Smart City Land Use Change Detection Using Multisource Remote Sensing and GNSS-Enriched GIS Framework

Project Abstract

The rapid pace of urban expansion in contemporary cities demands accurate, timely, and scalable approaches to monitor land use changes and inform planning decisions. This study presents a comprehensive framework that integrates multisource remote sensing data with GNSS-enriched GIS to detect, classify, and analyze land use change patterns in smart city environments. We combine high-resolution optical imagery, synthetic aperture radar (SAR), and LiDAR-derived terrain information with GNSS-based positioning to improve geometric fidelity, temporal consistency, and thematic accuracy across multi-temporal datasets. A multi-stage methodological pipeline is developed, beginning with data harmonization, geometric correction, and radiometric normalization to ensure cross-sensor comparability. An advanced feature extraction module leverages spectral indices, texture measures, radar backscatter characteristics, and LiDAR-derived height and canopy metrics to capture diverse land cover classes such as residential, commercial, industrial, green spaces, water bodies, and transportation networks. We introduce a robust change detection scheme that fuses abrupt and gradual transition signals using a probabilistic Bayes-based classifier and a time-series segmentation algorithm to reduce false positives induced by seasonal effects and sensor noise. The GNSS-enriched GIS component enhances feature geolocation accuracy, enables precise co-registration of multi-temporal datasets, and supports real-time updating of cadastral and infrastructure layers, thereby improving the reliability of change attribution to policy-driven versus natural drivers. The framework includes a spatiotemporal analytics module that quantifies urban growth intensity, infill versus spillover dynamics, and fragmentation indices, enabling scenario analysis under different planning interventions. To validate the approach, the framework is applied to a metropolitan region exhibiting rapid redevelopment, with ground truth generated from nationwide cadastral records, field surveys, and municipal planning documents. Evaluation metrics include overall accuracy, class-level F1 scores, kappa statistics, change detection rate, and geometric accuracy improvements attributable to GNSS corrections. Results demonstrate significant gains in detection performance when integrating GNSS-enriched GIS with multisource data, particularly in resolving boundary ambiguities near heterogeneous land use interfaces and in areas with dense infrastructure that challenge traditional single-sensor methods. The study also examines computational efficiency and scalability, outlining a modular architecture suitable for deployment in city-scale monitoring dashboards. Sensitivity analyses reveal the impact of sensor cadence, data fusion weights, and GNSS positioning accuracy on detection outcomes, informing best practices for operational monitoring. Policy implications are discussed in terms of urban resilience, land use policy evaluation, and sustainable smart city planning, highlighting how near-real-time change information can guide zoning updates, green space preservation, infrastructure investment, and disaster risk reduction. The research contributes a replicable, end-to-end framework that advances the state of the art in integration of multisource remote sensing and GNSS-enhanced GIS for dynamic urban land use management.

Project Overview

What This Project Is About

---

A straightforward study that looks at how city land use changes over time using photos and maps from different sources, combined with location data gathered from GPS. It shows how to detect where parks, new buildings, roads, and other uses appear or disappear in a city and what that means for planning.

---

---

The Problem It Addresses

---

Cities change quickly, and planners need up-to-date, accurate maps to make good decisions. Traditional methods can be slow or incomplete. This project addresses gaps in timely land use information by merging multiple data sources and precise location data to improve change detection.

---

---

Objectives of the Project

---

1. Identify and classify different land use types (e.g., residential, commercial, green spaces).
2. Detect changes in land use over a specified time period.
3. Integrate data from remote sensing, maps, and GPS for accurate results.
4. Develop a simple workflow that can be reused for other cities.
5. Assess the accuracy of detected changes and identify sources of error.

---

---

What You Will Do Step by Step

---

1. Collect multisource data: satellite images, aerial photos, and existing maps.
2. Gather accurate GPS coordinates for locations of interest.
3. Preprocess data to correct errors and align formats.
4. Classify land use in images and compare across time periods.
5. Merge datasets in a GIS to map changes.
6. Analyze results and assess accuracy with ground truth where possible.
7. Document the workflow and create a reusable plan.

---

---

Expected Outcome

A clear set of maps showing land use changes, a simple methodology for combining images and GPS data, and recommendations for city planning based on detected changes.