Stuck in the middle of ADS model process #775
-
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 2 replies
-
The TEM images are in the zip file. It's strange that I can not attach those Tiff files individually so I compress them to zip file. |
Beta Was this translation helpful? Give feedback.
-
Hi @Benson-Hsueh - it seems that because your resolution is an order of magnitude different from the resolution of the TEM training images we used, that the software is dividing your image into a lot of resampled "patches" (~224). You can adjust this type of thing with the "Zoom factor" in our settings, I just tried with a Zoom factor of 0.1 and it seems to have segmented your images decently well (you can adjust the options to see if you can do better; Zoom of 0.1 gave me 4 patches to segment, so increasing it slightly might improve things a a bit). As for your other image that you sent, the contrast between myelin and axon (and background) is very low (everything is bright), so either there was a staining issue or microscope issue there, it would be difficult to segment that type of image using the software without some manual image editing on your end to improve the contrast before trying our tool. Hope this helps! |
Beta Was this translation helpful? Give feedback.
Hi @Benson-Hsueh - it seems that because your resolution is an order of magnitude different from the resolution of the TEM training images we used, that the software is dividing your image into a lot of resampled "patches" (~224). You can adjust this type of thing with the "Zoom factor" in our settings, I just tried with a Zoom factor of 0.1 and it seems to have segmented your images decently well (you can adjust the options to see if you can do better; Zoom of 0.1 gave me 4 patches to segment, so increasing it slightly might improve things a a bit).
As for your other image that you sent, the contrast between myelin and axon (and background) is very low (everything is bright), so either …