A landslide hazard identification and extraction method, system and equipment based on GEE-SAR

Abstract

The present invention discloses a landslide hazard identification and extraction method, system and equipment based on GEE-SAR, which belongs to the technical field of landslide hazard monitoring, and includes the steps of: based on the intelligent optimization of remote sensing data of the GEE cloud platform, batch processing of images online, and establishment of a time series remote sensing image data set; in view of the characteristics of high mountain dense forest area, a detection method based on the difference change before and after the earthquake #imgabs0# is proposed, and the #imgabs1# index is constructed, and based on the #imgabs2# coherence change characteristics, the #imgabs3# backscattering rate difference of the before and after phase images is used to establish the #imgabs4# function; and the #imgabs5# index and #imgabs6# coherence are integrated to establish a landslide hazard comprehensive identification model #imgabs7#. The present invention aims at the problems of missed and misjudgment of remote sensing identification of landslide disasters in high mountain dense forest area, constructs a landslide intelligent identification model and method, realizes accurate identification of landslide disasters, and provides auxiliary decision-making for the release of early warning information of landslide potential points.

Description

Technical Field

[0001] The present invention relates to the technical field of landslide disaster monitoring, and in particular to a landslide disaster identification and extraction method, system and equipment based on GEE-SAR. Background Art

[0002] Landslide disaster refers to the natural phenomenon that the soil or rock on the slope slides down the slope as a whole or in a scattered manner along a certain weak surface or weak zone under the action of gravity. The occurrence of landslide disasters is usually related to multiple factors such as topography, geology, hydrology, and meteorology. Landslide disasters may cause serious casualties and property losses, and cause damage to infrastructure such as transportation, buildings, and farmland. In order to reduce the impact of landslide disasters, some prevention and response measures can be taken, such as strengthening geological monitoring, carrying out engineering protection, and formulating emergency plans.

[0003] With the current economic and technical conditions, potential landslides in high mountainous forest areas are often distributed in steep and densely covered areas with vegetation, and are highly sudden. Traditional geological disaster surveys cannot penetrate into this area, and the diversified and universal landslide remote sensing identification methods and models only consider the characteristics of single-period vegetation coverage changes, without considering the dynamic change characteristics of multiple phases. The development of early landslide identification technology has gradually matured in the field of remote sensing identification of geological disasters, from visual interpretation to pixel-based segmentation and object-oriented to machine learning and deep learning. However, the timeliness and accuracy of its landslide identification results are not enough to support the decision-making and deployment of geological disaster monitoring and early warning. With the development and application of SAR (Synthetic Aperture Radar) technology, most of them identify potential landslide disasters by studying surface settlement, and do not use a function value to quantify whether a landslide disaster has occurred. Summary of the invention

[0004] In order to solve the problems in the above background technology, the purpose of the present invention is to provide a landslide hazard identification and extraction method, system and equipment based on GEE-SAR, which can quickly obtain Data, effectively solving the time-consuming and memory-consuming problems of traditional acquisition methods. The data is different from optical remote sensing data in that it has good penetration and is not affected by weather factors. (Normalized Difference Vegetation Index) change difference characteristics and The logarithmic feature of the backscattering rate ratio is used to detect landslide hazards. The effective combination of the two realizes low-cost, high-efficiency and high-precision landslide hazard identification and extraction.

[0005] To achieve the above object, the present invention provides a landslide hazard identification and extraction method based on GEE-SAR, comprising the following steps:

[0006] Step S1, based on the intelligent optimization of high-resolution multi-temporal remote sensing data on the GEE cloud platform, batch process massive images online to establish a time series remote sensing image dataset;

[0007] Step S2: Based on the characteristics of high mountain dense forest area, The method of detecting the difference between before and after earthquakes is constructed. Index, based on The coherence variation characteristics are established by using the difference in SAR backscattering rate between the previous and next phase images. function;

[0008] Step S3, fusion Index and Establishing a comprehensive identification model for landslide hazards based on coherence .

[0009] Preferably, step S1 comprises the following steps:

[0010] S101. Given the vector data range of the target study area;

[0011] S102, defining a time period before a disaster occurs and a time period after a disaster occurs, wherein the time period before a disaster occurs is a pre-earthquake time period, and the time period after a disaster occurs is a post-earthquake time period;

[0012] S103. Use the function ee.ImageCollection encapsulated by the GEE cloud platform to call and view massive amounts of images before and after the earthquake. Description of data and metadata;

[0013] S104, constructing a SAR data declouding algorithm, acquiring SAR image data, selecting the QA60 band of the image data, the QA60 band containing opaque clouds and cirrus clouds, detecting the outer contour vector data of opaque clouds and cirrus clouds respectively, generating opaque cloud masks and cirrus cloud masks respectively, and finally merging and adding the two to crop the original input SAR image data to obtain a declouded image data result containing surface reflectivity;

[0014] S105. Based on the above steps, a time series remote sensing dataset within the vector data range of the study area is established.

[0015] Preferably, in S102, the time period before the disaster and the time period after the disaster are defined in a format of YYYY / MM / DD / HH:MM-YYYY / MM / DD / HH:MM;

[0016] In S103, before and after the earthquake The data include sensor type, image acquisition time, and temporal / spatial resolution.

[0017] Preferably, step S2 comprises the following steps:

[0018] S201. Calculation of pre-earthquake normalized difference vegetation index , the calculation formula is as follows:

[0019] ;

[0020] In the formula, Represents before the earthquake The reflectivity of the data in the near-infrared band, Represents before the earthquake The reflectivity of the red band of the data are band5 and band4 respectively;

[0021] S202. Calculate the post-earthquake normalized difference vegetation index , the calculation formula is as follows:

[0022] ;

[0023] In the formula, Representing the post-earthquake The reflectivity of the data in the near-infrared band, Representing the post-earthquake The reflectivity of the red band of the data are band5 and band4 respectively;

[0024] S203. Calculate the Normalized Difference Vegetation Index before and after the earthquake , the calculation formula is as follows:

[0025] ;

[0026] In the formula, is the post-earthquake normalized difference vegetation index, is the normalized difference vegetation index before the earthquake;

[0027] S204. Calculate before and after earthquakes Backscatter ratio , the calculation formula is as follows:

[0028] ;

[0029] In the formula, Before the earthquake Backscattering rate value, After the earthquake Backscatter value.

[0030] Preferably, step S3 comprises the following steps:

[0031] S301, the obtained The identification results and The identification results are exported to vector data format respectively;

[0032] S302, intersecting and overlapping the two vector data layers obtained in S301 to obtain a common part and a non-common part, removing the non-common part and retaining the common part to obtain a final landslide hazard identification and extraction result.

[0033] The present invention also provides a landslide hazard identification and extraction system based on GEE-SAR, which uses the above-mentioned landslide hazard identification and extraction method based on GEE-SAR, including:

[0034] Multi-temporal remote sensing data integration, optimization and database building module: Based on the vector data range of a given study area, integrate high-quality original remote sensing images from multiple sources, process them in batches to obtain cloud-free and noise-free images, and establish a time series remote sensing image dataset;

[0035] A method for detecting changes before and after earthquakes is proposed. Intensity contrast function building module: a method based on A method for detecting the difference between before and after earthquakes and establishing the image of the before and after phases Logarithmic function of the backscatter ratio;

[0036] Landslide hazard comprehensive identification model establishment module: integration and A comprehensive identification model for landslide hazards was established based on the coherence.

[0037] The present invention also provides a landslide disaster identification and extraction device based on GEE-SAR, comprising:

[0038] one or more processors;

[0039] A memory for storing one or more programs;

[0040] When one or more programs are executed by one or more processors, the one or more processors use the above-mentioned landslide hazard identification and extraction method based on GEE-SAR.

[0041] Therefore, the present invention adopts the above-mentioned landslide hazard identification and extraction method, system and equipment based on GEE-SAR, which has the following beneficial effects:

[0042] (1) The present invention matches and screens multi-temporal vectors within a given vector range through the GEE cloud platform. Data, based on cloud storage space, online batch processing of massive images, reducing the impact of pixel-level white noise, effectively solving the problem of time-consuming and memory-consuming image selection, downloading and preprocessing in the early stage;

[0043] (2) The present invention proposes a method based on The difference between the detection method of difference change before and after earthquake and the single-period NDVI is that: the difference between the NDVI before and after earthquake is The difference between the two values ​​can represent the change characteristics of surface vegetation, whether it is decreasing or unchanged, and based on this, it can be determined whether a landslide disaster has occurred; The coherence characteristics of the image are used to establish the logarithmic function of the ratio of the backscattering rate of the two images and quantify The difference characteristics of image coherence before and after changes;

[0044] (3) The present invention is comprehensive Index and The intensity contrast function is used to improve the precision and accuracy of landslide identification and extraction, which is conducive to avoiding disasters, reducing the safety of life and property caused by disasters, and providing effective support for disaster monitoring and early warning.

[0045] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0047] Figure 2 A technical roadmap of the method of the embodiment of the present invention;

[0048] Figure 3 A regional landslide hazard identification result diagram according to an embodiment of the present invention;

[0049] Figure 4 A schematic diagram of the system structure of an embodiment of the present invention;

[0050] Figure 5 This is a flow chart of a SAR data cloud removal algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention are further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions.

[0052] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or devices.

[0053] Like reference numerals and letters denote similar items in the following drawings, and thus, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0054] In the description of the present invention, it should be noted that the terms "upper", "lower", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings, or are directions or positional relationships in which the product of the invention is usually placed when in use. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present invention.

[0055] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "setting", "installation" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] like Figure 1 , Figure 2 As shown, the landslide hazard identification and extraction method based on GEE-SAR described in the present invention comprises the following steps:

[0057] Step S1: Based on the intelligent selection of high-resolution multi-temporal remote sensing data on the GEE cloud platform, batch process massive images online, including cloud removal and noise reduction, radiation correction, etc., to establish a time series remote sensing image dataset. The specific steps are as follows:

[0058] S101. Given the vector data range of the target study area ( .shp);

[0059] S102. Define the time period before the disaster occurs (YYYY / MM / DD / HH:MM-YYYY / MM / DD / HH:MM) and the time period after the disaster occurs (YYYY / MM / DD / HH:MM-YYYY / MM / DD / HH:MM), namely, the pre-event time period (PreTP) and the post-event time period (PostTP). YYYY / MM / DD / HH:MM-YYYY / MM / DD / HH:MM represents year / month / day / hour:minute.

[0060] S103. Use the function (ee.ImageCollection) encapsulated by the GEE cloud platform to call and view massive amounts of images before and after the earthquake. Description of the data and metadata (sensor type, image acquisition time, temporal / spatial resolution, etc.);

[0061] S104, such as Figure 5 As shown in the figure, a SAR data declouding algorithm is constructed, and its process is as follows: first, the SAR image data is acquired, and the "QA60" band of the image data, including opaque clouds and cirrus clouds, is selected. The outer contour vector data of opaque clouds and cirrus clouds are detected respectively, and opaque cloud masks and cirrus cloud masks are generated respectively. Finally, the two masks are merged and added to crop the original input SAR image data to obtain the declouded image data result including the surface reflectivity.

[0062] S105. Based on the above steps, a time series remote sensing dataset within the vector data range of the study area is established.

[0063] Step S2: Based on the characteristics of high mountain dense forest area, a The method of detecting the difference between before and after earthquakes is constructed. Index, based on The coherence variation characteristics are established by using the difference in SAR backscattering rate between the previous and next phase images. Function. The specific steps are as follows:

[0064] S201, calculate the normalized difference vegetation index before the earthquake ( ), the calculation formula is as follows:

[0065] ;

[0066] In the formula, Represents before the earthquake The reflectivity of the data in the near-infrared band, Represents before the earthquake The reflectivity of the red band of the data are band5 and band4 respectively.

[0067] S202, calculate the post-earthquake normalized difference vegetation index ( ), the calculation formula is as follows:

[0068] ;

[0069] In the formula, Representing the post-earthquake The reflectivity of the data in the near-infrared band, Representing the post-earthquake The reflectivity of the red band of the data are band5 and band4 respectively.

[0070] S203, calculate the normalized difference vegetation index before and after the earthquake ( ), the calculation formula is as follows:

[0071] ;

[0072] In the formula, is the post-earthquake normalized difference vegetation index, is the normalized difference vegetation index before the earthquake.

[0073] S204. Calculate before and after earthquakes Backscatter ratio ( ), the calculation formula is as follows:

[0074] ;

[0075] In the formula, Before the earthquake Backscattering rate value, After the earthquake Backscatter value.

[0076] Step S3: Fusion Index and The comprehensive identification model of landslide disasters is established based on the coherence The specific steps are as follows:

[0077] S301, the obtained The identification results and The identification results are exported to vector data format respectively;

[0078] S302, intersecting and overlapping the two vector data layers obtained in S301 to obtain the common part and the non-common part, removing the non-common part and retaining the common part, which is the final landslide hazard identification and extraction result; taking the edge area of ​​Zhangza Town, Jiuzhaigou as an example, the landslide extraction result of this embodiment is as follows: Figure 3 A typical sample area of ​​high mountain dense forest area was selected, and the remote sensing image of the area and the landslide identification results were superimposed and displayed. It can be seen that the local details of the landslide extracted in the typical sample area conform to the general laws of topography and landforms, and the extraction effect is good.

[0079] like Figure 4 As shown, the landslide hazard identification and extraction system based on GEE-SAR described in the present invention applies the above-mentioned landslide hazard identification and extraction method based on GEE-SAR, including

[0080] Multi-temporal remote sensing data integration, optimization and database building module: Based on the vector data range of a given study area, integrate high-quality original remote sensing images from multiple sources, process them in batches to obtain cloud-free and noise-free images, and establish a time series remote sensing image dataset;

[0081] A method for detecting changes before and after earthquakes is proposed. Intensity contrast function building module: a method based on A method for detecting the difference between before and after earthquakes and establishing the image of the before and after phases Logarithmic function of the backscatter ratio;

[0082] Landslide hazard comprehensive identification model establishment module: integration and A comprehensive identification model for landslide hazards was established based on the coherence.

[0083] The landslide disaster identification and extraction device based on GEE-SAR of the present invention comprises:

[0084] one or more processors;

[0085] A memory for storing one or more programs;

[0086] When one or more programs are executed by one or more processors, the one or more processors use the above-mentioned landslide hazard identification and extraction method based on GEE-SAR.

[0087] Therefore, the present invention adopts the above-mentioned landslide hazard identification and extraction method, system and equipment based on GEE-SAR to quickly obtain Data, effectively solving the time-consuming and memory-consuming problems of traditional acquisition methods. The data is different from optical remote sensing data in that it has good penetration and is not affected by weather factors. Variation characteristics and The logarithmic feature of the backscattering rate ratio is used to detect landslide hazards. The effective combination of the two realizes low-cost, high-efficiency and high-precision landslide hazard identification and extraction.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A landslide hazard identification and extraction method based on GEE-SAR, characterized by: The following steps are involved: Step S1, based on the intelligent optimization of high-resolution multi-temporal remote sensing data on the GEE cloud platform, batch process massive images online to establish a time series remote sensing image dataset; Step S2: Based on the characteristics of high mountain dense forest area, The method of detecting the difference between before and after earthquakes is constructed. Index, based on Coherence change characteristics, using the before and after phase images Backscatter rate difference establishment function; Step S3, fusion Index and Establishing a comprehensive identification model for landslide hazards based on coherence ; Step S2 includes the following steps: S201. Calculation of pre-earthquake normalized difference vegetation index , the calculation formula is as follows: ； In the formula, Represents before the earthquake The reflectivity of the data in the near-infrared band, Represents before the earthquake The reflectivity of the red band of the data are band5 and band4 respectively; S202. Calculate the post-earthquake normalized difference vegetation index , the calculation formula is as follows: ； In the formula, Representing the post-earthquake The reflectivity of the data in the near-infrared band, Representing the post-earthquake The reflectivity of the red band of the data are band5 and band4 respectively; S203. Calculate the Normalized Difference Vegetation Index before and after the earthquake , the calculation formula is as follows: ； In the formula, is the post-earthquake normalized difference vegetation index, is the normalized difference vegetation index before the earthquake; S204. Calculate before and after earthquakes Backscatter ratio , the calculation formula is as follows: ； In the formula, Before the earthquake Backscattering rate value, After the earthquake Backscatter rate value; Step S3 includes the following steps: S301, the obtained The identification results and The identification results are exported to vector data format respectively; S302, intersecting and overlapping the two vector data layers obtained in S301 to obtain a common part and a non-common part, removing the non-common part and retaining the common part to obtain a final landslide hazard identification and extraction result.

2. The landslide hazard identification and extraction method based on GEE-SAR according to claim 1, characterized in that: Step S1 includes the following steps: S101. Given the vector data range of the target study area; S102, defining a time period before a disaster occurs and a time period after a disaster occurs, wherein the time period before a disaster occurs is a pre-earthquake time period, and the time period after a disaster occurs is a post-earthquake time period; S103. Use the function ee.ImageCollection encapsulated by the GEE cloud platform to call and view massive amounts of images before and after the earthquake. Description of data and metadata; S104, Construction Data de-clouding algorithm, acquisition Image data, select the QA60 band of the image data, the QA60 band contains opaque clouds and cirrus clouds, detect the outer contour vector data of opaque clouds and cirrus clouds respectively, generate opaque cloud masks and cirrus cloud masks respectively, and finally merge and add the two to crop the original input Image data, to obtain cloud-removed image data results including surface reflectivity; S105. Based on the above steps, a time series remote sensing dataset within the vector data range of the study area is established.

3. The landslide hazard identification and extraction method based on GEE-SAR according to claim 2 is characterized in that: In S102, the format of the time period before the disaster and the time period after the disaster is defined is YYYY / MM / DD / HH:MM-YYYY / MM / DD / HH:MM; In S103, before and after the earthquake The data include sensor type, image acquisition time, and temporal / spatial resolution.

4. A landslide hazard identification and extraction system based on GEE-SAR, using the landslide hazard identification and extraction method based on GEE-SAR as claimed in any one of claims 1 to 3, characterized in that: include Multi-temporal remote sensing data integration, optimization and database building module: Based on the vector data range of a given study area, integrate high-quality original remote sensing images from multiple sources, process them in batches to obtain cloud-free and noise-free images, and establish a time series remote sensing image dataset; A method for detecting changes before and after earthquakes is proposed. Intensity contrast function building module: a method based on A method for detecting the difference between before and after earthquakes and establishing the image of the before and after phases Logarithmic function of the backscatter ratio; Landslide hazard comprehensive identification model establishment module: integration and A comprehensive identification model for landslide hazards was established based on the coherence.

5. A landslide hazard identification and extraction device based on GEE-SAR, characterized by: include one or more processors; A memory for storing one or more programs; When one or more programs are executed by one or more processors, the one or more processors use the landslide hazard identification and extraction method based on GEE-SAR as described in any one of claims 1 to 3.

Images