Movatterモバイル変換


[0]ホーム

URL:


CN120339743A - Method and device for constructing InSAR co-seismic deformation dataset for deep learning - Google Patents

Method and device for constructing InSAR co-seismic deformation dataset for deep learning

Info

Publication number
CN120339743A
CN120339743ACN202510350684.6ACN202510350684ACN120339743ACN 120339743 ACN120339743 ACN 120339743ACN 202510350684 ACN202510350684 ACN 202510350684ACN 120339743 ACN120339743 ACN 120339743A
Authority
CN
China
Prior art keywords
insar
seismic
data
crawling
earthquake
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202510350684.6A
Other languages
Chinese (zh)
Inventor
张迎峰
单新建
王振杰
刘旭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Geology
Original Assignee
Institute of Geology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of GeologyfiledCriticalInstitute of Geology
Priority to CN202510350684.6ApriorityCriticalpatent/CN120339743A/en
Publication of CN120339743ApublicationCriticalpatent/CN120339743A/en
Pendinglegal-statusCriticalCurrent

Links

Landscapes

Abstract

Translated fromChinese

本公开实施例涉及一种用于深度学习的InSAR同震形变数据集构建方法和装置,其中方法包括:响应于爬取请求获取爬取地震时间和爬取地震信息;从地震信息网站获取在爬取地震时间内与爬取地震信息匹配的地震事件信息存储至地震目录数据结构;从干涉图网站爬取帧标识文件,将地震目录数据结构中每个地震事件对应的震中坐标与帧标识文件进行匹配,获取每个地震事件对应的帧标识;基于每个地震事件对应的帧标识从干涉图网站中下载InSAR同震干涉图数据,对InSAR同震干涉图数据进行处理得到InSAR同震形变数据集。由此,能够有效扩展数据集的规模和多样性,为深度学习模型的训练提供高质量、多样化的数据支持。

The disclosed embodiment relates to a method and device for constructing an InSAR co-seismic deformation dataset for deep learning, wherein the method includes: obtaining crawled earthquake time and crawled earthquake information in response to a crawling request; obtaining earthquake event information matching the crawled earthquake information within the crawled earthquake time from an earthquake information website and storing it in an earthquake catalog data structure; crawling a frame identification file from an interference graph website, matching the epicenter coordinates corresponding to each earthquake event in the earthquake catalog data structure with the frame identification file, and obtaining a frame identification corresponding to each earthquake event; downloading InSAR co-seismic interference graph data from the interference graph website based on the frame identification corresponding to each earthquake event, and processing the InSAR co-seismic interference graph data to obtain an InSAR co-seismic deformation dataset. Thus, the scale and diversity of the dataset can be effectively expanded, providing high-quality and diversified data support for the training of deep learning models.

Description

InSAR (interferometric synthetic aperture radar) isoseism deformation data set construction method and device for deep learning
Technical Field
The disclosure relates to the technical field of data processing, in particular to an InSAR isoseism deformation dataset construction method and device for deep learning.
Background
The InSAR (interferometric synthetic aperture radar) technology is used as a high-precision ground deformation monitoring means and is widely applied to monitoring and evaluation of natural disasters such as earthquakes, volcanoes and the like. Especially after an earthquake occurs, the InSAR can provide accurate co-vibration deformation images, and through monitoring the ground deformation, the earthquake influence area is estimated, and important data support is provided for post-disaster estimation and recovery work. The advantages of InSAR technology are its higher spatial resolution and full coverage capability, making it an important tool in seismic research. Despite the significant advantages of InSAR technology in seismic monitoring and deformation analysis, how to efficiently utilize massive InSAR data sets remains a major challenge for existing InSAR data processing. Especially in deep learning applications, the size and quality of the InSAR dataset becomes a key factor limiting its application potential in the deep learning model.
Traditional methods for constructing the isoseism deformation data set from a large amount of InSAR data sets often rely on manual feature extraction and manual labeling, so that the data set is limited in scale and diversity. These methods typically require extensive manual intervention and rely on experience and domain knowledge for the extraction of deformation features. Therefore, with the rise of deep learning technology, in order to meet the demand of deep learning for a large number of data sets, researchers want to generate the co-seismic InSAR data set through an automated means, so as to train a deep learning model. The defects of the existing data set cause that the deep learning model is difficult to fully learn a complex deformation mode, and the performance and adaptability of the model are affected.
Disclosure of Invention
In order to solve the technical problems described above or at least partially solve the technical problems described above, the present disclosure provides an InSAR isoseism deformation dataset construction method and apparatus for deep learning.
The embodiment of the disclosure provides an InSAR isoseism deformation dataset construction method for deep learning, which comprises the steps of responding to a crawling request, obtaining crawling earthquake time and crawling earthquake information, obtaining earthquake event information matched with the crawling earthquake information in the crawling earthquake time from an earthquake information website based on a preset crawling technology, storing the earthquake event information into a preset earthquake catalog data structure, crawling a frame identification file from an interferogram website, matching a seism coordinate corresponding to each earthquake event in the earthquake catalog data structure with the frame identification file, obtaining a frame identification corresponding to each earthquake event, downloading InSAR isoseism interferogram data from the interferogram website based on the frame identification corresponding to each earthquake event, and processing the InSAR isoseism interferogram data to obtain an InSAR isoseism deformation dataset.
Optionally, the method further comprises the steps of receiving a starting time and an ending time input by a user, obtaining the crawling earthquake time, receiving a seismic level interval, a geographic range and a seismic source depth input by the user, obtaining the crawling earthquake information, and generating the crawling request based on the crawling earthquake time and the crawling earthquake information.
Optionally, the method further comprises the steps of obtaining data updating frequency and data change information of the seismic information website, and adjusting crawling intervals of the seismic information website based on the data updating frequency and the data change information.
Optionally, the method further comprises screening the seismic event information in the seismic catalog data structure according to a preset time range and/or a preset space range.
Optionally, in the process of downloading the InSAR co-vibration interferogram data, the method further comprises the steps of judging whether each InSAR co-vibration interferogram is stored in a database, if so, not downloading the InSAR co-vibration interferograms, and if any InSAR co-vibration interferograms are failed to be downloaded, carrying out re-downloading according to preset downloading times.
Optionally, the processing the InSAR co-vibration interferogram data to obtain an InSAR co-vibration deformation data set comprises the steps of screening the InSAR co-vibration interferogram data according to a preset image quality condition to obtain target InSAR co-vibration interferogram data, cutting each target InSAR co-vibration interferogram in the target InSAR co-vibration interferogram data according to a preset cutting image size to obtain to-be-processed InSAR co-vibration interferogram data, and carrying out resampling processing and normalization processing on each to-be-processed InSAR co-vibration interferogram to obtain the InSAR co-vibration deformation data set.
Optionally, the method further comprises performing image enhancement processing on each image in the InSAR isomorphous data set, wherein the image enhancement processing comprises one or more of image rotation, image flipping, image scaling and translation.
The embodiment of the disclosure also provides an InSAR isoseism deformation data set construction device for deep learning, which comprises a response acquisition module, an acquisition storage module and an image processing module, wherein the response acquisition module is used for responding to a crawling request, acquiring crawling earthquake time and crawling earthquake information, the acquisition storage module is used for acquiring earthquake event information matched with the crawling earthquake information in the crawling earthquake time from an earthquake information website based on a preset crawling technology and storing the earthquake event information into a preset earthquake catalog data structure, the crawling matching module is used for crawling frame identification files from an interferogram website, matching the coordinates of a seism corresponding to each earthquake event in the earthquake catalog data structure with the frame identification files, acquiring frame identifications corresponding to each earthquake event, the downloading module is used for downloading InSAR isoseism interferogram data from the interferogram website based on the frame identifications corresponding to each earthquake event, and the image processing module is used for processing the InSAR isoseism interferogram data to obtain an InSAR isoseism deformation data set.
The embodiment of the disclosure also provides electronic equipment, which comprises a processor, a memory for storing executable instructions of the processor, and the processor, wherein the processor is used for reading the executable instructions from the memory and executing the instructions to realize the InSAR isoseism deformation dataset construction method for deep learning.
The embodiment of the disclosure also provides a computer readable storage medium, wherein the storage medium stores a computer program, and the computer program is used for executing the InSAR isoseism deformation dataset construction method for deep learning.
The embodiment of the disclosure also provides a computer program product, which comprises a computer program, wherein the computer program, when being executed by a processor, realizes the InSAR isoseism deformation dataset construction method for deep learning.
Compared with the prior art, the technical scheme has the advantages that the InSAR isoseism deformation data set construction scheme for deep learning is provided, and comprises the steps of responding to a crawling request, obtaining crawling earthquake time and crawling earthquake information, obtaining earthquake event information matched with the crawling earthquake information in the crawling earthquake time from an earthquake information website based on a preset crawling technology, storing the earthquake event information into a preset earthquake catalog data structure, crawling a frame identification file from an interferogram website, matching the middle-of-earthquake coordinates corresponding to each earthquake event in the earthquake catalog data structure with the frame identification file, obtaining frame identification corresponding to each earthquake event, downloading InSAR isoseism interferogram data from the interferogram website based on the frame identification corresponding to each earthquake event, and processing the InSAR isoseism interferogram data to obtain an InSAR isoseism deformation data set. Therefore, the earthquake data can be efficiently grabbed and processed, the tedious process of manual screening and downloading is avoided, the speed and accuracy of data acquisition are obviously improved, the scale and diversity of a data set can be effectively expanded, and high-quality and diversified data support is provided for training of a deep learning model.
Drawings
The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale.
Fig. 1 is a schematic flow chart of an InSAR isoseism deformation dataset construction method for deep learning according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of another method for constructing InSAR isoseism deformation data sets for deep learning according to an embodiment of the present disclosure;
FIG. 3 is an exemplary diagram of an InSAR isoseism deformation dataset construction method for deep learning according to an embodiment of the present disclosure;
Fig. 4 is a schematic structural diagram of an InSAR isoseism deformation dataset construction device for deep learning according to an embodiment of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure have been shown in the accompanying drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure.
It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order and/or performed in parallel. Furthermore, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.
The term "including" and variations thereof as used herein are intended to be open-ended, i.e., including, but not limited to. The term "based on" is based at least in part on. The term "one embodiment" means "at least one embodiment," another embodiment "means" at least one additional embodiment, "and" some embodiments "means" at least some embodiments. Related definitions of other terms will be given in the description below.
It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.
It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.
The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.
Based on the description of the background art, the downloading and processing of InSAR isoseism deformation data often depend on manual intervention, and particularly when data is acquired from a public database, time, place and related data still need to be manually selected, so that the operation is complicated, the efficiency is low and the scale is small. For example, current InSAR data processing typically requires manual acquisition of a relevant seismic catalog and downloading of corresponding interferogram data, which may involve screening, clipping, resampling, etc. The mode depending on manual operation not only increases the workload of data processing, but also is easy to cause artificial errors, and influences the quality and consistency of the data set. Furthermore, the prior art lacks an efficient systematic approach to the automated processing of large-scale data sets, resulting in an inefficient and repetitive data generation and enhancement process. The lack of automated data capture and processing means greatly discounts the efficiency and scalability of InSAR data in large-scale, real-time applications.
Aiming at the problems, the invention provides an InSAR isoseism deformation dataset construction method for deep learning, which is characterized in that seismic information in a seismic information website such as GCMT website is automatically crawled, corresponding isoseism interferogram data is downloaded from an interferogram website such as LiCSAR website according to the seismic information, the InSAR dataset with large diversity and large scale is generated by combining processing steps such as data screening, clipping, resampling and normalization, and the scale and diversity of the dataset are further expanded by combining data enhancement technologies (such as translation, scaling, rotation and the like) so as to support efficient training of a deep learning model on complex isoseism deformation, and the problems that the data set is small in scale, sample diversity is insufficient, data processing efficiency is low and the deep learning model depends on high-quality data in the prior art are solved.
Fig. 1 is a schematic flow chart of an InSAR isomorphic deformation dataset construction method for deep learning, which is provided in an embodiment of the present disclosure, and the method may be executed by an InSAR isomorphic deformation dataset construction device for deep learning, where the device may be implemented by software and/or hardware, and may be generally integrated in an electronic device. As shown in fig. 1, the method includes:
step 101, responding to the crawling request, and acquiring crawling earthquake time and crawling earthquake information.
Step 102, acquiring the seismic event information matched with the crawled seismic information in the crawled seismic time from the seismic information website based on a preset crawling technology, and storing the seismic event information into a preset seismic catalog data structure.
In the embodiment of the disclosure, the starting time and the ending time input by a user can be received, the crawling earthquake time is acquired, the magnitude interval, the geographic range and the focus depth input by the user are received, the crawling earthquake information is acquired, and the crawling request is generated based on the crawling earthquake time and the crawling earthquake information.
In the embodiment of the disclosure, the starting time and the ending time provided by the user are received as crawling earthquake time, a magnitude interval, a geographic range (longitude and latitude), a focus depth and the like as crawling earthquake information, and the earthquake information website such as GCMT website is automatically accessed according to the crawling earthquake time and the crawling earthquake information.
In the embodiment of the disclosure, a crawling technology is preset to acquire seismic event information matched with crawling seismic information in crawling seismic time from a seismic information website, such as seismic event information of time, middle coordinates, magnitude, depth of a seismic source and the like of the occurrence of the earthquake, and the seismic event information (such as the time, middle coordinates (latitude, longitude) of the occurrence of the earthquake) is saved in a seismic catalog data structure, wherein the seismic catalog data structure at least comprises the time, middle coordinates (latitude, longitude) of the occurrence of the earthquake corresponding to each seismic event.
In some embodiments, the data update frequency and the data change information of the seismic information website can also be obtained, and the crawling interval of the seismic information website is adjusted based on the data update frequency and the data change information, for example, the higher the data update frequency and the more the data change information are, the shorter the crawling interval of the seismic information website is, so that the reasonable crawling interval is set according to the data update frequency and the data change condition of the seismic information website, and the integrity and the accuracy of the data are further ensured.
In some embodiments, the seismic event information in the seismic catalog data structure may be screened according to a preset time range and/or space range, that is, after the seismic data is captured, the seismic data may be initially screened, and the seismic data whose time or space range does not meet the requirements may be removed, so as to ensure that the subsequent InSAR data downloading request highly meets the actual requirements.
Step 103, crawling a frame identification file from the interferogram website, and matching the seismograph coordinates corresponding to each seismic event in the seismic catalog data structure with the frame identification file to obtain the frame identification corresponding to each seismic event.
Step 104, downloading InSAR co-vibration interference map data from the interference map website based on the frame identification corresponding to each seismic event, and processing the InSAR co-vibration interference map data to obtain an InSAR co-vibration deformation data set.
In the embodiment of the disclosure, after the seismic information website is crawled, the geographical area information (longitude and latitude) corresponding to each frame identifier FrameID in the automatic crawling interferogram website, such as LiCSAR website, is used for ensuring that the accurate positioning to a correct area can be achieved when the co-seismic interferograms are downloaded from the interferogram website later, and the geographical area information corresponding to each FrameID is stored and matched with the information in the seismic catalog data structure so as to provide accurate positioning data for the subsequent co-seismic interferogram data downloading.
Specifically, after the capturing of the seismic event information and the frame identification file is completed, the frame identification corresponding to the earthquake is obtained from the frame identification file according to the epicenter coordinate of each seismic event, and then a downloading request is automatically sent to the interferogram network station through the interferogram downloading control area, and the interferogram network station provides all InSAR interferograms, so that corresponding co-seismic interferogram data are accurately matched according to the information such as the earthquake starting time.
In some embodiments, in the process of downloading the InSAR co-vibration interferogram data, whether each InSAR co-vibration interferogram is stored in the database or not may be further determined, if so, the InSAR co-vibration interferograms are not downloaded, and if any InSAR co-vibration interferogram fails to be downloaded, the re-downloading process is performed according to the preset downloading times.
Specifically, the downloaded seismology interferograms are stored in a local data storage database, and the sources of the data (from which frame identification, interference time and other information) are marked, so that the traceability of the data is ensured. This process verifies that the data already exists, and if so, skips the download and continues to download the interferogram for the next earthquake. If the download fails or the data is incomplete, the automatic retry is performed, so that the data requested each time is ensured to be complete.
In the embodiment of the disclosure, inSAR isoseism interference map data are processed to obtain an InSAR isoseism deformation data set, which comprises the steps of screening the InSAR isoseism interference map data according to a preset image quality condition to obtain target InSAR isoseism interference map data, cutting each target InSAR isoseism interference map in the target InSAR isoseism interference map data according to a preset cutting image size to obtain to-be-processed InSAR isoseism interference map data, and carrying out resampling and normalization processing on each to-be-processed InSAR isoseism interference map to obtain the InSAR isoseism deformation data set.
Specifically, the obtained InSAR co-seismic interferogram data generally comprises image files with larger sizes, and interferograms of different seismic events may have different resolution, image size, noise and other problems, so that ineffective or poor quality images need to be rapidly screened and removed, the large-size InSAR interferograms are divided into small images, the purpose of improving the calculation efficiency and reducing the memory consumption is achieved, the interferograms of different sources may have different resolution, the cut images are subjected to resampling operation for unifying data formats and facilitating subsequent analysis, and in order to eliminate differences among different images, a system performs normalization processing on all the images, so that the data are compared under the same standard, and the uniformity of the data is guaranteed.
In the embodiment of the disclosure, each image in the InSAR isomorphism deformation data set is subjected to image enhancement processing, wherein the image enhancement processing comprises one or more of image rotation, image inversion, image scaling and translation.
Specifically, in order to increase the size and diversity of a data set when the processed data set is generated, the data enhancement technology is adopted in the present disclosure, and the data enhancement is a technology for generating new data by transforming existing data, so that the size of the data set can be expanded, a model can be adapted to more kinds of common-seismic deformation characteristics, and the specific enhancement method comprises one or more of image rotation, image inversion, image scaling and translation.
The InSAR isoseism deformation dataset construction scheme for deep learning is used for responding to a crawling request to acquire crawling earthquake time and crawling earthquake information, acquiring earthquake event information matched with the crawling earthquake information in the crawling earthquake time from an earthquake information website based on a preset crawling technology, storing the earthquake event information into a preset earthquake catalog data structure, crawling a frame identification file from an interferogram website, matching the seism coordinates corresponding to each earthquake event in the earthquake catalog data structure with the frame identification file to acquire a frame identification corresponding to each earthquake event, downloading InSAR isoseism interferogram data from the interferogram website based on the frame identification corresponding to each earthquake event, and processing the InSAR isoseism interferogram data to obtain an InSAR isoseism deformation dataset. Therefore, the earthquake data can be efficiently grabbed and processed, the tedious process of manual screening and downloading is avoided, the speed and accuracy of data acquisition are obviously improved, the scale and diversity of a data set can be effectively expanded, and high-quality and diversified data support is provided for training of a deep learning model.
Specifically, fig. 2 is a schematic flow chart of another method for constructing an InSAR isoseism deformation dataset for deep learning according to an embodiment of the present disclosure, where the method for constructing an InSAR isoseism deformation dataset for deep learning is further optimized on the basis of the above embodiment. As shown in fig. 2, the method includes:
Step 201, receiving a start time and an end time input by a user, acquiring crawling earthquake time, receiving a magnitude interval, a geographic range and a focus depth input by the user, acquiring crawling earthquake information, and generating a crawling request based on the crawling earthquake time and the crawling earthquake information.
Step 202, responding to the crawling request, acquiring crawling earthquake time and crawling earthquake information, acquiring earthquake event information matched with the crawling earthquake information in the crawling earthquake time from an earthquake information website based on a preset crawling technology, and storing the earthquake event information into a preset earthquake catalogue data structure.
Step 203, acquiring data updating frequency and data change information of the seismic information website, adjusting crawling intervals of the seismic information website based on the data updating frequency and the data change information, and screening the seismic event information in the seismic catalog data structure according to a preset time range and/or a preset space range.
Step 204, crawling the frame identification file from the interferogram website, and matching the seismograph coordinates corresponding to each seismic event in the seismic catalog data structure with the frame identification file to obtain the frame identification corresponding to each seismic event.
Step 205, downloading InSAR co-seismic interferogram data from the interferogram website based on the frame identifier corresponding to each seismic event, judging whether each InSAR co-seismic interferogram is stored in a database or not in the process of downloading the InSAR co-seismic interferogram data, if so, not downloading the InSAR co-seismic interferograms, and if any InSAR co-seismic interferogram fails to be downloaded, carrying out re-downloading according to preset downloading times.
Step 206, screening the InSAR co-vibration interferogram data according to the preset image quality condition to obtain target InSAR co-vibration interferogram data, cutting each target InSAR co-vibration interferogram in the target InSAR co-vibration interferogram data according to the preset cutting image size to obtain to-be-processed InSAR co-vibration interferogram data, and carrying out resampling and normalization on each to-be-processed InSAR co-vibration interferogram to obtain an InSAR co-vibration deformation data set.
And 207, performing image enhancement processing on each image in the InSAR isoseism deformation data set, wherein the image enhancement processing comprises one or more of image rotation, image inversion, image scaling and translation.
The present disclosure provides automatic generation and data enhancement of a seismic catalog crawling and co-seismic interferogram dataset based on a GCMT website as an example of a seismic information website and a LiCSAR website as an example of an interferogram website, as an example, as shown in fig. 3, including the following steps:
(1) In order to efficiently and automatically process large-scale InSAR data, the present disclosure first designs a data capture and processing module that captures seismic catalog data in GCMT websites to obtain relevant InSAR co-seismic interferograms.
Specifically, in the embodiment of the disclosure, an earthquake catalog data structure is preset, and the earthquake catalog data structure is used for storing the earthquake event information captured by GCMT websites, including the occurrence time, the seismocoordinate (latitude, longitude), the magnitude, the depth and the like of an earthquake, and through the earthquake catalog data structure, the earthquake event meeting the conditions can be screened out according to a given time range, so that a user can flexibly select an analysis time period, and finally only the occurrence time and the seismocoordinate (latitude, longitude) of the earthquake are stored.
Specifically, the embodiment of the disclosure further includes a data screening area, after capturing data such as seismic information, the data is subjected to preliminary screening, and data with non-satisfactory time or space ranges are removed, so that the subsequent InSAR data downloading request is ensured to highly agree with actual requirements.
Specifically, in the embodiment of the disclosure, the method further includes an interferogram downloading control area, and the interferogram downloading control area is mainly used for managing the co-vibration interferogram data requests downloaded from the LiCSAR website, and the data content of each request is ensured to be accurate and free through the interferogram downloading control area, so that invalid or repeated downloading is avoided.
Specifically, FIG. 3 also includes the steps of (2) crawling GCMT the seismic catalog, implementing an automated process of crawling the seismic catalog from the GCMT website. Firstly, receiving information such as starting time and ending time, a magnitude interval, a geographical range (longitude and latitude), a focus depth and the like provided by a user, automatically accessing GCMT websites according to the information, automatically extracting seismic event information meeting the conditions by utilizing a crawler technology, storing the seismic event information in a seismic catalog data structure, and setting reasonable grabbing intervals according to the updating frequency and the data change condition of the GCMT websites in order to ensure the integrity and the accuracy of data.
In this process, the seismic event information meeting the requirements is screened, the seismic events matched with the requirements of the user are captured and stored, the seismic events include but are not limited to key information such as time, seismocoordinate, magnitude, depth and the like of the occurrence of the earthquake, and the seismic catalog information (time, seismocoordinate (latitude and longitude) of the occurrence of the earthquake) meeting the requirements is saved in a document for subsequent downloading of InSAR seismogram data related to processing.
Specifically, FIG. 3 also includes the steps of (3) crawling FrameID the file, automatically crawling GCMT the seismic catalog, and then automatically crawling LiCSAR the geographic area information (longitude and latitude) corresponding to each FrameID in the website, wherein the process is used for ensuring that the accurate positioning to the correct area can be achieved when the co-seismic interferograms are downloaded from the LiCSAR website, and the geographic area information corresponding to each FrameID is stored and matched with the information in the seismic catalog so as to provide accurate positioning data for the subsequent co-seismic interferogram data downloading.
Specifically, fig. 3 further includes the steps of (4) downloading LiCSAR the seism interference pattern data, and after the completion of the capturing of the information of the seismic event and the FrameID file, obtaining FrameID corresponding to the earthquake from the FrameID file according to the seism coordinate of each seismic event in the disclosure, and further automatically sending a downloading request to the LiCSAR website through the interference pattern downloading control area. The LiCSAR website provides all InSAR interferograms of Sentinel-1 (Sentinel No. 1) satellite data, so that the system can be accurately matched with corresponding seismogram data according to the earthquake starting time information.
Specifically, once the related data is obtained, the downloaded co-seismic interferograms are stored in a local data storage module, the sources of the data (from which FrameID and interference time information and the like) are marked, so that the traceability of the data is ensured, the process can verify whether the data already exist, if so, the downloading is skipped, the next seismic interferogram is continuously downloaded, and if the downloading fails or the data is incomplete, the system can automatically retry, so that the data requested each time is ensured to be complete.
Specifically, FIG. 3 further includes the steps of (5) interferogram data processing, wherein the acquired InSAR co-seismic interferogram data generally comprises larger-sized image files, and interferograms of different seismic events may have different resolution, image size, noise and other problems. Therefore, after downloading the interferogram data, a series of preprocessing operations are performed on the data, including screening, and first, the system rapidly screens and eliminates invalid or poor quality images through a pop-up selection window. For example, if the image is seriously lost or has obvious noise, the system can automatically retain the image data with higher quality, ensure the data processed later to have higher accuracy, cut the large-size InSAR interferogram into small images, and aim to improve the calculation efficiency and reduce the memory consumption. By focusing on a specific area related to earthquake or deformation, the method can effectively reduce the interference of irrelevant areas and ensure the spatial consistency and scale standardization of data. In addition, the processing is helpful for improving the operability of data, providing higher-quality input data for the training of a subsequent deep learning model, resampling interference patterns of different sources possibly having different resolutions, resampling the cropped images by the system to meet the standard gauge dividing rate of the deep learning for the purpose of unifying data formats and facilitating subsequent analysis, resampling the resolutions of all images to target pixels such as 224×224 pixels, normalizing the images by the system for eliminating the difference between different images, comparing the data under the same standard, ensuring the uniformity of the data, and improving the stability and convergence speed of the deep learning model, wherein the data processing operation can ensure the consistency and comparability of the interference pattern data in the training of the subsequent deep learning model.
Specifically, FIG. 3 also includes the step (6) of data enhancement, which is employed in the present disclosure in order to increase the size and diversity of the data set when generating the processed data set. The data enhancement is a technology for generating new data by transforming the existing data, so that the size of a data set can be expanded, and a model can be adapted to different types of co-seismic deformation characteristics. The method comprises the steps of image rotation, image inversion, image horizontal or vertical inversion, image scaling and translation, simulation of different scales of seismic deformation, combination of various transformation forms, random combination of the transformation forms, and data enhancement, wherein the image rotation can simulate the same-seismic deformation under different visual angles through random rotation of interference images, the adaptation capability of a model to different seismic scenes is enhanced, the image inversion can be carried out, the images are horizontally or vertically inverted randomly, the diversity of samples is increased, the image scaling and translation can simulate different scales of the seismic deformation through random scaling or translation of the images, the multiple transformation forms are combined randomly, and the data enhancement step enables a data set with smaller original scale to be obviously expanded through the simple but effective image transformation, and can provide more training samples for deep learning models to learn.
Specifically, FIG. 3 further includes the steps of (7) data set generation and output, wherein through the data processing and enhancement process, the method finally generates a diversified and large-scale InSAR co-seismic deformation data set which contains co-seismic deformation images of a plurality of seismic events and covers seismic information of different geographic positions, different epicenter coordinates and different earthquake levels. The generated data set not only meets the requirement of the deep learning model on a large number of samples, but also has enough diversity, and can effectively support the training and verification of the model.
Therefore, by developing an integrated data grabbing and processing module, automatic grabbing and screening of GCMT website seismic event catalogues are realized, and the required seismic information is ensured to be accurately acquired by combining a time range designated by a user, so that the data grabbing efficiency is remarkably improved, and manual intervention is reduced. Meanwhile, through integration with LiCSAR websites, corresponding InSAR co-vibration interferogram data can be automatically matched and downloaded according to the earthquake event information, repeated downloading is avoided, storage resources are saved, and the overall operation efficiency of the system is improved. In order to meet the requirements of a deep learning model, the system performs multi-step preprocessing on downloaded co-seismic interferogram data, including data screening, clipping, resampling and normalization, ensures consistency of a data set and eliminates differences among different images. In addition, by combining various data enhancement technologies (such as image rotation, overturning, scaling and the like), the invention can effectively expand the scale and diversity of the data set and provide high-quality and diversified data support for training of the deep learning model.
Compared with the prior art, the method has the advantages that seismic data are efficiently captured and processed from GCMT and LiCSAR websites through an automatic crawler and data processing module, the complicated process of manual screening and downloading is avoided, the speed and accuracy of data acquisition are remarkably improved, the data screening, cutting, resampling and normalizing processing steps ensure that the generated data set has higher quality and consistency, in addition, the scale and diversity of the data set are expanded by combining data enhancement technologies such as image rotation, overturning and scaling, and the like, so that the robustness and adaptability of a deep learning model are improved, and the method is particularly applicable to the same-seismic deformation detection task.
Fig. 4 is a schematic structural diagram of an InSAR isomorphic deformation dataset construction device for deep learning according to an embodiment of the present disclosure, where the device may be implemented by software and/or hardware, and may be generally integrated in an electronic device. As shown in fig. 4, the apparatus includes:
A response acquisition module 401, configured to acquire crawling earthquake time and crawling earthquake information in response to a crawling request;
The acquisition and storage module 402 is configured to acquire, from a seismic information website, seismic event information that matches the crawled seismic information in the crawled seismic time based on a preset crawling technology, and store the seismic event information to a preset seismic catalog data structure;
The crawling matching module 403 is configured to crawl a frame identification file from an interferogram website, and match a seismographic coordinate corresponding to each seismic event in the seismic catalog data structure with the frame identification file to obtain a frame identification corresponding to each seismic event;
a downloading module 404, configured to download InSAR co-seismic interferogram data from the interferogram website based on the frame identifier corresponding to each seismic event;
and the image processing module 405 is configured to process the InSAR isoseism interferogram data to obtain an InSAR isoseism deformation dataset.
Optionally, the device further comprises a receiving module for receiving the starting time and the ending time input by a user, obtaining the crawling earthquake time, receiving the earthquake level interval, the geographical range and the earthquake source depth input by the user, obtaining the crawling earthquake information, and a generating module for generating the crawling request based on the crawling earthquake time and the crawling earthquake information.
Optionally, the device further comprises an acquisition adjustment module, which is used for acquiring the data updating frequency and the data change information of the seismic information website, and adjusting the crawling interval of the seismic information website based on the data updating frequency and the data change information.
Optionally, the method further comprises a screening processing module, which is used for screening the seismic event information in the seismic catalog data structure according to a preset time range and/or a preset space range.
Optionally, in the process of downloading the InSAR co-vibration interferogram data, the device further comprises a judging and processing module, wherein the judging and processing module is used for judging whether each InSAR co-vibration interferogram is in a storage database, if so, the InSAR co-vibration interferograms are not downloaded, and if any InSAR co-vibration interferogram is failed to be downloaded, the re-downloading processing is carried out according to preset downloading times.
Optionally, the image processing module is specifically configured to screen the InSAR isoseism interferogram data according to a preset image quality condition to obtain target InSAR isoseism interferogram data, perform clipping processing on each target InSAR isoseism interferogram in the target InSAR isoseism interferogram data according to a preset clipping image size to obtain to-be-processed InSAR isoseism interferogram data, and perform resampling processing and normalization processing on each to-be-processed InSAR isoseism interferogram to obtain the InSAR isoseism deformation data set.
Optionally, the image processing module is specifically further configured to perform image enhancement processing on each image in the InSAR isoseism deformation dataset, where the image enhancement processing includes one or more of image rotation, image inversion, and image scaling and translation.
The InSAR isoseism deformation data set construction device for deep learning provided by the embodiment of the disclosure can execute the InSAR isoseism deformation data set construction method for deep learning provided by any embodiment of the disclosure, and has the corresponding functional modules and beneficial effects of the execution method.
The embodiments of the present disclosure also provide a computer program product, including a computer program/instruction, which when executed by a processor, implements the method for constructing the InSAR isoseism deformation dataset for deep learning provided by any embodiment of the present disclosure.
It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of a computer-readable storage medium may include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to electrical wiring, fiber optic cable, RF (radio frequency), and the like, or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network protocol, such as HTTP (Hyper Text Transfer Protocol ), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.
The computer readable medium may be included in the electronic device or may exist alone without being incorporated into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the method described in the previous embodiments.
Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including, but not limited to, an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units involved in the embodiments of the present disclosure may be implemented by means of software, or may be implemented by means of hardware. Wherein the names of the units do not constitute a limitation of the units themselves in some cases.
The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic that may be used include Field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems-on-a-chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
According to one or more embodiments of the present disclosure, the present disclosure provides an electronic device comprising:
A processor;
a memory for storing the processor-executable instructions;
the processor is configured to read the executable instructions from the memory and execute the instructions to implement any one of the InSAR isomorphic deformation data set construction methods for deep learning provided in the present disclosure.
According to one or more embodiments of the present disclosure, the present disclosure provides a computer-readable storage medium storing a computer program for performing an InSAR isomorphous dataset construction method for deep learning as provided in any one of the present disclosure.
The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).
Moreover, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are example forms of implementing the claims.

Claims (10)

CN202510350684.6A2025-03-242025-03-24 Method and device for constructing InSAR co-seismic deformation dataset for deep learningPendingCN120339743A (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202510350684.6ACN120339743A (en)2025-03-242025-03-24 Method and device for constructing InSAR co-seismic deformation dataset for deep learning

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202510350684.6ACN120339743A (en)2025-03-242025-03-24 Method and device for constructing InSAR co-seismic deformation dataset for deep learning

Publications (1)

Publication NumberPublication Date
CN120339743Atrue CN120339743A (en)2025-07-18

Family

ID=96360962

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202510350684.6APendingCN120339743A (en)2025-03-242025-03-24 Method and device for constructing InSAR co-seismic deformation dataset for deep learning

Country Status (1)

CountryLink
CN (1)CN120339743A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN110929128A (en)*2019-12-112020-03-27北京启迪区块链科技发展有限公司Data crawling method, device, equipment and medium
CN113933838A (en)*2021-09-302022-01-14应急管理部国家自然灾害防治研究院InSAR deformation monitoring method and equipment for multiple geological disaster scenes
CN119087510A (en)*2024-08-272024-12-06中国地震局第二监测中心 A method for automatically and quickly generating co-seismic deformation fields based on focal mechanism solutions

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN110929128A (en)*2019-12-112020-03-27北京启迪区块链科技发展有限公司Data crawling method, device, equipment and medium
CN113933838A (en)*2021-09-302022-01-14应急管理部国家自然灾害防治研究院InSAR deformation monitoring method and equipment for multiple geological disaster scenes
CN119087510A (en)*2024-08-272024-12-06中国地震局第二监测中心 A method for automatically and quickly generating co-seismic deformation fields based on focal mechanism solutions

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LZJ把脉: "LiCSBAS/LiCSAR", pages 1 - 8, Retrieved from the Internet <URL:weibo.com/ttarticle/p/show?id=2309404994488765382824>*
XU LIU: "A Global Coseismic InSAR Dataset for Deep Learning : Automated Construction from Sentinel-1 Observations", MDPI, no. 2025, 23 May 2025 (2025-05-23), pages 1832*

Similar Documents

PublicationPublication DateTitle
US20230325402A1 (en)Server implemented geographic information system with graphical interface
US10372705B2 (en)Parallel querying of adjustable resolution geospatial database
CN108108499B (en)Face retrieval method, device, storage medium and equipment
EP3678088B1 (en)Method and system of convolution in neural network with variable dilation rate
CN112906638A (en)Remote sensing change detection method based on multi-level supervision and depth measurement learning
CN112906662B (en)Method, device and equipment for detecting change of remote sensing image and storage medium
CN114742707B (en)Multi-source remote sensing image splicing method and device, electronic equipment and readable medium
CN109491994B (en) The most simplified screening method for the selected remote sensing data set of Landsat-8 satellite
Masci et al.A new forced photometry service for the Zwicky Transient Facility
CN119671960A (en) Building detection method, device and equipment based on remote sensing image and laser point cloud
CN119598873A (en)Numerical value determination method and device for solar irradiance in preset time period
CN120339743A (en) Method and device for constructing InSAR co-seismic deformation dataset for deep learning
Zhang et al.Superresolution imaging with a deep multipath network for the reconstruction of satellite cloud images
Saleh et al.1996–2017 GPS position time series, velocities and quality measures for the CORS Network
CN117197278A (en) A satellite cloud image reconstruction method and system
Park et al.Significant winter Atlantic Niño effect on ENSO and its future projection
CN114677590B (en) Remote sensing image processing method, device, equipment and medium
CN114092762B (en) Remote sensing image classification training sample generation method, device and electronic equipment
CN113436074A (en)Meteorological image processing method, device, equipment and storage medium
Panidi et al.Cloud-Desktop remote sensing data management to ensure time series analysis, integration of QGIS and Google Earth Engine
CN113554747A (en)Unmanned aerial vehicle inspection data viewing method based on three-dimensional model
CN118394869B (en) A method and system for collecting geographic information for national land space planning
CN119624766B (en)Global precipitation data splicing and synthesizing method and device
Jin et al.Analysis of the Modeling Method and Application of 3D City Model based on the CityEngine
LeonardHydroTerre: Towards an expert system for scaling hydrological data and models from hill-slopes to major-river basins

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination

[8]ページ先頭

©2009-2025 Movatter.jp