Remote sensing assessment method and system for mining intensity of ion adsorption rare earth mining areas

Abstract

The present invention discloses a remote sensing evaluation method and system for the mining intensity of an ion adsorption type rare earth mining area, including: constructing a remote sensing evaluation model for the mining intensity of an ion adsorption type rare earth mining area; obtaining remote sensing satellite images of the mining area and long-time series remote sensing satellite images; making the pre-processed remote sensing images into a data set; constructing a YOLOv8 target detection model and improving it, using the data set to train the improved YOLOv8 target detection model, using the trained model to identify typical landmarks in the mining area, obtaining the center point of the typical landmark and the coordinates of the detection frame; performing segmentation based on the detection frame according to the recognition result, obtaining the pixel-level segmentation result of the typical landmark and calculating the area; constructing a mining intensity evaluation index system; calculating the mining intensity evaluation index weight of the mining area; and calculating and evaluating the mining intensity of the ion adsorption type rare earth mining area. The present invention realizes a simple, efficient and real-time remote sensing evaluation of the mining intensity of the ion adsorption type rare earth mining area.

Description

Technical Field

[0001] The invention belongs to the technical field of mining assessment, and in particular relates to a remote sensing assessment method and system for the mining intensity of an ion adsorption type rare earth mining area. Background Art

[0002] Ion-adsorbed rare earths are important strategic resources in the 21st century. With their unique physical and chemical properties, they are not only widely used in traditional fields such as metallurgy, petrochemicals, glass and ceramics, and agriculture, but also play an irreplaceable role in high-tech fields such as aerospace, new energy vehicles, and artificial intelligence. Since the late 1970s, China's rare earth industry has risen rapidly, with rare earth reserves and supply both leading the world, and its smelting technology has been at the world's first-class level for many years. However, while developing rapidly, China's ion-adsorbed rare earth reserves are also declining sharply. Excessive exploitation of rare earths has also led to serious damage to the ecological environment. The rare earth mining process will seriously damage the surface vegetation and cause soil erosion. The large amount of ammonia nitrogen and heavy metal pollutants produced in the leaching process seriously pollute the soil, surface water and groundwater, thereby affecting the growth of vegetation and crops; in addition, excessive exploitation of rare earth resources will also lead to landslides, river blockages, water pollution, and even more serious accidents and disasters, causing serious damage to the public's life and health and the ecological environment.

[0003] On this basis, it is essential to monitor ion-adsorbed rare earth mining areas and evaluate mining intensity in mining areas. Most ion-adsorbed rare earths are located in remote mountainous areas with high mountains and dense forests, scattered mining areas and numerous mining sites. This is a typical small-scale manual mining model. The supervision of traditional rare earth mining activities is costly, difficult, inefficient and time-effective. How to achieve efficient and timely monitoring of rare earth mining activities and how to formulate a reasonable evaluation system to achieve mining intensity assessment have become issues that need to be urgently addressed. Summary of the invention

[0004] The purpose of the present invention is to provide a remote sensing assessment method for the mining intensity of an ion adsorption rare earth mining area in view of the deficiencies of the prior art. The method realizes a simple, efficient and real-time remote sensing assessment of the mining intensity of an ion adsorption rare earth mining area.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A remote sensing assessment method for the mining intensity of an ion adsorption rare earth mining area comprises the following steps:

[0007] Step 1: Construct a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas;

[0008] Step 2, obtaining spatial resolution remote sensing satellite images and long time series remote sensing satellite images of the ion adsorption type rare earth mining area in a specified year, and preprocessing them;

[0009] Step 3: The pre-processed spatial resolution remote sensing images are made into a data set that annotates typical features in the mining area;

[0010] Step 4: Build and improve the YOLOv8 target detection model. Use the data set in step 3 to train the improved YOLOv8 target detection model. Use the trained model to identify typical landmarks in the mining area and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box.

[0011] Step 5: According to the recognition result of step 4, use the Segment Anything segmentation model to perform segmentation based on the detection frame, obtain the pixel-level segmentation result of the typical landmark and calculate the area;

[0012] Step 6: Based on the results of Step 4 and Step 5 and the long-term remote sensing satellite images, a mining intensity evaluation index system is constructed;

[0013] Step 7: Calculate the mining intensity evaluation index weight of the mining area according to the constructed mining intensity evaluation index system;

[0014] Step 8: Based on the constructed remote sensing evaluation model for mining intensity of ion adsorption type rare earth mining areas, the evaluation index system constructed in step 6, and the weights of each evaluation index obtained in step 7, the mining intensity of ion adsorption type rare earth mining areas is calculated and evaluated.

[0015] Furthermore, the remote sensing evaluation model for mining intensity of ion adsorption type rare earth mining areas constructed in step 1 is:

[0016]

[0017] In the formula, The change of mining intensity per unit area in a specified time period for ion adsorption type rare earth mining areas; is the unit area; The length of time between the beginning and end of the study; The duration is a single year; Characterize the number of total mining activities; For the The weight of each mining activity characterization type; , The status of the mining activity characterization per unit area at the beginning and end of the study, respectively; It is the ecological environment adjustment coefficient.

[0018] Furthermore, the preprocessing method in step 2 is: declouding the spatial resolution remote sensing image, and performing geometric correction, radiation calibration, and atmospheric correction, and then cropping the spatial resolution remote sensing image using the mining area boundary vector data.

[0019] Furthermore, typical landmarks in mining areas include sedimentation ponds and high-level ponds.

[0020] Furthermore, the method for constructing the mining intensity evaluation index system in step 6 includes:

[0021] The analytic hierarchy process is used to construct an evaluation index system for mining intensity, and a progressive structure of target layer, criterion layer and indicator layer is established. The target layer is the mining intensity of ion adsorption type rare earth mining areas, and the criterion layer includes development and utilization indicators and natural environment indicators.

[0022] Furthermore, the development and utilization indicators specifically include: the number density of sedimentation tanks, the number density of high-level tanks, the area density of sedimentation tanks, the area density of high-level tanks, the area density of sedimentation tank buffer zones, and the area density of high-level tank buffer zones;

[0023] Natural environment indicators specifically include: vegetation index, bare soil index, and land desertification index, which are calculated based on long-term remote sensing satellite images.

[0024] Furthermore, the calculation method of each indicator is:

[0025] The study area is divided into grids, and the area of ​​each grid is the unit area. The density of sedimentation tanks is obtained by summing up the number of sedimentation tank center points in each grid; the density of high-level tanks is obtained by summing up the number of high-level tank center points in each grid; the area density of sedimentation tanks is obtained by summing up the area of ​​sedimentation tanks in each grid; the area density of high-level tanks is obtained by summing up the area of ​​high-level tanks in each grid;

[0026] A buffer zone is created for each sedimentation tank center point, and the area density of the sedimentation tank buffer zone is obtained by counting the sum of the areas of the sedimentation tank buffer zones in each grid; a buffer zone is created for each high-level pool center point, and the number density of the high-level pool is obtained by counting the sum of the areas of the high-level pool buffer zones in each grid;

[0027] Vegetation Index The calculation formula is:

[0028]

[0029] Where: , Indicates the reflectivity of the red band and near-infrared band in Landsat8 images;

[0030] Calculate the mean value of vegetation index in each grid;

[0031] Bare soil index The calculation formula is:

[0032]

[0033] In the formula, , , , , , Respectively represent the reflectance of the blue band, green band, red band, near infrared band, mid-infrared band 1, and thermal infrared band 1 in the image;

[0034] Calculate the mean value of bare soil index in each grid;

[0035] First, the surface albedo is calculated according to the existing empirical formula , the calculation formula is:

[0036]

[0037] In the formula, , , , , They represent the reflectance of the blue band, red band, near infrared band, mid-infrared band 1, and thermal infrared band 1 in Landsat8 images respectively;

[0038] Secondly, after normalization and Linear regression analysis was performed and the calculation formula was:

[0039]

[0040] In the formula, for and The slope of the curve fitted in the feature space, is the intercept of the curve fitted in the feature space;

[0041] Finally calculate the land desertification index , the calculation formula is:

[0042]

[0043] Calculate the average value of land desertification index in each grid.

[0044] Furthermore, the analytic hierarchy process and entropy weight method are used to assign weights to the mining intensity evaluation indicators of the mining area, specifically:

[0045] ①Use the hierarchical analysis method to determine the weight of the index

[0046] Based on the mining intensity evaluation index system in step 6, a judgment matrix between criterion layers and a judgment matrix between indicator layers under each criterion are constructed, and the importance of each object is compared two by two:

[0047]

[0048] In the formula, is the judgment matrix, is the number of objects in the criterion layer or the number of objects in the standard layer;

[0049] Then perform hierarchical single sorting and consistency test, and the calculation formula is:

[0050]

[0051]

[0052] In the formula, is the maximum eigenvalue of the judgment matrix, is the consistency indicator, is the consistency ratio, is a random consistency index, and its value is determined by looking up the corresponding average random number consistency index table. is the number of standard layer objects or the number of standard layer objects, ensuring ;

[0053] Then use the arithmetic mean method to calculate the weight, and the calculation formula is:

[0054]

[0055] In the formula, For the The weight of an object, is the number of objects in the criterion layer or the number of objects in the standard layer;

[0056] Finally, the weights of each indicator based on the hierarchical analysis method are calculated using the following formula:

[0057]

[0058] In the formula, For the The weight of the indicators calculated based on the hierarchical analysis method, For the The weight of the criteria, For the The weight of this indicator under each criterion layer, is the number of criterion layer objects;

[0059] ②Use entropy weight method to determine the weight of indicators

[0060] First, the data is normalized, and then the information entropy is calculated. The calculation formula is as follows:

[0061]

[0062]

[0063] In the formula, For the The information entropy of the indicator, is the number of samples;

[0064] Since information entropy is negatively correlated with weight, the weight calculation formula is:

[0065]

[0066] In the formula, For the The weight of the indicator calculated based on the entropy weight method, is the number of indicators;

[0067] ③ Combine the analytic hierarchy process and entropy weight method to determine the comprehensive weight of the indicators. The calculation formula is:

[0068]

[0069] In the formula, is the comprehensive weight of the indicator, is the weight of the indicator calculated based on the hierarchical analysis method, is the weight of the indicator calculated based on the entropy weight method.

[0070] Furthermore, the mining intensity remote sensing evaluation index is calculated based on the intensity remote sensing evaluation model and the obtained weights, and then graded and divided into four levels using the threshold method:

[0071] The value range of the first-level index is , indicating that there is no ion adsorption type rare earth mining activity in the area, or the mining activity has weakened;

[0072] The secondary index value range is , indicating that there is low-intensity ion adsorption-type rare earth mining activity in the area;

[0073] The value range of the third-level index is , indicating that there is relatively low intensity of ion adsorption type rare earth mining activities in the area;

[0074] The four-level index value range is , indicating that there is a high intensity of ion adsorption type rare earth mining activities in the area;

[0075] The five-level index value range is , indicating that there is a high intensity of ion adsorption type rare earth mining activities in the area.

[0076] Another object of the present invention is to provide a system for implementing the above-mentioned ion adsorption type rare earth mining area mining intensity remote sensing assessment method, comprising:

[0077] Evaluation model building module, used to build a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas;

[0078] The data acquisition module is used to acquire high spatial resolution remote sensing satellite images and long time series remote sensing satellite images of ion adsorption type rare earth mining areas in a specified year and pre-process them;

[0079] The data set production module is used to produce the pre-processed high spatial resolution remote sensing images into a data set annotating typical features in the mining area;

[0080] YOLOv8 target detection model building module, used to build and improve the YOLOv8 target detection model, and use the data set to train the improved YOLOv8 target detection model, use the trained model to identify typical landmarks in the mining area, and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box;

[0081] The area estimation module is used to establish the module recognition results according to the YOLOv8 target detection model, use the SegmentAnything segmentation model to perform detection box-based segmentation, obtain the pixel-level segmentation results of typical landmarks, and calculate the area;

[0082] An evaluation index system construction module is used to construct a mining intensity evaluation index system for a mining area based on the estimation results of the area estimation module and the long-time series remote sensing satellite images obtained by the data acquisition module;

[0083] A weight calculation module is used to calculate the weight of the mining intensity evaluation index of the mining area according to the constructed mining intensity evaluation index system of the mining area;

[0084] The mining intensity assessment module is used to calculate and assess the mining intensity of ion-adsorption rare earth mining areas based on the constructed remote sensing assessment model for mining intensity of ion-adsorption rare earth mining areas and the obtained weights.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] 1. The present invention innovatively proposes an evaluation method for the mining intensity of ion adsorption rare earth mining areas based on remote sensing technology. First, remote sensing and deep learning technology are used to accurately locate the mining points in the mining area and estimate the area of ​​typical landforms in the mining activities based on the typical remote sensing interpretation marks of the ion adsorption rare earth mining area. Secondly, according to the mining process and principle of ion adsorption rare earth, objects related to development and utilization and natural environment that can accurately and effectively characterize mining behavior and intensity are selected as evaluation indicators. Finally, a set of evaluation system for the mining intensity of ion adsorption rare earth mining areas based entirely on remote sensing data is constructed to realize simple, efficient and real-time remote sensing evaluation of the mining intensity of rare earth mining areas.

[0087] 2. There are few evaluation bodies for the mining intensity of ion adsorption rare earth mining areas, and there is no evaluation system that regards the development and utilization process of the mining area as the cumulative effect of human mining activities on the mining elements of the mining area to evaluate the mining intensity of the mining area. Based on the theoretical knowledge of the impulse theorem of physics, the present invention regards all human mining activities on the rare earth resources in the study area as the force applied to the study area. The cumulative change process of all mining states of rare earth resources in the study area within a certain time range is the impulse of the force of human mining activities in the study area. The total cumulative change of all mining states of rare earth resources in the study area within a certain time and space range is the mining intensity within this range. A new remote sensing evaluation system for the mining intensity of ion adsorption rare earth mining areas is constructed;

[0088] 3. In view of the problems that the evaluation indicators involved in the existing ion adsorption rare earth mining intensity evaluation system are complex and diverse, and the relevant data are difficult to obtain, the present invention constructs an evaluation system for the mining intensity of ion adsorption rare earth mining areas that is completely based on remote sensing data. The system uses the characteristic that sedimentation tanks and high-level tanks are key indicators of ion adsorption rare earth mining activities, and cleverly uses the number density of sedimentation tanks and high-level tanks, the area density of sedimentation tanks and high-level tanks, and the buffer zone of sedimentation tanks and high-level tanks as development and utilization indicators. At the same time, the vegetation index, bare soil index, and land desertification index are selected as environmental impact indicators. The hierarchical analysis method and the entropy weight method are combined to give weights to each indicator, making the evaluation system more objective and accurate, and finally realizing the remote sensing evaluation of the mining intensity of ion adsorption rare earth mining areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 The present invention is a flowchart of a remote sensing evaluation method for ion adsorption type rare earth mining intensity according to an embodiment of the present invention. DETAILED DESCRIPTION

[0090] The technical solutions in the embodiments of the present invention will be described clearly and completely below in combination with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0092] The present invention will be further described below in conjunction with specific embodiments, but the present invention is not limited thereto.

[0093] like Figure 1As shown, an embodiment of the present invention provides a remote sensing assessment method for mining intensity of an ion adsorption type rare earth mining area. The research object of this embodiment is a typical ion adsorption type rare earth mining area in 2013, 2016, 2018, 2021 and 2023. This embodiment aims to establish a mining intensity assessment model through remote sensing and deep learning technology, monitor and track the mining activity points of the rare earth mining area in each year, and realize the visualization of the long time series and the whole process of mining area development, further analyze the changes and development trends of mining behavior in the rare earth mining area over the past 10 years, and then realize the quantitative assessment of the development degree and resource reserves of the mining area, so as to assist relevant departments in more scientific and reasonable planning and development of rare earth mining areas. It includes the following steps:

[0094] Step 1: Construct a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas;

[0095] The resources and location advantages of the mining area make it a place where the development and utilization of local mineral resources take place. Under this artificially imposed "thrust", after a certain period of time, the shape, internal structure and development and utilization status of the mining area will change. The degree of change in the properties of the mining area caused by human factors can be understood as the mining intensity of the mining area, and the quantitative description of the degree of change is the mining intensity assessment of the mining area.

[0096] The mining process of ion adsorption rare earth mining areas can be regarded as the cumulative effect of human mining activities on various mining elements in the rare earth mining areas. The mining intensity of ion adsorption rare earth mining areas can be defined as the weighted total amount of cumulative changes in various mining activities per unit area of ​​the mining area within a certain time range. The formula is as follows:

[0097]

[0098] In the formula, is the mining intensity of the mining area; , , ... respectively characterize the types of mining activities; , , ...respectively, the relationship between mining intensity in a mining area and the cumulative changes in the types of mining activity characterization; , , ...are weighting coefficients respectively.

[0099] In physics, the momentum theorem states that the change in momentum of an object at the beginning and end of a process is equal to the impulse of the force it is subjected to during the process, that is, the product of the force and the time over which the force acts. The momentum theorem formula is as follows:

[0100]

[0101] Pressure refers to the pressure per unit area of ​​an object. The relationship between pressure and force is as follows:

[0102]

[0103] According to the momentum theorem and the relationship between pressure and force, if all human mining activities on rare earth resources in the study area are regarded as the force exerted on the study area , then the cumulative change process of all mining states of rare earth resources in the study area within a certain time range is the impulse of the force of human mining activities in the study area The total cumulative change of all mining states of rare earth resources in the study area within a certain time and space range is the mining intensity within this range. .

[0104] The mining momentum formula is as follows:

[0105]

[0106] The mining intensity formula is as follows:

[0107]

[0108] For ion adsorption rare earth mining areas, the quality It can be understood as the weight of a certain type of mining activity. , speed difference It can be understood as the difference in the rate of change of the mining activity characterization types in a mining area within a certain time range.

[0109] In summary, the calculation formula for the mining momentum of ion adsorption rare earth mining areas in a specified year is:

[0110]

[0111] In the formula, It is the mining momentum per unit area of ​​ion adsorption type rare earth mining area in a specified year; The duration is a single year; Characterize the number of total mining activities; For the The weight of each mining activity characterization type; For the a state that characterizes mining activities; is the ecological environment adjustment coefficient, and the optimal value is determined through multiple experiments in this embodiment.

[0112] The calculation formula for the mining intensity of ion adsorption rare earth mining areas within a specified time period is:

[0113]

[0114] In the formula, The change of mining intensity per unit area in a specified time period for ion adsorption type rare earth mining areas; is the unit area; The length of time between the beginning and end of the study; The duration is a single year; Characterize the number of total mining activities; For the The weight of each mining activity characterization type; , The status of the mining activity characterization per unit area at the beginning and end of the study, respectively; is the ecological environment adjustment coefficient, and the optimal value is determined through multiple experiments in this embodiment.

[0115] Step 2, obtaining high spatial resolution remote sensing satellite images and long time series remote sensing satellite images of the ion adsorption type rare earth mining area in a specified year, and preprocessing them;

[0116] The Google Earth satellite map base map and multiple Landsat-8 images of a typical ion adsorption rare earth mining area in 2013, 2016, 2018, 2021 and 2023 were obtained, the remote sensing images were declouded, and geometric correction, radiation calibration and atmospheric correction were performed. Then, the remote sensing images were cropped using the vector data of the study area.

[0117] Step 3: The pre-processed high spatial resolution remote sensing images are made into a data set for marking typical landforms in the mining area;

[0118] In this step, the preprocessed high-spatial-resolution remote sensing images are cropped into image slices to produce a typical mining area landform dataset for target detection. The sample image size in the dataset is 512 pixels * 512 pixels, and the sample label categories include sedimentation tanks, high-level pools, buildings, and farmland. The samples are randomly divided into training sets, validation sets, and test sets.

[0119] Step 4: Build and improve the YOLOv8 target detection model. Use the data set in step 3 to train the improved YOLOv8 target detection model. Use the trained model to identify typical landmarks in the mining area and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box.

[0120] In this embodiment, the improved YOLOv8 target detection model structure is improved as follows:

[0121] (1) Introduce the Asymptotic Feature Pyramid Network (AFPN) structure to improve the original neck network structure and realize the mutual fusion of high-level semantic information and low-level spatial position information; specifically, connect the third CSPLayer_2Conv module and the last SPPF module in the feature extraction network structure to two ASFF2 modules; then connect these two ASFF2 modules to the upsampling structure composed of three consecutive CSPLayer_2Conv modules, and at the same time, connect the second CSPLayer_2Conv module in the feature extraction network structure to the last CSPLayer_2Conv module in the upsampling structure. Then connect the three CSPLayer_2Conv modules in the upsampling structure to three ASFF3 modules. Then connect these three ASFF3 modules to the downsampling structure composed of three consecutive CSPLayer_2Conv modules;

[0122] Through the above improvements, the network starts to fuse adjacent low-level features and gradually introduces high-level features, achieving more comprehensive and rich information transmission and fusion between different levels, while alleviating the inconsistency of multi-object information through adaptive spatial fusion operations;

[0123] (2) Use the Convolutional Block Attention Module (CBAM) to integrate into the original neck network structure to improve the detection performance of the network. Specifically, the three CSPLayer_2Conv modules in the downsampling structure composed of three consecutive CSPLayer_2Conv modules are connected to three CBAM attention modules respectively, and then the three CBAM attention modules are connected to the Dectect detection head modules for different target sizes respectively;

[0124] Through the above improvements, the CBMA attention module infers the attention map from the channel and spatial dimensions in turn, and then multiplies the obtained attention map with the input feature map to perform adaptive feature refinement;

[0125] (3) The Slicing Aided Hyper inference (SAHI) module is introduced in the model reasoning phase. This module is designed specifically for small object detection. The module first divides the image to be inferred into M*N slices with overlapping parts, performs target detection on each slice, sets the intersection-over-union ratio threshold, uses non-maximum suppression to eliminate redundant overlapping detection frames, and finally merges the inference results back to the original image size.

[0126] Using SAHI for auxiliary reasoning can effectively improve the model's detection effect on small targets and help capture smaller sedimentation ponds in remote sensing images.

[0127] The training set is used to train the YOLOv8 target detection model and the validation set is used to verify it. Finally, the test set is used to test and obtain the YOLOv8 target detection model with the best performance.

[0128] After obtaining the trained YOLOv8 target detection model, the Google Earth satellite map base maps of the study area in 2013, 2016, 2018, 2021, and 2023 were cropped into image slices according to the size of 512pixel*512pixel, and the above slices were put into the trained YOLOv8 target detection model for recognition. When the YOLOv8 target detection model recognizes that there are sedimentation ponds or high-level ponds in the image slices, the identified object (i.e., sedimentation ponds or high-level ponds) will be marked with a rectangular detection frame at the corresponding position of the image slice. At the same time, the pixel coordinates of the center point of the identified object based on the image slice pixel coordinate system and the pixel coordinates of the upper left and lower right corners of the rectangular detection frame, as well as the category label corresponding to the identified object are output in text file format.

[0129] The pixel coordinate system coordinates of the center point of the identified object generated by the improved YOLOv8 target detection model are converted into the WGS_1984_UTM_zone_47N plane coordinate system coordinates representing the real space position based on the pixel coordinate system coordinates of the image slice; the conversion formula between pixel coordinates and plane coordinates is:

[0130]

[0131]

[0132] In the formula, , Represents the pixel coordinates of the center point of the detection box; , Indicates the pixel coordinates of the upper left corner of the detection box; , Indicates the pixel coordinates of the lower right corner of the detection box; , express The pixel coordinates of the point; , express The WGS_1984_UTM_zone_47N plane coordinates of the point; , Represents the WGS_1984_UTM_zone_47N plane coordinate of the upper left corner of the slice; , It represents the spatial resolution of the remote sensing image in the horizontal and vertical directions, which is 0.284 (unit: m) in this embodiment.

[0133] Step 5: According to the recognition result of step 4, use the Segment Anything segmentation model to perform segmentation based on the detection frame, obtain the pixel-level segmentation result of the typical landmark and calculate the area;

[0134] Based on the sedimentation pool and high-level pool identified by the above YOLOv8 target detection model, the sedimentation pool and high-level pool inside the detection frame are segmented at the pixel level as follows:

[0135] 1) Load a specific segmentation model from the "Segment Anything" model registry based on a predefined or user-specified model type and checkpoint path;

[0136] 2) The input of the model is a directory containing image slices and corresponding detection result text files. For each image slice, the corresponding detection result is read from the text file, including the detection box and label of the identified object in the image slice;

[0137] 3) Use the "Segment Anything" model to segment the image slices within the detection box of the identified object; each segmented region is then converted into a binary mask, which is vectorized into polygons using the vectorize_mask function, which uses raster geometry to convert pixel-based masks into vectorized shapes. Each polygon is assigned a label based on the detection result, creating a labeled vector dataset;

[0138] 4) Use the GeoPandas package to compile the vectorized and labeled polygons into a GeoDataFrame, which includes the original coordinate system of the image slices, namely the WGS_1984_UTM_zone_47N plane coordinate system; then save it as a Shapefile file in the specified path, making the vector data accessible to GIS applications;

[0139] Through the above operations, the Segment Anything segmentation model is implemented to perform pixel-level segmentation on the main recognition object only within the detection frame, and the corresponding category label is assigned to the segmentation result according to the category label to which the detection frame belongs, and then the pixel-level segmentation results of the sedimentation pool and the high-level pool are output in .shp format.

[0140] After pixel-level segmentation of the sedimentation tank and the high-level pool, the area of ​​the sedimentation tank and the high-level pool is estimated as follows:

[0141] Load a specific segmentation model from the "Segment Anything" model registry based on predefined or user-specified model type and checkpoint path;

[0142] Read the image slices and use OpenCV to convert the image from BGR format to RGB format to adapt to the model input requirements. Use the "Segment Anything" model to segment the image slices within the detection box of the identified object. For each segmented object, calculate its pixel area and convert it to the actual area using the given spatial resolution. The area calculation formula is:

[0143] ;

[0144] In the formula, It is the area of ​​sedimentation tank or high-level tank in the detection frame; is the number of pixels of the segmentation result in the detection frame; , It represents the spatial resolution of the remote sensing image in the horizontal and vertical directions, which is 0.284 (unit: m) in this embodiment.

[0145] After calculating the area of ​​the sedimentation tank or high-level pool within the detection frame, the area is accumulated and counted according to the category of the identified object to obtain the total area of ​​the sedimentation tank and the total area of ​​the high-level pool in the study area. Then, the pastemask2img function is used to superimpose the segmentation mask on the original image and mark the detection frame and calculated area of ​​each identified object.

[0146] Step 6: Based on the results of step 4 and step 5, construct an evaluation index system for mining intensity in the mining area;

[0147] According to the construction concept of the mining intensity model of ion adsorption rare earth mining areas, objects that can accurately reflect the current mining status of rare earth mining areas are selected as the representation of mining activities, namely, mining intensity evaluation indicators. The hierarchical analysis method is used to construct the mining intensity evaluation index system, and a progressive structure of target layer, criterion layer, and indicator layer is established: the target layer is the mining intensity of ion adsorption rare earth mining areas; the criterion layer includes development and utilization indicators and natural environment indicators; the development and utilization indicators specifically include: the number density of sedimentation ponds, the number density of high-level ponds, the area density of sedimentation ponds, the area density of high-level ponds, the area density of 200m buffer zone of sedimentation ponds, and the area density of 200m buffer zone of high-level ponds; the natural environment indicators specifically include: vegetation index, bare soil index, and land desertification index.

[0148] The calculation methods of each index are as follows. The study area is divided into grids of 100m, and the area of ​​each grid is the unit area:

[0149] ① Sedimentation tank number density

[0150] According to the results of step 4 and step 5, count the total number of sedimentation tank center points in each grid (unit: piece);

[0151] ②High-level pool number density

[0152] According to the results of step 4 and step 5, count the total number of high-level pool center points in each grid (unit: number);

[0153] ③ Sedimentation tank area density

[0154] According to the results of step 4 and step 5, the total area of ​​sedimentation tanks in each grid is calculated (unit: m 2 );

[0155] ④High-level pool area density

[0156] According to the results of step 4 and step 5, the total area of ​​high-level pools in each grid is calculated (unit: m 2 );

[0157] ⑤ Area density of 200m buffer zone of sedimentation tank

[0158] Create a buffer zone with a radius of 200m for each sedimentation pond center point, and calculate the total area of ​​the sedimentation pond buffer zone in each grid (unit: m 2 );

[0159] ⑥ High-level pool density

[0160] Based on the results of steps 4 and 5, a buffer zone with a radius of 200 m is created for each high-level pool center point, and the total area of ​​the high-level pool buffer zone in each grid is calculated (unit: m 2 );

[0161] ⑦ Vegetation index

[0162] Vegetation Index The calculation formula is:

[0163]

[0164] Where: , Indicates the reflectivity of the red band and near-infrared band in Landsat8 images;

[0165] Calculate the mean value of vegetation index in each grid;

[0166] ⑧ Bare soil index

[0167] Bare soil index The calculation formula is:

[0168]

[0169] In the formula, , , , , , They represent the reflectance of the blue band, green band, red band, near infrared band, mid-infrared band 1, and thermal infrared band 1 in Landsat8 images respectively;

[0170] Calculate the mean value of bare soil index in each grid;

[0171] ⑨Land desertification index

[0172] First, the surface albedo is calculated according to the existing empirical formula , the calculation formula is:

[0173]

[0174] In the formula, , , , , They represent the reflectance of the blue band, red band, near infrared band, mid-infrared band 1, and thermal infrared band 1 in Landsat8 images respectively;

[0175] Secondly, after normalization and Linear regression analysis was performed and the calculation formula was:

[0176]

[0177] In the formula, for and The slope of the curve fitted in the feature space, is the intercept of the curve fitted in the feature space;

[0178] Finally calculate the land desertification index , the calculation formula is:

[0179]

[0180] Calculate the average value of land desertification index in each grid.

[0181] In order to unify the dimensions of each mining intensity evaluation index, each index in each grid is reclassified according to the value range. The reclassified values ​​of the mining intensity evaluation index of the ion adsorption type rare earth mining area in this experimental example are shown in Table 1.

[0182] Table 1 Reclassification value table of mining intensity evaluation indexes for ion adsorption rare earth mining areas

[0183]

[0184] Step 7: Calculate the mining intensity evaluation index weight of the mining area according to the constructed mining intensity evaluation index system;

[0185] In the embodiment, in order to reflect the statistical laws between indicators and reflect the importance of the selected indicators in a specific field, the hierarchical analysis method and entropy weight method are used to assign weights to the mining intensity evaluation indicators of the mining area.

[0186] ①Use the hierarchical analysis method to determine the weight of the indicators.

[0187] First, based on the mining intensity evaluation index system in step 6, a judgment matrix between criterion layers and a judgment matrix between indicator layers under each criterion are constructed, and the importance of each object is compared two by two:

[0188]

[0189] In the formula, is the judgment matrix, is the number of objects in the criterion layer or the number of objects in the standard layer;

[0190] Then perform hierarchical single sorting and consistency test, and the calculation formula is:

[0191]

[0192] In the formula, is the maximum eigenvalue of the judgment matrix, is the consistency indicator, is the consistency ratio, is a random consistency index, and its value is determined by looking up the corresponding average random number consistency index table. is the number of standard layer objects or the number of standard layer objects, ensuring ;

[0193] Then use the arithmetic mean method to calculate the weight, and the calculation formula is:

[0194]

[0195] In the formula, For the The weight of an object, is the number of objects in the criterion layer or the number of objects in the standard layer;

[0196] Finally, the weights of each indicator based on the hierarchical analysis method are calculated using the following formula:

[0197]

[0198] In the formula, For the The weight of the indicators calculated based on the hierarchical analysis method, For the The weight of the criteria, For the The weight of this indicator under each criterion layer, is the number of criteria layer objects;

[0199] ②Use entropy weight method to determine the weight of indicators

[0200] First, the data is normalized, and then the information entropy is calculated. The calculation formula is as follows:

[0201]

[0202]

[0203] In the formula, For the The information entropy of the indicator, is the number of samples;

[0204] Since information entropy is negatively correlated with weight, the weight calculation formula is:

[0205]

[0206] In the formula, For the The weight of the indicator calculated based on the entropy weight method, is the number of indicators;

[0207] ③ Combine the analytic hierarchy process and entropy weight method to determine the comprehensive weight of the indicators. The calculation formula is:

[0208]

[0209] In the formula, is the comprehensive weight of the indicator, is the weight of the indicator calculated based on the hierarchical analysis method, is the weight of the indicator calculated based on the entropy weight method.

[0210] The weight values ​​of the evaluation indexes for the mining intensity of the ion adsorption type rare earth mining area in this experimental example are shown in Table 2.

[0211] Table 2 Weight values ​​of evaluation indexes for mining intensity of ion adsorption rare earth mining areas

[0212]

[0213] Step 8: Based on the constructed remote sensing evaluation model for mining intensity of ion-adsorption type rare earth mining areas and the weights obtained in step 7, the mining intensity of ion-adsorption type rare earth mining areas is calculated and evaluated.

[0214] In the embodiment, the remote sensing evaluation index value of mining intensity of the mining area is calculated based on the constructed ion adsorption type rare earth mining area mining intensity remote sensing evaluation model, the evaluation index obtained in step 6, and the evaluation index weight obtained in step 7. Among them, the various mining intensity evaluation indicators calculated in step 6 are the states of the mining activities per unit area in the calculation formula of mining intensity of ion adsorption type rare earth mining area in step 1. The weights of various mining intensity evaluation indicators calculated in step 7 are the weights of various mining activity characterization types in the calculation formula for the mining intensity of ion adsorption rare earth mining areas in step 1. After calculation, the remote sensing evaluation index of mining intensity of ion adsorption type rare earth mining area is graded according to the actual situation of the embodiment, and is divided into four levels using the threshold method: the first level value interval is , indicating that there is no ion adsorption type rare earth mining activity in the area, or the mining activity has weakened; the secondary value range is , indicating that there is low-intensity ion adsorption rare earth mining in the area; the third-level value range is , indicating that there is a relatively low intensity of ion adsorption type rare earth mining in the area; the fourth level value range is , indicating that there is a high intensity of ion adsorption rare earth mining activities in the area; the five-level value range is , indicating that there is a high intensity of ion adsorption rare earth mining in the area. Based on the above classification results, the mining intensity of the ion adsorption rare earth mining area is evaluated and analyzed.

[0215] The embodiment of the present invention further provides a system for implementing the above-mentioned ion adsorption type rare earth mining area mining intensity remote sensing assessment method, comprising:

[0216] Evaluation model building module, used to build a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas;

[0217] The data acquisition module is used to acquire high spatial resolution remote sensing satellite images and long time series remote sensing satellite images of ion adsorption type rare earth mining areas in a specified year and pre-process them;

[0218] The data set production module is used to produce the pre-processed high spatial resolution remote sensing images into a data set annotating typical features in the mining area;

[0219] YOLOv8 target detection model building module, used to build and improve the YOLOv8 target detection model, and use the data set to train the improved YOLOv8 target detection model, use the trained model to identify typical landmarks in the mining area, and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box;

[0220] The area estimation module is used to establish the module recognition results according to the YOLOv8 target detection model, use the SegmentAnything segmentation model to perform detection box-based segmentation, obtain the pixel-level segmentation results of typical landmarks, and calculate the area;

[0221] An evaluation index system construction module is used to construct a mining intensity evaluation index system for a mining area based on the estimation results of the area estimation module and the long-time series remote sensing satellite images obtained by the data acquisition module;

[0222] A weight calculation module is used to calculate the weight of the mining intensity evaluation index of the mining area according to the constructed mining intensity evaluation index system of the mining area;

[0223] The mining intensity assessment module is used to calculate and assess the mining intensity of ion-adsorption rare earth mining areas based on the constructed remote sensing assessment model for mining intensity of ion-adsorption rare earth mining areas and the obtained weights.

[0224] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention constructs an ion adsorption type rare earth mining area mining intensity evaluation model based on the momentum theorem of physics through multi-source remote sensing data, selects objects that can accurately reflect the current mining status of rare earth mining areas as mining intensity evaluation indicators, and uses a combination of hierarchical analysis method and entropy weight method to give weights to evaluation indicators, making the evaluation system more objective and comprehensive. Finally, a simple, efficient and real-time remote sensing evaluation of the mining intensity of ion adsorption type rare earth mining areas is realized.

[0225] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the contents of the specification of the present invention should be included in the protection scope of the present invention.

Claims

1. A remote sensing assessment method for the mining intensity of an ion adsorption rare earth mining area, characterized in that: The steps include: Step 1: Construct a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas; Step 2, obtaining spatial resolution remote sensing satellite images and long time series remote sensing satellite images of the ion adsorption type rare earth mining area in a specified year, and preprocessing them; Step 3: The pre-processed spatial resolution remote sensing images are made into a data set that annotates typical features in the mining area; Step 4: Build and improve the YOLOv8 target detection model. Use the data set in step 3 to train the improved YOLOv8 target detection model. Use the trained model to identify typical landmarks in the mining area and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box. Step 5: According to the recognition result of step 4, use the Segment Anything segmentation model to perform segmentation based on the detection frame, obtain the pixel-level segmentation result of the typical landmark and calculate the area; Step 6: Based on the results of Step 4 and Step 5 and the long-term remote sensing satellite images, a mining intensity evaluation index system is constructed; Step 7: Calculate the mining intensity evaluation index weight of the mining area according to the constructed mining intensity evaluation index system; Step 8: Based on the constructed remote sensing evaluation model for mining intensity of ion adsorption type rare earth mining areas, the evaluation index system constructed in step 6, and the weights of each evaluation index obtained in step 7, the mining intensity of ion adsorption type rare earth mining areas is calculated and evaluated; Among them, the remote sensing evaluation model for the mining intensity of the ion adsorption type rare earth mining area constructed in step 1 is: Where P is the change in mining intensity per unit area in a specified time period of an ion adsorption rare earth mining area; S is the unit area; ΔT is the time interval between the beginning and the end of the study; Δt is the length of a single year; N is the total number of mining activity representations; λ i is the weight of the i-th mining activity characterization type; x i1 、x i2 are the states of mining activities per unit area at the beginning and end of the study, respectively; b is the ecological environment adjustment coefficient.

2. The remote sensing assessment method for ion adsorption type rare earth mining intensity according to claim 1 is characterized in that: The preprocessing method in step 2 is: to remove clouds from the spatial resolution remote sensing image, and perform geometric correction, radiation calibration, and atmospheric correction, and then use the mining area boundary vector data to crop the spatial resolution remote sensing image.

3. The remote sensing assessment method for ion adsorption type rare earth mining area mining intensity according to claim 1 is characterized in that: Typical landmarks in mining areas include sedimentation ponds and high-level ponds.

4. The remote sensing assessment method for ion adsorption type rare earth mining intensity according to claim 3 is characterized in that: The method for constructing the mining intensity evaluation index system in step 6 includes: The analytic hierarchy process is used to construct an evaluation index system for mining intensity, and a progressive structure of target layer, criterion layer and indicator layer is established. The target layer is the mining intensity of ion adsorption type rare earth mining areas, and the criterion layer includes development and utilization indicators and natural environment indicators.

5. The remote sensing assessment method for ion adsorption type rare earth mining area mining intensity according to claim 4 is characterized in that: The specific development and utilization indicators include: the number density of sedimentation tanks, the number density of high-level tanks, the area density of sedimentation tanks, the area density of high-level tanks, the area density of sedimentation tank buffer zones, and the area density of high-level tank buffer zones; Natural environment indicators specifically include: vegetation index, bare soil index, and land desertification index, which are calculated based on long-term remote sensing satellite images.

6. The remote sensing assessment method for ion adsorption type rare earth mining intensity according to claim 5 is characterized in that: The calculation method of each indicator is: The study area is divided into grids, and the area of ​​each grid is the unit area. The density of sedimentation tanks is obtained by summing up the number of sedimentation tank center points in each grid; the density of high-level tanks is obtained by summing up the number of high-level tank center points in each grid; the area density of sedimentation tanks is obtained by summing up the area of ​​sedimentation tanks in each grid; the area density of high-level tanks is obtained by summing up the area of ​​high-level tanks in each grid; A buffer zone is created for each sedimentation tank center point, and the area density of the sedimentation tank buffer zone is obtained by counting the sum of the areas of the sedimentation tank buffer zones in each grid; a buffer zone is created for each high-level pool center point, and the number density of the high-level pool is obtained by counting the sum of the areas of the high-level pool buffer zones in each grid; The calculation formula of vegetation index NDVI is: Where: B red , B nir Indicates the reflectivity of the red band and near-infrared band in Landsat8 images; Calculate the mean value of vegetation index in each grid; The bare soil index SI calculation formula is: In the formula, B blue , B red , B nir Respectively represent the reflectance of the blue band, red band, and near-infrared band in the image; Calculate the mean value of bare soil index in each grid; First, the surface albedo is calculated according to the existing empirical formula. The calculation formula is: Albedo=0.356B blue +0.130B red +0.373B nir +0.0858B swir1 +0.072B swir2 +0.0018 In the formula, B blue , B red , B nir Respectively represent the reflectance of the blue band, red band, and near-infrared band in the Landsat8 image; Secondly, the normalized NDVI and Albedo were subjected to linear regression analysis, and the calculation formula was: NDVI=a·Albedo+b Where a is the slope of the curve fitted by NDVI and Albedo in the feature space, and b is the intercept of the curve fitted in the feature space; Finally, the land desertification index DDI is calculated using the following formula: Calculate the average value of land desertification index in each grid.

7. The remote sensing assessment method for ion adsorption type rare earth mining intensity according to claim 1 is characterized in that: The analytic hierarchy process and entropy weight method are used to assign weights to the mining intensity evaluation indicators of the mining area, specifically: ①Use the hierarchical analysis method to determine the weight of the index Based on the mining intensity evaluation index system in step 6, a judgment matrix between criterion layers and a judgment matrix between indicator layers under each criterion are constructed, and the importance of each object is compared two by two: In the formula, A is the judgment matrix, n is the number of objects in the criterion layer or the number of objects in the standard layer; Then perform hierarchical single sorting and consistency test, and the calculation formula is: In the formula, λ max is the maximum eigenvalue of the judgment matrix, CI is the consistency index, CR is the consistency ratio, RI is the random consistency index, and the value is determined by looking up the corresponding average random number consistency index table, n is the number of objects in the criterion layer or the number of objects in the standard layer, ensuring that CR < 0.1; Then use the arithmetic mean method to calculate the weight, and the calculation formula is: In the formula, w i is the weight of the ith object, n is the number of objects in the criterion layer or the number of objects in the standard layer; Finally, the weights of each indicator based on the hierarchical analysis method are calculated using the following formula: Where W iAHP is the weight of the jth indicator calculated based on the hierarchical analysis method, W iGuidelines is the weight of the ith criterion, W iIndicators is the weight of the indicator under the i-th criterion layer, and n is the number of objects in the criterion layer; ②Use entropy weight method to determine the weight of indicators First, the data is normalized, and then the information entropy is calculated. The calculation formula is as follows: In the formula, E j is the information entropy of the jth indicator, and m is the number of samples; The weight calculation formula is: Where W jEWM is the weight of the jth indicator calculated based on the entropy weight method, and n is the number of indicators; ③ Combine the analytic hierarchy process and entropy weight method to determine the comprehensive weight of the indicators. The calculation formula is: W=W AHP ·IN EWM In the formula, W is the comprehensive weight of the indicator, W AHP is the weight of the index calculated based on the hierarchical analysis method, W EWM is the weight of the indicator calculated based on the entropy weight method.

8. The remote sensing assessment method for ion adsorption type rare earth mining intensity according to claim 1 is characterized in that: According to the intensity remote sensing evaluation model and the obtained weight, the mining intensity remote sensing evaluation index is calculated and graded, and divided into four levels using the threshold method: The value range of the first-level index is (-∞,0], which means that there is no ion adsorption type rare earth mining activity in the area, or the mining activity has weakened; The value range of the secondary index is (0,0.15], indicating that there is low-intensity ion adsorption type rare earth mining activity in the region; The value range of the third-level index is (0.15, 0.2], indicating that there is a relatively low intensity of ion adsorption type rare earth mining activities in the region; The value range of the fourth-level index is (0.2, 0.3], indicating that there is a high intensity of ion adsorption rare earth mining activities in the region; The five-level index value range is (0.3, +∞), indicating that there is high-intensity ion adsorption-type rare earth mining activities in the region.

9. A system for implementing the remote sensing assessment method for ion adsorption type rare earth mining area mining intensity according to any one of claims 1 to 8, characterized in that: include: Evaluation model building module, used to build a remote sensing evaluation model for the mining intensity of ion adsorption rare earth mining areas; The data acquisition module is used to acquire the spatial resolution remote sensing satellite images and long-time series remote sensing satellite images of the ion adsorption type rare earth mining area in a specified year and pre-process them; The data set production module is used to produce the pre-processed spatial resolution remote sensing images into a data set annotating typical landforms in the mining area; YOLOv8 target detection model building module, used to build and improve the YOLOv8 target detection model, and use the data set to train the improved YOLOv8 target detection model, use the trained model to identify typical landmarks in the mining area, and obtain the coordinates of the center point of the typical landmark and the coordinate data of the detection box; The area estimation module is used to establish the module recognition results according to the YOLOv8 target detection model, use the SegmentAnything segmentation model to perform detection box-based segmentation, obtain the pixel-level segmentation results of typical landmarks, and calculate the area; An evaluation index system construction module is used to construct a mining intensity evaluation index system for a mining area based on the estimation results of the area estimation module and the long-time series remote sensing satellite images obtained by the data acquisition module; A weight calculation module is used to calculate the weight of the mining intensity evaluation index of the mining area according to the constructed mining intensity evaluation index system of the mining area; The mining intensity assessment module is used to calculate and assess the mining intensity of ion-adsorption rare earth mining areas based on the constructed remote sensing assessment model for mining intensity of ion-adsorption rare earth mining areas and the obtained weights.

Images