CREMI

displayName: CREMI
license:
- Unknown
taskTypes:
- Semantic Segmentation
- Neuroscience
mediaTypes:
- Image
labelTypes:
- Semantic_seg_map
tags:
- attrs: null
id: 552
name:
en: Drosophila Melanogaster
zh: 黑腹果蝇
publisher:
- HHMI Janelia
publishDate: '2016-05-01'
publishUrl: https://cremi.org/
paperUrl: ''

---
# 数据集介绍
## 简介
这一挑战的目标是评估从串行切片电子显微镜数据中自动重建神经元和神经元连接的算法。不仅通过评估神经元分割的质量，还通过评估检测突触和识别突触伙伴的准确性来进行比较。该挑战是在来自成年黑腹果蝇大脑组织的三个大型且多样化的数据集上进行的，包括神经元分割基础事实和突触连接的注释。一个成功的解决方案将证明其效率和普遍性，并具有巨大的潜力，可以减少在电子显微镜体积中手动重建神经回路所花费的时间。

## Download dataset
:modelscope-code[]{type="git"}

displayName: CREMI
license:
- Unknown
taskTypes:
- Semantic Segmentation
- Neuroscience
mediaTypes:
- Image
labelTypes:
- Semantic_seg_map
tags:
- attrs: null
id: 552
name:
en: Drosophila melanogaster
publisher:
- HHMI Janelia
publishDate: '2016-05-01'
publishUrl: https://cremi.org/
paperUrl: ''
---
# Dataset Introduction
## Introduction
This challenge aims to evaluate algorithms for automated reconstruction of neurons and neuronal connections from serial-section electron microscopy (ssEM) data. Evaluations and comparisons are conducted not only by assessing the quality of neuronal segmentation, but also by evaluating the accuracy of synaptic detection and synaptic partner identification. The challenge is carried out using three large, diverse datasets sourced from adult *Drosophila melanogaster* brain tissue, which include ground truth annotations for neuronal segmentation and synaptic connections. A successful solution will demonstrate its efficiency and generalizability, with great potential to reduce the time spent on manual neural circuit reconstruction within electron microscopy volumes.

## Download Dataset
:modelscope-code[]{type="git"}

搜集汇总

背景与挑战

背景概述

CREMI是一个神经科学领域的语义分割数据集，专注于成年黑腹果蝇大脑组织的电子显微镜图像分析，包含神经元分割和突触连接注释，旨在评估自动重建算法以减少手动神经回路重建时间。数据集由HHMI Janelia于2016年发布，大小为35.43GB，适用于语义分割任务。