five

CSIRO Australian Phytoplankton Microscopy Dataset (CAPMD) 2022

收藏
Research Data Australia2025-12-20 收录
下载链接:
https://researchdata.edu.au/csiro-australian-phytoplankton-capmd-2022/3654223
下载链接
链接失效反馈
官方服务:
资源简介:
The CSIRO Australian Phytoplankton Microscopy Dataset (CAPMD) 2022 is a comprehensive collection of high-quality microscopy images documenting the morphological diversity of phytoplankton species from the Australian National Algae Culture Collection (ANACC).\nLineage: Images were acquired using ZEISS Axio Observer and Axio Plan microscopes under standardised laboratory conditions at the CSIRO Battery Point site in Hobart, Tasmania. Live specimens were prepared in suspension and systematically imaged by capturing multiple short videos (50-200 frames) using a focal rolling technique that involved maneuvering the objective lens through different focal planes of the sample to capture key taxonomic features and cellular structures. The imaging protocol implemented multiple modalities including bright field, differential interference contrast, and phase contrast microscopy at magnifications ranging from 100x to 1000x. Specimens were selectively treated with Tylose to immobilise highly motile cells while maintaining their structural integrity, or Lugol's (Iodine) solution as a fixing agent to capture fixed cell imagery, which is a standard approach in phytoplankton identification. The imaging protocol was specifically designed to support the development of automated phytoplankton identification systems using machine learning techniques, with careful attention given to image quality, consistency, and comprehensive coverage of taxonomically significant features across different imaging conditions. This systematic imaging campaign successfully balanced the need for high-quality, representative samples with efficient, high-volume image capture, creating a robust dataset for advancing automated phytoplankton identification methods.
提供机构:
Commonwealth Scientific and Industrial Research Organisation
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作