An Optimized Smart City Model for Advancing Urban Management Goals Using Remote Sensing Techniques and MLC & SVM Models (Case Study: Shiraz Metropolis)

An Optimized Smart City Model for Advancing Urban Management Goals Using Remote Sensing Techniques and MLC & SVM Models (Case Study: Shiraz Metropolis)

aminfakhar1 Marzieh Mogholi2 Abdolrasool Ghanbari3

1) Amin fakhar : Ph.D. Student of Geography, Larestan Unit, Islamic Azad University, Larestan, Iran
2) Dr Marzieh Mogholi1: Associate Professor, Department of Geography, Larestan Branch, Islamic Azad University, Larestan, Iran
3) Dr Abdolrasool Ghanbari: Assistant Professor, Department of Geography, Larestan Branch, Islamic Azad University, Larestan, Iran

IOI Code : AACA-ABAZHC
Publication : 10th.International Conference on Researches in Science & Engineering & 7th.International Congress on Civil, Architecture and Urbanism in Asia(10icrsie.com)
Abstract :
    In the contemporary era, the rapid growth of cities has created multiple challenges in areas such as environment, transportation, water and energy resources, waste management, and urban green spaces. Addressing these challenges requires innovative, technology-driven approaches that can enhance urban decision-making processes and pave the way for the realization of smart cities and long-term sustainability.In this study, a proposed model for intelligent urban management in the metropolis of Shiraz is presented, based on the application of remote sensing technologies and two advanced machine learning methods: Maximum Likelihood Classification (MLC) and Support Vector Machines (SVM).The MLC method, as one of the classical approaches, was employed for multi-level data analysis and the classification of complex and multidimensional information. This method enables the identification of urban areas based on criteria such as population density, access to services, green space conditions, and energy consumption. Despite its simplicity and initial efficiency, the main limitation of MLC lies in handling heterogeneous and high-dimensional data, which may reduce classification accuracy.In contrast, the SVM method, as a powerful and flexible algorithm, demonstrates high capability in managing complex and nonlinear data. By establishing optimal decision boundaries, SVM improves classification accuracy and provides better differentiation among various urban patterns. The comparative results indicated that SVM outperforms MLC in terms of precision and stability when identifying land use, transportation patterns, and natural resource conditions.Overall, the findings of this study suggest that while MLC can serve as a baseline method for urban data classification, SVM, with its higher accuracy and ability to manage complex datasets, is a more suitable option for developing intelligent urban management models in metropolitan areas.

Keywords : smart city urban management remote sensing shiraz city