Method and system for unmanned aerial vehicle to shoot and identify soil mineral composition and storage medium

Abstract

The application discloses a kind of unmanned aerial vehicle shooting identification soil mineral component method, comprising the following steps: S1, obtains the several acquisition data of unmanned aerial vehicle in target area collection, each acquisition data corresponds at least one sub-region in target area, at least include the surface multispectral image data of target area, geographic information;S2, the surface multispectral image data is uniformly corrected and handled, generates standard multispectral data;S3, based on the geographic information in the acquisition data, the standard multispectral data of each acquisition data is integrated, generates the standard multispectral data of target area;S4, based on the standard multispectral data of target area, obtains the soil mineral component of this target area.The application is identified by using unmanned aerial vehicle to soil mineral component, simplifies the acquisition method of multispectral image data, reduces cost.

Description

Technical Field

[0001] This invention relates to the field of resource and environmental information technology, and in particular to a method, system and storage medium for identifying soil mineral composition by taking pictures with a drone. Background Technology

[0002] The key to the utilization of soil mineral resources lies in the accurate identification of their mineral composition and content. Traditional laboratory analysis methods (such as X-ray diffraction and X-ray fluorescence spectroscopy) are highly accurate, but they suffer from problems such as complex sample preparation, high cost, and long cycle time, making it difficult to meet the needs of large-scale and rapid screening.

[0003] Identification methods based on imaging technology have attracted attention due to their convenience and speed. However, most identification methods currently rely on hyperspectral imaging, which provides continuous and rich spectral information. However, the equipment is expensive and the data processing is complex. Although it can be deployed on drones, it lacks the ability to integrate the entire region's spectrum based on drone-related information. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for identifying soil mineral composition by drone photography.

[0005] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: to provide a method for identifying soil mineral composition by drone photography, comprising the following steps:

[0006] S1. Acquire several data points collected by the UAV in the target area, each data point corresponding to at least one sub-region in the target area, and the data points include at least multispectral image data of the surface of the target area and geographic information.

[0007] S2. Perform unified correction processing on the surface multispectral image data to generate standard multispectral data;

[0008] S3. Based on the geographic information in the collected data, the standard multispectral data of each collected data are integrated to generate standard multispectral data of the target area;

[0009] S4. Based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

[0010] Furthermore, the geographic information includes at least one or more of the following: UAV flight trajectory information, UAV flight altitude information, and the location information of the UAV when the data was collected.

[0011] Furthermore, the acquisition of soil mineral composition in the target area based on standard multispectral data of the target area includes:

[0012] S41. Input the standard multispectral data into the multispectral mapping model, and obtain the initial distribution result of mineral content based on pixel-by-pixel calculation;

[0013] S42. Based on the initial distribution results of mineral content, the multispectral mapping model is subjected to response consistency correction to generate optimized distribution results of mineral content;

[0014] S43. Based on the optimized distribution results of the mineral content, perform statistical analysis on the mineral types and their spatial distribution within the target area to generate the soil mineral composition of the target area.

[0015] Furthermore, the initial distribution result of mineral content obtained in step S41 based on pixel-by-pixel calculation includes:

[0016] For each pixel in the standard multispectral data, its reflectance values ​​in different bands are extracted to form a feature vector; the feature vector is input into the multispectral mapping model to obtain the predicted mineral content value at that pixel location; after traversing all pixels, the predicted values ​​at each location are arranged according to the spatial coordinates of the original image to form the initial distribution result of the mineral content in matrix form; during this process, invalid pixels caused by cloud shadows, water bodies, etc. are marked and excluded from the calculation.

[0017] Furthermore, the response consistency correction in step S42 includes:

[0018] S421. Based on the initial distribution results of the mineral content, the moving window method is used to calculate the statistical characteristics of the mineral content of each pixel and its surrounding neighboring pixels, including the local mean and standard deviation.

[0019] S422. Identify abnormal areas in the initial distribution results of mineral content where local statistical characteristics significantly conflict with global distribution patterns or prior geological knowledge; the judgment condition for significant conflict is that the local mean deviates from the global mean by more than three times the local standard deviation.

[0020] S423. For the identified abnormal regions, a distance-weighted inverse interpolation method is used to correct them based on the pixel values ​​of the spatially adjacent normal regions, generating a more smooth and continuous optimized distribution result of the mineral content in spatial distribution.

[0021] Furthermore, the statistical analysis in step S43 includes:

[0022] S431. Based on the numerical range of the optimized distribution results of the mineral content, the natural breakpoint method is used to divide it into multiple content level intervals.

[0023] S432. For each content level range, calculate its spatial distribution area, average content, and spatial aggregation index.

[0024] S433. Based on preset content thresholds and spatial aggregation thresholds, determine the location and range of mineral enrichment areas, and in conjunction with the laws of mineral symbiotic association, determine the main mineral types and their resource enrichment levels within the target area.

[0025] Further, step S4 obtains the soil mineral composition of the target area, which is further used for:

[0026] Generate a visual thematic map of mineral distribution, identifying enrichment areas for different mineral types;

[0027] Based on mineral content, enrichment area area and mining feasibility, the potential resource value of the target area is quantitatively assessed.

[0028] Integrating the assessment results with a geographic information system provides decision support for the classification and recycling of construction waste soil, transportation route planning, and site selection for resource utilization plants.

[0029] Furthermore, the step of integrating the standard multispectral data from each collected data point based on the geographic information in the collected data to generate standard multispectral data for the target area includes:

[0030] S31. Acquire UAV flight trajectory information and corresponding standard multispectral data;

[0031] S32. Based on the UAV flight trajectory, the standard multispectral data collected by different UAVs at different locations are fitted to generate standard multispectral data corresponding to the target area.

[0032] This invention also provides a system for identifying soil mineral composition by taking pictures with an unmanned aerial vehicle (UAV), comprising:

[0033] The drone module integrates a spectral image acquisition function, which is used to collect multispectral image data of the ground surface and geographic information over the target area;

[0034] The processing module is used to perform unified correction processing on the surface multispectral image data to generate standard multispectral data; and based on the geographic information in the collected data, integrate the standard multispectral data of each collected data to generate standard multispectral data of the target area; and based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

[0035] The present invention also provides a storage medium storing a computer program that can run on a processor and, when executed by the processor, can implement the above-described method.

[0036] Compared with existing technologies, this invention simplifies the acquisition of multispectral image data and reduces costs by using drones to identify soil mineral composition, and integrates the spectrum of the entire region based on drone-related information. Attached Figure Description

[0037] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0038] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0039] One technical solution adopted by this invention is to provide a method for identifying soil mineral composition by taking pictures with a drone, comprising the following steps:

[0040] S1. Acquire several data points collected by the UAV in the target area, each data point corresponding to at least one sub-region in the target area, and the data points include at least multispectral image data of the surface of the target area and geographic information.

[0041] S2. Perform unified correction processing on the surface multispectral image data to generate standard multispectral data;

[0042] S3. Based on the geographic information in the collected data, the standard multispectral data of each collected data are integrated to generate standard multispectral data of the target area;

[0043] S4. Based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

[0044] The UAV is equipped with multispectral acquisition equipment, specifically an imaging sensor containing multiple discrete band filters. The preset flight path is pre-set in the flight control software during the mission planning phase based on the boundary range of the target area and the required ground resolution (i.e., spatial resolution).

[0045] During the data acquisition process, the flight control system automatically adjusts the UAV's flight altitude according to the preset route through programmed control to maintain a constant scale of ground observation by the sensor, thereby ensuring that the generated sequence of images has a consistent spatial resolution.

[0046] Meanwhile, by precisely adjusting the attitude of the drone or the angle of the gimbal through programmed control, the imaging plane of the multispectral acquisition device is kept perpendicular to the target surface area or at a preset fixed tilt angle, thereby stabilizing the acquisition angle and ensuring that the imaging geometry is consistent at all acquisition times, avoiding distortion of ground reflectivity caused by differences in viewing angle.

[0047] During each exposure, the system records the solar elevation angle in real time using an integrated solar position sensor or by combining GPS time and location information. In subsequent unified correction processing, the recorded solar elevation angle information is used to perform radiation compensation calculations on the original image data using a radiative transfer model or empirical model. This eliminates differences in surface illumination radiation caused by variations in the sun's position throughout the day, ensuring that the final generated standard multispectral data are comparable in reflectance values.

[0048] Through the programmed control of acquisition height and angle, as well as the recording and compensation of solar altitude angle information, the acquired multispectral data are ensured to meet strict consistency requirements in three dimensions: spatial scale, geometric shape, and illumination radiation, providing a high-quality input foundation for subsequent model transfer and quantitative identification.

[0049] In addition, for each selected sampling area, two tasks are carried out simultaneously: first, hyperspectral data of the area is acquired using a ground-based or airborne hyperspectral imager, which has continuous band reflectance information with nanometer-level spectral resolution; second, representative rock or soil samples are collected in the area and sent to the laboratory. Using precise quantitative analysis methods such as X-ray diffraction analysis and inductively coupled plasma mass spectrometry, the target mineral content value corresponding to each sample is obtained, which is the laboratory's precise quantitative detection result.

[0050] The core of establishing a hyperspectral quantitative mapping model is correlation analysis.

[0051] The specific implementation process is as follows: First, the hyperspectral data obtained from each sampling area is preprocessed (such as radiometric calibration, atmospheric correction, and bad line repair), and the spectral curve corresponding to each sampling point is extracted, which is the feature vector composed of continuous band reflectance values. This feature vector is used as the set of independent variables (X matrix) for model input.

[0052] Simultaneously, the precise quantitative test results (mineral content percentage) of the corresponding sampling points in the laboratory are used as the dependent variable (Y matrix) of the model. Subsequently, partial least squares regression algorithm is used to model the X and Y matrices. This algorithm overcomes the multicollinearity problem among independent variables while effectively reducing dimensionality by finding the latent variables (principal components) in the X space that have the highest correlation with the Y space, thus establishing a robust mapping relationship from hyperspectral continuous band reflectance to mineral content, i.e., a quantitative prediction model.

[0053] The model ultimately outputs a mathematical function or a set of coefficients that can map the input hyperspectral reflectance feature vector to a continuous mineral content prediction value, thus providing a benchmark model with clear physical meaning for subsequent model transfer.

[0054] Optionally, the geographic information includes at least one or more of the following: UAV flight trajectory information, UAV flight altitude information, and the location information of the UAV when the data was collected.

[0055] The surface multispectral image data includes multiple spectral image frames. The UAV flight trajectory information corresponds to the trajectory of the UAV during the acquisition of the target area. Based on the UAV flight trajectory information, the specific position information of the UAV in the UAV flight trajectory corresponding to each spectral image frame is confirmed. The position information includes the horizontal position information and altitude information of the UAV when the spectral image frame was captured.

[0056] Specifically, the step of integrating the standard multispectral data from each collected data point based on the geographic information in the collected data to generate standard multispectral data for the target area includes:

[0057] S31. Acquire UAV flight trajectory information and corresponding standard multispectral data;

[0058] S32. Based on the UAV flight trajectory, the standard multispectral data collected by different UAVs at different locations are fitted to generate standard multispectral data corresponding to the target area.

[0059] Specifically, S32 includes:

[0060] S321. Determine the specific position information of the UAV in the UAV flight trajectory and the corresponding spectral image frame. Set a standard flight altitude value. Based on the standard flight altitude value and the altitude information of the UAV in each position information, perform scaling processing on each spectral image frame to obtain a corrected spectral image frame.

[0061] S322. Based on the horizontal position information of the UAV when it captures the spectral image frame, the specific position of the UAV in the horizontal position information is taken as the center of its corresponding spectral image frame. Each corrected spectral image frame is fitted and matched with the geographical coordinates of the target area to generate standard multispectral data corresponding to the target area.

[0062] In an optional embodiment of the present invention, S321 further includes:

[0063] S3211. Determine the specific location information of the UAV in the UAV flight trajectory and the corresponding spectral image frame, and set a standard flight altitude value;

[0064] S3212. Based on the distortion of the spectral image acquisition device carried by the UAV, set processing parameters and perform image processing on the acquired spectral image frames to eliminate image distortion.

[0065] S3213. Based on the standard flight altitude value and the altitude information of the UAV in each position information, a corrected spectral image frame is obtained by scaling each spectral image frame of the UAV.

[0066] Optionally, obtaining the soil mineral composition of the target area based on standard multispectral data of the target area includes:

[0067] S41. Input the standard multispectral data into the multispectral mapping model, and obtain the initial distribution result of mineral content based on pixel-by-pixel calculation;

[0068] S42. Based on the initial distribution results of mineral content, the multispectral mapping model is subjected to response consistency correction to generate optimized distribution results of mineral content;

[0069] S43. Based on the optimized distribution results of the mineral content, perform statistical analysis on the mineral types and their spatial distribution within the target area to generate the soil mineral composition of the target area.

[0070] Specifically, the initial distribution result of mineral content obtained in step S41 based on pixel-by-pixel calculation includes:

[0071] For each pixel in the standard multispectral data, its reflectance values ​​in different bands are extracted to form a feature vector; the feature vector is input into the multispectral mapping model to obtain the predicted mineral content value at that pixel location; after traversing all pixels, the predicted values ​​at each location are arranged according to the spatial coordinates of the original image to form the initial distribution result of the mineral content in matrix form; during this process, invalid pixels caused by cloud shadows, water bodies, etc. are marked and excluded from the calculation.

[0072] Specifically, the response consistency correction in step S42 includes:

[0073] S421. Based on the initial distribution results of the mineral content, the moving window method is used to calculate the statistical characteristics of the mineral content of each pixel and its surrounding neighboring pixels, including the local mean and standard deviation.

[0074] S422. Identify abnormal areas in the initial distribution results of mineral content where local statistical characteristics significantly conflict with global distribution patterns or prior geological knowledge; the judgment condition for significant conflict is that the local mean deviates from the global mean by more than three times the local standard deviation.

[0075] S423. For the identified abnormal regions, a distance-weighted inverse interpolation method is used to correct them based on the pixel values ​​of the spatially adjacent normal regions, generating a more smooth and continuous optimized distribution result of the mineral content in spatial distribution.

[0076] Specifically, the statistical analysis in step S43 includes:

[0077] S431. Based on the numerical range of the optimized distribution results of the mineral content, the natural breakpoint method is used to divide it into multiple content level intervals.

[0078] S432. For each content level range, calculate its spatial distribution area, average content, and spatial aggregation index.

[0079] S433. Based on preset content thresholds and spatial aggregation thresholds, determine the location and range of mineral enrichment areas, and in conjunction with the mineral symbiotic combination rules, determine the main mineral types and their resource enrichment levels within the target area.

[0080] Optionally, step S4, obtaining the soil mineral composition of the target area, is further used for:

[0081] Generate a visual thematic map of mineral distribution, identifying enrichment areas for different mineral types;

[0082] Based on mineral content, enrichment area area and mining feasibility, the potential resource value of the target area is quantitatively assessed.

[0083] Integrating the assessment results with a geographic information system provides decision support for the classification and recycling of construction waste soil, transportation route planning, and site selection for resource utilization plants.

[0084] In one embodiment of the present invention, the application of the regional mineral identification results generated in step S4 mainly includes three aspects:

[0085] Generate a thematic map of visualized mineral distribution:

[0086] Based on the statistical analysis conclusions obtained in step S4, namely the division of multiple content level intervals, the location and range of the determined mineral enrichment areas, and the inferred main mineral types and their resource enrichment levels, a visualized mineral distribution thematic map is generated in GIS software or a dedicated mapping platform.

[0087] In practice, the geographic coordinates of standard multispectral data are used as a spatial reference, and the two-dimensional matrix of the optimized mineral content distribution results (i.e., the mineral content value of each pixel) is used as the basic raster data. First, based on the content level intervals divided in step S41, different color gradients are assigned to different content levels (e.g., from low content to high content, the color transitions from cool to warm tones), and a level distribution map reflecting the continuous change of mineral content is generated by rendering.

[0088] Next, the spatial boundaries of each mineral enrichment zone identified in step S43 are overlaid on the grade distribution map and highlighted with prominent outlines (such as dashed or solid lines) or filled patterns. Finally, the map is clearly marked with legends, labels, or annotations to indicate the inferred main mineral type of each enrichment zone (such as "kaolinite enrichment zone" or "calcite enrichment zone") and the resource enrichment level assessed in step S43 (such as "high", "medium", or "low"). The resulting thematic map visually displays the spatial distribution pattern, enrichment level, and resource potential area of ​​different mineral types within the target region.

[0089] Quantitatively assess the potential resource value:

[0090] For each identified mineral enrichment area, a quantitative assessment of its potential resource value is conducted. The core parameters used in the assessment are directly derived from the calculation results of step S4:

[0091] Mineral content: The average content of the corresponding content grade range in this enrichment area, calculated using S42. , which serves as a core indicator for characterizing the mineral abundance of the region.

[0092] Area of ​​enriched region: The spatial distribution area of ​​the enriched region corresponding to the content level range calculated in step S42. .

[0093] Based on the above two basic parameters, and combined with the feasibility of mining (such as the type of surface cover, terrain slope, accessibility and other external constraints, which need to be obtained through additional on-site surveys or geographical data, but the evaluation framework itself does not introduce new data types, but only as comprehensive consideration factors), a comprehensive "resource value index" is constructed for quantitative evaluation.

[0094] One feasible quantification method is:

[0095]

[0096] in, It is a weighting coefficient assigned based on the comprehensive conditions of mining feasibility (which can be simplified into three levels: high, medium, and low). (For example, high feasibility is...) , middle is , low This index comprehensively reflects the total resource volume of a specific mineral-rich area (compared to...). and The resource value index is calculated based on the relative availability and ease of resource utilization. Different resource-rich areas can be ranked and classified according to their calculated value, providing a quantitative basis for decisions on priority resource development and utilization.

[0097] Integrate with geographic information systems and provide decision support:

[0098] The generated visualized mineral distribution thematic maps and resource value assessment results for each enrichment area are imported into a Geographic Information System (GIS) as thematic layers. Within the GIS platform, this information can be overlaid and analyzed with existing basic geographic data (such as road networks, settlements, water systems, and land use maps) and engineering geological data, providing direct decision support for the refined management and resource utilization of construction waste soil. Specifically, this includes:

[0099] Support for classified recycling: Based on the spatial location of mineral-rich areas and their corresponding main mineral types, combined with soil type maps, it can guide the source classification and targeted collection of soils containing minerals of different values ​​(such as specific clay minerals and carbonate minerals) during building demolition or land reclamation, thereby improving the efficiency and purity of subsequent resource utilization.

[0100] Support for transportation route planning: Based on the location of each enrichment area, the assessed resource value, and road network data, GIS can perform optimal route analysis to plan efficient and economical transportation routes from multiple enrichment areas (as raw material collection points) to pre-designated resource utilization plant sites or temporary storage yards, thereby optimizing logistics costs.

[0101] Provide decision support for the site selection of resource utilization plants: By combining the spatial distribution of various rich areas, resource quantity assessment results, surrounding environmentally sensitive areas, infrastructure (water, electricity, transportation) conditions, land use planning and other multi-source spatial information, suitable sites for the construction of resource utilization facilities (such as soil conditioner production plants and aggregate processing plants) are selected through multi-criteria evaluation and analysis in GIS, so as to realize the on-site conversion of resources, reduce transportation costs and improve overall efficiency.

[0102] Through the specific implementation of the above three aspects, the mineral identification results generated in step S4 will be effectively transformed from information level into spatial management solutions and quantitative decision-making basis that can directly guide production practices.

[0103] This invention also provides a system for identifying soil mineral composition by taking pictures with an unmanned aerial vehicle (UAV), comprising:

[0104] The drone module integrates a spectral image acquisition function, which is used to collect multispectral image data of the ground surface and geographic information over the target area;

[0105] The processing module is used to perform unified correction processing on the surface multispectral image data to generate standard multispectral data; and based on the geographic information in the collected data, integrate the standard multispectral data of each collected data to generate standard multispectral data of the target area; and based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

[0106] The drone module includes a drone and a drone controller. The drone is equipped with a spectral image acquisition device, which is used to collect spectral data of the soil below the drone during flight.

[0107] The processing module is specifically a server with a multispectral mapping model installed, used to process the data collected by the drone to obtain soil composition.

[0108] The present invention also provides a storage medium storing a computer program that can run on a processor and, when executed by the processor, can implement the above-described method.

[0109] When the storage medium is running, the computer program on it can implement the above-described embodiments.

[0110] The above are merely embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for identifying soil mineral composition using drone photography, characterized in that: Includes the following steps: S1. Acquire several data points collected by the UAV in the target area, each data point corresponding to at least one sub-region in the target area, and the data points include at least multispectral image data of the surface of the target area and geographic information. S2. Perform unified correction processing on the surface multispectral image data to generate standard multispectral data; S3. Based on the geographic information in the collected data, the standard multispectral data of each collected data are integrated to generate standard multispectral data of the target area; S4. Based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

2. The method for identifying soil mineral composition by drone photography according to claim 1, characterized in that: The geographic information includes at least one or more of the following: UAV flight trajectory information, UAV flight altitude information, and the location information of the UAV when the data was collected.

3. The method for identifying soil mineral composition by drone photography according to claim 1, characterized in that: The soil mineral composition of the target area is obtained based on standard multispectral data of the target area, including: S41. Input the standard multispectral data into the multispectral mapping model, and obtain the initial distribution result of mineral content based on pixel-by-pixel calculation; S42. Based on the initial distribution results of mineral content, the multispectral mapping model is subjected to response consistency correction to generate optimized distribution results of mineral content; S43. Based on the optimized distribution results of the mineral content, perform statistical analysis on the mineral types and their spatial distribution within the target area to generate the soil mineral composition of the target area.

4. The method for identifying soil mineral composition by drone photography according to claim 3, characterized in that: The initial distribution of mineral content obtained in step S41 based on pixel-by-pixel calculation includes: For each pixel in the standard multispectral data, its reflectance values ​​in different bands are extracted to form a feature vector; the feature vector is input into the multispectral mapping model to obtain the predicted mineral content value at that pixel location; after traversing all pixels, the predicted values ​​at each location are arranged according to the spatial coordinates of the original image to form the initial distribution result of the mineral content in matrix form; during this process, invalid pixels caused by cloud shadows, water bodies, etc. are marked and excluded from the calculation.

5. The method for identifying soil mineral composition by drone photography according to claim 4, characterized in that: The response consistency correction in step S42 includes: S421. Based on the initial distribution results of the mineral content, the moving window method is used to calculate the statistical characteristics of the mineral content of each pixel and its surrounding neighboring pixels, including the local mean and standard deviation. S422. Identify abnormal areas in the initial distribution results of mineral content where local statistical characteristics significantly conflict with global distribution patterns or prior geological knowledge; the judgment condition for significant conflict is that the local mean deviates from the global mean by more than three times the local standard deviation. S423. For the identified abnormal regions, a distance-weighted inverse interpolation method is used to correct them based on the pixel values ​​of the spatially adjacent normal regions, generating a more smooth and continuous optimized distribution result of the mineral content in spatial distribution.

6. The method for identifying soil mineral composition by drone photography according to claim 5, characterized in that: The statistical analysis in step S43 includes: S431. Based on the numerical range of the optimized distribution results of the mineral content, the natural breakpoint method is used to divide it into multiple content level intervals. S432. For each content level range, calculate its spatial distribution area, average content, and spatial aggregation index. S433. Based on preset content thresholds and spatial aggregation thresholds, determine the location and range of mineral enrichment areas, and in conjunction with the mineral symbiotic combination rules, determine the main mineral types and their resource enrichment levels within the target area.

7. The method for identifying soil mineral composition by drone photography according to claim 1, characterized in that: Step S4 obtains the soil mineral composition of the target area, which is further used for: Generate a visual thematic map of mineral distribution, identifying enrichment areas for different mineral types; Based on mineral content, enrichment area area and mining feasibility, the potential resource value of the target area is quantitatively assessed. Integrating the assessment results with a geographic information system provides decision support for the classification and recycling of construction waste soil, transportation route planning, and site selection for resource utilization plants.

8. The method for identifying soil mineral composition by drone photography according to claim 2, characterized in that: The step of integrating the standard multispectral data from each collected data point based on the geographic information in the collected data to generate standard multispectral data for the target area includes: S31. Acquire UAV flight trajectory information and corresponding standard multispectral data; S32. Based on the UAV flight trajectory, the standard multispectral data collected by different UAVs at different locations are fitted to generate standard multispectral data corresponding to the target area.

9. A system for identifying soil mineral composition by taking pictures with an unmanned aerial vehicle (UAV), characterized in that: The system includes: The drone module integrates a spectral image acquisition function, which is used to collect multispectral image data of the ground surface and geographic information over the target area; The processing module is used to perform unified correction processing on the surface multispectral image data to generate standard multispectral data; and based on the geographic information in the collected data, integrate the standard multispectral data of each collected data to generate standard multispectral data of the target area; and based on the standard multispectral data of the target area, obtain the soil mineral composition of the target area.

10. A storage medium, characterized in that, The storage medium stores a computer program that can run on a processor, characterized in that, when the computer program is executed by the processor, it can implement the method as claimed in any one of claims 1 to 9.

Images