of the monitored area based on the conflicting real-time monitoring data and the priority value of each real-time monitoring data, and marks it as the final risk level, including: The central server marks the real-time monitoring data corresponding to the warning level of soil moisture value, underground acoustic signal, rainfall, and aerial image as target data. The central server marks the real-time monitoring data with the highest priority value among the target data as priority monitoring data; The central server will prioritize the warning level of the monitoring data as the final risk level.

6. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 2, characterized in that: The drone also includes a controller for controlling the flight of the drone and activating the camera; the data acquisition assembly also includes an indicator, which includes a connecting rod and an indicator light disposed on the top of the connecting rod; the connecting rod is configured to be vertically embedded underground in the area to be monitored; the indicator light is elevated above the ground; the indicator lights of the indicator lights of the indicator lights of the indicator lights of the indicator lights of the same sub-area are different colors; the area to be monitored is divided into a plurality of square sub-areas of uniform area; a plurality of indicator lights are disposed in a matrix arrangement in each sub-area, and the number of indicator lights in each sub-area is uniform; The controller can control the start and stop of each indicator light through wireless communication; The central server determines the warning level corresponding to the aerial image, including: The central server sends the aerial location points of each sub-area to the drone; The controller controls the drone to fly to the aerial photography position point of the sub-area every fourth preset time period to perform a photography task, including: When the drone flies to the aerial photography location of a sub-area and hovers, the controller turns on the indicator lights of the indicator components in the sub-area and turns off the indicator lights of all other indicator components in the sub-areas. The drone then starts the camera to take a bird's-eye view of the sub-area to obtain an aerial image, which is then sent to the central server. The central server marks the most recently received aerial image as the current image, and marks the aerial image acquired in the most recent shooting mission of the same sub-area as the reference image; The central server determines the warning level corresponding to the aerial image based on the current image and the comparison image.

7. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 6, characterized in that: The central server determines the warning level corresponding to the aerial image based on the current image and the reference image, including: The central server performs image recognition on the current image to obtain the position points of each indicator light in the current image, and performs image recognition on the reference image to obtain the position points of each indicator light in the reference image; The central server determines the position of each indicator light in the current image based on the color of each indicator light and marks it as a first position point; The central server sorts the first position points in the current image and determines a unique serial number, wherein the serial number of the first position point corresponds to the color of the corresponding indicator light; The central server determines the position point of each indicator light in the reference image based on the color of each indicator light, and marks it as a second position point; The central server sorts the second position points in the reference image and determines a unique serial number, wherein the sorting method of the second position points is consistent with the sorting method of the first position points; The central server obtains the distance between each first position point in the current image and the first base point, and the distance between each second position point in the reference image and the second base point, wherein the first base point is the upper left corner position point in the current image, and the second base point is the upper left corner position point in the reference image; The central server calculates the average relative displacement of the sub-regions: ， Where, is the average relative displacement of the sub-region; is the distance between the i-th first position point and the first base point in the current image; is the distance between the i-th second position point and the second base point in the reference image, , N is the total number of first position points, which is also the total number of second position points; When the average relative displacement of the sub-region is greater than or equal to a third preset value, the central server determines that the warning level corresponding to the aerial image is high risk; When the average relative displacement of the sub-region is greater than the fourth preset value and less than the third preset value, the central server determines that the warning level corresponding to the aerial image is medium risk; When the average relative displacement of the sub-areas is less than a fourth preset value, the central server determines that the warning level corresponding to the aerial image is low risk.

8. A multi-source data processing method for intelligent monitoring of geological disasters according to claim 7, characterized in that: The central server stores the altitude of the location where each indicator is embedded in each sub-area; the central server obtains the distance between each first position point and the first base point in the current image, and the distance between each second position point and the second base point in the reference image, and then further includes: The central server calculates the distance difference corresponding to each indicator light: ， Where, is the distance difference corresponding to the i-th indicator light, which represents the difference between the distance between the i-th first position point and the first base point in the current image and the distance between the i-th second position point and the second base point in the reference image; The central server marks the indicator light corresponding to the distance difference greater than the fifth preset value as the target indicator light; The central server marks other indicator lights whose altitudes are lower than the altitude of the location where the target indicator light is embedded as indicator lights to be analyzed; The central server obtains the distance difference corresponding to the indicator light to be analyzed; When the distance difference corresponding to the indicator light to be analyzed is greater than or equal to a fifth preset value, the central server determines that the warning level corresponding to the aerial image is high risk.

9. A multi-source data processing system for intelligent monitoring of geological disasters, characterized in that: A multi-source data processing method for intelligent monitoring of geological disasters as described in any one of claims 1 to 8 is applied; the system includes a central server and a data acquisition component communicatively connected to the central server; the number of the data acquisition components is multiple; and the data acquisition components are arranged in the area to be monitored.

Images