project-oceania/zoocamnet

ZooCAMNet数据集包含使用ZooCAM捕获的浮游生物图像。这些图像是通过在比斯开湾的4米水下使用连续水下鱼卵采样器（CUFES，网眼尺寸为315微米）和在100米到水面的WP2网（网眼尺寸为200微米）收集的。图像使用ZooCAM软件和Matrox Imaging Library进行处理，生成每个独立对象的感兴趣区域（ROIs）和一系列特征。然后，使用scikit-image库重新处理相同的对象以计算特征。结果得到的1,286,590个对象根据有限的操作员，遵循一个共同的分类指南，分为93个分类，使用EcoTaxa网络应用程序。为了训练机器学习分类器，每个类别的图像被分为训练、验证和测试集，比例分别为70%、15%和15%。

The ZooCAMNet dataset contains plankton images captured with the ZooCAM. These images were collected using a Continuous Underway Fish Egg Sampler (CUFES, 315µm mesh size) at 4 meters below the surface and a WP2 net (200µm mesh size) from 100 meters to the surface or 5 meters above the sea floor when the depth was less than 100 meters in the Bay of Biscay. The images were processed with the ZooCAM software and the embedded Matrox Imaging Library to generate regions of interest (ROIs) around each individual object and a set of features measured on the object. The same objects were re-processed with the scikit-image library to compute features. The resulting 1,286,590 objects were sorted into 93 taxa by a limited number of operators following a common taxonomic guide using the EcoTaxa web application. For the purpose of training machine learning classifiers, the images in each class were split into training, validation, and test sets with proportions of 70%, 15%, and 15% respectively.