Fast Improvement of TEM Images with Low-Dose Electrons by Deep Learning
收藏NIAID Data Ecosystem2026-03-13 收录
下载链接:
https://zenodo.org/record/7024736
下载链接
链接失效反馈官方服务:
资源简介:
This is a dataset of High-Dose-Electron (HDE) images and Low-Dose-Electron (LDE) images taken with a transmission electron microscopy used in H. Katsuno, Y. Kimura, T. Yamazaki and I Takigawa, Microsc. Microanal. 28 (2022), pp 138--144 (arXiv:2106.01718).
There are two LDE images for each HDE image.
cf) HDE image is 0001.tif and corresponding LDE images are 0002.tif and 0003.tif.
Equipment of TEM:
field-emission gun (JEM-2100F, JEOL, Tokyo)
OneView IS (Gatan, Inc., Pleasanton, CA, USA)
Typical magnification was 25,000x and 30,000x.
HDE image
The resolution was 4096 x 4096 pixels and its exposure time was 5 s.
The typical total doses was 1010 e-.
LDE image
The resolution was 512 x 512 pixels and its exposure time was 3.3 ms.
The typical total doses was 106 e-.
Filename
Material
Total number of images
Total number of a pair of HDE and LDE
train1_Ni.zip
Ni
336
224
train2_FeNi.zip
FeNi
390
260
train3_SiC.zip
SiC
210
140
train4_Silicate
Silicate
264
176
train5_Alumina
Alumina
300
200
val1_Ni.zip
Ni
54
36
val2_FeNi.zip
FeNi
60
40
val3_SiC.zip
SiC
30
20
val4_Silicate
Silicate
36
24
val5_Alumina
Alumina
60
40
The ipynb file and model parameters for machine learning are located in the GitHub page.
创建时间:
2022-09-06



