Scholarlink

Urban‍‌ megacities worldwide have been severely impacted by rapid urbanization to the point where they are suffering from limited land availability, ecological degradation, and inefficient spatial governance. Therefore, the implementation of intelligent, scalable systems is a prerequisite for the successful tackling of such issues, as these systems should be able to monitor urban dynamics and optimize land reuse. With this in view, this research introduces the Urban Spatial Optimization and Monitoring Framework (proposed framework) - an AI-powered, remote sensing-based platform that can easily track changes in GIS data and simulate the best strategies for urban space reuse. proposed framework combines semantic segmentation and probabilistic modeling to attain ultrafine building footprint detection and the functional shift in the areas of construction and non-construction zones. Through multi-stage pipeline, proposed framework is deeply involved: Initially, it uses the deep learning model U-Net for multi-temporal ZY-3 satellite imaging to get the building footprints which help in the accurate locating of new constructions and the structures that are demolished. Then, it deploys Random Forest regression for spatial reuse evaluation based on land use, population, and transport accessibility indices. Besides, the system makes use of planning permit data and the ever-changing demographics to get the model ready and check the urban planning reforms and social-environment conditions. With the Beijing example, China, the framework investigated local changes in building spaces. The analyses pointed out that the total building area was decreased by 76.24 km2 with most of the area being environmentally sensitive zones where the reduction was made. The optimization for non-construction areas resulted in 79.99% actual reforestation and 89.21% suitability for recultivation. While public service facilities the reuse simulations in the construction zones were 81.47% accurate, industrial and residential allocations achieved 78.63 and 74.92% accuracy correspondingly. The proposed framework introduced here considerably promotes urban management, ecological restoration and spatial effectiveness, thus being a model that can be repeated for sustainable development in global megacities.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An Optimized Framework to Monitor Remotely Gis Data in Urban Cities

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sandeep Kumar Thota,
  • Venkata Ramana Gudelli,
  • G. Sunil,
  • Neeraj Varshney,
  • Anandakumar Haldorai,
  • Minu Balakrishnan

摘要

Urban‍‌ megacities worldwide have been severely impacted by rapid urbanization to the point where they are suffering from limited land availability, ecological degradation, and inefficient spatial governance. Therefore, the implementation of intelligent, scalable systems is a prerequisite for the successful tackling of such issues, as these systems should be able to monitor urban dynamics and optimize land reuse. With this in view, this research introduces the Urban Spatial Optimization and Monitoring Framework (proposed framework) - an AI-powered, remote sensing-based platform that can easily track changes in GIS data and simulate the best strategies for urban space reuse. proposed framework combines semantic segmentation and probabilistic modeling to attain ultrafine building footprint detection and the functional shift in the areas of construction and non-construction zones. Through multi-stage pipeline, proposed framework is deeply involved: Initially, it uses the deep learning model U-Net for multi-temporal ZY-3 satellite imaging to get the building footprints which help in the accurate locating of new constructions and the structures that are demolished. Then, it deploys Random Forest regression for spatial reuse evaluation based on land use, population, and transport accessibility indices. Besides, the system makes use of planning permit data and the ever-changing demographics to get the model ready and check the urban planning reforms and social-environment conditions. With the Beijing example, China, the framework investigated local changes in building spaces. The analyses pointed out that the total building area was decreased by 76.24 km2 with most of the area being environmentally sensitive zones where the reduction was made. The optimization for non-construction areas resulted in 79.99% actual reforestation and 89.21% suitability for recultivation. While public service facilities the reuse simulations in the construction zones were 81.47% accurate, industrial and residential allocations achieved 78.63 and 74.92% accuracy correspondingly. The proposed framework introduced here considerably promotes urban management, ecological restoration and spatial effectiveness, thus being a model that can be repeated for sustainable development in global megacities.