Scholarlink

<p>This study examines the integration of vein and fracture characterization into traditional rock mass classification systems and highlights how advances in imaging technologies and artificial intelligence (AI) optimize this process. Conventional approaches, such as the Rock Mass Rating (RMR) system and the Q index, have been widely used to assess rock mass stability. However, these methodologies exhibit limitations when addressing parameters such as morphology, spatial variability, and mineralogical composition. Emerging technologies, such as hyperspectral imaging and AI-based models, offer innovative solutions to overcome these challenges. Hyperspectral imaging enables high-resolution, non-invasive mineralogical mapping, providing detailed information on composition and alteration zones associated with veins. Simultaneously, AI algorithms, such as the YOLO series and Mask R-CNN, facilitate the automated detection and segmentation of veins, reducing subjectivity and processing time while improving characterization accuracy. This work reviews recent advances in these technologies, emphasizing their ability to bridge gaps in traditional methodologies. The discussion focuses on their application to enhance the evaluation of geotechnical properties and contribute to more efficient and reliable workflows. Finally, the need for continued research on data integration, algorithm performance, and sensor accuracy is emphasized to fully exploit these tools, promoting safer and more sustainable mining operations.</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

Characterization of Fractures and Veins from Images of Drill Core: A review of imaging technologies

  • 期刊论文

  • 发表日期：

    2026年3月2日

查看全文
  • Thiare Salazar,
  • Kimie Suzuki Morales

摘要

This study examines the integration of vein and fracture characterization into traditional rock mass classification systems and highlights how advances in imaging technologies and artificial intelligence (AI) optimize this process. Conventional approaches, such as the Rock Mass Rating (RMR) system and the Q index, have been widely used to assess rock mass stability. However, these methodologies exhibit limitations when addressing parameters such as morphology, spatial variability, and mineralogical composition. Emerging technologies, such as hyperspectral imaging and AI-based models, offer innovative solutions to overcome these challenges. Hyperspectral imaging enables high-resolution, non-invasive mineralogical mapping, providing detailed information on composition and alteration zones associated with veins. Simultaneously, AI algorithms, such as the YOLO series and Mask R-CNN, facilitate the automated detection and segmentation of veins, reducing subjectivity and processing time while improving characterization accuracy. This work reviews recent advances in these technologies, emphasizing their ability to bridge gaps in traditional methodologies. The discussion focuses on their application to enhance the evaluation of geotechnical properties and contribute to more efficient and reliable workflows. Finally, the need for continued research on data integration, algorithm performance, and sensor accuracy is emphasized to fully exploit these tools, promoting safer and more sustainable mining operations.