Land-use maps are instrumental to inform urban planning and environmental research. Street view images (SVIs) have shown great potential for automated land-use classification for land-use mapping. However, previous studies overlooked SVI-derived location contextual information that may help improve land-use classification. This study proposes a novel land-use classification method that synthesizes location semantics from SVIs to account for contextual information from SVIs, land parcels and roads around the SVIs. The proposed method first generates land-use scene images (LUSIs) by using an SVI-derived straightforward algorithm. The LUSIs are then relocated to land parcels by using a displacement strategy and classified into land-use types by using a deep learning network. This study determines the land-use types of land parcels with classified LUSIs. Two case studies, consisting of LUSIs for five land-use types, show that introducing location semantics of SVIs can remarkably improve the classification accuracy of land-use types.

Severe air pollution has become a major risk to human health from a global environmental perspective. It has been recognized that human mobility is an essential component in individual exposure assessment. Activity structure reflects the characteristics of human mobility. Thus, a better understanding of the relationship between human activity structure and individual exposure level is of crucial relevance. This study examines this relationship using a large cell-phone GPS dataset in Wuhan, China. The results indicate that there is a strong linear relationship between people’s activity structures and exposures to PM_2.5. Inter-group comparisons based on the four activity structure groups obtained with K-means clustering found that groups with different activity structures do experience different levels of PM_2.5 exposure. Furthermore, differences in detailed characteristics of activity structure were also found at different exposure levels at the intra-group level. These results show that people’s activity structures do influence their exposure levels. The paper provides a new perspective for understanding individual exposure through human activity structure, which helps move the perspective of research on individual exposure from the semantic of physical location to the semantic of human activity pattern.

The primary objective of vectorial road network matching is to identify homonymous roads from two different data sources. Previous methods usually focus on matching road networks with the same coordinate system but rarely with different or unknown coordinate systems, which may lead to nontrivial and nonsystematic deviations (e.g., rotation angle) between homonymous objects. To fill this gap, this study proposes a novel hierarchical road network matching method based on Delaunay triangulation (DTRM). First, the entire urban road network is divided into three levels (L1, L2, L3) by using the principle of stroke. Then, the triangular meshes are constructed from L2, and the minimum matching unit (MMU) in the triangular mesh is used instead of the traditional “node-arc” unit to measure the similarity for the matching of L2. Lastly, a hierarchical matching solution integrating the probabilistic relaxation method and MMU similarity is yielded to identify the matching relationships of the three-level road network. Experiments conducted in Wuhan, China, and Auckland, New Zealand, show that the MMU similarity metrics can effectively calculate the similarity value with different rotation angles, and DTRM has higher precision than the benchmark probability-relaxation-matching method (PRM) and can correctly identify the most matching-relationships with an average accuracy of 89.63%. This study provides a matching framework for road networks with different or even unknown coordinate systems and contributes to the integration and updating of urban road networks.

Community roads are crucial to community navigation. There are automatic methods to obtain community roads using trajectories, but the sparsity and uneven density distribution of community trajectories present significant challenges in identifying community roads. To overcome these challenges, we propose a conditional generative adversarial network (MAC-GAN) supervised by pedestrian trajectories and neighborhood building footprints for road generation. MAC-GAN packs the “road trajectory–building footprint” pairs into images to characterize implicit ternary relations and sets up a multi-scale skip-connected and asymmetric convolution-based generator to incorporate such a relationship, in which the generator and discriminator mutually learn to optimize the network parameters and then derive approximate optimal results. Experiments on 37 real-world community datasets in Wuhan, China, are conducted to verify the effectiveness of the proposed model. The experimental results show that the F1 score of our model increases by 1.7–6.8%, and the IOU of our model increases by 2.2–7.5% compared with three baselines (i.e., Pix2pix, GANmapper, and DLinkGAN (configured by DLinknet)). In areas with sparse and missing trajectory data, the generated fine roads have high accuracy with the supervision of building footprints.