Life beneath the ice: jellyfish and ctenophores from the Ross Sea, Antarctica, with an image-based training set for machine learning.
Life beneath the ice: jellyfish and ctenophores from the Ross Sea, Antarctica, with an image-based training set for machine learning.
复制标题
冰下的生命：来自南极罗斯海的水母和栉水母，以及基于图像的机器学习训练集。
DOI：
10.3897/bdj.9.e69374
复制
发表时间：
2021
影响因子：
1.3
通讯作者：
Lindsay D
中科院分区：
文献类型：
--
作者：
Verhaegen G;Cimoli E;Lindsay D
南大洋生态系统目前正在经历越来越多的环境变化和人为压力，这促使科学家们报告其生物多样性和海洋地理学。两个主要的分类学上的多样性和营养重要的凝胶状浮游动物群体，然而，在很大程度上，直到现在仍然研究不足的刺胞水母和栉水母。这种数据的稀缺主要是由于许多这些脆弱的软体生物很容易破碎和/或破坏传统的净采样方法。替代调查方法的进展，包括例如基于光学的方法，正在慢慢开始克服这些障碍。由于人类观察者的视频注释既耗时又经济昂贵，因此应该开发机器学习技术来分析基于原位/水中图像的数据集。这就需要分类学上准确的训练集正确的物种识别和本文件是第一个提供这样的数据。在这项研究中，我们两次进行了为期三周的原位光学为基础的调查水母和栉水母下发现的冰在南极洲的麦克默多声音。我们的研究构成了第一个基于光学的调查凝胶状浮游动物在罗斯海和第一个研究使用原位/在水中观察，以描述分类和一些营养和行为特征的凝胶状浮游动物从南大洋。尽管我们的研究的地理和时间尺度小，我们提供了新的未描述的形态特征，所有观察到的凝胶状浮游动物物种（8刺胞动物和4栉水母物种）。三种栉水母和一种细水母可能代表未描述的物种。此外，沿着的摄影和摄像，我们准备了一个共同的对象在上下文（COCO）数据集，使这项研究是第一个提供一个分类学家批准的图像训练集，为未来的机器学习算法开发有关南大洋凝胶浮游动物物种。
Southern Ocean ecosystems are currently experiencing increased environmental changes and anthropogenic pressures, urging scientists to report on their biodiversity and biogeography. Two major taxonomically diverse and trophically important gelatinous zooplankton groups that have, however, stayed largely understudied until now are the cnidarian jellyfish and ctenophores. This data scarcity is predominantly due to many of these fragile, soft-bodied organisms being easily fragmented and/or destroyed with traditional net sampling methods. Progress in alternative survey methods including, for instance, optics-based methods is slowly starting to overcome these obstacles. As video annotation by human observers is both time-consuming and financially costly, machine-learning techniques should be developed for the analysis of in situ /in aqua image-based datasets. This requires taxonomically accurate training sets for correct species identification and the present paper is the first to provide such data. In this study, we twice conducted three week-long in situ optics-based surveys of jellyfish and ctenophores found under the ice in the McMurdo Sound, Antarctica. Our study constitutes the first optics-based survey of gelatinous zooplankton in the Ross Sea and the first study to use in situ / in aqua observations to describe taxonomic and some trophic and behavioural characteristics of gelatinous zooplankton from the Southern Ocean. Despite the small geographic and temporal scales of our study, we provided new undescribed morphological traits for all observed gelatinous zooplankton species (eight cnidarian and four ctenophore species). Three ctenophores and one leptomedusa likely represent undescribed species. Furthermore, along with the photography and videography, we prepared a Common Objects in Context (COCO) dataset, so that this study is the first to provide a taxonomist-ratified image training set for future machine-learning algorithm development concerning Southern Ocean gelatinous zooplankton species.