A snow disaster monitoring and early warning system

Abstract

This invention discloses a snow disaster monitoring and early warning system, comprising: a data acquisition module, a data preprocessing module, a wireless transmission module, an analysis center module, and an early warning module. The data acquisition module collects raw data. The data preprocessing module, connected to the data acquisition module, preprocesses the raw data to obtain preprocessed data. The wireless transmission module is connected to both the data preprocessing module and the analysis center module, transmitting the preprocessed data to the analysis center module. The analysis center module performs comprehensive analysis on the preprocessed data to obtain monitoring results. The early warning module, connected to the analysis center module, triggers early warning information based on the monitoring results. This invention provides intelligent monitoring of geological disasters, improving the accuracy of snow disaster monitoring and early warning.

Description

Technical Field

[0001] This invention relates to the field of disaster monitoring and early warning technology, and in particular to a snow disaster monitoring and early warning system. Background Technology

[0002] Snow disasters are among the most severe meteorological disasters in my country's pastoral areas during winter and spring. They often cause livestock to have difficulty foraging or become unable to forage, resulting in varying degrees of livestock injury and death. They can also be accompanied by frostbite among herders, traffic congestion, and power and communication line outages, causing enormous losses to the national economy and people's lives and property. Therefore, establishing a long-term regional snow disaster monitoring and early warning database and management information system, dynamically monitoring the environment of target areas, and constructing a snow disaster monitoring and early warning system are prerequisites for the operationalization of a snow disaster early warning system.

[0003] Snow cover is mostly located in high-altitude and frigid regions, and estimating its parameters is of significant research importance for hydrological resource statistics and crop yield forecasting. However, due to the harsh environment in which snow cover exists, it is difficult to obtain large-scale and highly accurate data using traditional manual measurement methods. Current snow cover monitoring methods mainly include conventional ground observation and satellite remote sensing monitoring. Conventional ground observation is a point-based observation, and the uneven distribution of stations makes it difficult to accurately reflect the spatial distribution of large-scale snow disasters. Compared with ground monitoring, remote sensing monitoring is more adaptable, acquires data faster, and provides a larger amount of information. It can conduct timely dynamic analysis and long-term dynamic monitoring of snow disasters. Remote sensing has a wider and continuous detection range, which can compensate for the data gaps caused by discrete observations from meteorological stations.

[0004] However, current snow disaster monitoring based on remote sensing technology is often affected by clouds and fog, leading to insufficient snow cover extraction and reducing the accuracy of snow disaster prediction. Therefore, this invention proposes a snow disaster monitoring and early warning system. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to overcome the shortcomings of the prior art and to provide a snow disaster monitoring and early warning system.

[0006] To achieve the above-mentioned technical objectives, the present invention provides the following technical solution: a snow disaster monitoring and early warning system, comprising: a data acquisition module, a data preprocessing module, a wireless transmission module, an analysis center module, and an early warning module;

[0007] The data acquisition module is used to acquire raw data.

[0008] The data preprocessing module is connected to the data acquisition module, and the data preprocessing module is used to preprocess the raw data to obtain preprocessed data.

[0009] The wireless transmission module is connected to the data preprocessing module and the analysis center module respectively, and the wireless transmission module is used to transmit the preprocessed data to the analysis center module.

[0010] The analysis center module is used to comprehensively analyze the preprocessed data to obtain monitoring results;

[0011] The early warning module is connected to the analysis center module, and the early warning module is used to trigger early warning information based on the monitoring results.

[0012] Preferably, the data acquisition module includes a first acquisition unit, a second acquisition unit, and a third acquisition unit;

[0013] The first acquisition unit is used to acquire raw remote sensing images;

[0014] The second acquisition unit is used to acquire raw meteorological data;

[0015] The third acquisition unit is used to acquire raw radar images.

[0016] Preferably, the data preprocessing module includes a cloud image separation unit, a meteorological data processing unit, and a radar image processing unit;

[0017] The cloud image separation unit is connected to the first acquisition unit and is used to separate the cloud image from the original remote sensing image to obtain cloud image and ground object image.

[0018] The meteorological data processing unit is connected to the second acquisition unit. The meteorological data processing unit is used to perform outlier detection and correction on the raw meteorological data to obtain preprocessed meteorological data.

[0019] The radar image processing unit is connected to the third acquisition unit, and the radar image processing unit is used to crop and mosaic the original radar image to obtain a panoramic radar image of the target area.

[0020] Preferably, the cloud image separation unit includes a deep learning model subunit and a preprocessed image subunit;

[0021] The deep learning model subunit is used to perform cloud detection on the original remote sensing image based on an improved generative adversarial network to obtain cloud feature vectors and ground feature vectors.

[0022] The preprocessed image subunit is used to obtain cloud images and ground object images based on the cloud feature vector and ground object feature vector.

[0023] Preferably, the improved generative adversarial network includes a first network structure, a second network structure, and a residual calculation structure;

[0024] The first network structure includes a vector builder, a generator, and a discriminator; the second network structure includes an encoder network and a decoder network.

[0025] The process of cloud detection based on the improved generative adversarial network includes:

[0026] The original remote sensing image is input into the vector builder to generate an original spectral vector set; the original spectral vector set is then input into the generator for encoding training to obtain an encoded feature vector; the original spectral vector set is sampled to obtain a sampled feature vector; the encoded feature vector and the sampled feature vector are input into the discriminator for vector authenticity discrimination to obtain an optimized image feature vector;

[0027] The optimized image feature vector is input into the second network structure, and a deep feature vector is obtained through the encoding network and the decoding network.

[0028] The optimized image feature vector and the deep feature vector are input into the residual calculation structure to obtain the residual value. Based on the residual value, the feature vector is obtained, wherein the feature vector includes: cloud feature vector and ground feature vector.

[0029] Preferably, the analysis center module includes: a snow depth inversion unit, a snowfall monitoring unit, a terrain change monitoring unit, and a snow disaster monitoring unit;

[0030] The snow depth inversion unit is connected to the cloud image separation unit, and the snow depth inversion unit is used to obtain the snow depth of the target area based on the ground image.

[0031] The snowfall monitoring unit is connected to the cloud image separation unit and the meteorological data processing unit respectively. The snowfall monitoring unit is used to predict the snowfall amount in the target area within a target time based on the cloud image and the preprocessed meteorological data.

[0032] The terrain change monitoring unit is connected to the radar image processing unit, and the terrain change monitoring unit is used to monitor the amount of surface subsidence displacement based on the panoramic radar image.

[0033] The snow disaster monitoring unit is connected to the snow depth inversion unit, the snowfall monitoring unit and the terrain change monitoring unit respectively. The snow disaster monitoring unit is used to obtain snow disaster monitoring results based on the snow depth in the target area, the snowfall within the target time and the ground subsidence displacement.

[0034] Preferably, the early warning module includes: a triggering unit, a buzzer, and a hazard information sending unit;

[0035] The triggering unit is used to determine whether the monitoring result exceeds a safety threshold.

[0036] The buzzer is used to issue early warning information to monitoring personnel;

[0037] The information sending unit is used to send early warning information to residents in the target area.

[0038] The present invention has the following technical effects:

[0039] This invention enables intelligent monitoring of geological disasters, improving the accuracy of snow disaster monitoring and early warning.

[0040] This invention proposes an improved adversarial network for cloud detection in original remote sensing images, resulting in cloud and ground object images. The generated cloudless snow images and cloud images can be used to retrieve snow depth and snow cover under various weather conditions and to predict snowfall in target areas.

[0041] This invention uses radar images acquired by spaceborne SAR to monitor surface displacement, taking surface displacement as one of the important factors in the occurrence of snow disasters, thus improving the accuracy of snow disaster early warning. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a system flowchart in an embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Example 1

[0046] like Figure 1 As shown, this embodiment provides a snow disaster monitoring and early warning system, including: a data acquisition module, a data preprocessing module, a wireless transmission module, an analysis center module, and an early warning module;

[0047] The data acquisition module is used to collect raw data.

[0048] The data preprocessing module is connected to the data acquisition module. The data preprocessing module is used to preprocess the raw data to obtain preprocessed data.

[0049] The wireless transmission module is connected to both the data preprocessing module and the analysis center module, and is used to transmit preprocessed data to the analysis center module.

[0050] The analysis center module is used to perform comprehensive analysis on preprocessed data to obtain monitoring results;

[0051] The early warning module is connected to the analysis center module, and the early warning module is used to trigger early warning information based on monitoring results.

[0052] As a preferred embodiment, the data acquisition module includes a first acquisition unit, a second acquisition unit, and a third acquisition unit.

[0053] As a preferred embodiment, the data acquisition process based on the data acquisition module includes:

[0054] The first acquisition unit uses ENVI software to preprocess the time-series-based remote sensing images, extracts the reflected and radiated waves from the remote sensing images for band fusion, and then uses shp files to crop the remote sensing images of the target area.

[0055] The second acquisition unit acquires the time series of remote sensing images from the first acquisition unit, and obtains raw meteorological data using meteorological stations, meteorological satellites, etc., including ground data. Ground data includes temperature, air pressure, air humidity, wind direction, wind speed, clouds, visibility, evaporation, sunshine duration, and low temperature, among other things.

[0056] The third acquisition unit, based on the aforementioned time series, uses spaceborne synthetic aperture radar to acquire raw radar images of the current state of the earth.

[0057] As a preferred embodiment, the data preprocessing module includes a cloud image separation unit, a meteorological data processing unit, and a radar image processing unit.

[0058] As a preferred embodiment, the process of preprocessing the acquired raw data based on the data processing module includes:

[0059] First, the cloud image separation unit separates the original remote sensing image into cloud images and ground object images;

[0060] This embodiment improves the generative adversarial network to detect clouds in remote sensing images and performs image layering to obtain cloud images and ground object images;

[0061] The improved generative adversarial network includes a first network structure, a second network structure, and a residual computation structure;

[0062] The first network structure includes a vector builder, a generator, and a discriminator; the second network structure includes an encoder network and a decoder network.

[0063] The process of image separation based on the improved generative adversarial network model includes:

[0064] The original remote sensing image is input into a vector builder to generate an original spectral vector set; the original spectral vector set is then input into the generator for encoding training to obtain encoded feature vectors; the original spectral vector set is sampled to obtain sampled feature vectors; the encoded feature vectors and sampled feature vectors are input into a discriminator for vector authenticity discrimination to obtain optimized image feature vectors.

[0065] The optimized image feature vector is input into the second network structure, and deep feature vectors are obtained through the encoding and decoding networks.

[0066] The optimized image feature vector and deep feature vector are input into the residual calculation structure to obtain cloud feature vector and ground feature vector.

[0067] Cloud images and ground object images are obtained based on cloud feature vectors and ground object feature vectors, respectively.

[0068] Then, the meteorological data processing unit performs outlier detection and correction on the raw meteorological data to obtain preprocessed meteorological data; the preprocessing methods include: direct substitution method, correlation interpolation method, direct deletion method and deviation correction method;

[0069] Finally, the radar image processing unit crops and mosaics the original radar images to obtain N radar images of the target area.

[0070] As a preferred embodiment, the analysis center module includes a snow depth inversion unit, a snowfall monitoring unit, a terrain change monitoring unit, and a snow disaster monitoring unit.

[0071] As a preferred embodiment, the snow depth inversion unit is connected to the cloud image separation unit to obtain the snow depth of the target area;

[0072] The process includes: first, based on land use cover data and the microwave radiation characteristics of the underlying surface in China, four main land cover types are classified: forest, farmland, grassland, and bare land; second, snow depth inversion algorithms are established for relatively pure pixels of these four main land cover types; then, a high-precision snow depth inversion algorithm for microwave pixels is established using linear hybrid pixel decomposition technology; finally, the high-precision snow depth inversion algorithm is applied to the land cover images to obtain the snow depth of the target area.

[0073] The snowfall monitoring unit is used to predict snowfall in a target area within a target time period. Specifically, it includes: first, acquiring upper-air meteorological data based on cloud images; and second, establishing a snowfall regression prediction model based on the upper-air meteorological data and surface meteorological data using the logistic regression method.

[0074] The topographic change monitoring unit is used to monitor the amount of surface subsidence and displacement.

[0075] The snow disaster monitoring unit conducts a comprehensive analysis based on data on snow depth, snowfall, and ground subsidence displacement to obtain snow disaster monitoring results;

[0076] The monitoring results of snow disasters are classified into three levels: Level 1, Level 2, and Level 3.

[0077] As a preferred embodiment, the early warning module includes a buzzer and a hazard information transmission unit.

[0078] As a preferred option in this implementation, the buzzer issues early warning information to monitoring personnel based on the snow disaster level;

[0079] The information sending unit sends early warning information to residents in the target area based on the current snow disaster level.

[0080] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A snow disaster monitoring and early warning system, characterized in that, include: The system includes a data acquisition module, a data preprocessing module, a wireless transmission module, an analysis center module, and an early warning module. The data acquisition module is used to acquire raw data; the data acquisition module includes a first acquisition unit, a second acquisition unit, and a third acquisition unit; wherein the first acquisition unit is used to acquire raw remote sensing images; the second acquisition unit is used to acquire raw meteorological data; and the third acquisition unit is used to acquire raw radar images. The data preprocessing module is connected to the data acquisition module. The data preprocessing module is used to preprocess the raw data to obtain preprocessed data. The data preprocessing module includes a cloud image separation unit, a meteorological data processing unit, and a radar image processing unit. The cloud image separation unit is connected to the first acquisition unit and is used to separate cloud images from the raw remote sensing image to obtain cloud layer images and ground feature images. The meteorological data processing unit is connected to the second acquisition unit and is used to perform outlier detection and correction on the raw meteorological data to obtain preprocessed meteorological data. The radar image processing unit is connected to the third acquisition unit and is used to crop and mosaic the raw radar image to obtain a panoramic radar image of the target area. The wireless transmission module is connected to the data preprocessing module and the analysis center module respectively, and the wireless transmission module is used to transmit the preprocessed data to the analysis center module. The analysis center module is used for comprehensive analysis of the preprocessed data to obtain monitoring results; The early warning module is connected to the analysis center module, and the early warning module is used to trigger early warning information based on the monitoring results; The analysis center module includes: a snow depth inversion unit, a snowfall monitoring unit, a terrain change monitoring unit, and a snow disaster monitoring unit. The snow depth inversion unit is connected to the cloud image separation unit and is used to obtain the snow depth of the target area based on the ground feature image. The snowfall monitoring unit is connected to both the cloud image separation unit and the meteorological data processing unit and is used to predict the snowfall amount in the target area within a target time period based on the cloud image and the preprocessed meteorological data. The terrain change monitoring unit is connected to the radar image processing unit and is used to monitor the ground subsidence displacement based on the panoramic radar image. The snow disaster monitoring unit is connected to the snow depth inversion unit, the snowfall monitoring unit, and the terrain change monitoring unit and is used to obtain snow disaster monitoring results based on the snow depth in the target area, the snowfall amount within the target time period, and the ground subsidence displacement.

2. The snow disaster monitoring and early warning system according to claim 1, characterized in that, The cloud image separation unit includes a deep learning model subunit and a preprocessed image subunit; The deep learning model subunit is used to perform cloud detection on the original remote sensing image based on an improved generative adversarial network to obtain cloud feature vectors and ground feature vectors. The preprocessed image subunit is used to obtain cloud images and ground object images based on the cloud feature vector and ground object feature vector.

3. The snow disaster monitoring and early warning system according to claim 2, characterized in that, The improved generative adversarial network includes a first network structure, a second network structure, and a residual computation structure; The first network structure includes a vector builder, a generator, and a discriminator; the second network structure includes an encoder network and a decoder network. The process of cloud detection based on the improved generative adversarial network includes: The original remote sensing image is input into the vector builder to generate an original spectral vector set; the original spectral vector set is then input into the generator for encoding training to obtain an encoded feature vector; the original spectral vector set is sampled to obtain a sampled feature vector; the encoded feature vector and the sampled feature vector are input into the discriminator for vector authenticity discrimination to obtain an optimized image feature vector; The optimized image feature vector is input into the second network structure, and a deep feature vector is obtained through the encoding network and the decoding network. The optimized image feature vector and the deep feature vector are input into the residual calculation structure to obtain the cloud feature vector and the ground feature vector.

4. The snow disaster monitoring and early warning system according to claim 1, characterized in that, The early warning module includes: a triggering unit, a buzzer, and a hazard information sending unit; The triggering unit is used to determine whether the monitoring result exceeds a safety threshold; The buzzer is used to issue early warning information to monitoring personnel; The information sending unit is used to send early warning information to residents in the target area.

Images