P1: 21.3 Test Util Scripts

[saloynton]

21.3 Test Util Scripts - Insight AND web

• Scenario implemented for 4.3.0

There are 3 scripts that make up Test Util Scripts - Maybe run a couple of times on one client and a couple of times on the other client:

• Channel_Offsets.py
• Combine_Images.py
• Images_From_ROIs.py

* Prerequisite image to import - test_images_good/lsm/2chzt.lsm

1) Channel Offsets

Prerequisite: Image with multiple z sections (multiple time points is a bonus) and 2 or 3 channels available from test_images_good/lsm/2chzt.lsm

• Select single Image in Insight.
• Open Util Scripts > Channel Offsets - 'Image' & ID should be populated
• Choose to combine 2 or more channels from the input image into the output image
  • Specify various X, Y, and Z offsets as desired for each channel
• As a test case, try to include a Channel that does not exist in input image (E.g. Channel 4).
• Leave dataset name blank
• Run the Script
• On completion of the script, confirm it went OK and message is displayed with 'View' link to new Image.
• Click on the info label and view > the text file of the script variables is shown > check this corresponds to the information entered.

If possible, repeat above choosing several Images as input with a mix of images of different dimensions. Add a name for new dataset to put images in. This should create a new dataset.

2) Combine Images

Prerequisites: 2 or more Z-stack images in a Dataset - Each Z-stack should have same X, Y, Z size.
T and C sizes over 1 will be ignored.

• Select Datset or the images within it you want to combine.
• Open Util Scripts > Combine Images
• Follow the url in description to get more info on how to run the script.
• Untick auto define dimensions
• Tick manually define dimensions
• Choose 1 Dimension to stack the planes over. E.g. Channel
• Run the script
• Once completed click 'View' in the activities window.
• Check the image has combined a Z stack from each image into a multi-channel image.

Now repeat this again using 'Auto Define Dimensions' - TODO: Requires test dataset of correctly named images.

3) Images from ROIs

Prerequisite
Multi-Channel, multi-Z image(s) with multiple rectangle ROIs through z and time. ROIs should span a range of Z and/or T sections, including some single-Shape ROIs and some that span the entire Z / T range

• Select an Image(s) with ROIs in Insight - Include an image with no ROIs (test).
• Open Util Scripts > Images from ROI's
• Run the script with the default values.
• On the completion of the script check any messages and links.
• Refresh the data manager to view new images in a new Dataset (named 'From_ROIs').
• Images should be a 'crop' of the original image with all Channels spanning the T, Z range of each ROI.

Now repeat this again, combining a single plane from each ROI into a Z-stack (EM use case).
This is designed to work on a single plane input image, with each ROI having a single rectangle, but it should also work on the multi-dimensional image(s) above. Simply repeat as above, choosing the "Make Image Stack" option.

[mlinkert]

I am not able to VPN in to connect to the Windows server, so testing this is difficult.