Scholarlink

<p>Satellite images have been and continue to be a major area of study due to their significant contribution to the monitoring and assessment of environmental conditions using numerous indices of different categories (e.g., vegetation, water). However, processing large-scale, multitemporal datasets remains time-consuming and technically complex. To address this, an open-source QGIS Python-based plugin named Remote Sensing Toolkit (RST) was developed to automate the processing of satellite images and the computation of 100 biophysical indices from five satellite missions (i.e., AVHRR, MODIS, Landsat-8/9, Sentinel-2, ASTER). RST includes key preprocessing tools such as cloud masking, scaling, and clipping by mask layers, in addition to an AI-based outlier detection module that is integrated to enhance result reliability, all with a programming-free interface and customizable parameters. The plugin’s utility was demonstrated through an application to the 2024 Spain flash flood using Sentinel-2 data from 2022 to 2024. A SARIMA model was used to detect temporal anomalies in NDVI-derived water extent time series, revealing a significant deviation coinciding with the flood event. Moreover, spatial maps of flood extent were generated to visualize and quantify the affected areas, and a comparative assessment with Copernicus Emergency Management Service Rapid Mapping (CEMS RM) products was conducted to evaluate the accuracy of detected flood extents. This analysis highlighted the complementary value of NDVI-derived flood maps and demonstrated the plugin’s effectiveness in automating satellite data workflows for environmental monitoring and rapid disaster assessment.</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

Remote Sensing Toolkit (RST) plugin for automated multitemporal remote sensing analysis: application to Spain’s 2024 flash flood

  • 期刊论文

  • 发表日期：

    2025年7月11日

查看全文
  • Fatima Ezahrae Ezzaher,
  • Nizar Ben Achhab,
  • Hafssa Naciri,
  • Naoufal Raissouni,
  • Boutaina Sebbah,
  • Abdelilah Azyat

摘要

Satellite images have been and continue to be a major area of study due to their significant contribution to the monitoring and assessment of environmental conditions using numerous indices of different categories (e.g., vegetation, water). However, processing large-scale, multitemporal datasets remains time-consuming and technically complex. To address this, an open-source QGIS Python-based plugin named Remote Sensing Toolkit (RST) was developed to automate the processing of satellite images and the computation of 100 biophysical indices from five satellite missions (i.e., AVHRR, MODIS, Landsat-8/9, Sentinel-2, ASTER). RST includes key preprocessing tools such as cloud masking, scaling, and clipping by mask layers, in addition to an AI-based outlier detection module that is integrated to enhance result reliability, all with a programming-free interface and customizable parameters. The plugin’s utility was demonstrated through an application to the 2024 Spain flash flood using Sentinel-2 data from 2022 to 2024. A SARIMA model was used to detect temporal anomalies in NDVI-derived water extent time series, revealing a significant deviation coinciding with the flood event. Moreover, spatial maps of flood extent were generated to visualize and quantify the affected areas, and a comparative assessment with Copernicus Emergency Management Service Rapid Mapping (CEMS RM) products was conducted to evaluate the accuracy of detected flood extents. This analysis highlighted the complementary value of NDVI-derived flood maps and demonstrated the plugin’s effectiveness in automating satellite data workflows for environmental monitoring and rapid disaster assessment.