Scholarlink

<p>Fungal diseases pose significant threats to forestry species such as <i>Tectona grandis</i> (teak), as well as other important forestry and agricultural crops, highlighting the need for early and accurate identification of fungal spores, which serve as primary agents of dissemination and infection. Although Artificial Intelligence (AI) has enabled automated spore recognition, its effectiveness depends on the availability of large, diverse, and well-annotated datasets. However, publicly available datasets targeting fungal taxa associated with commercially valuable timber species remain scarce. To address this gap, we introduce a microscopic image dataset of fungal spores isolated from symptomatic teak foliage, including <i>Olivea tectonae</i>, <i>Colletotrichum siamense</i>, and <i>Neopestalotiopsis</i> sp. The dataset was developed through systematic field sampling, direct microscopic observation, and axenic culturing, followed by high-resolution imaging and manual annotation by experts. This annotated dataset serves as a foundational resource for AI-assisted spore detection across both field-based and atmospheric surveillance workflows, supporting applications such as sample-based analysis, air-based monitoring, and real-time diagnostics. Its cross-species utility and future extensibility enhance its value for plant disease management.</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

Microscopic image dataset of fungal spores for computer vision applications in Tectona grandis and other taxa

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月11日

查看全文
  • Syeda Munjiba Islam,
  • Md. Shariar Hossain Sazzad,
  • Md. Faysal Ahamed,
  • Maaz Ahmed,
  • Sadikul Islam Nayon,
  • Toufiq Islam,
  • Sabrina Saniat Ayon,
  • Julfikar Haider,
  • M. Abdullah-Al-Wadud,
  • S. M. Riazul Islam

摘要

Fungal diseases pose significant threats to forestry species such as Tectona grandis (teak), as well as other important forestry and agricultural crops, highlighting the need for early and accurate identification of fungal spores, which serve as primary agents of dissemination and infection. Although Artificial Intelligence (AI) has enabled automated spore recognition, its effectiveness depends on the availability of large, diverse, and well-annotated datasets. However, publicly available datasets targeting fungal taxa associated with commercially valuable timber species remain scarce. To address this gap, we introduce a microscopic image dataset of fungal spores isolated from symptomatic teak foliage, including Olivea tectonae, Colletotrichum siamense, and Neopestalotiopsis sp. The dataset was developed through systematic field sampling, direct microscopic observation, and axenic culturing, followed by high-resolution imaging and manual annotation by experts. This annotated dataset serves as a foundational resource for AI-assisted spore detection across both field-based and atmospheric surveillance workflows, supporting applications such as sample-based analysis, air-based monitoring, and real-time diagnostics. Its cross-species utility and future extensibility enhance its value for plant disease management.