Scholarlink

<p>Temporal remote-sensing satellite data are useful for natural resources and environmental monitoring. Environmental landscape modelling, urban planning, and historical land cover change studies benefit from accurate representations of landscape attributes and precise assessments of spatio-temporal changes. Manual methods for analysing temporal satellite data to identify changes are time-consuming. Automated procedures analysing remote-sensing satellite data using online platforms can be an efficient solution for land-use and land-cover (LU/LC) studies. The goal of this study is to assess LU/LC changes using online geospatial processing tools and multi-temporal remote sensing satellite data in an efficient manner. Open-source Landsat data from 1988 to 2021 are used to identify changes. Different regions, including urban, semi-urban, and rural areas, are selected to assess the performance of the proposed procedures. The regions Kanpur, Uttar Pradesh (Metropolitan), Farrukhabad, Uttar Pradesh (Urban), Nainital, Uttarakhand (Semi-Urban), and Pauri Garhwal, Uttarakhand (Rural) have been studied to quantify long-term LU/LC changes and identify the driving reasons behind them. The LU/LC classification has been implemented on the Google Earth Engine (GEE) platform, and results have been compared for Random Forest, Support Vector Machine, and Classification and Regression Tree (CART) classifiers for each region. The RF classifier produced better results than the CART and SVM classifiers across all four regions. The quantitative analysis of LU/LC maps reveals divergent trends across classes over a 33-year period. Prediction of Future LU/LC Changes is executed for the year 2030 for each region using the QGIS MOLUSCE Plugin. Furthermore, we discussed the potential applications and limitations of the land cover classification method in other areas.</p>

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

Geospatial analytics for urbanization monitoring using multi-temporal satellite imageries

  • 期刊论文

  • 发表日期：

    2026年5月22日

查看全文
  • Tushita Singh,
  • Vinod K. Sharma,
  • Sitender Malik,
  • S. K. Srivastav,
  • Jatin Dahiya

摘要

Temporal remote-sensing satellite data are useful for natural resources and environmental monitoring. Environmental landscape modelling, urban planning, and historical land cover change studies benefit from accurate representations of landscape attributes and precise assessments of spatio-temporal changes. Manual methods for analysing temporal satellite data to identify changes are time-consuming. Automated procedures analysing remote-sensing satellite data using online platforms can be an efficient solution for land-use and land-cover (LU/LC) studies. The goal of this study is to assess LU/LC changes using online geospatial processing tools and multi-temporal remote sensing satellite data in an efficient manner. Open-source Landsat data from 1988 to 2021 are used to identify changes. Different regions, including urban, semi-urban, and rural areas, are selected to assess the performance of the proposed procedures. The regions Kanpur, Uttar Pradesh (Metropolitan), Farrukhabad, Uttar Pradesh (Urban), Nainital, Uttarakhand (Semi-Urban), and Pauri Garhwal, Uttarakhand (Rural) have been studied to quantify long-term LU/LC changes and identify the driving reasons behind them. The LU/LC classification has been implemented on the Google Earth Engine (GEE) platform, and results have been compared for Random Forest, Support Vector Machine, and Classification and Regression Tree (CART) classifiers for each region. The RF classifier produced better results than the CART and SVM classifiers across all four regions. The quantitative analysis of LU/LC maps reveals divergent trends across classes over a 33-year period. Prediction of Future LU/LC Changes is executed for the year 2030 for each region using the QGIS MOLUSCE Plugin. Furthermore, we discussed the potential applications and limitations of the land cover classification method in other areas.