Geometric calibration of an airborne imager by natural angular targets: rainbow, glory, and hot spot - Université de Lille Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Geometric calibration of an airborne imager by natural angular targets: rainbow, glory, and hot spot

Résumé

Among future multi-directional polarimetric spaceborne missions that will advance cloud and aerosol studies, the Multiviewing, Multi-channel, Multi-polarisation Imager (3MI) on the MetOp Second Generation – A (MetOp-SG-A) satellite is a direct successor of the classic and successful design of Polarisation and Directionality of the Earth’s Reflectance (POLDER) sensors. The POLDER design comprises of three essential elements: a wide field-of-view lens, a rotating wheel holding spectral and polarizer filters, and a two-dimensional imaging sensor. This presentation demonstrates a new geometric calibration technique for an airborne prototype of the 3MI and other POLDER-like sensors. Geometric calibration is often challenging for an airborne sensor because it is frequently mounted to and dismounted from the aircraft. The rotation of the instrument with respect to the aircraft reference frame must be calibrated every time when the instrument is reinstalled, but the opportunity for a stable flight leg over ground control points is limited because of weather and air-traffic conditions. In addition, the precise aircraft position is often more difficult to obtain than the precise aircraft attitude. To mitigate this difficulty, our new technique derives the instrument rotation by detecting three natural angular features: rainbow, glory, and hot spot. These features are often observed during scientific data acquisitions over clouds and land surface, thus enable the direct validation of the geometric calibration quality for every eligible image under the actual acquisition condition. We present the automatic detection technique of these angular features in in-flight data and derived rotation angles and errors for the AEROCLO-SA field campaign.
Fichier non déposé

Dates et versions

hal-04548640 , version 1 (16-04-2024)

Identifiants

Citer

Souichiro Hioki, Jerome Riedi, Mohamed Salah Djellalli. Geometric calibration of an airborne imager by natural angular targets: rainbow, glory, and hot spot. Sensors, Systems, and Next-Generation Satellites XXVI, 2022, Berlin, Allemagne. ⟨10.1117/12.2636206⟩. ⟨hal-04548640⟩
9 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More