The remote sensing classification of urban functional zones provides scientific support for urban planning, land resource optimization, and ecological environment protection. However, urban functional zone classification encounters significant challenges in accuracy and efficiency due to complicated image structures, ambiguous critical features, and high computational complexity. To tackle these challenges, this work proposes a novel XT-SECA algorithm employing a strengthened efficient channel attention mechanism (SECA) to integrate the feature-extraction XGBoost branch and the feature-enhancement Transformer feedforward branch. The SECA optimizes the feature-fusion process through dynamic pooling and adaptive convolution kernel strategies, reducing feature confusion between various functional zones. XT-SECA is characterized by sufficient learning of complex image structures, effective representation of significant features, and efficient computational performance. The Futian, Luohu, and Nanshan districts in Shenzhen City are selected to conduct urban functional zone classification by XT-SECA, and they feature administrative management, technological innovation, and commercial finance functions, respectively. XT-SECA can effectively distinguish diverse functional zones such as residential zones and public management and service zones, which are easily confused by current mainstream algorithms. Compared with the commonly adopted algorithms for urban functional zone classification, including Random Forest (RF), Long Short-Term Memory (LSTM) network, and Multi-Layer Perceptron (MLP), XT-SECA demonstrates significant advantages in terms of overall accuracy, precision, recall, F1-score, and Kappa coefficient, with an accuracy enhancement of 3.78%, 42.86%, and 44.17%, respectively. The Kappa coefficient is increased by 4.53%, 51.28%, and 52.73%, respectively.

Mine exploitation and recovery reflect environmental destruction and remediation, and mine change detection is critical to environmental sustainability and management. Numerous mines are distributed worldwide and exhibit complicated variations; thus, there is an urgent need to develop automatic detection of mine changes. This work proposes a novel three-dimensional (3D) multi-scale attention-based full convolutional neural network (FCNN), named as 3D MSA-UNet, for mine exploitation and regreenization detection. The new neural network features 3 advantages. (1) It constructs 3D filters in the FCNN of U-Net; thus, the spatial and temporal changing features can be synchronously extracted, and the mine changes during different periods can be simultaneously and automatically detected. (2) It establishes a multi-scale convolutional layer to capture abundant global and local features and achieves a good generalization ability when applying in multiple mining regions. (3) It introduces an attention mechanism of Convolutional Block Attention Module (CBAM) to learn significant and useful feature representation and to catch long-range and comprehensive feature dependency. Thus, the detection accuracy can be effectively improved. Therefore, due to the obvious advantages in spatiotemporal synchronization, multiscale useful feature learning, and good generalization, the suggested 3D MSA-UNet can relatively accurately reveal the progress of mine exploitation and regreenization. The novel network is applied to 3 large-scale mining areas, and the overall accuracy of change detection attains 97.47, 97.12, and 90.25%, respectively. Moreover, 3D MSA-UNet outperforms the state-of-the-art algorithms of U-Net, convolutional neural network (CNN), patch-based 3D CNN, XGBoost, and random forest. The novel network provides a technical support for the rational development and planning of mines and environment.