NcCut

Documentation: NcCut Docs

NcCut is a GUI for getting pixel data from images and data values from NetCDF files along linear transects. This simplifies the analysis of images with linear features where the brightness of the pixel can be used to gauge some physical value. Some examples include measuring sea ice floe concentration or characterizing ice sheet fractures from satellite imagery. For NetCDF files, NcCut serves as an easy way to view datasets quickly along any set of dimensions and to extract data along transects. Using NcCut users can view vertical transect images of their data along a third dimension and the marked out line. In addition, NcCut is designed to make the measurement of linear features as automatic as possible through it's "Transect Marker" tool where linear features can be marked out and have transects automatically be made perpendicular to the feature. Multiple features can be marked at once on a file and saved all together as a “project” that can be uploaded and continued or edited.

Users can display a plot of the data as well as package the data into a downloadable JSON file. When extracting values from the image/dataset, the program uses linear interpolation to interpolate between the values of the pixels to ensure an accurate portrayal of the line drawn.

This is an open source project. For contribution guidelines please refer to the contribution section of the docs.

Installation

1. It is recommended to first create a virtual environment before installing packages on your system to prevent package compatibility issues. From the terminal at your desired directory use:

   • For Linux and Mac:

     python3 -m venv nccut-venv
     source nccut-venv/bin/activate

   • For Windows:

     python3 -m venv nccut-venv
     nccut-venv\Scripts\activate

2. Then install NcCut using PIP:

pip install nccut

3. To run the app execute the following Python code:

   from nccut.nccut import NcCut
   NcCut().run()

4. To exit virtual environment when finished:

deactivate

5. To open the same virtual environment again in the future simply execute:

   • For Linux and Mac:

     source nccut-venv/bin/activate

   • For Windows:

     nccut-venv\Scripts\activate

If you are having trouble installing NcCut please refer to the Troubleshooting section of the docs

Using the App

• If you would like to try out the app there are example image and NetCDF files in the support folder of this repository

  • Relative file paths:

    support/example.jpg\ support/test_img.png\ support/example_4v.nc\ support/example_3d.nc\

  • There is an example project file for upload as well:

    support/project_example.json

Running the App

1. To install the app, refer to installation instructions for your OS
2. Execute the following python code:

   from nccut.nccut import NcCut
   NcCut().run()

Loading a File

1. To load an image file or NetCDF file type the relative or absolute file path to the file into the file entry text box and select "Go"

   • If you are loading a NetCDF file a popup window will appear with additional selections:
     • Select which variable from your file you would like to see.
     • Confirm or adjust which dimensions you would like to use as the X and Y axes for your variable.
     • If your variable has data in a third dimension select the Z dimension and an initial Z value to display. You will be able to toggle between z values later.
     • Select "Go" to load the selected data

2. You can scroll to zoom in and out of the image and can click and drag the image to move it around.

3. From the "View" menu in the settings bar you can rotate or flip the image as well as change graphic settings for the tools.

4. If you are loading a NetCDF file, from the "NetCDF" menu in the settings bar you can change which variable or z value you'd like to see as well as change the color map or contrast of the image.

Tools

• To clear widgets from the viewer at any point click any of the tool buttons

  Transects

• This tool will make multiple transects between two points

1. Hit "Transect" to enter transect mode
2. Click two points you'd like to make a transect between
3. Repeat for as many transects as you'd like
4. When done, select the "Plot" button and a popup will appear with a plot of all transects and downloading options.

Transect Markers

• This tool allows you to click along a feature and have transects automatically made perpendicular to the feature.
• You can mark out multiple features and save them all together as a single "project".
• You can upload previous projects to edit or continue them.

Working on a project

1. Hit the "Transect Marker" button to enter transect mode
2. Click points along the feature you'd like transects of. Dots will appear on either side of the line drawn indicating the start and end of the transects that will be made
   • To change the width of the transects being made you may enter the number of pixels into the "Width" text box. Select "Go" to change the width for all future transects.
3. Select "New Line" to begin a new marker and repeat for as many markers as you'd like
4. When done select the "Plot" button and a popup will appear with a plot of all transects from the first marker and downloading options.
   • If you want to continue working on your project at a later time, select all the transects and save the data to JSON format. Refer to section below for reloading that data.

Uploading a project

1. Load the same dataset/image you worked on previously.
2. Hit the "Transect Marker" button to enter transect mode.
3. Instead of clicking new points, select the "Upload Project" button.
4. Enter the file name of the transect data you saved previously and select "Ok".
5. All markers from the file will load onto the viewer and you can continue working on the project.

For more on these tools please refer to usage section of the docs.