Skip to content

hyperspectral image analysis project - part 2

Notifications You must be signed in to change notification settings

djjayeeta/ML_KGP

 
 

Repository files navigation

ML_KGP

Run the commands in requirements.txt to install the requirements

To cythonize run the command

python setup_seg.py build_ext --inplace

This will generate mrf.so,mrf.c,ss_fuzzy.so,ss_fuzzy.c

Now to run segmentation on a image you need a .pickle of the image which will have a 3 dimensional numpy array. the 3rd dimension denoting the spectral dimension of the image.

To run segmentaion change the following parameters in run_seg.py

  1. algo (can be fuzzy_c/mrf/db_scan)
  2. pickle_file (.pickle file for the input image)
  3. cluster_number (no of classes , will be used in fuzzy_c and mrf )
  4. output_file (file to store the output, in case of mrf and fuzzy_c outputs in form of .jpeg and .pickle will be stored for every iteration,for dbscan final output will be stored)

Now run the following command python run_seg.py

About

hyperspectral image analysis project - part 2

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%