Mineral exploitation provides essential material and energy resources for socio-economic development, but also affects the surrounding ecological environment of mining areas. Evaluating the ecological environmental quality in mining areas is vital to balance resource development and ecological environmental protection. This study attempts to review existing practices of ecological environmental quality evaluation in mining areas using remote sensing in terms of 1) requirements for ecological environmental quality evaluation in mining areas in relevant national laws, regulations and standards, 2) research progress in ecological environmental quality evaluation in mining areas based on remote sensing, 3) existing research gaps of ecological environmental quality evaluation in mining areas using remote sensing and their implications for future research. We found that existing studies have made progress in indicator acquisition, establishment and improvement of evaluation models, yet are still limited in 1) the acquisition capacity and accuracy of remote sensing indicators, including difficulties in acquiring information about the subsurface environment in mining areas, insufficient temporal and spatial resolution of observations, and low accuracy of models for monitoring remote sensing parameters, 2) the generalization ability of existing evaluation models by remote sensing, including low applicability of indicators in different mining areas, excessively complex model implementation and difficulties in automation. Potential research opportunities include expanding accessible indicators by remote sensing, constructing a framework for integrated data collection on the surface and underground, improving the data quality of remote sensing indicators, constructing a new indicator system for evaluation using remote sensing, and developing cloud computing algorithms.

To practice the development concept of "lucid waters and lush mountains are invaluable assets", and balance mineral development and ecological environment protection, it is urgent to carry out dynamic monitoring and scientific evaluation of the ecological environment in mining areas.Based on the characteristics of the ecological environment in mining areas, this paper analyzes the temporal and spatial change characteristics and differences of the ecological environment elements in the mining area, the impact mechanism of the mining and restoration activities on the ecological environment elements, and the cooperative evolution law of each element.Guided by the requirements on ecological environment monitoring and evaluation in mining areas in the new era, a technology framework for quantitative remote sensing-based monitoring and evaluation of the ecological environment in mining areas is proposed from the perspective of 'data-monitoring-evaluation-application'.The framework makes full use of multi-source big data in mining areas with remote sensing images as the main body and takes advantage of the emerging technologies such as artificial intelligence and quantitative remote sensing.This research archives quantitative monitoring and evaluation of the ecological environment elements in the mining areas with the features of high frequency, large-extent, long-term, continuous, all-factor observation and quantitative inversion, which can support the operational applications including mining activity monitoring, ecological environment diagnosis and early warning and restoration effect evaluation in mining areas.Finally, two real-world cases are introduced to illustrate the effectiveness and application flow of the proposed framework.

To enhance the accuracy of land use classification in mining areas, the Object-based Convolutional Neural Network (OCNN) method has been widely used. However, existing researches tend to neglect the importance of decision-level fusion, focusing only on feature-level fusion. This study proposes a new classification framework with Multi-level Fusion of Object-based analysis and CNN (MFOCNN) to achieves high-accuracy land use classification in mining areas. First, Simple Linear Iterative Cluster (SLIC) is employed to generate image objects, which serve as the basic unit for classification. Second, an improved DenseNet is proposed to extract deep features from image patches, which represent image objects, and provide the classification result. Third, handcrafted features including spectral, textural and geometric of the image objects are extracted and fused with the deep features to obtain the classification result with random forest classifier. Finally, the Dempster-Shafer (DS) evidence theory is applied to fuse the two previously described classification results at the decision-level to obtain the final result. Experiments conducted in the mining areas of Erdos using Gaofen-6 images demonstrate that the proposed MFOCNN achieves the best visual performance and accuracy among all tested methods. The MFOCNN, with its feature-level fusion and decision-level fusion, significantly improves the accuracy of land use classification in mining areas. The results suggest that the proposed MFOCNN is a promising method for achieving high-accuracy land use classification in mining areas.