Files
aimsight-libredesk/snippets/textes/which-diagnostic-signals-are-available-in-data-exploration.md
T
maximeandClaude Opus 5.5 10ed887417 Noms plus clairs : snippets/, scripts de test numérotés, verifier_ollama.py
- base_connaissances/ → snippets/ (corpus/ → textes/, instructions.md → consignes_redaction.md, decoupage.py → fichiers.py, importer.py → envoyer.py) ; python3 -m snippets.envoyer.
- instance-de-test/ : 1_demarrer.sh, 2_creer_boite_et_cle.py, 3_brancher_ia.py, 4_essayer_brouillons.py, voir_empreintes.sh, connexion_admin.py.
- ollama/verifier.py → ollama/verifier_ollama.py ; tests, tâches VS Code et docs alignés.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-12 10:27:06 +01:00

35 lines
2.4 KiB
Markdown

# Which diagnostic signals are available in Data Exploration?
In addition to the principal displacement values, Data Exploration provides signals describing
target visibility, image quality, registration reliability, and camera movement. These indicators are
calculated by the station and are intended to be interpreted together with the corresponding
images and as trends over time.
The principal signals are:
• Relative Horizontal / Vertical: the main target-displacement results in millimetres, after
calibration and, when enabled, camera-motion correction.
• Raw Horizontal / Vertical: the target position in the image in pixels, before conversion to
millimetres and before camera-motion correction.
• Error Value: the difference between the current target image and its reference after
registration. It is the main general-purpose recognition-quality indicator and is used by the
obstruction filters.
• Sharpness: indicates how concentrated the target image is. A decreasing trend can
suggest defocus, vibration, atmospheric disturbance, or another loss of image quality.
• FWHM: estimates the apparent width of the target image in pixels. An increasing trend can
indicate image spreading caused by defocus, turbulence, motion blur, or vibration.
• Dominance: indicates whether another bright feature is becoming comparable with the
tracked reflector. A value approaching 1 indicates an increasing risk of competition between
image features.
• Peak Intensity: the maximum pixel value in the target image. It helps identify major changes
in illumination or target response, as well as possible saturation.
• Flux: the total target signal after estimating the image background. It is useful for detecting
gradual changes in illumination, reflector condition, alignment, or obstruction.
• SNR: compares the target signal with variations in the local background. A decreasing value
means that the target is becoming less distinct from its surroundings.
• Camera Movement: the global movement estimated from the reference targets and used to
compensate for movement of the camera or its support.
No single metric should normally be treated as a universal pass/fail criterion. The combination of
several trends is generally more informative. The Available Signals and Diagnostic Metrics
section of the station manual provides the complete definitions and interpretation guidance.
Source : AIMsight FAQ, rév. 1.5