Urban Land Use Mapping Using Deep Learning

Dear Colleagues,

Urban Land Use Mapping (ULUM) plays a pivotal role in reconciling cities’ dual identities as engines of economic growth and focal points of socio-environmental challenges. Its applications span 3D urban modeling, building height estimation, point cloud semantic/instance segmentation, traffic pattern analysis, land use change detection, ecological–environmental monitoring, local climate zoning, nightlight-based economic mapping, and fine-grained functional zoning, all of which are critical for sustainable urban governance. However, conventional remote sensing approaches face inherent limitations in ULUM due to spectral ambiguities (e.g., same-object-different-spectra phenomena in mixed-use areas) and resolution constraints. Crucially, purely spectral data lack human activity signatures—despite land use being intrinsically defined by spatio-temporal socio-economic behaviors—necessitating the integration of multi-modal social sensing data (open geospatial layers, crowdsourced OSM/Google/Baidu maps, street views, mobile signaling, social media, Wi-Fi hotspots, and taxi trajectories).

Advancements in deep learning (DL), particularly foundation models like SAM (universal image segmentation), GPT (cross-modal reasoning), and Deepseek (multi-task optimization), are revolutionizing ULUM by enabling robust urban object recognition and multi-source heterogeneous data fusion (e.g., SAR–optical–social alignment). These innovations address persistent gaps in scalability and semantic granularity while reducing reliance on labor-intensive annotations. Looking ahead, ULUM will increasingly adopt automated foundation model frameworks to support dynamic applications such as real-time zoning, emission tracking, and disaster resilience modeling—bridging AI-driven methodologies with urban sustainability imperatives.

This Special Issue aims to advance ULUM by leveraging deep learning to address critical challenges in spectral ambiguity, multi-scale dynamics, and human activity integration. It focuses on synergizing multi-modal remote sensing (optical, SAR, LiDAR) with social sensing (mobile signaling, geosocial data, trajectories) to achieve high-precision, dynamic, and interpretable ULUM. Key applications include 3D/4D urban modeling, emission tracking (PM2.5, carbon), local climate zoning, and disaster resilience assessment, aligning with the journal’s emphasis on sensor-driven innovations and sustainable urban solutions.

This Special Issue invites submissions that synergize multi-modal data, advanced algorithms, and urban-specific applications. Topics include, but are not limited to, the following:

• SAM-enhanced urban object segmentation with LiDAR–optical fusion;
• Weakly supervised ULUM using crowdsourced OSM and SAR data;
• CLIP-driven cross-modal alignment for social media–street view fusion;
• Domain adaptation between satellite imagery and mobile signaling data;
• Self-supervised pretraining for sparse UAV–thermal urban datasets;
• Knowledge graph-guided fusion of taxi trajectories and InSAR data;
• Reinforcement learning for dynamic traffic-emission mapping;
• Generative models for synthetic multi-sensor urban datasets;
• Lightweight SAM variants for edge-device urban 3D mapping;
• Multi-temporal InSAR-NDVI fusion for soil erosion monitoring;
• PM2.5 mapping via satellite–social media multimodal fusion;
• Local climate zoning using SAR–thermal–UAV data integration;
• Wi-Fi hotspot clustering for mixed land-use pattern analysis;
• OSM-GPT fusion for informal settlement identification;
• Nightlight–VHR data fusion for economic activity zoning;
• Three-dimensional building height mapping with LiDAR–street view alignment;
• Urban flood risk modeling via crowdsourced–geospatial fusion;
• Urban man-made soil erosion mapping via multimodal data;
• Fine-scale green space mapping using street view–social sensing;
• Taxi trajectory mining for carbon emission hotspot detection;
• SAR–optical fusion for cross-city LU change detection;
• Open multi-modal ULUM datasets (SAR, trajectories, OSM);
• Benchmarks for urban 3D point cloud semantic segmentation.

Prof. Dr. Chang Li
Dr. Rongjun Qin
Prof. Dr. Ruisheng Wang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• urban land use mapping
• deep learning
• multi-modal data integration
• knowledge graphs
• remote sensing
• urban environmental monitoring
• semantic segmentation
• three-dimensional urban modeling
• dynamic urban change detection