MicSim_FluoMT
收藏资源简介:
MicSim_FluoMT数据集是由法国雷恩遗传与发展研究所和摩洛哥苏丹穆莱斯利曼大学科学和技术学院的研究人员创建的,包含数百个合成荧光标记微管图像,用于细胞内微管的分割。数据集包括均匀荧光和逐渐减弱荧光两种情况,以模拟真实显微镜图像中的噪声。数据集的每个图像都带有精确的分割标签,是评估分割算法在真实噪声和结构复杂性条件下的性能的理想基准。
The MicSim_FluoMT dataset was developed by researchers from the Institute of Genetics and Development of Rennes (France) and the Faculty of Sciences and Technologies at Sultan Moulay Slimane University (Morocco). It comprises hundreds of synthetic fluorescence-labeled microtubule images intended for intracellular microtubule segmentation tasks. The dataset covers two imaging conditions: uniform fluorescence and gradually attenuated fluorescence, both of which simulate the noise inherent in real microscopic images. Every image is accompanied by precise segmentation annotations, making it an ideal benchmark for assessing the performance of segmentation algorithms under conditions involving real-world noise and structural complexity.
MicSim_FluoMT: Two synthetic datasets of images of fluorescent microtubules
数据集概述
- 发布日期: January 20, 2025
- 版本: v1
- 资源类型: Image
- 语言: English
- DOI: 10.5281/zenodo.14696280
- 许可证: Creative Commons Attribution 4.0 International
创作者与贡献者
创作者
- Bouvrais, Hélène (Project leader) - Institut de génétique et de développement de Rennes
- Crespo, Mewen (Project member)
贡献者
项目负责人
- ELMOURABIT, yousef - Technology and Sciences Faculty, Sultan Moulay Slimane University
- Bouvrais, Hélène - Institut de génétique et de développement de Rennes
项目成员
- Ait Laydi, Achraf - Technology and Sciences Faculty, Sultan Moulay Slimane University
- Crespo, Mewen - Institut de recherche mathématique de Rennes
数据集描述
- 数据集支持Ait Laydi等人(2025)的论文。
- 包含两个合成图像数据集,模拟荧光标记的微管显微镜图像。
- 用于训练深度学习架构进行微管分割。
- 包括输入图像(称为“noisy”)和对应的地面真实图像(称为“binary”)。
- easy数据集: 1192张图像,微管荧光均匀。
- hard数据集: 1192张图像,地面真实与easy数据集相同,但微管荧光向末端递减,分割更具挑战性。
文件
- dataset_easy.zip (112.8 MB) - MD5: f973f1a03fd2a6a6a83cff8127d7d01c
- dataset_hard.zip (110.2 MB) - MD5: b3c862e0ea3cbf816809eddc1cdbb483
生成工具
- Cytosim: https://gitlab.com/f-nedelec/cytosim
- confocalGN: https://github.com/SergeDmi/ConfocalGN
引用
Bouvrais, H., & Crespo, M. (2025). MicSim_FluoMT: Two synthetic datasets of images of fluorescent microtubules (Ait Laydi et al., 2025) (Version v1). Zenodo. https://doi.org/10.5281/zenodo.14696280
资金支持
- Agence Nationale de la Recherche (MICENN ANR-22-CE45-0016-01)
- Université de Rennes (Défis scientifiques 2020, Soutien Collaborations Internationales 2024)
- Campus France (PHC Toubkal 2024 49945RE)
关键词
- filament segmentation
- synthetic image dataset
- noisy fluorescent images
MeSH主题词
- Microtubules
- Deep Learning/classification
- Datasets as Topic/classification

- 1Adaptive Attention Residual U-Net for curvilinear structure segmentation in fluorescence microscopy and biomedical imagesCNRS, IGDR (Institute Genetics and Development of Rennes) – UMR 6290, Rennes, France and TIAD Laboratory, Sciences and Technology Faculty, Sultan Moulay Slimane Univ., Morocco · 2025年



