Scholarlink

<p>Unexploded ordnance (UXO) detection remains a critical humanitarian challenge affecting over 60 countries worldwide. This systematic review examines unmanned aerial vehicle (UAV) remote sensing technologies for UXO detection, analyzing 92 relevant publications through PRISMA methodology. Our analysis reveals that ground penetrating radar (31%) and magnetometry (26%) were the most prevalent detection methods, followed by imaging-based approaches (21%), with 14% of studies employing integrated multi-sensor systems. Multi-rotor platforms dominated the field, accounting for approximately 80% of all implementations due to their hovering capability and low-altitude operational advantages. Despite technological advances, no single detection method has proven universally effective. Our analysis categorizes UAV-based approaches into passive methods (magnetometry, visible-light imaging, multispectral, hyperspectral, and thermal imaging) and active methods (electromagnetic surveys and ground penetrating radar). We find that while individual techniques offer distinct advantages in specific conditions, integrated multi-platform and multi-sensor approaches yield superior results by compensating for methodological limitations. The emerging integration of artificial intelligence (AI) with sensing technologies demonstrates particular promise for enhancing detection accuracy, speed, and cost-effectiveness, with deep learning approaches achieving detection rates exceeding 97% in several studies. Our structured analysis of current UAV-based UXO detection technologies identifies promising research directions, offering valuable insights for safer and more efficient clearance operations in post-conflict regions.</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

Unmanned Aerial Remote Sensing for Unexploded Ordnance Detection: Concepts, Methods, Current Status and Trends

  • 期刊论文

  • 发表日期：

    2025年10月18日

查看全文
  • Farhad Samadzadegan,
  • Farzad Qaderi,
  • Tahereh Layegh Haghighi,
  • Ahmad Toosi,
  • Seyed Mahdi Mousavi

摘要

Unexploded ordnance (UXO) detection remains a critical humanitarian challenge affecting over 60 countries worldwide. This systematic review examines unmanned aerial vehicle (UAV) remote sensing technologies for UXO detection, analyzing 92 relevant publications through PRISMA methodology. Our analysis reveals that ground penetrating radar (31%) and magnetometry (26%) were the most prevalent detection methods, followed by imaging-based approaches (21%), with 14% of studies employing integrated multi-sensor systems. Multi-rotor platforms dominated the field, accounting for approximately 80% of all implementations due to their hovering capability and low-altitude operational advantages. Despite technological advances, no single detection method has proven universally effective. Our analysis categorizes UAV-based approaches into passive methods (magnetometry, visible-light imaging, multispectral, hyperspectral, and thermal imaging) and active methods (electromagnetic surveys and ground penetrating radar). We find that while individual techniques offer distinct advantages in specific conditions, integrated multi-platform and multi-sensor approaches yield superior results by compensating for methodological limitations. The emerging integration of artificial intelligence (AI) with sensing technologies demonstrates particular promise for enhancing detection accuracy, speed, and cost-effectiveness, with deep learning approaches achieving detection rates exceeding 97% in several studies. Our structured analysis of current UAV-based UXO detection technologies identifies promising research directions, offering valuable insights for safer and more efficient clearance operations in post-conflict regions.