CN116403093A - Multi-source remote sensing interpretation method and device for potential landslide along railway - Google Patents
Multi-source remote sensing interpretation method and device for potential landslide along railway Download PDFInfo
- Publication number
- CN116403093A CN116403093A CN202310678080.5A CN202310678080A CN116403093A CN 116403093 A CN116403093 A CN 116403093A CN 202310678080 A CN202310678080 A CN 202310678080A CN 116403093 A CN116403093 A CN 116403093A
- Authority
- CN
- China
- Prior art keywords
- information
- remote sensing
- sensing data
- railway
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013178 mathematical model Methods 0.000 claims abstract description 59
- 238000004458 analytical method Methods 0.000 claims abstract description 41
- 230000004927 fusion Effects 0.000 claims abstract description 22
- 238000007499 fusion processing Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 70
- 238000000605 extraction Methods 0.000 claims description 53
- 230000008569 process Effects 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 23
- 230000008859 change Effects 0.000 claims description 22
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 230000002068 genetic effect Effects 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 abstract description 7
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 230000010354 integration Effects 0.000 description 9
- 238000012360 testing method Methods 0.000 description 9
- 230000010287 polarization Effects 0.000 description 8
- 238000006073 displacement reaction Methods 0.000 description 7
- 238000007619 statistical method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000000551 statistical hypothesis test Methods 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B15/00—Measuring arrangements characterised by the use of electromagnetic waves or particle radiation, e.g. by the use of microwaves, X-rays, gamma rays or electrons
- G01B15/06—Measuring arrangements characterised by the use of electromagnetic waves or particle radiation, e.g. by the use of microwaves, X-rays, gamma rays or electrons for measuring the deformation in a solid
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1793—Remote sensing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a multisource remote sensing interpretation method and device for potential landslide along a railway, which relate to the technical field of engineering geological survey and comprise the steps of acquiring first information and second information; constructing a target recognition mathematical model according to the second information, and recognizing SAR remote sensing data according to the target recognition mathematical model to obtain third information; interpreting the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information; according to the visible light, near infrared and thermal infrared remote sensing data, fifth information is identified and extracted by analyzing the temperature characteristics of different ground objects at different times; and carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information. The invention improves the identification accuracy and the comprehensiveness of potential landslide through multi-source data fusion and comprehensive analysis, and provides effective monitoring and evaluation means for the safe operation of railway lines.
Description
Technical Field
The invention relates to the technical field of engineering geological investigation, in particular to a multisource remote sensing interpretation method and device for potential landslide along a railway.
Background
Along with the rapid development of remote sensing technology, the remote sensing data are increasingly widely applied to the field of monitoring and evaluating geological disasters along the railway. The remote sensing interpretation is taken as an effective method, and the earth surface information and the landform characteristics can be extracted through processing and analyzing the remote sensing data, so that the monitoring and the prediction of geological disasters are realized. However, existing remote sensing interpretation methods have some limitations in facing the challenges of complex geological environments and multi-source data fusion. The existing remote sensing interpretation method usually adopts a single data source to analyze, lacks comprehensive utilization of multi-source data, and is not capable of comprehensively capturing important information such as surface deformation, ground cracks and the like by only relying on the single remote sensing data for identifying and evaluating potential landslide along a railway. In addition, the traditional method is often based on feature extraction and classification at the pixel level, lacks deep analysis on micro-landform features, ground surface change information and the like, and limits the accurate identification and prediction capability of potential landslide.
Therefore, in order to overcome the limitations of the existing remote sensing interpretation method, the invention provides a multisource remote sensing interpretation method for potential landslide along a railway.
Disclosure of Invention
The invention aims to provide a multisource remote sensing interpretation method and device for potential landslide along a railway so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
on one hand, the application provides a multisource remote sensing interpretation method for potential landslide along a railway, which comprises the following steps:
acquiring first information and second information, wherein the first information comprises SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area, the second information comprises historical SAR remote sensing data, and the historical SAR remote sensing data comprises structural information along a railway;
constructing a target recognition mathematical model according to the second information, and carrying out recognition processing on the SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of a railway structure;
performing interpretation processing on the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of potential landslide and key earth surface deformation information;
According to the visible light remote sensing data, the near infrared remote sensing data, the thermal infrared remote sensing data and the geological data of the target area, fifth information is identified and extracted by analyzing the temperature characteristics of different ground features at different times, wherein the fifth information comprises surface variation information along a railway, ground crack information of a potential landslide and micro-landform feature interpretation results;
and carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of potential landslide.
On the other hand, the application also provides a multisource remote sensing interpretation device facing the potential landslide along the railway, which comprises:
the acquisition module is used for acquiring first information and second information, wherein the first information comprises SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area, the second information comprises historical SAR remote sensing data, and the historical SAR remote sensing data comprises structural information along a railway;
the construction module is used for constructing a target recognition mathematical model according to the second information, and carrying out recognition processing on the SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of a railway structure;
The interpretation module is used for performing interpretation processing on the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of the potential landslide and key earth surface deformation information;
the extraction module is used for identifying and extracting fifth information according to the visible light remote sensing data, the near infrared remote sensing data, the thermal infrared remote sensing data and the geological data of the target area by analyzing the temperature characteristics of different ground features at different times, wherein the fifth information comprises ground surface change information along a railway, ground crack information of a potential landslide and micro-landform feature interpretation results;
and the fusion module is used for carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of the potential landslide.
The beneficial effects of the invention are as follows:
the invention realizes the comprehensive monitoring of the railway line by utilizing multi-source remote sensing data including SAR remote sensing data, multi-platform and multi-time-phase visible light, near infrared and thermal infrared remote sensing data. By utilizing the historical SAR remote sensing data and combining with the extraction of the information of the railway structures along the railway, the railway structures can be identified and the influence of the railway structures can be accurately estimated. Meanwhile, the phase difference in SAR remote sensing data to be interpreted is analyzed, so that the earth surface deformation along the railway can be interpreted, and the evolution process of the potential landslide and key earth surface deformation information can be obtained. The invention improves the identification accuracy and the comprehensiveness of potential landslide through multi-source data fusion and comprehensive analysis, and provides effective monitoring and evaluation means for the safe operation of railway lines.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a multi-source remote sensing interpretation method for potential landslide along a railway, which is described in the embodiment of the invention;
fig. 2 is a schematic structural diagram of a multi-source remote sensing interpretation device for potential landslide along a railway in an embodiment of the invention.
The marks in the figure: 1. an acquisition module; 2. constructing a module; 21. a first processing unit; 22. a first extraction unit; 221. a first clustering unit; 222. a first analysis unit; 223. a second extraction unit; 224. a first integration unit; 23. a first training unit; 24. a first identification unit; 3. an interpretation module; 31. a second processing unit; 32. a first calculation unit; 33. a third processing unit; 34. a second analysis unit; 4. an extraction module; 41. a third analysis unit; 42. a fourth analysis unit; 43. a third extraction unit; 44. a fourth extraction unit; 5. a fusion module; 51. a second calculation unit; 52. a first optimizing unit; 53. a first fusion unit.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1:
the embodiment provides a multisource remote sensing interpretation method for potential landslide along a railway.
Referring to fig. 1, the method is shown to include steps S100, S200, S300, S400, and S500.
Step S100, acquiring first information and second information, wherein the first information comprises SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area, the second information comprises historical SAR remote sensing data, and the historical SAR remote sensing data comprises structural information along a railway.
It will be appreciated that in this step, the source of the first information includes multi-source, multi-platform, multi-phase remote sensing data such as SAR, visible light, near infrared, thermal infrared, etc. to provide a comprehensive information base. The second information is historical SAR remote sensing data, which contains information of structures along the railway. These historical data record the status and changes of structures along the railway and provide an important reference for subsequent identification and interpretation. By acquiring the information, a comprehensive data base can be established, and a foundation is laid for analysis and processing of subsequent steps.
And step 200, constructing a target recognition mathematical model according to the second information, and carrying out recognition processing on SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of the railway structure.
It can be understood that in this step, the mathematical model for target recognition is to extract the characteristic parameters of the structural information of the railway in the historical SAR remote sensing data by analyzing and processing the structural information of the railway, and take the characteristic parameters as the input of the model. Through training and optimizing the model, the railway structure in SAR remote sensing data to be interpreted can be accurately identified by the model. Through the step, the railway structure can be automatically identified and extracted, the requirement of manual intervention is reduced, and the efficiency and accuracy are improved. The step 200 includes step S210, step S220, step S230, and step S240.
And step S210, preprocessing the historical SAR remote sensing data in the second information to obtain preprocessed data.
It will be appreciated that in this step, the pre-processing is to optimise the quality of the data and to accommodate the need for subsequent processing. A series of preprocessing operations, such as noise removal, geometric distortion correction, atmospheric interference removal, etc., are often required before processing the historical SAR remote sensing data. These preprocessing steps aim to improve the usability and accuracy of the data and provide a more reliable data basis for subsequent object recognition and surface deformation analysis. Through preprocessing operation, necessary data processing and correction can be carried out on the historical SAR remote sensing data, unnecessary interference and error are eliminated, and the data is more reliable and accurate.
And S220, identifying and extracting SAR strong scattering characteristics of structures along the railway according to the preprocessing data to obtain a characteristic set.
It is understood that in this step, the strong scattering characteristic of SAR means that in the SAR image, the structure has higher intensity for scattering reflection of radar waves. By analyzing and processing the pre-processed data, the strong scattering features exhibited by the railway line structures in the SAR image can be identified and extracted using image processing and signal processing techniques. First, structures along railways typically have more regular geometries such as retaining walls, piling slabs, concrete skeleton revetments, and the like. The edges, corner points and surface features of these structures may cause radar waves to reflect and scatter when interacting therewith. These structures present more pronounced edges and corner points in the SAR image than natural features, resulting in strong scattering signals. Second, structures along railways often employ materials with higher reflectivity and scattering, such as metal, concrete, and the like. The materials have good electromagnetic wave reflection performance, and can effectively reflect incident radar waves back to form strong scattering signals. In contrast, naturally, materials such as soil, vegetation, and the like have relatively weak reflective and scattering properties. And the structures along the railway are generally distributed according to a certain rule, such as retaining walls, pile plate walls, concrete skeleton revetments and the like with fixed intervals. This regular arrangement results in repeatability and consistency between structures such that they form a distinct repeated reflection pattern in the SAR image, further enhancing the strong scattering signal. The step S220 includes a step S221, a step S222, a step S223, and a step S224.
And S221, carrying out clustering processing according to the structure remote sensing data in the historical SAR remote sensing data to obtain clustering results of different structure categories.
It will be appreciated that the structures along the railway are classified in this step into different categories. Clustering is an unsupervised learning method that classifies data points with similar characteristics into the same category by performing similarity analysis and clustering on the data. In this step, the structure remote sensing data in the history SAR remote sensing data is used as input data, and the structure data is grouped by a clustering algorithm such as K-means clustering or hierarchical clustering. In this way, structures along the railway can be classified into different categories according to their characteristics and properties. The clustering process can classify the structures according to the characteristics of reflection and scattering characteristics, shape, size and the like of the structures, so that a basis is provided for subsequent characteristic extraction and target identification.
Step S222, carrying out descriptive statistical analysis on the clustering result according to a preset Krueschel-Wallis test mathematical model to determine strong scattering feature standards of various structures, and determining statistical significance and stability of various features by using hypothesis test and confidence interval to obtain feature extraction standards, wherein the feature extraction standards comprise polarization feature extraction standards, interference phase feature extraction standards and reflection intensity feature extraction standards.
It will be appreciated that the objective in this step is to perform a feature analysis on different classes of structures based on the clustering results, and determine the criteria for their strong scattering characteristics. First, the clustering result is subjected to descriptive statistical analysis through a KluyKarl-Wolis test mathematical model. The model is a non-parametric statistical method for comparing differences between multiple independent samples. In this step, different classes of structures were taken as independent samples, and their differences in strong scattering characteristics were evaluated by the kruekarl-vorax test. Through descriptive statistical analysis, statistical indexes such as average value, variance, distribution and the like among different types of structures can be known, and a basis is provided for subsequent feature extraction. Statistical significance and stability of the feature is then determined using the hypothesis test and confidence interval. Hypothesis testing is a statistical inference method used to verify whether a hypothesis is true. In this step, it is compared by hypothesis testing whether the differences in the strong scattering features of the different classes of structures are statistically significant. Meanwhile, the confidence interval is used for estimating the true value range of the feature so as to evaluate the stability and reliability of the feature. And finally, obtaining a feature extraction standard according to the result of descriptive statistical analysis, hypothesis test and confidence interval inference. The feature extraction criteria include a polarization feature extraction criteria, an interference phase feature extraction criteria, and a reflection intensity feature extraction criteria. Through statistical analysis and inference, the obvious difference and stability of each feature in different types of structures are determined, and an accurate reference standard is provided for the subsequent feature extraction process.
Step S223, carrying out feature extraction on each pixel according to feature extraction standard preprocessing data to obtain feature data, wherein the feature data comprises polarization feature data, interference phase feature data and intensity feature data.
It can be understood that in this step, for each pixel, corresponding polarization characteristics, interference phase characteristics, and reflection intensity characteristics are extracted according to parameters and algorithms determined by the feature extraction criteria. The polarization characteristic data represents characteristic information on different polarization channels, the interference phase characteristic data represents phase difference information between a plurality of pieces of phase data, and the reflection intensity characteristic data represents reflection intensity information in the SAR image. These feature data will be used in subsequent analysis and processing to effect feature extraction and comprehensive information acquisition of the railway line structures.
And step 224, performing feature integration processing on the feature data according to a preset entropy weight multi-objective decision mathematical model to obtain a feature set.
It can be understood that in this step, the weights of different features in feature integration are determined according to a preset entropy weight multi-objective decision mathematical model. The weights reflect the importance and contribution degree of each feature in comprehensive analysis, and the accuracy and reliability of feature integration are ensured through calculation and optimization of a mathematical model. And then, carrying out weighted integration on the characteristic data according to the determined weight. In the process of feature integration, the feature data of each pixel point is weighted and summed according to the weight of each feature, so that a comprehensive feature value is obtained, and the comprehensive feature of the pixel point is represented. The feature set will be used for subsequent analysis and processing, such as object recognition, surface deformation analysis, etc., to enable a comprehensive assessment of potential landslides.
And step S230, training by using a supervised learning algorithm according to the feature set to obtain a target recognition mathematical model for recognizing the railway line structures.
Preferably, the present embodiment uses a support vector machine algorithm for training. The feature set is first required as a training data set, wherein each data sample contains feature data from structures along the railway and corresponding labels for representing the categories of structures. At the beginning of the training process, feature data is converted into numerical values with the same scale and distribution by preprocessing and normalizing the feature set, which helps to avoid excessive influence of certain features on model training. Next, a Gaussian kernel function is selected to map the feature space and optimize the performance of the model by adjusting the hyper-parameters. Preferably, the selection of these superparameters may be determined by cross-validation or the like. Model training is then performed using the training dataset. The support vector machine algorithm maximizes the distance of samples of different classes from the hyperplane as far as possible by mapping the feature data to a high-dimensional space and finding an optimal hyperplane. This allows for accurate classification of structures. In the training process, the support vector machine algorithm can also adjust the decision boundary of the model according to the relation between the sample characteristics and the labels in the training data set so as to achieve a better classification effect, and find the optimal hyperplane through solving the optimization problem so as to separate structures of different categories to the greatest extent. Finally, through repeated iterative training and parameter adjustment, a vector machine algorithm generates a target recognition mathematical model capable of accurately classifying new feature data. When the model faces new characteristic data, accurate classification judgment can be carried out according to the characteristic value and the similarity between the model and a trained sample, and important data support is provided for subsequent analysis and processing.
And step 240, according to the obtained target recognition mathematical model, recognizing the SAR remote sensing data to be interpreted to obtain a recognition result of the railway structure.
It can be understood that in this step, feature extraction is performed on each pixel of the data to be interpreted, the extracted feature data is input into a target recognition mathematical model for classification, and the model determines whether each pixel belongs to a railway structure according to knowledge and decision boundaries learned in the training process. Through the step, the recognition result of the railway structure aiming at the SAR remote sensing data to be interpreted can be obtained, the recognition result can indicate which pixel points belong to the railway structure, and an important information basis is provided for subsequent surface deformation analysis and potential landslide prediction.
And step S300, performing interpretation processing on the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of the potential landslide and key earth surface deformation information.
It can be understood that in this step, the historical SAR remote sensing data and the SAR remote sensing data to be interpreted in the third information are used to perform the phase difference processing. And calculating the phase difference between adjacent time points by carrying out interference processing on SAR data at different times. The phase differences can reflect deformation conditions of the earth surface, including horizontal displacement, vertical displacement, deformation rate and other information. And then, according to the position and the shape of the railway structure provided in the third information, combining the phase difference image in the SAR remote sensing data to be interpreted, and performing interpretation processing on the surface deformation along the railway. By analyzing the spatial distribution and the change trend of the phase difference, the evolution process of the potential landslide can be identified, wherein the evolution process comprises key information such as expansion, sliding direction, landslide speed and the like of a landslide body. And simultaneously, carrying out quantitative analysis on the surface deformation by combining calculation of the reflection intensity and the phase difference of the railway structure provided in the third information. The reflection intensity of the structure can be used as a reference to further deduce the numerical variation of the surface deformation, such as the magnitude and direction of horizontal displacement, the elevation variation of vertical displacement, and the like. Through the interpretation processing of step S300, key information of the deformation of the surface along the railway can be obtained, including the evolution process of the potential landslide and important surface deformation characteristics. The step S300 includes a step S310, a step S320, a step S330, and a step S340.
And step S310, carrying out interference superposition processing on SAR data at different times in SAR remote sensing data to be interpreted, and obtaining earth surface deformation information through integrating the phase difference.
It can be understood that in this step, the SAR data at different time points are subjected to interference processing, and the phase difference between adjacent time points is calculated. The phase difference reflects the change condition of the surface deformation, including horizontal displacement, vertical displacement and the like. And integrating the phase difference into one image through interference superposition processing to form a surface deformation information graph. This image shows the relative displacement and deformation of different regions or sites over different time periods. Preferably, the surface deformation information may be represented by color coding or gray values, different colors or gray values corresponding to different degrees of deformation. The surface deformation information is used for revealing the evolution process of the potential landslide, determining key surface deformation characteristics and providing basis for risk assessment and engineering decision of the potential landslide.
Step S320, according to the ground surface deformation information and the third information, analyzing and counting the reflection intensity of different railway structures, and combining with the calculation of the phase difference to obtain the evolution process of the potential landslide, wherein the evolution process comprises the weak deformation information of the potential landslide.
It will be appreciated that the reflection strength remains unchanged to some extent due to deformation of the ground surface in the potential landslide area, while the railway structure is relatively stable. Therefore, by comparing and analyzing the reflection intensity data of different time periods, the evolution process of the potential landslide area and key surface deformation information can be identified. By analyzing and extracting the strong scattering characteristics of these structures along the railway, a more robust extraction of the weak deformation information of the potential landslide can be achieved.
And step S330, according to the third information and the space coordinate information of the SAR remote sensing data to be interpreted, carrying out space superposition on the deformation information to obtain comprehensive earth surface deformation information.
Preferably, in this embodiment, the surface deformation information is spatially superimposed by means of a GIS (geographic information system) technique, so as to obtain comprehensive surface deformation information. Firstly, the SAR remote sensing data and the deformation data to be interpreted are loaded into a GIS environment through a space analysis tool in GIS software, and are matched according to the space positions of the SAR remote sensing data and the deformation data. In a GIS environment, interpolation processing is carried out on the deformation data by utilizing a spatial interpolation technology so as to fill the defect or uneven distribution of the deformation data in space. By interpolation, a more continuous and accurate deformed surface can be obtained. And then, overlapping the interpolated deformation data with SAR remote sensing data to be interpreted. The method can be realized by a space superposition analysis tool in GIS software, the space positions of the deformation data and the SAR remote sensing data are matched, and the numerical value of the deformation data is given to the corresponding SAR remote sensing pixel. Through the space superposition process, the deformation information can be associated with the surface characteristics, and comprehensive surface deformation information can be obtained in a GIS environment. The comprehensive surface deformation information reflects deformation conditions of different areas along the railway, and reveals the evolution process of potential landslide.
And S340, comprehensively evaluating and analyzing the comprehensive surface deformation information based on a preset worst case analysis set mathematical model, and screening to obtain key surface deformation information.
It will be appreciated that in this step, the least favorable case analysis gathers mathematical models taking into account a number of possible least favorable cases and gathers them together to obtain comprehensive surface deformation information. In the analysis process, the model comprehensively considers different surface deformation factors such as strength, spatial distribution, time variation and the like, and the relation between the surface deformation factors and structures and geological conditions along the railway. The comprehensive surface deformation information can be quantized and compared through operation and analysis of a mathematical model, and the surface deformation characteristics with important significance can be found out. The screened key surface deformation information has important geological, engineering and safety significance. Such information may be used to guide engineering design, monitoring, and risk assessment, as well as to formulate corresponding countermeasures.
And step 400, identifying and extracting fifth information according to visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area by analyzing temperature characteristics of different ground features at different times, wherein the fifth information comprises ground surface change information along a railway, ground crack information of a potential landslide and micro-landform feature interpretation results.
It can be understood that in the step, by utilizing an image processing and feature extraction algorithm, the surface change information, the ground crack information and the micro-landform feature interpretation result of the potential landslide along the railway can be identified, and an important data basis is provided for subsequent analysis and processing. The step S400 includes a step S410, a step S420, a step S430, and a step S440.
And step S410, analyzing and processing according to the visible light remote sensing data, and comparing the change conditions of the earth surface in the optical images at different times to obtain earth surface change information along the railway.
It can be understood that in this step, the visible light remote sensing image data of different time periods are collected, the images cover the target area along the railway, the image of each time period reflects the state of the earth surface in different time periods, and the change condition of the earth surface along the railway is observed and analyzed by comparing the images of different time periods and utilizing the image processing and feature extraction algorithm. The changes that occur include changes in the type of land used (e.g., farmland converted to construction land), changes in vegetation cover, expansion or contraction of the body of water, etc. By quantitatively analyzing and counting the changes in the images, the surface change information along the railway can be obtained, and the information comprises changed areas, changed degrees, changed trends and the like.
And S420, analyzing the temperature change modes of the ground object of the target area in different time periods according to the near infrared remote sensing data and the thermal infrared remote sensing data to obtain temperature characteristic data.
It can be understood that in this step, temperature data of different features in the target area in different time periods are analyzed by using multi-platform and multi-phase near infrared and thermal infrared remote sensing data. Through analysis of the temperature change mode, temperature characteristic data of the ground object of the target area can be obtained. The characteristic data comprise information such as temperature change trend, temperature fluctuation range, temperature spatial distribution and the like of the ground object in different seasons and different time periods.
And step S430, identifying and extracting the multi-visible light remote sensing data according to the temperature characteristic data to obtain the ground crack information of the potential landslide along the railway.
It will be appreciated that the identification and interpretation of surface features may be plagued by vegetation shadows in the remote sensing image, which may mask or obscure the presence of cracks, making them difficult to accurately extract and identify. In the step, the vegetation shadows in the images are detected and removed by analyzing the multi-platform and multi-time-phase visible light remote sensing data and the temperature characteristic data, so that the positions and the forms of the ground cracks are accurately extracted. Preferably, the ground fault information is extracted using a threshold segmentation algorithm in this embodiment. Firstly, converting a visible light remote sensing image with multiple platforms and multiple time phases into a binary image, and distinguishing a vegetation area from other ground objects by selecting a proper threshold value, wherein the vegetation area is marked as white, and a non-vegetation area is marked as black. And then carrying out corrosion and expansion treatment on the binary image. The erosion operation may reduce the size of the vegetation area and the expansion operation may fill the voids of the vegetation area. Therefore, the influence of vegetation shadows can be eliminated, and the ground cracks are more obvious and clear. And finally, extracting the ground crack information by comparing the processed image with the original remote sensing image. The ground fissures typically exhibit linear or fissured features, and thus edge detection algorithms are utilized to further enhance and extract morphological features of the ground fissures.
And S440, extracting the landform features of the visible light remote sensing data according to a preset cavity convolution mathematical model, and carrying out superposition processing on the extracted potential landslide micro-landform features and a three-dimensional landform model in geological data of a target area to obtain a micro-landform feature interpretation result.
It can be understood that in this step, the visible light remote sensing data of multiple platforms and multiple time phases is processed by using a preset cavity convolution mathematical model, so as to identify and extract the potential landslide micro-landform features. These features include closed depressions, bulge, fissures, vegetation anomalies, spring water, and the like. The cavity convolution model is an algorithm specially designed for extracting specific relief elements, and the relief features with specific forms and scales are highlighted by sliding cavity windows with different sizes on the image. And then, superposing the extracted potential landslide micro-relief features with the three-dimensional topographic model of the working area. The three-dimensional terrain model can provide elevation information of the earth surface and reflect the fluctuation of the terrain. By superposing the micro-relief features with the three-dimensional relief model, more accurate spatial positioning and morphological feature description of the micro-relief features can be realized. In the interpretation process, other auxiliary data and geological data, such as topographic maps of different scales, regional geological data, hydrogeological information, seismic records, mineral products distribution and the like, and an iterative training method are combined, so that the interpretation precision is gradually improved. Through comprehensive analysis and comprehensive interpretation of the multi-source data, detailed information about the micro-geomorphic features of potential landslide along the railway can be obtained.
And S500, carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of the potential landslide.
It can be understood that in this step, through the fusion process, the sixth information comprehensively considers each feature in the fourth information and the fifth information, and combines analysis and decision rules of the sign fusion mathematical model to obtain comprehensive information on the potential landslide. This information, including the comprehensive trade-offs and evaluation results between different features, may provide a more comprehensive and accurate description of potential landslide conditions. The step S500 includes a step S510, a step S520, and a step S530.
And S510, performing feature extraction processing according to the fourth information and the fifth information, and performing fuzzy membership calculation on the extracted feature information to obtain a fuzzy feature set.
It can be understood that the fuzzy membership calculation is performed for each feature on the basis of feature extraction. The fuzzy membership calculation is to convert the numerical type characteristic into a membership function form and is used for describing the fuzzy relation between the characteristic and the potential landslide. Through fuzzy membership calculation, the characteristic value can be associated with membership degrees of different degrees to reflect the relative importance or correlation of the characteristic value in potential landslide. Finally, obtaining a fuzzy feature set through feature extraction and fuzzy membership calculation. The fuzzy membership value for each feature is included in this set to describe the degree of contribution or impact of that feature on the potential landslide. Such fuzzy feature sets can provide more information, helping an analyst to better understand the features of potential landslides and their role in the overall.
And step S520, carrying out genetic optimization processing according to the fuzzy feature set, and obtaining a feature combination by carrying out optimization selection and cross variation on the fuzzy features.
It can be understood that the genetic optimization algorithm is a heuristic search algorithm based on the biological evolution theory, simulates the biological evolution process, and searches for the optimal solution through operations such as superior and inferior elimination, crossover, mutation and the like. In this step, a series of different feature combinations can be obtained by continuously selecting, intersecting and mutating features through the iterative process of the genetic optimization algorithm. The combinations are subjected to fitness evaluation, each feature combination is ranked and screened according to specific evaluation criteria, such as an objective function or a performance index, and finally the optimal feature combination is selected.
And step S530, training the feature combination according to a preset neural network mathematical model, and obtaining the integrated information of the fused potential landslide through the association relation between learning and insufficient features.
It can be understood that in this step, the preset neural network mathematical model is used to train the feature combinations, and the association relationship between the features is learned and captured to obtain comprehensive information of the potential landslide more comprehensively and accurately. In the training process, the feature combination is fed into a neural network model as input data, and the neural network model carries out forward propagation and backward propagation through neurons and weight parameters of a plurality of layers so as to gradually adjust and optimize the parameters of the model, so that the relevance among the features can be better captured, and the integrated information of the fused potential landslide is generated. Through neural network training, feature combinations are effectively learned and integrated so that various aspects of potential landslide are comprehensively considered and analyzed. The integrated information of the potential landslide after fusion can describe the characteristics, evolution process and relevance of the potential landslide more comprehensively and accurately, and provide more valuable information basis for further decision making and analysis. Furthermore, the integrated information of the fused potential landslide can be combined with expert decisions to finally obtain the planar information of the potential landslide of the working area. Expert decisions are tradeoffs and judgments based on expertise and experience for evaluating and analyzing potential landslides. And providing the integrated comprehensive information of the potential landslide to an expert for comprehensive evaluation and decision making so as to further verify and confirm the existence and degree of the potential landslide. Expert decisions involve a number of factors including the geological conditions of the work area, engineering construction requirements, environmental impact, etc. And the expert judges and deduces the planar information of the potential landslide according to the expert knowledge and experience of the expert and by combining the integrated comprehensive information of the potential landslide. This may include spatial distribution of potential landslides, boundary features, risk levels, etc. By combining with expert decision, the influence of different factors can be comprehensively considered, so that the planar information of the potential landslide of the working area is more accurate and reliable. Such information can provide important basis for relevant decision makers to take corresponding control measures, adjust engineering design or formulate reasonable risk management strategies, thereby reducing risks and influences of potential landslides on engineering and environment.
Example 2:
as shown in fig. 2, the present embodiment provides a multi-source remote sensing interpretation device for potential landslide along a railway, the device includes:
the acquisition module 1 is configured to acquire first information and second information, where the first information includes SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data, and geological data of a target area, the second information includes historical SAR remote sensing data, and the historical SAR remote sensing data includes railway line structure information.
And the construction module 2 is used for constructing and obtaining a target recognition mathematical model according to the second information, and carrying out recognition processing on SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of the railway structure.
And the interpretation module 3 is used for performing interpretation processing on the surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of the potential landslide and key surface deformation information.
The extraction module 4 is configured to identify and extract fifth information according to the visible light remote sensing data, the near infrared remote sensing data and the thermal infrared remote sensing data by analyzing temperature characteristics of different features at different times, where the fifth information includes surface change information along a railway, land crack information of a potential landslide, and a micro-landform feature interpretation result.
And the fusion module 5 is used for carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of the potential landslide.
In one embodiment of the present disclosure, the build module 2 includes:
the first processing unit 21 is configured to pre-process the historical SAR remote sensing data in the second information to obtain pre-processed data.
The first extraction unit 22 is configured to identify and extract, according to the preprocessing data, the SAR strong scattering characteristics of the structures along the railway, and obtain a characteristic set.
The first training unit 23 is configured to train by using a supervised learning algorithm to obtain a mathematical model for identifying the object along the railway structure according to the feature set.
The first identifying unit 24 is configured to identify the remote sensing data of the SAR to be interpreted according to the obtained target identification mathematical model, so as to obtain an identification result of the railway structure.
In one embodiment of the present disclosure, the first extraction unit 22 includes:
the first clustering unit 221 is configured to perform clustering processing according to the structure remote sensing data in the historical SAR remote sensing data, so as to obtain clustering results of different structure categories.
The first analysis unit 222 is configured to perform descriptive statistical analysis on the clustering result according to a preset krueschel-wales test mathematical model to determine strong scattering feature standards of each structure, and determine statistical significance and stability of each feature by using hypothesis test and confidence interval to obtain feature extraction standards, where the feature extraction standards include a polarization feature extraction standard, an interference phase feature extraction standard and a reflection intensity feature extraction standard.
The second extracting unit 223 is configured to perform feature extraction on each pixel according to feature extraction standard preprocessing data to obtain feature data, where the feature data includes polarization feature data, interference phase feature data, and intensity feature data.
The first integration unit 224 is configured to perform feature integration processing on the feature data according to a preset entropy weight multi-objective decision mathematical model, so as to obtain a feature set.
In one embodiment of the present disclosure, the interpretation module 3 includes:
the second processing unit 31 is configured to perform interference superposition processing on SAR data of different times in the SAR remote sensing data to be interpreted, and obtain surface deformation information by performing integration processing on the phase difference.
The first calculating unit 32 is configured to obtain an evolution process of the potential landslide by analyzing and counting the reflection intensities of different railway structures according to the surface deformation information and the third information and combining with the calculation of the phase difference, where the evolution process includes weak deformation information of the potential landslide.
And the third processing unit 33 is configured to spatially superimpose the deformation information according to the third information and the spatial coordinate information of the SAR remote sensing data to be interpreted to obtain comprehensive surface deformation information.
The second analysis unit 34 performs comprehensive evaluation and analysis on the comprehensive surface deformation information based on the preset worst case analysis set mathematical model, and screens to obtain key surface deformation information.
In one embodiment of the present disclosure, the extraction module 4 includes:
the third analysis unit 41 is configured to perform analysis processing according to the remote sensing data of visible light, and obtain surface variation information along the railway by comparing the surface variation conditions in the optical images at different times.
And a fourth analysis unit 42, configured to analyze the temperature change patterns of the ground object in the target area according to the near infrared remote sensing data and the thermal infrared remote sensing data, so as to obtain temperature characteristic data.
And the third extraction unit 43 is used for identifying and extracting the visible light remote sensing data according to the temperature characteristic data to obtain the ground crack information of the potential landslide along the railway.
And a fourth extraction unit 44, configured to extract the landform features of the visible light remote sensing data according to a preset cavity convolution mathematical model, and perform superposition processing on the extracted potential landslide micro-landform features and the three-dimensional topographic model of the working area, so as to obtain a micro-landform feature interpretation result.
In one embodiment of the present disclosure, the fusion module 5 includes:
the second calculating unit 51 is configured to perform feature extraction processing according to the fourth information and the fifth information, and perform fuzzy membership calculation on the extracted feature information to obtain a fuzzy feature set.
The first optimizing unit 52 is configured to perform genetic optimization processing according to the fuzzy feature set, and obtain a feature combination by performing optimization selection and cross variation on the fuzzy feature.
The first fusion unit 53 is configured to train the feature combination according to a preset neural network mathematical model, and obtain integrated information of the fused potential landslide through learning and the association relationship between the insufficient features.
It should be noted that, regarding the apparatus in the above embodiments, the specific manner in which the respective modules perform the operations has been described in detail in the embodiments regarding the method, and will not be described in detail herein.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. A multisource remote sensing interpretation method for potential landslide along railway is characterized by comprising the following steps:
acquiring first information and second information, wherein the first information comprises SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area, the second information comprises historical SAR remote sensing data, and the historical SAR remote sensing data comprises structural information along a railway;
constructing a target recognition mathematical model according to the second information, and carrying out recognition processing on the SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of a railway structure;
performing interpretation processing on the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of potential landslide and key earth surface deformation information;
according to the visible light remote sensing data, the near infrared remote sensing data, the thermal infrared remote sensing data and the geological data of the target area, fifth information is identified and extracted by analyzing the temperature characteristics of different ground features at different times, wherein the fifth information comprises surface variation information along a railway, ground crack information of a potential landslide and micro-landform feature interpretation results;
And carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of potential landslide.
2. The multi-source remote sensing interpretation method for potential landslide along railway line according to claim 1, wherein constructing a target recognition mathematical model according to the second information, and performing recognition processing on the SAR remote sensing data according to the target recognition mathematical model to obtain third information, comprises:
preprocessing the historical SAR remote sensing data in the second information to obtain preprocessed data;
identifying and extracting SAR strong scattering characteristics of structures along the railway according to the preprocessing data to obtain a characteristic set;
training by using a supervised learning algorithm according to the feature set to obtain a target recognition mathematical model for recognizing the structures along the railway;
and according to the obtained target recognition mathematical model, carrying out recognition processing on the SAR remote sensing data to be interpreted to obtain a recognition result of the railway structure.
3. The method for multi-source remote sensing interpretation of potential landslide along railway line according to claim 1, wherein the interpreting the surface deformation along railway line according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information comprises:
Carrying out interference superposition processing on SAR data at different times in the SAR remote sensing data to be interpreted, and obtaining earth surface deformation information by integrating phase differences;
according to the ground surface deformation information and the third information, analyzing and counting the reflection intensity of different railway structures, and combining with the calculation of phase difference to obtain the evolution process of the potential landslide, wherein the evolution process comprises the weak deformation information of the potential landslide;
according to the third information and the space coordinate information of the SAR remote sensing data to be interpreted, carrying out space superposition on the deformation information to obtain comprehensive earth surface deformation information;
and comprehensively evaluating and analyzing the comprehensive surface deformation information based on a preset worst case analysis set mathematical model, and screening to obtain key surface deformation information.
4. The method for multi-source remote sensing interpretation of potential landslide along railway line according to claim 1, wherein fifth information is identified and extracted by analyzing temperature characteristics of different ground objects at different times according to the visible light remote sensing data, the near infrared remote sensing data, the thermal infrared remote sensing data and the geological data of the target area, and the fifth information comprises surface variation information, ground crack information and micro-relief feature interpretation result of potential landslide along railway line, and comprises the following steps:
Analyzing and processing according to the visible light remote sensing data, and comparing the change conditions of the earth surface in the optical images at different times to obtain earth surface change information along the railway;
according to the near infrared remote sensing data and the thermal infrared remote sensing data, analyzing temperature change modes of ground objects in a target area in different time periods to obtain temperature characteristic data;
identifying and extracting the visible light remote sensing data according to the temperature characteristic data to obtain ground crack information of potential landslide along the railway;
and extracting the landform features of the visible light remote sensing data according to a preset cavity convolution mathematical model, and carrying out superposition processing on the extracted potential landslide micro-landform features and a three-dimensional landform model in the geological data of the target area to obtain a micro-landform feature interpretation result.
5. The multi-source remote sensing interpretation method for potential landslide along railway line according to claim 1, wherein the fusion processing is performed according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, comprising:
performing feature extraction processing according to the fourth information and the fifth information, and performing fuzzy membership calculation on the extracted feature information to obtain a fuzzy feature set;
Performing genetic optimization processing according to the fuzzy feature set, and obtaining feature combinations by performing optimization selection and cross variation on fuzzy features;
training the feature combination according to a preset neural network mathematical model, and obtaining the integrated information of the fused potential landslide through the association relation between learning and insufficient features.
6. A multi-source remote sensing interpretation device for potential landslide along a railway line, comprising:
the acquisition module is used for acquiring first information and second information, wherein the first information comprises SAR remote sensing data to be interpreted, visible light remote sensing data, near infrared remote sensing data, thermal infrared remote sensing data and geological data of a target area, the second information comprises historical SAR remote sensing data, and the historical SAR remote sensing data comprises structural information along a railway;
the construction module is used for constructing a target recognition mathematical model according to the second information, and carrying out recognition processing on the SAR remote sensing data according to the target recognition mathematical model to obtain third information, wherein the third information comprises a recognition result of a railway structure;
the interpretation module is used for performing interpretation processing on the earth surface deformation along the railway according to the third information and the phase difference in the SAR remote sensing data to be interpreted to obtain fourth information, wherein the fourth information comprises the evolution process of the potential landslide and key earth surface deformation information;
The extraction module is used for identifying and extracting fifth information according to the visible light remote sensing data, the near infrared remote sensing data, the thermal infrared remote sensing data and the geological data of the target area by analyzing the temperature characteristics of different ground features at different times, wherein the fifth information comprises ground surface change information along a railway, ground crack information of a potential landslide and micro-landform feature interpretation results;
and the fusion module is used for carrying out fusion processing according to the fourth information, the fifth information and a preset sign fusion mathematical model to obtain sixth information, wherein the sixth information is comprehensive information of the potential landslide.
7. The multi-source remote sensing interpretation device for potential landslide along a railway line of claim 6, wherein the construction module comprises:
the first processing unit is used for preprocessing the historical SAR remote sensing data in the second information to obtain preprocessed data;
the first extraction unit is used for identifying and extracting SAR strong scattering characteristics of structures along the railway according to the preprocessing data to obtain a characteristic set;
the first training unit is used for training by using a supervised learning algorithm according to the feature set to obtain a target recognition mathematical model for recognizing the structures along the railway;
And the first recognition unit is used for recognizing the SAR remote sensing data to be interpreted according to the obtained target recognition mathematical model to obtain a recognition result of the railway structure.
8. The multi-source remote sensing interpretation device for potential landslide along a rail of claim 6, wherein the interpretation module comprises:
the second processing unit is used for carrying out interference superposition processing on SAR data at different times in the SAR remote sensing data to be interpreted, and obtaining earth surface deformation information through integrating the phase difference;
the first calculation unit is used for analyzing and counting the reflection intensity of different railway structures according to the ground surface deformation information and the third information, and calculating the phase difference to obtain the evolution process of the potential landslide, wherein the evolution process comprises weak deformation information of the potential landslide;
the third processing unit is used for carrying out space superposition on the deformation information according to the third information and the space coordinate information of the SAR remote sensing data to be interpreted to obtain comprehensive earth surface deformation information;
and the second analysis unit is used for comprehensively evaluating and analyzing the comprehensive surface deformation information based on a preset worst case analysis set mathematical model and screening to obtain key surface deformation information.
9. The multi-source remote sensing interpretation device for potential landslide along railway line of claim 6, wherein the extraction module comprises:
the third analysis unit is used for carrying out analysis processing according to the visible light remote sensing data and obtaining the earth surface change information along the railway by comparing the earth surface change conditions in the optical images at different times;
the fourth analysis unit is used for analyzing the temperature change modes of the ground objects of the target area in different time periods according to the near infrared remote sensing data and the thermal infrared remote sensing data to obtain temperature characteristic data;
the third extraction unit is used for identifying and extracting the visible light remote sensing data of the multiple platforms and the multiple time phases according to the temperature characteristic data to obtain the ground crack information of the potential landslide along the railway;
and the fourth extraction unit is used for extracting the landform features of the visible light remote sensing data according to a preset cavity convolution mathematical model, and carrying out superposition processing on the extracted potential landslide micro-landform features and the three-dimensional landform model in the geological data of the target area to obtain a micro-landform feature interpretation result.
10. The multi-source remote sensing interpretation device for potential landslide along a railway line of claim 6, wherein the fusion module comprises:
The second computing unit is used for carrying out feature extraction processing according to the fourth information and the fifth information, and carrying out fuzzy membership calculation on the extracted feature information to obtain a fuzzy feature set;
the first optimizing unit is used for carrying out genetic optimization processing according to the fuzzy feature set, and obtaining a feature combination through optimization selection and cross variation of fuzzy features;
the first fusion unit is used for training the feature combination according to a preset neural network mathematical model, and obtaining the integrated information of the fused potential landslide through the association relationship between learning and insufficient features.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310678080.5A CN116403093B (en) | 2023-06-09 | 2023-06-09 | Multi-source remote sensing interpretation method and device for potential landslide along railway |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310678080.5A CN116403093B (en) | 2023-06-09 | 2023-06-09 | Multi-source remote sensing interpretation method and device for potential landslide along railway |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116403093A true CN116403093A (en) | 2023-07-07 |
CN116403093B CN116403093B (en) | 2023-08-22 |
Family
ID=87014698
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310678080.5A Active CN116403093B (en) | 2023-06-09 | 2023-06-09 | Multi-source remote sensing interpretation method and device for potential landslide along railway |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116403093B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117671508A (en) * | 2024-02-01 | 2024-03-08 | 中国水利水电第七工程局有限公司 | SAR image-based high-steep side slope landslide detection method and system |
CN118115887A (en) * | 2024-01-09 | 2024-05-31 | 中铁工程设计咨询集团有限公司 | Multi-source remote sensing dynamic identification method and device for undesirable geological hidden trouble along railway line |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030154060A1 (en) * | 2003-03-25 | 2003-08-14 | Damron James J. | Fusion of data from differing mathematical models |
WO2016169699A1 (en) * | 2015-04-23 | 2016-10-27 | Forest Vision As | A system, an apparatus and a method for determining mass change in a study area using remote sensing data |
CN107621636A (en) * | 2017-11-13 | 2018-01-23 | 河海大学 | A kind of Large-scale Railway Method of Bridge Health Monitoring based on PSI |
AU2020103096A4 (en) * | 2020-10-29 | 2021-01-07 | Bajrang Tapase, Anand DR | Movements/shifts/displacements monitoring SMART box of Earth Retaining Structures in Landslides Mitigation |
CN113885025A (en) * | 2021-09-03 | 2022-01-04 | 宁夏大学 | Landslide deformation monitoring method and visual service platform |
CN115453526A (en) * | 2022-08-02 | 2022-12-09 | 中国科学院空天信息创新研究院 | Landslide monitoring method and device |
-
2023
- 2023-06-09 CN CN202310678080.5A patent/CN116403093B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030154060A1 (en) * | 2003-03-25 | 2003-08-14 | Damron James J. | Fusion of data from differing mathematical models |
WO2016169699A1 (en) * | 2015-04-23 | 2016-10-27 | Forest Vision As | A system, an apparatus and a method for determining mass change in a study area using remote sensing data |
CN107621636A (en) * | 2017-11-13 | 2018-01-23 | 河海大学 | A kind of Large-scale Railway Method of Bridge Health Monitoring based on PSI |
AU2020103096A4 (en) * | 2020-10-29 | 2021-01-07 | Bajrang Tapase, Anand DR | Movements/shifts/displacements monitoring SMART box of Earth Retaining Structures in Landslides Mitigation |
CN113885025A (en) * | 2021-09-03 | 2022-01-04 | 宁夏大学 | Landslide deformation monitoring method and visual service platform |
CN115453526A (en) * | 2022-08-02 | 2022-12-09 | 中国科学院空天信息创新研究院 | Landslide monitoring method and device |
Non-Patent Citations (5)
Title |
---|
MOHSEN POURKHOSRAVANI ETC.: "Monitoring of Maskun landslide and determining its quantitative relationship to different climatic conditions using D-InSAR and PSI techniques", GEOMATICS, NATURAL HAZARDS AND RISK, vol. 13, no. 01 * |
YANING YI ETC.: "Landslide Detection Using Tme-Series In SAR Method along the Kangding-Batang Section of Shanghai-Nyalam Road", ROCKFALL HAZARD ANALYSIS USING REMOTE SENSING TECHNIQUES * |
刘晓云: "浅谈遥感技术的发展与应用", 三晋测绘, no. 01 * |
张勤等: "地质灾害监测预警中的精密空间对地观测技术", 测绘学报, no. 10 * |
谢猛: "基于多源数据的伊阿铁路工程地质遥感解译及选线", 铁道勘察, no. 05 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118115887A (en) * | 2024-01-09 | 2024-05-31 | 中铁工程设计咨询集团有限公司 | Multi-source remote sensing dynamic identification method and device for undesirable geological hidden trouble along railway line |
CN117671508A (en) * | 2024-02-01 | 2024-03-08 | 中国水利水电第七工程局有限公司 | SAR image-based high-steep side slope landslide detection method and system |
CN117671508B (en) * | 2024-02-01 | 2024-04-05 | 中国水利水电第七工程局有限公司 | SAR image-based high-steep side slope landslide detection method and system |
Also Published As
Publication number | Publication date |
---|---|
CN116403093B (en) | 2023-08-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Nhat-Duc et al. | Automatic recognition of asphalt pavement cracks using metaheuristic optimized edge detection algorithms and convolution neural network | |
CN116403093B (en) | Multi-source remote sensing interpretation method and device for potential landslide along railway | |
Tonini et al. | Rockfall detection from terrestrial LiDAR point clouds: A clustering approach using R | |
O’Byrne et al. | Texture analysis based damage detection of ageing infrastructural elements | |
Cano-Ortiz et al. | Machine learning algorithms for monitoring pavement performance | |
Southgate et al. | Analysis of field data of coastal morphological evolution over yearly and decadal timescales. Part 2: Non-linear techniques | |
CN117576581B (en) | Geological exploration remote sensing monitoring method based on image processing | |
CN113343563A (en) | Landslide susceptibility evaluation method based on automatic sample selection and surface deformation rate | |
CN113009447A (en) | Road underground cavity detection early warning method based on deep learning and ground penetrating radar | |
CN117876381B (en) | AI visual detection method and system for identifying and analyzing concrete structure cracks | |
CN116415843A (en) | Multi-mode remote sensing auxiliary mine ecological environment evaluation method for weak network environment | |
CN115330664A (en) | Image recognition-based surrounding rock weathering degree full-automatic recognition method and device | |
CN117664245B (en) | Dam safety real-time monitoring system | |
CN116992522A (en) | Deep foundation pit support structure deformation prediction method, device, equipment and storage medium | |
Chen et al. | A hierarchical DCNN-based approach for classifying imbalanced water inflow in rock tunnel faces | |
CN117741808A (en) | Geological crack detection method based on geological exploration | |
CN116719241A (en) | Automatic control method for informationized intelligent gate based on 3D visualization technology | |
CN112084553A (en) | Surveying method for tunnel planning | |
Liu et al. | Advances in automatic identification of road subsurface distress using ground penetrating radar: State of the art and future trends | |
Collins et al. | Automated Extraction of a Depth-Defined Wave Runup Time Series From Lidar Data Using Deep Learning | |
MacMillan | Experiences with applied DSM: protocol, availability, quality and capacity building | |
CN112767365A (en) | Flaw detection method | |
Wei et al. | Application of crack detection algorithm using convolutional neural network in concrete pavement construction | |
Olsen | Automated interpretation of depth to bedrock from airborne electromagnetic data using machine learning techniques | |
Momtaz | Damage Life Cycle Analysis for Present and Future Condition Assessments Using Statistical and Machine Learning Techniques |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |