With the advent of high-spatial resolution (HSR) satellite imagery, urban land use/land cover (LULC) mapping has become one of the most popular applications in remote sensing. Due to the importance of context information (e.g., size/shape/texture) for classifying urban LULC features, Geographic Object-Based Image Analysis (GEOBIA) techniques are commonly employed for mapping urban areas. Regardless of adopting a pixel- or object-based framework, the selection of a suitable classifier is of critical importance for urban mapping. The popularity of deep learning (DL) (or deep neural networks (DNNs)) for image classification has recently skyrocketed, but it is still arguable if, or to what extent, DL methods can outperform other state-of-the art ensemble and/or Support Vector Machines (SVM) algorithms in the context of urban LULC classification using GEOBIA. In this study, we carried out an experimental comparison among different architectures of DNNs (i.e., regular deep multilayer perceptron (MLP), regular autoencoder (RAE), sparse, autoencoder (SAE), variational autoencoder (AE), convolutional neural networks (CNN)), common ensemble algorithms (Random Forests (RF), Bagging Trees (BT), Gradient Boosting Trees (GB), and Extreme Gradient Boosting (XGB)), and SVM to investigate their potential for urban mapping using a GEOBIA approach. We tested the classifiers on two RS images (with spatial resolutions of 30 cm and 50 cm). Based on our experiments, we drew three main conclusions: First, we found that the MLP model was the most accurate classifier. Second, unsupervised pretraining with the use of autoencoders led to no improvement in the classification result. In addition, the small difference in the classification accuracies of MLP from those of other models like SVM, GB, and XGB classifiers demonstrated that other state-of-the-art machine learning classifiers are still versatile enough to handle mapping of complex landscapes. Finally, the experiments showed that the integration of CNN and GEOBIA could not lead to more accurate results than the other classifiers applied.

Multi-objective land allocation (MOLA) can be regarded as a spatial optimization problem that allocates appropriate use to certain land units subjecting to multiple objectives and constraints. This article develops an improved knowledge-informed non-dominated sorting genetic algorithm II (NSGA-II) for solving the MOLA problem by integrating the patch-based, edge growing/decreasing, neighborhood, and constraint steering rules. By applying both the classical and the knowledge-informed NSGA-II to a simulated planning area of 30 × 30 grid, we find that: when compared to the classical NSGA-II, the knowledge-informed NSGA-II consistently produces solutions much closer to the true Pareto front within shorter computation time without sacrificing the solution diversity; the knowledge-informed NSGA-II is more effective and more efficient in encouraging compact land allocation; the solutions produced by the knowledge-informed have less scattered/isolated land units and provide a good compromise between construction sprawl and conservation land protection. The better performance proves that knowledge-informed NSGA-II is a more reasonable and desirable approach in the planning context.

The monthly ground-level ozone (O₃) concentrations in eastern China (EC) are estimated using the surface NO₂ and CH₂O concentrations derived from the ozone monitoring instrument (OMI) according to the geographically weighted regression (GWR) method. The accuracy assessment shows that the proposed GWR model performed well for estimating monthly O₃ concentrations (R² = 0.81, absolute error (AE) = 7.38 μg/m³ for the fitted results, and R ²= 0.77, AE = 8.20 μg/m³). The estimated yearly average O₃ concentration is 56.20 μg/m³, with a standard deviation of 3.68 μg/m³ in 2014 across EC. The O₃ concentrations show great spatial variation, with hotspots extending from north-EC to the Yangtze River Delta and the Pearl River Delta. The monthly O₃ concentrations are generally low during the cold season and high during the warm season, but local areas show different seasonal variations.