Skip to content

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow

License

Notifications You must be signed in to change notification settings

eminalparslan/Mask_RCNN

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Radio Galaxy Morphoogy Classification with Mask R-CNN

This is an application of Mask R-CNN to Radio Galaxy Morphology Classification using data from Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies.

Paper

Presented at the 2020 2nd International Workshop on Signal Processing and Machine Learning (WSPML 2020), November 20-22, 2020.

Published in the proceeedings of the 2020 4th International Conference on Vision, Image, and Signal Processing, ICVISP 2020.

Paper: Radio Galaxy Morphology Classification with Mask R-CNN

Installation

  • Clone repo
  • Download and unzip images from here

Dataset

The top image below is an original flux density image obtained from the open databased maintained by Space Telescope Science Institute (third.ucllnl.org) with linear scaling while the bottom image shows the derived image with log scaling as used in CLARAN and this paper.

Results

Annotated Radio Galaxy images with bounding boxes used for training Mask R-CNN.

Upper left 1C_1P, upper right 1C_2P, lower left 2C_2P and lower right 3C_3P

Classification results with Mask-RCNN for the above images.

Author

Emin Alp Arslan eminalparslan@gmail.com

About

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%