Geological disaster analysis method and system based on high mountain terrain remote sensing image

Abstract

The invention discloses a geological disaster analysis method and system based on a high mountain terrain remote sensing image. The method comprises the following steps: S1, acquiring high mountain terrain data of a historical predetermined region; s2, constructing a geological disaster analysis model according to the mountain terrain data of the historical predetermined region; and S2, inputting the high mountain terrain data of the target area into the geological disaster analysis model for geological disaster prediction. By adopting the technical scheme of the invention, the problems of low executive performance, high risk and large workload are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of geological disasters, and in particular relates to a geological disaster analysis method and system based on remote sensing images of high mountain terrain. Background Art

[0002] The mountainous terrain is undulating, the engineering geological environment is fragile, human engineering activities are intense, the rainfall in the rainy season is large and concentrated, extreme weather events occur frequently, and geological disasters such as collapse, landslides, and mudslides develop, posing a huge threat to people's lives and property. Previous researchers have conducted in-depth research on the mechanism of geological disasters, the stability evaluation of geological disasters, and post-disaster management. Traditional geological disaster identification methods rely on only a few sensors or manual measurements to obtain mountain terrain information, which has the problems of low executability, high risk, and heavy workload. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for analyzing geological disasters based on remote sensing images of high mountain terrain.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A geological disaster analysis method based on remote sensing images of high mountain terrain, comprising:

[0006] Step S1, obtaining high mountain terrain data of a historical predetermined area;

[0007] Step S2, constructing a geological disaster analysis model based on the historical mountain terrain data of the predetermined area;

[0008] Step S2: input the high mountain terrain data of the target area into the geological disaster analysis model to perform geological disaster prediction.

[0009] Preferably, the mountain terrain data of the historical predetermined area includes: satellite remote sensing image data and meteorological data, wherein the satellite remote sensing image data includes the terrain structure and mountain slope; the meteorological data includes rainfall information.

[0010] Preferably, step S2 comprises:

[0011] Preprocessing of high mountain terrain data in historically scheduled areas;

[0012] A convolutional neural network is trained based on the pre-processed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

[0013] Preferably, preprocessing the high mountain terrain data of the historical predetermined area includes: performing data cleaning on the high mountain terrain data of the historical predetermined area.

[0014] The present invention also provides a geological disaster analysis system based on high mountain terrain remote sensing images, comprising:

[0015] An acquisition module is used to obtain high mountain terrain data of a historical predetermined area;

[0016] A training module is used to construct a geological disaster analysis model based on the historical mountain terrain data of the predetermined area;

[0017] The prediction module is used to input the high mountain terrain data of the target area into the geological disaster analysis model to predict geological disasters.

[0018] Preferably, the mountain terrain data of the historical predetermined area includes: satellite remote sensing image data and meteorological data, wherein the satellite remote sensing image data includes the terrain structure and mountain slope; the meteorological data includes rainfall information.

[0019] Preferably, the training module includes:

[0020] A processing unit for preprocessing the high mountain terrain data of the historical predetermined area;

[0021] A construction unit is used to train a convolutional neural network based on the pre-processed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

[0022] Preferably, the processing unit is used to clean the high mountain terrain data of the historical predetermined area.

[0023] The present invention constructs a geological disaster analysis model based on the high mountain terrain data of the historical predetermined area; the high mountain terrain data of the target area is input into the geological disaster analysis model to predict geological disasters. The technical solution of the present invention solves the problems of low executability, high risk and large workload. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0025] Figure 1The present invention is a flowchart of an embodiment of the present invention for analyzing geological hazards based on remote sensing images of high mountain terrain. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0028] Embodiment 1:

[0029] like Figure 1 As shown, an embodiment of the present invention provides a method for analyzing geological disasters based on remote sensing images of high mountain terrain, including:

[0030] Step S1, obtaining high mountain terrain data of a historical predetermined area;

[0031] Step S2, constructing a geological disaster analysis model based on the historical mountain terrain data of the predetermined area;

[0032] Step S2: input the high mountain terrain data of the target area into the geological disaster analysis model to perform geological disaster prediction.

[0033] As an implementation method of an embodiment of the present invention, the historical mountain terrain data of the predetermined area includes: satellite remote sensing image data and meteorological data, wherein the satellite remote sensing image data includes the terrain structure and the mountain slope; the meteorological data includes rainfall information.

[0034] As an implementation of the embodiment of the present invention, step S2 includes:

[0035] Step S21, pre-processing the high mountain terrain data of the historical predetermined area;

[0036] Step S22, training a convolutional neural network based on the pre-processed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

[0037] Furthermore, the convolutional neural network uses Resnet50 as the backbone network, and the preprocessed historical predetermined area high mountain terrain data first passes through the PSP-Net network structure, that is, the PSP-Net network model for scene analysis, and then uses the K-Net network structure, that is, the K-Net network model for image segmentation, and finally obtains the prediction result.

[0038] More specifically, the PSP-Net network model uses a "pyramid pooling module" to enhance the scene parsing capability of semantic segmentation by fusing contextual information from different regions; the K-Net network model updates the convolution kernel weights in a dynamic manner so that the convolution kernel weights can adaptively activate feature map content to ensure that each convolution kernel accurately responds to different geological features in the image. Finally, by using this adaptive convolution kernel update strategy in an iterative manner, the K-Net network model significantly improves the recognition and segmentation performance of the convolution kernel. Moreover, the K-Net network model uses a mutually matching strategy to assign learning objectives to each convolution kernel. This training method is superior to traditional training strategies.

[0039] As an implementation of the embodiment of the present invention, step S21, pre-processing the high mountain terrain data of the historical predetermined area includes: performing data cleaning on the high mountain terrain data of the historical predetermined area. Specifically, it includes:

[0040] Screening out garbled data and offside data from the high mountain terrain data of the historical predetermined area to obtain screening index data;

[0041] Performing vectorization operation on the screening index data to obtain a screening index vector set, and adding a time-space vector to each screening index vector in the screening index vector set to obtain a time-space vector set;

[0042] Performing feature clustering on the space-time vector set to obtain a cluster center data set;

[0043] Fill each empty value data in the screening index data according to the cluster center data set.

[0044] As an implementation method of an embodiment of the present invention, in step S22, the mountain terrain data of the historical predetermined area is input into the deep learning convolutional neural network model for training to obtain a geological disaster analysis model.

[0045] Training process: After data cleaning, the high mountain terrain data of the historical predetermined area is divided into training set and test set in a ratio of 8:2, and then the convolutional neural network model is trained with the training set; the relevant parameters of the convolutional neural network model are debugged, and the quality of the trained convolutional neural network model is judged according to the model evaluation indicators until the geological disaster analysis model is obtained.

[0046] The embodiment of the present invention can further use the coefficient of determination R2 and the mean square error MSE (mean squared error) to evaluate the training and testing results of the model to obtain the optimal geological disaster analysis model.

[0047] Embodiment 2:

[0048] The embodiment of the present invention provides a geological disaster analysis system based on high mountain terrain remote sensing images, including:

[0049] An acquisition module is used to obtain high mountain terrain data of a historical predetermined area;

[0050] A training module is used to construct a geological disaster analysis model based on the historical mountain terrain data of the predetermined area;

[0051] The prediction module is used to input the high mountain terrain data of the target area into the geological disaster analysis model to predict geological disasters.

[0052] As an implementation method of an embodiment of the present invention, the historical mountain terrain data of the predetermined area includes: satellite remote sensing image data and meteorological data, wherein the satellite remote sensing image data includes the terrain structure and the mountain slope; the meteorological data includes rainfall information.

[0053] As an implementation of an embodiment of the present invention, the training module includes:

[0054] A processing unit for preprocessing the high mountain terrain data of the historical predetermined area;

[0055] A construction unit is used to train a convolutional neural network based on the pre-processed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

[0056] As an implementation manner of the embodiment of the present invention, the processing unit is used to clean the high mountain terrain data of the historical predetermined area.

[0057] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for analyzing geological hazards based on remote sensing images of high mountain terrain, characterized in that: include: Step S1, obtaining high mountain terrain data of a historical predetermined area; Step S2, constructing a geological disaster analysis model based on the historical mountain terrain data of the predetermined area; Step S2: input the high mountain terrain data of the target area into the geological disaster analysis model to perform geological disaster prediction.

2. The method for analyzing geological hazards based on remote sensing images of high mountain terrain according to claim 1, characterized in that: The high mountain terrain data of the historical predetermined area includes: satellite remote sensing image data and meteorological data; among which, the satellite remote sensing image data includes the terrain structure and high mountain slope; the meteorological data includes rainfall information.

3. The method for analyzing geological hazards based on remote sensing images of high mountain terrain according to claim 2, characterized in that: Step S2 includes: Preprocessing of high mountain terrain data in historically scheduled areas; A convolutional neural network is trained based on the pre-processed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

4. The method for analyzing geological hazards based on remote sensing images of high mountain terrain according to claim 3, characterized in that: Preprocessing the high mountain terrain data in the historical predetermined area includes: data cleaning the high mountain terrain data in the historical predetermined area.

5. A geological disaster analysis system based on remote sensing images of high mountain terrain, characterized in that: include: An acquisition module is used to obtain high mountain terrain data of a historical predetermined area; A training module is used to construct a geological disaster analysis model based on the historical mountain terrain data of the predetermined area; The prediction module is used to input the high mountain terrain data of the target area into the geological disaster analysis model to predict geological disasters.

6. The geological disaster analysis system based on high mountain terrain remote sensing images as claimed in claim 5, characterized in that: The high mountain terrain data of the historical predetermined area includes: satellite remote sensing image data and meteorological data; among which, the satellite remote sensing image data includes the terrain structure and high mountain slope; the meteorological data includes rainfall information.

7. The geological disaster analysis system based on high mountain terrain remote sensing images as claimed in claim 6, characterized in that: The training modules include: A processing unit, used for preprocessing the high mountain terrain data of the historical predetermined area; A construction unit is used to train a convolutional neural network based on the preprocessed historical mountain terrain data of the predetermined area to obtain a geological disaster analysis model; wherein PReLu is used instead of ReLu as the activation function in the convolutional neural network, and the convolutional neural network includes a K-Net+PSP-Net network model.

8. The geological disaster analysis system based on high mountain terrain remote sensing images as claimed in claim 7, characterized in that: The processing unit is used to clean the high mountain terrain data of the historical predetermined area.

Images