M5-Malaria Dataset (Multi Micrscope Multi Magnification Malaria Dataset)
收藏OpenDataLab2026-05-24 更新2024-05-09 收录
下载链接:
https://opendatalab.org.cn/OpenDataLab/M5-Malaria_Dataset_Multi_Micrscope_etc
下载链接
链接失效反馈官方服务:
资源简介:
M5-Malaria数据集 (多Micrscope多倍率疟疾数据集) 包含2x3x1257图像包含疟疾血细胞。我们有来自薄血涂片的1257图像,并且在两种不同显微镜的三种不同放大倍数上跟踪和捕获了相同的区域。我们使用HCM (高成本显微镜) 在三个放大倍率 (100x,400x,1000x) 上捕获图像,然后使用LCM (低成本显微镜) 跟踪和捕获相同的位置。图像在。png格式和注释为pascal_voc格式。如果您需要我们数据集的COCO格式,则可以使用文件PascalToCoco.py生成coco格式注释。
M5-Malaria Dataset (Multi-Microscope, Multi-Magnification Malaria Dataset) contains 2×3×1257 images of malaria-infected blood cells. There are 1257 image sets derived from thin blood smears, where the same field of view was tracked and captured using two different microscopes across three distinct magnification levels. Specifically, images were captured at three magnifications (100x, 400x, 1000x) using a high-cost microscope (HCM), followed by capturing the exact same locations with a low-cost microscope (LCM). All images are stored in .png format, with annotations provided in Pascal VOC format. If you require COCO-format annotations for this dataset, you can generate them using the provided script PascalToCoco.py.
提供机构:
OpenDataLab
创建时间:
2023-02-13
搜集汇总
数据集介绍

背景与挑战
背景概述
M5-Malaria数据集是一个用于疟疾检测的多显微镜多倍率图像集合,包含1257张来自薄血涂片的疟疾血细胞图像,使用高成本和低成本显微镜在100x、400x和1000x三种放大倍数下捕获。图像以.png格式存储,注释为Pascal VOC格式,并可转换为COCO格式。
以上内容由遇见数据集搜集并总结生成



