• Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
    Dedicated community for Japanese speakers
  • 한국 커뮤니티
    Dedicated community for Korean speakers

Automating image segmentations from multiple input and mask files

New Here ,
Oct 18, 2022 Oct 18, 2022

Copy link to clipboard

Copied

Hi all,

 

I am working on some image analyses for my research that are based on colors found in specific parts of flowers. I used the program Ilastik to create segmentations from a large training set of image. As far as I can tell, Ilastik can only output black and white images from the resulting segmentation training.

 

Our pipeline currently looks like this:

The original image 104414921.jpeg

the segmentation showing just the petals104414921_Simple Segmentation1,2.jpg

the segmentation showing all the parts of the flower we want104414921_Simple Segmentation2,3.jpg

Ultimately we just want the petals in one image and the anthers in another preferably as jpgs or pngs (like shown below) but these were created manually using the load selection tool, deleting the selection then saving the file. I tried recording those actions but was unable to get it to automate the selection of the masking file.

104414921_anthers.png104414921_petals.png

The difficult part is automating this across nearly 500,000 images. We have a dedicated computer to do this on we just aren't sure how to automate this. I did come across this script, but have been unable to get it working. https://github.com/MarshySwamp/Combine-Image-Mask-Images-to-PSD 

 

Any help is appreciated.

TOPICS
Actions and scripting , Windows

Views

2

Likes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
no replies

Have something to add?

Join the conversation