hv_image_processing

Repository to process images from the KIT HV electrode test setup. For more information, see the wiki page

Basic usage

Download

• Make sure you have access to this repository: your github account needs to be a member of KIT-Dark-Matter and you need an associated ssh key
• Open your favorite command line interface in your workspace
• Type git clone [email protected]:KIT-Dark-Matter/hv_image_processing.git
• cd hv_image_processing
• Open the files run_handler.py and image_handler.py in your favorite python IDE and change the two global path variables to make them point to the directories cache and run_data of this module (Complaints about having to do this and suggestions on how to do this better will not be accepted.)
• Now every one of the notebooks in the notebooks directory should work out-of-the-box. If not, you might not have access to the data directories. See the data structure section

Naming conventions

The process of ramping up the voltage, observing some glow or sparks or breakdowns and then ramping the voltage back down is called a run. One run ideally has the same conditions in terms of lighting and Ar environment. Multiple runs from the same day can be collected in a test measurement, or measurement campaign. The runs in one campaign dont have to have the same conditions, but in general they will at least share a wiki page.

During one run there will be several pictures taken. The contour finding procedure in this module tries to detect areas of increased brightness, in the ideal case areas where the mesh glows from electron emission. These areas are called clusters. A cluster has an area and a perimeter. These are the number of pixels contained in the cluster, and the cluster edge respectively. A cluster also has a mass, that is the sum of all pixels within the cluster.

Config files

The config files are meant to enable loading and processing an entire measurement campaign without any additional input. They are JSON files and examples can be found in the configs directory. They currently have these keys:

• id: A unique identifier for this campaign. If it is not unique, you risk overwriting other data, or loading wrong data from cache.
• date: The date of this campaign in the format YYYY-MM-DD
• image_path: The absolute path to the images. If the images of individual runs are in separate subdirectories, this should be the parent of those subdirs.
• image_subdirs: In case runs are separate, add the relative path from image_path as a list (same length as number of runs)
• bright_images: Sometimes we want to plot features on a bright image with the same positional cofiguration. These are the images that will be used. List of length 1 or same length as number of runs.
• hv_path: Absolute path to the corresponding HV output.

Data structure

Where to find the data?

The data is (at time of writing this) stored in Y:\Cathode\LocalHVTest. If you dont have access to this directory, ask someone for help (I forgot how to get access for this, ask Vera).

I made a new test measurement, how should I store the data?

Somehow similar to the previous campaigns: Historically grown, the folder structure like this

• Main directory for this campaign. Examples are SpareMesh or SpareMesh_2nd_230502.
  • hv_output_data: the folder where we put the HV data
    • Individual HV files. Some of the HV files will have headers and others will not. There is no pattern to this and I have not yet made the run handler be able to load both. If a HV csv file still has its headers, delete them or adjust the run handler to ignore headers where present.
  • Directory for NIKON camera images (the jpg ones)
    • Misc diretories for calibration, for edited files, for everything you want
    • One directory per run. You can call them what you want. The config file for the campaign just has to reflect what you chose
  • Directory for CANON camera images (the cr2 raw ones)
    • Same as the NIKON directory

Module overview

processing.py

Contains all the functions to process images, like brightness and contrast adjustment or scaling.

image_handler.py

Contains the class and functions to work with individual images.

run_handler.py

Contains the classes and functions to work with runs and campaigns.