dataArtist - …scientific data processing made easy.

Download for Windows 7-10

About

dataArtist is a graphical program for interactive data analysis and processing. It is currently specialized image processing tasks in combination with electroluminescence imaging of photovoltaic devices.

It is written in Python (2.7) and is released under open source.

dataArtist is written to be platform independent. It is known to run under Windows 7-10 and Ubuntu Linux 14.10 (soon).

Please cite *dataArtist* as follows:

K.G. Bedrich et al., “Electroluminescence Imaging of PV Devices: Camera Calibration and Image Correction”, Proc. 43rd IEEE Photovoltaic Specialists Conference, Portland, 2016.

Manuals

USER manual

DEVELOPERS manual

Online Tutorials

General usage, camera calibration and image correction are explained in youtube screencast sessions, see

Supported file types

Data is imported through drag n’drop.

1. Images

• Common used (TIF, BMP, PNG, JPG, CSV, TXT, MAT)
• RAW, if data type and image shape are known
• Numpy arrays

1. Plots

• CSV, TXT, numpy arrays

Installation

a) Portable version

• dataArtist runs out of the box. No installation needed (currently Windows only). See section Releases.

b) Installation into existing Python installation using pip

• After installing Python2.7 and ensuring that 'pip' is a recognised command in the command shell (Windows: CMD.exe). you need to install the following packages BEFORE installing dataArtist:

  • PyQt4
  • OpenCV2.4
  • OPTIONAL: RabbitMQ (for inter-program communication)

  Please refer to the USER manual for more information.

• Once these packages are installed, open a command shell and type:

pip install dataArtist

• Now start dataArtist from the command shell:

dataArtist

• To launch dataArtist into the system tray type:

dataArtist -s

• To open a dataArtist session directly type:

dataArtist -o /path/to/my/session.da

Scripting, Automation, Modding

dataArtist comes with a built-in python shell. Data can be examplary accessed though d1.l3 (display 1, data layer 3) and tools e.g. through d.tools['Axes'].click() (in current display execute tool ‘Axes’).

dataArtist allows adding own tools, displays and importers, for examples, see dataArtist/modding.

Main dependencies

Package	Description
pyqtgraph_karl	Scientific Graphics and GUI Library based on Qt (Fork)
imgProcessor	General propose image processing libary
appBase	Base packages for apps, based on Qt
fancyWidgets	A collection of fancy widgets, based on Qt
fancyTools	A collection of useful not-GUI tools

Example: Electroluminescence imaging

Camera calibration

For camera calibration all needed images are dropped into dataArtist and the matching tool is executed. The calibration results are hereinafter saved to a calibration file. The determination of the point spread function is exemplary shown in the following figure:

dataArtist screenshot - toolbar ‘calibration’. a: Best focus determination; b: noise-level-function measurement; c: Dark current mapping; d: Flat field mapping; e: PSF estimation (selected): f: lens distortion measurement

Image correction

The correction of EL image is shown in the following figure. Perspective correction (red box) can be done either using the outline of the PV device (automatically detected or manually defined) or using a reference image.

dataArtist screenshot - a: tool ‘CalibrationFile’; b: tool ‘CorrectCamera’; c: tool ‘PerspectiveCorrection’

green line: Camera correction; red line: Perspective correction