Geological surveying and mapping system for multi-source data fusion

Abstract

The invention relates to the technical field of geological surveying and mapping, in particular to a multi-source data fusion geological surveying and mapping system which comprises a multi-dimensional data collection module, a data calibration and synchronization module, a geological image enhancement module, a data fusion and model generation module and a visualization and interaction module. The data calibration and synchronization module carries out timestamp calibration and geographic position processing on data, the geological image enhancement module enhances geological images, the data fusion and model generation module fuses the processed data and constructs a geological model by using a modeling technology, and the visualization and interaction module provides a multi-dimensional data interaction platform. Through the geological texture recovery sub-module, visual quality of images is improved, reliability and accuracy during geological data and resource evaluation are enhanced, polarized light data are collected through the polarized image processing sub-module, key information such as surface material roughness is revealed, and surface and underground geological processes can be better understood and explained.

Description

Technical Field

[0001] The present invention relates to the technical field of geological surveying and mapping, and in particular to a geological surveying and mapping system for multi-source data fusion. Background Art

[0002] In the current field of geological surveying and mapping technology, although there are various geological image processing technologies and data analysis methods, there are still some important problems that need to be solved. Traditional geological image processing technology is limited by resolution and quality, and it is difficult to fully capture and characterize surface and underground geological characteristics, especially under complex terrain and landform conditions. This limitation leads to inaccuracy and incompleteness of geological data, affecting the efficiency and accuracy of geological exploration.

[0003] Traditional geological data processing methods still have the problem of monitoring micro-terrain changes. Current micro-terrain change monitoring methods are usually based on a single data source or hand-made terrain models, lacking high precision, real-time and multi-dimensional analysis capabilities. This makes accurate monitoring and early warning of tiny terrain changes a challenge, limiting the ability of geological hazard risk assessment and resource management.

[0004] In addition, existing geological image processing technologies are often unable to fully understand the characteristics of surface materials. Traditional geological image processing methods mainly rely on optical images, and have limited ability to obtain information such as the roughness, humidity, and reflective properties of surface materials. This leads to a lack of multi-dimensional perspective and in-depth analysis of surface materials, limiting the accuracy and comprehensiveness of geological exploration and resource assessment. Summary of the invention

[0005] Based on the above purpose, the present invention provides a geological surveying and mapping system for multi-source data fusion.

[0006] A geological surveying and mapping system with multi-source data fusion, comprising a multi-dimensional data collection module, a data calibration and synchronization module, a geological image enhancement module, a data fusion and model generation module, and a visualization and interaction module, wherein;

[0007] The multi-dimensional data collection module receives geological data from satellite remote sensing, drone images, ground sensors and underground geological radar, and the geological data includes original geological images, surface and underground data to provide a comprehensive geological analysis basis;

[0008] The data calibration and synchronization module performs timestamp calibration and geographic location standardization on data from different devices and platforms to ensure the consistency of all data in space and time, and provide accurate basic data for subsequent module analysis;

[0009] The geological image enhancement module includes a geological texture restoration submodule, a terrain micro-change analysis submodule and a polarization image processing submodule, wherein;

[0010] The geological texture restoration submodule uses deep learning technology to restore details and textures in low-resolution or poor-quality original geological images, thereby improving the usability and accuracy of the images;

[0011] The terrain micro-change analysis submodule uses high-precision terrain measurement technology to detect and analyze small changes in terrain;

[0012] The polarization image processing submodule uses polarization technology to enhance geological images, revealing the roughness, moisture and reflective properties of surface materials, providing additional perspectives and data layers for geological analysis;

[0013] The data fusion and model generation module fuses the data processed by the geological image enhancement module and constructs a geological model using three-dimensional geological modeling technology;

[0014] The visualization and interaction module provides a multi-dimensional data interaction platform, allowing users to view and interact with geological models through virtual reality technology, thereby enhancing users' understanding and decision support capabilities.

[0015] Furthermore, the multi-dimensional data collection module specifically includes:

[0016] Satellite remote sensing data receiving submodule: configured to receive remote sensing sensor data from geosynchronous orbit or polar orbit satellites. The collected data types include optical images, infrared and radar data, which are used to capture a wide range of geological features and changes on the surface of the earth;

[0017] UAV image collection submodule: It is designed to be a drone equipped with a multispectral camera and a lidar to collect high-resolution surface data from low altitudes. The drone is used for intensive monitoring of surface details and small-scale geological changes;

[0018] Ground sensor network submodule: includes a distributed sensor network deployed at key geological monitoring points, such as fault lines and mining areas. The data collected include ground vibration, temperature, and humidity, which are helpful for analyzing geological activities and environmental changes under the surface.

[0019] Underground geological radar receiving submodule: using geological radar technology, it receives reflected wave data at a depth of at least tens of meters underground to reveal underground structures and unexposed geological layers.

[0020] Furthermore, the data calibration and synchronization module specifically includes:

[0021] Timestamp calibration submodule: configured to receive the original data from the multi-dimensional data collection module, where each data item is accompanied by an original timestamp, and use a unified time server or network time protocol to perform time calibration on all data items to ensure the consistency of the data items in time and eliminate the time error caused by clock deviation between devices;

[0022] Geographic location standardization submodule: designed to analyze and process the geographic location information of geological data, uniformly format and calibrate the geographic coordinates from satellites, drones, ground sensors and underground geological radars, and use high-precision geographic information systems and satellite positioning systems to adjust and standardize geographic location data;

[0023] Data synchronization processing submodule: It is configured to further synchronize the update frequency and sampling interval of each data source after the timestamp and geographic location are standardized, and sort and align the data.

[0024] Furthermore, the geological texture restoration submodule specifically includes:

[0025] Data reception and preliminary evaluation: Receive raw geological image data, perform preliminary evaluation of received images, and identify low-resolution or poor-quality images, which are usually degraded due to transmission errors, environmental interference, or equipment limitations;

[0026] Deep learning training: Use the convolutional neural network deep learning architecture to restore image texture and details, and learn to reconstruct high-resolution details from low-resolution or low-quality images;

[0027] Texture restoration processing: responsible for applying convolutional neural networks to actual low-resolution or poor-quality geological images to improve clarity, contrast, and color saturation, as well as reconstructing missing or blurred geological features in the images;

[0028] Quality Control and Output: After texture and detail are restored, a quality control check is performed on the restored image to ensure that the restoration process did not introduce unnatural artifacts or over-smoothing.

[0029] Furthermore, the convolutional neural network includes:

[0030] Data preparation and preprocessing: Collect high-resolution geological images and their corresponding low-resolution or poor-quality versions for training convolutional neural network models, scale the images to a uniform size, and normalize them to make the input data fit the network and accelerate the learning process;

[0031] Design a convolutional neural network model: Design a network with multiple convolutional layers, activation layers, and upsampling layers. The convolutional layers are used to capture the features of the image. The activation layers (such as ReLU) are used to introduce nonlinearity and improve the expressiveness of the model. The upsampling layers are used to increase the resolution of the image and use convolutional kernels with refined features to extract details and texture information. The convolutional kernels are used to learn to reconstruct high-resolution details from low-resolution images.

[0032] Training model: Use mean square error (MSE) to measure the difference between the reconstructed image and the original high-resolution image. Perceptual loss is particularly helpful in preserving texture and detail information. The Adam optimizer is used to adjust the weights of the convolutional neural network model to minimize the loss function.

[0033] Enhancement of texture and detail: Convolutional neural networks improve the clarity and contrast of images by learning to increase local contrast and sharpen edges. By adjusting the processing method of the convolutional neural network model in the color channel, the color saturation of the image is enhanced to make the geological features more distinct. Convolutional neural networks predict and reconstruct geological features that are missing or blurred in low-quality images by learning complex patterns and textures in high-resolution images.

[0034] Testing and deployment: Verify the restoration effect of the convolutional neural network model on unseen low-resolution or poor-quality test data, deploy the trained convolutional neural network model into the geological mapping system, process the collected low-resolution geological images in real time, and provide clear, high-resolution geological images for further analysis.

[0035] Furthermore, the terrain micro-change analysis submodule specifically includes:

[0036] Application of high-precision topographic measurement technology: Using laser radar (LiDAR) technology and imaging system (ODIS) as topographic measurement tools. LiDAR technology calculates the distance between the surface and the equipment by emitting laser pulses and measuring the time difference of the reflected signal, thereby generating high-resolution three-dimensional topographic maps;

[0037] Detection and analysis of small changes: Designed to continuously monitor the topographic data of the surveyed area, identify small topographic changes by comparing data records at consecutive time points, and use change detection algorithms to analyze changes in topographic data.

[0038] Furthermore, the change detection algorithm includes multi-scale difference and structural similarity index SSIM;

[0039] The multi-scale difference extracts terrain change information through filters of different scales, and uses a Gaussian pyramid for multi-level analysis, with the low level capturing rough large-scale changes and the high level capturing detailed changes. The multi-scale difference specifically includes:

[0040] Image pyramid construction: Create an image pyramid for each time point. For terrain images I1 and I2, create multiple resolution levels by gradually downsampling, calculated as;

[0041] G i (x,y)=downsample(G i-1 (x,y)),i=1,2,…,n;

[0042] Where G0(x,y)=((x,y), represents the original image, and downsample represents the subsampling after Gaussian filtering;

[0043] Layer-by-layer difference calculation: Calculate the difference between two time points at each scale level, calculated as;

[0044]

[0045] in and are the images at the i-th layer at time t1 and t2 respectively;

[0046] Change assessment: Thresholds are set at each scale level to identify significant changes, and small-scale changes at high levels are combined with large-scale changes at low levels to accurately locate and describe terrain changes;

[0047] The structural similarity index SSIM is used to evaluate terrain changes. For images of the same location at different times, the structural similarity index is calculated as follows:

[0048]

[0049] where μ x and μ y is the average of the pixel values ​​in the two windows x and y, σ x and σ y is the standard deviation, σ xy is the covariance, c1 and c2 are small constants added to avoid the denominator being zero;

[0050] Small change detection: By comparing the structural similarity index, we can find out the areas where the structural similarity is significantly reduced. The structural similarity index value close to 1 indicates that the change is almost small, and the value far from 1 indicates a large change.

[0051] Furthermore, the polarization image processing submodule specifically includes:

[0052] Collection of polarized light data: Use a polarized light camera combined with geological imaging equipment to capture polarized light images of the surface;

[0053] Polarization data analysis: Stokes parameters are used to quantify the polarization characteristics of the Earth's surface. Stokes parameters include a detailed description of the intensity and direction of polarized light and are calculated as:

[0054] S0=I0+I 90 ,S1=I0-I 90 ,S2=I 45 -I 135 ;

[0055] Among them, S0 is the first parameter in the Stokes parameter, representing the total light intensity, which is obtained by adding the light intensities in different polarization directions. S1 is the second parameter in the Stokes parameter, describing the light intensity difference between two perpendicular polarization directions, usually 0° and 90°. S2 is the third parameter, describing the light intensity difference between the polarized light in the directions of 45° and 135°. The two angles of 45° and 135° are measured relative to a certain reference direction, including the horizontal direction. I0,I 45 ,I 90 ,I 135 They are the light intensities collected from the polarization directions of 0°, 45°, 90°, and 135° respectively;

[0056] Analysis of surface characteristics: The scattered light distribution pattern in the polarization image is used to identify the surface roughness. The rough surface will cause the polarized light to be scattered more widely and evenly. The humidity of the surface is assessed by analyzing the change in the degree of polarization of the polarized light. The degree of polarization and phase delay are calculated, providing information on the optical properties of the surface material, including reflectivity and refractive index.

[0057] Further, the calculating of polarization degree and phase delay specifically includes:

[0058] Degree of polarization: The degree of polarization is a measure that describes the degree to which the polarization state of incident light changes after it is reflected by an object. It is defined as the ratio of the polarization component of the light after it is reflected by the object to the total light intensity and is calculated as:

[0059]

[0060] Among them, DoP degree of polarization, S0 is the total light intensity, S1, S2, S3 represent the intensity difference of polarized light in different directions, S3 represents the intensity difference of circularly polarized light, and the value range of degree of polarization is from 0 to 1, where 0 represents unpolarized light and 1 represents fully polarized light. In geological mapping, the measurement of degree of polarization helps identify material types and surface conditions.

[0061] Phase delay: Phase delay is used to describe the phase difference between two orthogonal components of polarized light. During the reflection process, light waves with different polarization directions have different propagation speeds due to the different optical properties of the medium, resulting in phase differences. It is calculated as:

[0062]

[0063]

[0064] Where δ is the phase difference between the two orthogonal components, and S1, S2, and S3 are values ​​obtained from the Stokes parameters.

[0065] Furthermore, the data fusion and model generation module specifically includes:

[0066] Data preprocessing and standardization: Receive processed data from the geological image enhancement module, enhance the geological image data through the polarization image processing submodule, geological texture restoration submodule and terrain micro-change analysis submodule, and standardize all input data to ensure that the formats and scales of different data sources are consistent, so as to facilitate efficient data fusion;

[0067] Multi-source data fusion: Use multi-scale fusion technology, feature-level fusion technology and decision-level fusion technology to integrate geological information from satellite remote sensing data, drone images and ground sensor data;

[0068] Application of 3D geological modeling technology: Utilize the fused data, adopt computer-aided design and geographic information system tools to construct 3D geological models, and use ArcGIS 3D Analyst geological modeling software to refine and optimize the models. During the model generation process, the 3D representation of the geological characteristics of strata, faults, and sedimentary structures will be automatically constructed according to the spatial distribution and attributes of the geological data.

[0069] Beneficial effects of the present invention:

[0070] The present invention significantly improves the clarity and details of low-resolution or poor-quality geological images by applying the geological texture restoration submodule and using convolutional neural network (CNN) technology. The deep learning architecture of CNN can effectively learn and reconstruct missing textures and details from existing high-resolution images, thereby providing more accurate geological data. This not only improves the visual quality of the images, but also enhances the reliability and accuracy of geological data in structural analysis and resource assessment.

[0071] The present invention adopts high-precision topographic measurement technology through the terrain micro-change analysis submodule, and combines multi-scale difference and structural similarity index (SSIM) change detection algorithm to effectively monitor and analyze tiny terrain changes caused by groundwater flow or small-scale crustal movement. This refined monitoring capability enables the system to detect potential risks in the early stages of geological activities, providing a scientific basis for the prevention of geological disasters such as earthquakes and landslides, and greatly improving the response time of the early warning system and the effectiveness of preventive measures.

[0072] The present invention collects and analyzes polarized light data through the polarization image processing submodule, revealing key information such as the roughness, humidity and reflective characteristics of surface materials. This information provides an unprecedented perspective and data level for geological analysis, and helps to more accurately identify and classify surface materials and their environmental conditions. Combined with this detailed surface characteristic information, geological experts can better understand and interpret surface and underground geological processes, thereby making more informed decisions in areas such as resource exploration, environmental protection and land management. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0074] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;

[0075] Figure 2 Schematic diagram of multi-scale difference steps according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0077] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0078] like Figure 1-Figure 2 As shown, a geological mapping system with multi-source data fusion includes a multi-dimensional data collection module, a data calibration and synchronization module, a geological image enhancement module, a data fusion and model generation module, and a visualization and interaction module, wherein;

[0079] The multi-dimensional data collection module receives geological data from satellite remote sensing, drone images, ground sensors and underground geological radar. The geological data includes original geological images, surface and underground data to provide a comprehensive geological analysis basis;

[0080] The data calibration and synchronization module performs timestamp calibration and geographic location standardization on data from different devices and platforms to ensure the consistency of all data in space and time, and provide accurate basic data for subsequent module analysis;

[0081] The geological image enhancement module includes a geological texture restoration submodule, a terrain micro-change analysis submodule and a polarization image processing submodule, among which;

[0082] The geological texture restoration submodule uses deep learning technology to restore details and textures in low-resolution or poor-quality original geological images, improving the usability and accuracy of the images;

[0083] The terrain micro-change analysis submodule uses high-precision terrain measurement technology to detect and analyze small changes in terrain, such as those caused by groundwater flow or small-scale crustal movement, for earthquake precursor research and soil erosion assessment;

[0084] The polarization image processing submodule uses polarized light technology to enhance geological images, revealing the roughness, moisture and reflective properties of surface materials, providing additional perspectives and data layers for geological analysis;

[0085] The data fusion and model generation module fuses the data processed by the geological image enhancement module and constructs a geological model using 3D geological modeling technology. The data fusion and model generation module integrates multi-source data and enhanced image data to generate a highly accurate geological model.

[0086] The visualization and interaction module provides a multi-dimensional data interaction platform, allowing users to view and interact with geological models through virtual reality technology, enhancing users' understanding and decision support capabilities.

[0087] The multi-dimensional data collection module specifically includes:

[0088] Satellite remote sensing data receiving submodule: configured to receive remote sensing sensor data from geosynchronous orbit or polar orbit satellites. The collected data types include optical images, infrared and radar data, which are used to capture a wide range of geological features and changes on the surface of the earth;

[0089] UAV image collection submodule: It is designed to be a drone equipped with a multispectral camera and a lidar to collect high-resolution surface data from low altitudes. The drone is used for intensive monitoring of surface details and small-scale geological changes;

[0090] Ground sensor network submodule: includes a distributed sensor network deployed at key geological monitoring points, such as fault lines and mining areas. The data collected include ground vibration, temperature, and humidity, which are helpful for analyzing geological activities and environmental changes under the surface.

[0091] Underground geological radar receiving submodule: Using geological radar technology, it receives reflected wave data at a depth of at least tens of meters underground to reveal underground structures and unexposed geological layers. Geological radar is particularly suitable for detecting groundwater layers, caves, fissures and other hidden geological structures.

[0092] The data calibration and synchronization module specifically includes:

[0093] Timestamp calibration submodule: configured to receive the original data from the multi-dimensional data collection module, where each data item is accompanied by an original timestamp, and use a unified time server or network time protocol to perform time calibration on all data items to ensure the consistency of the data items in time and eliminate the time error caused by clock deviation between devices;

[0094] Geographic location standardization submodule: designed to analyze and process the geographic location information of geological data, uniformly format and calibrate the geographic coordinates from satellites, drones, ground sensors and underground geological radars, use high-precision geographic information systems and satellite positioning systems, adjust and standardize geographic location data, and ensure that the collected geographic information is consistent and accurate in spatial coordinates;

[0095] Data synchronization processing submodule: It is configured to further synchronize the update frequency and sampling interval of each data source after the timestamp and geographic location are standardized, and sort and align the data.

[0096] The geological texture restoration submodule specifically includes:

[0097] Data reception and preliminary evaluation: Receive raw geological image data, perform preliminary evaluation of received images, and identify low-resolution or poor-quality images, which are usually degraded due to transmission errors, environmental interference, or equipment limitations;

[0098] Deep learning training: Use the convolutional neural network deep learning architecture to restore image texture and details, and learn to reconstruct high-resolution details from low-resolution or low-quality images;

[0099] Texture restoration processing: responsible for applying convolutional neural networks to actual low-resolution or poor-quality geological images to improve clarity, contrast, and color saturation, as well as reconstructing missing or blurred geological features in the images;

[0100] Quality Control and Output: After texture and detail are restored, a quality control check is performed on the restored image to ensure that the restoration process did not introduce unnatural artifacts or over-smoothing.

[0101] Convolutional neural networks include:

[0102] Data preparation and preprocessing: Collect high-resolution geological images and their corresponding low-resolution or poor-quality versions for training convolutional neural network models, scale the images to a uniform size, and normalize them to make the input data fit the network and accelerate the learning process;

[0103] Design a convolutional neural network model: Design a network with multiple convolutional layers, activation layers, and upsampling layers. The convolutional layers are used to capture the features of the image. The activation layers (such as ReLU) are used to introduce nonlinearity and improve the expressiveness of the model. The upsampling layers are used to increase the resolution of the image and use convolutional kernels with refined features to extract details and texture information. The convolutional kernels are used to learn to reconstruct high-resolution details from low-resolution images.

[0104] Training model: The mean square error (MSE) is used to measure the difference between the reconstructed image and the original high-resolution image. The perceptual loss is particularly helpful in preserving texture and detail information. The Adam optimizer is used to adjust the weights of the convolutional neural network model to minimize the loss function, thereby learning how to effectively restore the details and texture of geological images.

[0105] Enhancement of texture and detail: Convolutional neural networks improve the clarity and contrast of images by learning to increase local contrast and sharpen edges. By adjusting the processing method of the convolutional neural network model in the color channel, the color saturation of the image is enhanced to make the geological features more distinct. Convolutional neural networks predict and reconstruct geological features that are missing or blurred in low-quality images by learning complex patterns and textures in high-resolution images.

[0106] Testing and deployment: Verify the recovery effect of the convolutional neural network model on unseen low-resolution or poor-quality test data to ensure that the model has good generalization ability. Deploy the trained convolutional neural network model to the geological mapping system to process the collected low-resolution geological images in real time and provide clear, high-resolution geological images for further analysis.

[0107] The terrain micro-change analysis submodule specifically includes:

[0108] Application of high-precision topographic measurement technology: Using laser radar (LiDAR) technology and imaging system (ODIS) as topographic measurement tools, providing high-resolution surface and underground topographic data, accurate to the centimeter level. LiDAR technology calculates the distance between the surface and the device by emitting laser pulses and measuring the time difference of the reflected signal, thereby generating a high-resolution three-dimensional topographic map;

[0109] Detection and analysis of small changes: Designed to continuously monitor the topographic data of the surveyed area, it identifies small topographic changes by comparing data records at consecutive time points. This includes soil erosion, subsidence, landslides or any small height changes caused by geological activities. Change detection algorithms are used to analyze changes in topographic data. Change detection algorithms can identify very small topographic changes and provide early warnings of geological activities.

[0110] Change detection algorithms include multi-scale difference and structural similarity index SSIM;

[0111] Multi-scale difference extracts terrain change information through filters of different scales, and uses Gaussian pyramid for multi-level analysis. The low level captures the rough large-scale changes, and the high level captures the detailed changes. Multi-scale difference specifically includes:

[0112] Image pyramid construction: Create an image pyramid for each time point. For terrain images I1 and I2, create multiple resolution levels by gradually downsampling, calculated as;

[0113] G i (x,y)=downsample(G i-1 (x,y)),i=1,2,…,n;

[0114] Where G0(x,y)=I(x,y) represents the original image, and downsample represents the subsampling after Gaussian filtering;

[0115] Layer-by-layer difference calculation: Calculate the difference between two time points at each scale level, calculated as;

[0116]

[0117] in and are the images at the i-th layer at time t1 and t2 respectively;

[0118] Change assessment: Thresholds are set at each scale level to identify significant changes, and small-scale changes at high levels are combined with large-scale changes at low levels to accurately locate and describe terrain changes;

[0119] The structural similarity index SSIM is used to evaluate terrain changes. The structural similarity index SSIM takes into account the brightness, contrast and structure of the image, and can better reflect the visual changes. For images of the same location at different times, the structural similarity index is calculated as;

[0120]

[0121] where μ x and μ y is the average of the pixel values ​​in the two windows x and y, σ x and σ y is the standard deviation, σ xy is the covariance, c1 and c2 are small constants added to avoid the denominator being zero;

[0122] Small change detection: By comparing the structural similarity index, we can find out the areas where the structural similarity is significantly reduced. The structural similarity index value close to 1 indicates that the change is almost small, and the value far from 1 indicates a large change.

[0123] The polarization image processing submodule specifically includes:

[0124] Polarization data collection: Polarization cameras and geological imaging equipment are used to capture polarized images of the surface. These devices can record light polarized at different angles to analyze the characteristics of surface materials.

[0125] Polarization data analysis: Stokes parameters are used to quantify the polarization characteristics of the Earth's surface. Stokes parameters include a detailed description of the intensity and direction of polarized light and are calculated as:

[0126] S0=I0+I 90 ,S1=I0-I 90 ,S2=I 45 -I 135 ;

[0127] Among them, S0 is the first parameter in the Stokes parameter, representing the total light intensity, which is obtained by adding the light intensities in different polarization directions. S1 is the second parameter in the Stokes parameter, describing the light intensity difference between two perpendicular polarization directions, usually 0° and 90°. S2 is the third parameter, describing the light intensity difference between the polarized light in the directions of 45° and 135°. The two angles of 45° and 135° are measured relative to a certain reference direction, including the horizontal direction. I0,I 45 ,I 90 ,I 135 They are the light intensities collected from the polarization directions of 0°, 45°, 90°, and 135° respectively;

[0128] Analysis of surface characteristics: The scattered light distribution pattern in the polarization image is used to identify the surface roughness. The rough surface will cause the polarized light to be scattered more widely and evenly. The humidity of the surface is assessed by analyzing the change in the degree of polarization of the polarized light. The light reflected from the wet surface often has a lower degree of polarization because the reflection from the water surface reduces the degree of polarization of the polarized light. The degree of polarization and phase delay are calculated to provide information on the optical properties of the surface material, including reflectivity and refractive index.

[0129] Calculating the degree of polarization and phase delay specifically includes:

[0130] Degree of polarization: The degree of polarization is a measure that describes the degree to which the polarization state of incident light changes after it is reflected by an object. It is defined as the ratio of the polarization component of the light after it is reflected by the object to the total light intensity and is calculated as:

[0131]

[0132] Among them, DoP degree of polarization, S0 is the total light intensity, S1, S2, S3 represent the intensity difference of polarized light in different directions, S3 represents the intensity difference of circularly polarized light, and the value range of degree of polarization is from 0 to 1, where 0 represents unpolarized light and 1 represents fully polarized light. In geological mapping, the measurement of degree of polarization helps identify material types and surface conditions.

[0133] Phase delay: Phase delay is used to describe the phase difference between two orthogonal components of polarized light. During the reflection process, light waves with different polarization directions have different propagation speeds due to the different optical properties of the medium, resulting in phase differences. It is calculated as:

[0134]

[0135]

[0136] Where σ is the phase difference between the two orthogonal components, and S1, S2, and S3 are values ​​obtained from the Stokes parameters.

[0137] The data fusion and model generation module specifically includes:

[0138] Data preprocessing and standardization: Receive processed data from the geological image enhancement module, enhance the geological image data through the polarization image processing submodule, geological texture restoration submodule and terrain micro-change analysis submodule, and standardize all input data to ensure that the formats and scales of different data sources are consistent, so as to facilitate efficient data fusion;

[0139] Multi-source data fusion: Use multi-scale fusion technology, feature-level fusion technology and decision-level fusion technology to integrate geological information from satellite remote sensing data, drone images and ground sensor data;

[0140] Application of 3D geological modeling technology: Utilize the fused data, adopt computer-aided design and geographic information system tools to construct a 3D geological model, which includes a comprehensive model of surface topography, underground structure and geological layers. Use ArcGIS 3D Analyst geological modeling software to refine and optimize the model. During the model generation process, the 3D representation of the geological characteristics of strata, faults and sedimentary structures will be automatically constructed according to the spatial distribution and attributes of the geological data.

[0141] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the above aspects of the present invention, which are not provided in detail for the sake of simplicity.

[0142] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A geological surveying and mapping system with multi-source data fusion, characterized in that: It includes multi-dimensional data collection module, data calibration and synchronization module, geological image enhancement module, data fusion and model generation module and visualization and interaction module, among which; The multi-dimensional data collection module receives geological data from satellite remote sensing, drone images, ground sensors and underground geological radar, and the geological data includes original geological images, surface and underground data; The data calibration and synchronization module performs timestamp calibration and geographic location standardization on data from different devices and platforms to ensure the consistency of all data in space and time; The geological image enhancement module includes a geological texture restoration submodule, a terrain micro-change analysis submodule and a polarization image processing submodule, wherein; The geological texture restoration submodule uses deep learning technology to restore details and textures in low-resolution or poor-quality original geological images; The terrain micro-change analysis submodule uses high-precision terrain measurement technology to detect and analyze small changes in terrain; The polarization image processing submodule uses polarization light technology to enhance geological images and reveal the roughness, humidity and reflective properties of surface materials; The data fusion and model generation module fuses the data processed by the geological image enhancement module and constructs a geological model using three-dimensional geological modeling technology; The visualization and interaction module provides a multi-dimensional data interaction platform, allowing users to view and interact with geological models through virtual reality technology.

2. A geological surveying and mapping system for multi-source data fusion according to claim 1, characterized in that: The multi-dimensional data collection module specifically includes: Satellite remote sensing data receiving submodule: configured to receive remote sensing sensor data from geosynchronous orbit or polar orbit satellites. The collected data types include optical images, infrared and radar data, which are used to capture a wide range of geological features and changes on the surface of the earth; UAV image collection submodule: It is designed to be a drone equipped with a multispectral camera and a lidar to collect high-resolution surface data from low altitudes. The drone is used for intensive monitoring of surface details and small-scale geological changes; Ground sensor network submodule: includes a distributed sensor network deployed at key geological monitoring points, including fault lines and mining areas, and collects data including ground vibration, temperature, and humidity; Underground geological radar receiving submodule: uses geological radar technology to receive underground reflected wave data to reveal underground structures and unexposed geological layers.

3. A geological surveying and mapping system with multi-source data fusion according to claim 1, characterized in that: The data calibration and synchronization module specifically includes: A timestamp calibration submodule: configured to receive raw data from the multi-dimensional data collection module, wherein each data item is accompanied by an original timestamp, and to perform time calibration on all data items using a unified time server or a network time protocol; Geographic location standardization submodule: designed to analyze and process the geographic location information of geological data, uniformly format and calibrate the geographic coordinates from satellites, drones, ground sensors and underground geological radars, and use high-precision geographic information systems and satellite positioning systems to adjust and standardize geographic location data; Data synchronization processing submodule: It is configured to further synchronize the update frequency and sampling interval of each data source after the timestamp and geographic location are standardized, and sort and align the data.

4. A geological surveying and mapping system for multi-source data fusion according to claim 1, characterized in that: The geological texture restoration submodule specifically includes: Data reception and preliminary evaluation: Receive raw geological image data, conduct preliminary evaluation of received images, and identify images with low resolution or poor quality; Deep learning training: Use the convolutional neural network deep learning architecture to restore image texture and details, and learn to reconstruct high-resolution details from low-resolution or low-quality images; Texture restoration processing: responsible for applying convolutional neural networks to actual low-resolution or poor-quality geological images to improve clarity, contrast, and color saturation, as well as reconstructing missing or blurred geological features in the images; Quality Control and Output: After texture and detail are restored, a quality control check is performed on the restored image to ensure that the restoration process did not introduce unnatural artifacts or over-smoothing.

5. A geological surveying and mapping system with multi-source data fusion according to claim 4, characterized in that: The convolutional neural network comprises: Data preparation and preprocessing: Collect high-resolution geological images and their corresponding low-resolution or poor-quality versions for training convolutional neural network models, scale the images to a uniform size, and normalize them to make the input data fit the network and accelerate the learning process; Design a convolutional neural network model: Design a network with multiple convolutional layers, activation layers, and upsampling layers. The convolutional layers are used to capture the features of the image, the activation layers are used to introduce nonlinearity, the upsampling layers are used to increase the resolution of the image, and convolutional kernels with refined features are used to extract details and texture information. The convolutional kernels are used to learn to reconstruct high-resolution details from low-resolution images. Training model: Use mean square error to measure the difference between the reconstructed image and the original high-resolution image, and use Adam optimizer to adjust the weights of the convolutional neural network model to minimize the loss function; Enhancement of texture and detail: Convolutional neural networks improve the clarity and contrast of images by learning to increase local contrast and sharpen edges. The color saturation of images is enhanced by adjusting the processing method of the convolutional neural network model in the color channel. Convolutional neural networks predict and reconstruct geological features that are missing or blurred in low-quality images by learning complex patterns and textures in high-resolution images. Testing and deployment: Verify the restoration effect of the convolutional neural network model on unseen low-resolution or poor-quality test data, and deploy the trained convolutional neural network model into the geological mapping system to process the collected low-resolution geological images in real time.

6. A geological surveying and mapping system for multi-source data fusion according to claim 1, characterized in that: The terrain micro-change analysis submodule specifically includes: Application of high-precision topographic measurement technology: Using LiDAR technology and imaging systems as topographic measurement tools. LiDAR technology calculates the distance between the surface and the equipment by emitting laser pulses and measuring the time difference of the reflected signals, thereby generating high-resolution three-dimensional topographic maps. Detection and analysis of small changes: Designed to continuously monitor the topographic data of the surveyed area, identify small topographic changes by comparing data records at consecutive time points, and use change detection algorithms to analyze changes in topographic data.

7. A geological surveying and mapping system with multi-source data fusion according to claim 6, characterized in that: The change detection algorithm includes multi-scale difference and structural similarity index SSIM; The multi-scale difference extracts terrain change information through filters of different scales, and uses a Gaussian pyramid for multi-level analysis, with the low level capturing rough large-scale changes and the high level capturing detailed changes. The multi-scale difference specifically includes: Image pyramid construction: Create an image pyramid for each time point. For terrain images I1 and I2, create multiple resolution levels by gradually downsampling, calculated as; G i (x,y)=downsample(G i-1 (x,y)),i=1,2,…,n; Where G0(x,y)=I(x,y) represents the original image, and downsample represents the subsampling after Gaussian filtering; Layer-by-layer difference calculation: Calculate the difference between two time points at each scale level, calculated as; in and are the images at time t1 and t2 at layer i; Change assessment: Thresholds are set at each scale level to identify significant changes, and small-scale changes at high levels are combined with large-scale changes at low levels to accurately locate and describe terrain changes; The structural similarity index SSIM is used to evaluate terrain changes. For images of the same location at different times, the structural similarity index is calculated as follows: where μ x and μ y is the average of the pixel values ​​in the two windows x and y, σ x and σ y is the standard deviation, σ xy is the covariance, c1 and c2 are small constants added to avoid the denominator being zero; Small change detection: By comparing the structural similarity index, we can find out the areas where the structural similarity is significantly reduced. The structural similarity index value close to 1 indicates that the change is almost small, and the value far from 1 indicates a large change.

8. The multi-source data fusion geological surveying and mapping system according to claim 1, characterized in that: The polarization image processing submodule specifically includes: Collection of polarized light data: Use a polarized light camera combined with geological imaging equipment to capture polarized light images of the surface; Polarization data analysis: Stokes parameters are used to quantify the polarization characteristics of the Earth's surface. Stokes parameters include a detailed description of the intensity and direction of polarized light and are calculated as: S0=I0+I 90 ,S1=I0-I 90 ,S2=I 45 -I 135 ; Among them, S0 is the first parameter in the Stokes parameter, representing the total light intensity, which is obtained by adding the light intensities in different polarization directions. S1 is the second parameter in the Stokes parameter, describing the light intensity difference between two perpendicular polarization directions, usually 0° and 90°. S2 is the third parameter, describing the light intensity difference between the polarized light in the directions of 45° and 135°. The two angles of 45° and 135° are measured relative to a certain reference direction, including the horizontal direction. I0,I 45 ,I 90 ,I 135 They are the light intensities collected from the polarization directions of 0°, 45°, 90°, and 135° respectively; Analysis of surface characteristics: The scattered light distribution pattern in the polarization image is used to identify the surface roughness. The rough surface will cause the polarized light to be scattered more widely and evenly. The humidity of the surface is assessed by analyzing the change in the degree of polarization of the polarized light. The degree of polarization and phase delay are calculated, providing information on the optical properties of the surface material, including reflectivity and refractive index.

9. A geological surveying and mapping system with multi-source data fusion according to claim 8, characterized in that: The calculation of polarization degree and phase delay specifically includes: Degree of polarization: The degree of polarization is a measure that describes the degree to which the polarization state of incident light changes after it is reflected by an object. It is defined as the ratio of the polarization component of the light after it is reflected by the object to the total light intensity and is calculated as: Among them, DoP degree of polarization, S0 is the total light intensity, S1, S2, S3 represent the intensity difference of polarized light in different directions, S3 represents the intensity difference of circularly polarized light, and the value range of degree of polarization is from 0 to 1, where 0 represents unpolarized light and 1 represents fully polarized light. In geological mapping, the measurement of degree of polarization helps identify material types and surface conditions. Phase delay: Phase delay is used to describe the phase difference between two orthogonal components of polarized light. During the reflection process, light waves with different polarization directions have different propagation speeds due to the different optical properties of the medium, resulting in phase differences. It is calculated as: Where δ is the phase difference between the two orthogonal components, and S1, S2, and S3 are values ​​obtained from the Stokes parameters.

10. A geological surveying and mapping system with multi-source data fusion according to claim 1, characterized in that: The data fusion and model generation module specifically includes: Data preprocessing and standardization: Receive processed data from the geological image enhancement module, enhance the geological image data through the polarization image processing submodule, geological texture restoration submodule and terrain micro-change analysis submodule, and standardize all input data to ensure that the formats and scales of different data sources are consistent, so as to facilitate efficient data fusion; Multi-source data fusion: Use multi-scale fusion technology, feature-level fusion technology and decision-level fusion technology to integrate geological information from satellite remote sensing data, drone images and ground sensor data; Application of 3D geological modeling technology: Utilize the fused data, adopt computer-aided design and geographic information system tools to construct 3D geological models, and use ArcGIS 3D Analyst geological modeling software to refine and optimize the models. During the model generation process, the 3D representation of the geological characteristics of strata, faults, and sedimentary structures will be automatically constructed according to the spatial distribution and attributes of the geological data.

Images