Drone-Based Remote Sensing Technologies for Monitoring Urban Biodiversity and Habitat Fragmentation in Smart Cities

Abstract

 Abstract: With the speed of urbanization, increasing habitat fragmentation and unplanned land use change, urban biodiversity is under more threat than ever, leading to a need for ongoing ecological monitoring for sustainable smart city development. In this paper, we propose a framework based on remote sensing with a drone to monitor the biodiversity of the urban landscape and to estimate the fragmentation of the habitat, using images acquired by the drone from high resolution and geospatial analysis. The proposed approach consists of integrating unmanned aerial vehicles (UAVs), multispectral sensing, orthomosaic generation, vegetation index computation, and artificial intelligence (AI) based land cover classification in order to generate accurate maps of biodiversity and distribution of habitats. Drone High-Resolution (HR) imagery allows for the detailed detection of vegetation patches, ecological corridors and fragmented habitats that might not be detectable using standard satellite monitoring. The framework also includes spatial metrics that measure the habitat connectivity, habitat fragmentation and landscape heterogeneity to inform Urban ecological planning. Experimental results prove high classification accuracy, better spatial resolution and efficient assessment of biodiversity in comparison to the traditional remote sensing methods. The proposed system is timely, cost-effective and scalable that can be used for smart city applications, conservation planning and ecosystem management. In addition, the study explores the challenges associated with flight regulations, sensor limitations and environmental variability, as well as future opportunities with the use of autonomous drones equipped with AI capabilities, edge computing and digital twin technologies for real-time monitoring and assessment of biodiversity in urban areas and supporting evidence-based environmental decision-making.