Comparative Analysis of Statistical Thresholding Techniques in Nighttime Light Segmentation Nurin Izzati Azmi and Norzailawati Mohd Noor*
Department of Urban and Regional Planning, Kulliyyah of Architecture and Built Environment, International Islamic University Malaysia, 53100 Kuala Lumpur, Malaysia
*norzailawati[at]iium.edu.my
Abstract
Nighttime light (NTL) data has become an effective instrument for tracking urban growth, providing vital information on the temporal and spatial development patterns. Nighttime light accurately and systematically depicts human activities and affected areas. This study analyses differences in statistical threshold methods in defining nighttime light patterns. The threshold methods^- Otsu, Tsallis and Kapur, were employed to extract built-up areas from nighttime light data, allowing for the identification of key urban expansion corridors relative to the city centre. Assessing the temporal dynamics of urbanisation is critical for understanding urban growth, infrastructure demands, and planning needs. By comparing the outputs of these thresholding methods, this study highlights how the choice of statistical technique can influence the accuracy and interpretation of built-up areas using high-resolution nighttime light data. The Chinese Academy of Sciences engineered a satellite in 2021 that offered higher 10-m spatial resolution nighttime light data from the Sustainable Development Science Satellite-1 (SDGSAT-1), providing the opportunity for more in-depth analysis. With the latest and higher-resolution nighttime light imagery, the study also evaluates the effectiveness of multiple thresholding methods in detecting dynamic urban development patterns in Selangor, with a particular focus on rural areas. Selangor has increasingly focused on expanding suburban districts to promote economic equality and reduce dependency on the city centre. This growth reflects policy-driven efforts to decentralise economic opportunities. Understanding these dynamics is essential for aligning urban expansion with the SDGs, particularly Goal 11- Sustainable Cities and Communities, which highlights inclusive, safe, resilient, and sustainable urbanisation. The study highlights how remote sensing techniques, particularly nighttime light analysis, can serve as a cost-effective and scalable method for monitoring urban sprawl, informing policymakers in their efforts to enhance infrastructure planning, optimise resource allocation, and promote sustainable land use practices.
