Skip to content

quickstart.md

Latest commit

 

History

History
56 lines (44 loc) · 2.44 KB

quickstart.md

File metadata and controls

56 lines (44 loc) · 2.44 KB
Raw

Desktop Quickstart

The packaged quickstart produces a first relative-DN Camera characterization without writing Python. It uses a deterministic synthetic campaign containing four dark frames and four flat exposure levels with four frames per level. All frames are 96 x 128 pixels and share one static synthetic sensor.

The dark frames represent shutter-closed readout and expose bias, pixel/row/ column fixed-pattern offsets, lower-right amplifier glow, read noise, and dead/hot pixels. The flat frames represent an illuminated uniform field, not a scene: exposure-dependent photocharge is modulated by vignetting, three soft dust shadows, PRNU, shot noise, and the same sensor defects. Metadata records the frame role and the enabled physical components.

Run the Example

  1. Start DataLab with the Camera & Detector Characterization plugin installed.
  2. Choose Plugins > Camera & Detector Characterization > Open quickstart example.
  3. Confirm replacement if the current workspace already contains objects. The 20 example images are loaded in five groups and selected automatically.
  4. Choose Plugins > Camera & Detector Characterization > Run camera characterization....
  5. Review the dark/flat assignments and recipe parameters, then accept both dialogs.

Titles beginning with Dark are preassigned in the compact Dark frames checklist. Unchecked images are assigned to Flat, so every selected image has exactly one role before execution.

Expected Result

The Signal panel receives one response curve. The Image panel keeps the 20 inputs and receives a mean dark image and a mean flat image. The response curve anchors a table containing the response slope, intercept, temporal noise, linearity residual, maximum unsaturated signal, relative dynamic range, and selected flat exposure.

The example and current workflow are alpha, synthetic, relative-DN tools. They do not claim EMVA 1288 compliance, calibrated radiometry, or a normative camera assessment. The visible structures are pedagogical deterministic surrogates, not evidence that the simulator reproduces a particular camera or optical bench.

Regenerate the Resource

Maintainers may regenerate the native DataLab workspace from the versioned simulator parameters:

$env:PYTHONPATH="src;../DataLab;../Sigima;../guidata;../PlotPy;../PythonQwt"
python scripts/generate_quickstart.py

The generated resource is src/datalab_camera_characterization/examples/camera_quickstart.h5.