Skip to content
/ KACC Public

Identifying of clouds and aurora in Kiruna All-sky camera pictures.

License

Notifications You must be signed in to change notification settings

magnific0/KACC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

KACC - Kiruna All-sky Camera Cloudiness

Description

The script uses ImageMagick for identifying clouds in pictures of the Kiruna All-sky camera. The script tries to differentiate between clouds and aurora in images and computes a cloudiness fraction which is tested against a threshold value.

Requirements

  • ImageMagick (tested with ImageMagick 6.9.1-2 Q16 x86_64)

Usage

  1. Download a number of images ready for processing.
  2. Place the script into the same folder.
  3. From the terminal execute ./bacc.sh (make sure the script is executable).
  4. The script will output which images are cloudy and which are not and report the cloudiness factor.

Settings

You can change parameters such as the thresold in the script itself.

Threshold

Threshold value above which an image is considered cloudy. Keep in mind that this is a circle in a rectangular black image, therefore a good portion of the image remains black even when fully clouded. Fraction values range from 0.00 to 0.30ish (default 0.0004).

HSV range settings for clouds and aurora

Based on the HSV values clouds and aurora can be differentiated. These values can be obtained by manually inspecting an image using the color picker in any popular image procesing program (e.g., GIMP). Make sure to get the HSL values (not RGB, HSV, HSB etc).

Name Range Aurora Clouds
(H)ue [0,360] [80,95] [20,40]
(S)aturation [0,100] [30,70] [50,95]
(L)ightness [0,100] [10,100] [10,100]

It is adviced to use broad ranges and tweak accordingly by inspecting the intermediate results (result1.png/result2.png).

File pattern

All files in the same folder as the script that match *.JPG (see for f in *.JPG statement) will be processed by the script. Adapt this expression to your needs.

Acknowledgements

Todo

Possible improvements include:

  • Improving the cloudiness score, by correcting for the margins around the circular image.
  • Calcultating an edge cloudiness score, by applying a circular mask (smaller than the picture by an edge distance) and again correcting the score.
  • Better handling of input files through commandline options.

Legal notice

Copyright 2015 by Jacco Geul jacco@geul.net

Licensed under GNU General Public License 3.0 or later.

About

Identifying of clouds and aurora in Kiruna All-sky camera pictures.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages