-
Notifications
You must be signed in to change notification settings - Fork 10
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Zoo Classifier Workflow Update #197
Comments
|
The zoo workflow is significantly less intuitive than the coastsat workflow. It doesn't let you run the model or extract shorelines for multiple ROIs at the same time unlike the coastsat workflow. I think the workflow needs to be updated so that the user can select a session that contains multiple ROIs and run the model on that directory. Unfortunately I can't make this change in time for tomorrow. Given that I still need to work on #198 and #179.
|
|
After a bit of digging I found out that the reason why the cloud mask shows up in the image segmentation even when the cloud mask is off is because the cloud mask was already applied to the downloaded jpgs the model is running the segmentations on. The reason they don't show up in the final extracted shoreline diagrams is because those diagrams are created from the tifs being converted to jpgs without the cloud mask. So if we want to see the cloud mask in the outputted segmentations then we would need to re-read the tifs, create a new RGB jpg, then run the model on these making the model even slower. I can make this a feature but its going to take some extra work. |
You should probably fix these future warning in coastsat package while you do this
|
Jan Tasks
|
@dbuscombe-usgs Do you think that its necessary to have the zoo workflow works with multiple ROIs at a time like coastseg or can we skip this feature until after the paper release? |
Depends. It will be necessary eventually, so just depends on how tricky it is to implement. I'm guessing it is more tricky than you'd like to tackle right now, and if so, we can wait |
|
…dicition segmentation png to good/bad & npz to good/bad
The zoo classifier notebook is not fully up to date with the workflows present in the coastsat classifier Notebook. We are making a push to improve the workflows in that notebook in the coming months so I've created this issue to compile the changes that will need to be made.
Bugs
Cloud Mask Application Issue
simplified_find_contours
Modification Issuesimplified_find_contours
function, shorelines now appear as closed line string segments.Reference Shoreline Buffer Issue
simplified_find_contours
doesn't undergo a dilation of 5 pixels, contrary to the workflow in the Coastsat classifier.Related Issues
We aim to address these issues in the coming months to ensure seamless and efficient workflows. Feedback and contributions are welcome.
The text was updated successfully, but these errors were encountered: