Mining area land utilization classification method based on unmanned aerial vehicle photogrammetry

Abstract

The invention discloses a mining area land utilization classification method based on unmanned aerial vehicle photogrammetry, and the method comprises the steps: integrating an unmanned aerial vehicle photogrammetry technology, data dimension reduction processing, multi-scale image segmentation, support vector machine classification, particle swarm optimization and other advanced technical means; the invention provides an efficient and accurate dynamic monitoring method for land utilization of a surface mining area. The beneficial effects of the invention are that the method overcomes the defects of the prior art, improves the timeliness and accuracy of monitoring, and contributes to the sustainable development and environmental protection of mining areas.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering survey, and in particular to a mining area land use classification method based on unmanned aerial vehicle photogrammetry. Background Art

[0002] At present, dynamic monitoring of land use in surface mining areas is an important part of resource management and environmental protection. Its accuracy and timeliness directly affect the ecological restoration of mining areas, resource planning, and the formulation and implementation of environmental protection policies. Traditional monitoring methods mostly rely on manual field surveys, which are not only time-consuming and labor-intensive, but also difficult to achieve large-scale and high-frequency monitoring. In recent years, with the rapid development of remote sensing technology, especially the widespread application of unmanned aerial vehicle photogrammetry, new technical means have been provided for the dynamic monitoring of land use in mining areas. However, there are still some problems with existing remote sensing monitoring technology, such as high complexity of data processing, serious redundancy of multi-period data, and difficulty in optimizing classification model parameters, which limit the accuracy and timeliness of monitoring results. Summary of the invention

[0003] The purpose of the present invention is to provide an efficient and accurate method for dynamic monitoring of land use in surface mining areas, aiming to solve the problems of difficult data collection, high processing complexity, serious multi-period data redundancy and difficulty in optimizing classification model parameters in current mining area land use monitoring. By integrating advanced technical means such as drone photogrammetry technology, data dimensionality reduction processing, multi-scale image segmentation, support vector machine classification and particle swarm algorithm optimization, an efficient and accurate method for dynamic monitoring of land use in surface mining areas is provided to overcome the shortcomings of existing technologies, improve the timeliness and accuracy of monitoring, and contribute to the sustainable development and environmental protection of mining areas.

[0004] The technical solution of the present invention is as follows: A mining area land use classification method based on drone photogrammetry comprises the following steps: (1) Data collection: Using drone photogrammetry, we obtain aerial photos and corresponding POS data of the mining area in four periods of each quarter of the year. We generate digital orthophotos (DOM) and digital surface models (DSM) of the surface mining area through aerial triangulation software, ensuring that the data coordinate system is CGCS2000, providing a high-quality data basis for subsequent analysis. (2) Sample selection and division: Based on digital topographic maps of previous years, historical image data, and field surveys, we selected classified samples of various land use types, including grassland, woodland, impervious surface, and water bodies, and randomly divided the samples into training and validation sets to prepare for model training and validation; (3) Data preprocessing: Calculate the spectral index of the RGB channels of the DOM data and synthesize the analysis results with the original DOM and DSM data; (4) Image segmentation: Multi-scale image segmentation technology is used to segment the preprocessed image into multiple objects, providing accurate object boundaries for subsequent classification model construction; (5) Model construction and optimization: A mining area land use classification model was constructed based on support vector machine (SVM). The radial basis function was selected as the kernel function and initialized by randomly introducing default hyperparameters. The SVM hyperparameters were then optimized using the particle swarm algorithm. The model accuracy was evaluated using validation set samples to ensure model performance. (6) Generation of classification results: When the optimized model reaches the predetermined accuracy requirement, it is combined with the image segmentation results to generate the land use classification results in the form of a raster for the entire surface mining area, thus achieving fast and accurate dynamic monitoring of land use.

[0005] Step (2) described in the above scheme utilizes the minimum noise separation technology to achieve data dimensionality reduction, reduce data redundancy, and improve the efficiency of subsequent analysis.

[0006] The beneficial effect of the present invention is that it significantly improves the accuracy and timeliness of dynamic monitoring of land use in surface mining areas by integrating technical means such as unmanned aerial vehicle photogrammetry, minimum noise separation, multi-scale image segmentation, support vector machine classification and particle swarm algorithm optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a technical flow chart of the present invention. DETAILED DESCRIPTION

[0008] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in 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.

[0009] like Figure 1 As shown, the present invention proposes a high-resolution dynamic monitoring method for land use in surface mining areas based on UAV photogrammetry and multi-period long-sequence remote sensing data, aiming to achieve efficient and accurate monitoring through the following steps: (1) Data collection: Using drone photogrammetry, we obtain aerial photos and corresponding POS data of the mining area in four periods of each quarter of the year. We generate digital orthophotos (DOM) and digital surface models (DSM) of the surface mining area through aerial triangulation software, ensuring that the data coordinate system is CGCS2000, providing a high-quality data basis for subsequent analysis. (2) Sample selection and division: Based on digital topographic maps of previous years, historical image data, and field surveys, we selected classified samples of various land use types, including grassland, woodland, impervious surface, and water bodies, and randomly divided the samples into training and validation sets to prepare for model training and validation; (3) Data preprocessing: Calculate the RGB channel spectral index of DOM data, synthesize the analysis results with the original DOM and DSM data, and use the minimum noise separation technology to achieve data dimensionality reduction, reduce data redundancy, and improve the efficiency of subsequent analysis; (4) Image segmentation: Multi-scale image segmentation technology is used to segment the preprocessed image into multiple objects, providing accurate object boundaries for subsequent classification model construction; (5) Model construction and optimization: A mining area land use classification model was constructed based on support vector machine (SVM). The radial basis function was selected as the kernel function and initialized by randomly introducing default hyperparameters. The SVM hyperparameters were then optimized using the particle swarm algorithm. The model accuracy was evaluated using validation set samples to ensure model performance. (6) Generation of classification results: When the optimized model reaches the predetermined accuracy requirement, it is combined with the image segmentation results to generate the land use classification results in the form of a raster for the entire surface mining area, thus achieving fast and accurate dynamic monitoring of land use.

[0010] The present invention significantly improves the accuracy and timeliness of dynamic monitoring of land use in surface mining areas by integrating unmanned aerial vehicle photogrammetry, minimum noise separation, multi-scale image segmentation, support vector machine classification and particle swarm algorithm optimization. Specifically: Reduce data redundancy: Use minimum noise separation technology to effectively reduce data redundancy of multi-period orthophotos and improve data processing efficiency.

[0011] Optimize model parameters: Automatically optimize SVM hyperparameters through particle swarm algorithm, which improves the performance and generalization ability of the classification model.

[0012] Rapid classification: A land use classification model for surface mining areas based on SVM was established, which achieved rapid and accurate classification of land use in the entire area and generated raster products that are easy to interpret.

[0013] Enhanced practicality: The entire monitoring process is highly automated, which reduces manual intervention and monitoring costs, and provides strong technical support for mining resource management and environmental protection.

[0014] In summary, the present invention has significant advantages in improving the accuracy and timeliness of dynamic monitoring of land use in surface mining areas, and has broad application prospects and social and economic benefits.

[0015] In the description of this specification, the description with reference to the term "embodiment" means that the specific features or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment. Moreover, the specific features or characteristics described can be combined in any one or more embodiments in a suitable manner.

[0016] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A mining area land use classification method based on drone photogrammetry, characterized by: The following steps are involved: (1) Data collection: Using drone photogrammetry, we obtain aerial photos and corresponding POS data of the mining area in four periods of each quarter of the year. We generate digital orthophotos (DOM) and digital surface models (DSM) of the surface mining area through aerial triangulation software, ensuring that the data coordinate system is CGCS2000, providing a high-quality data basis for subsequent analysis. (2) Sample selection and division: Based on digital topographic maps of previous years, historical image data, and field surveys, we selected classified samples of various land use types, including grassland, woodland, impervious surface, and water bodies, and randomly divided the samples into training and validation sets to prepare for model training and validation; (3) Data preprocessing: Calculate the spectral index of the RGB channels of the DOM data and synthesize the analysis results with the original DOM and DSM data; (4) Image segmentation: Multi-scale image segmentation technology is used to segment the preprocessed image into multiple objects, providing accurate object boundaries for subsequent classification model construction; (5) Model construction and optimization: A mining area land use classification model was constructed based on support vector machine (SVM). The radial basis function was selected as the kernel function and initialized by randomly introducing default hyperparameters. The SVM hyperparameters were then optimized using the particle swarm algorithm. The model accuracy was evaluated using validation set samples to ensure model performance. (6) Generation of classification results: When the optimized model reaches the predetermined accuracy requirement, it is combined with the image segmentation results to generate the land use classification results in the form of a raster for the entire surface mining area, thus achieving fast and accurate dynamic monitoring of land use.

2. The mining area land use classification method based on drone photogrammetry according to claim 1 is characterized in that: The step (2) uses the minimum noise separation technology to achieve data dimension reduction, reduce data redundancy, and improve the efficiency of subsequent analysis.

Images