Vis Ghost Position Model Product¶

Metadata¶

Data product name	DpdVisGhostModel
Data product custodian	VIS
Name of the Schema file	euc-vis-GhostModel.xsd
Processing Element(s) creating/using the data product	Ghost Position Model - VIS Processfield and Ghost pipelines
Processing function using the data product	VIS

Proposed for inclusion in EAS/SAS¶

Yes

Data product elements¶

Header	object of type sys:genericHeader
Data	object of type vis:visFileContainerData
QualityFlags	object of type dqc:sqfPlaceHolder
Parameters	object of type ppr:genericKeyValueParameters

Detailed description of the data product¶

Optical ghosts are artefacts created in VIS images by very bright sources (saturated stars). They are caused by internal reflections of light inside the dichroic plate which separates light between the VIS and NISP instruments. Infrared wavelengths are transmitted through the dichroic plate, while visible light is reflected by the front face of the dichroic plate. But visible light reflection is not perfect and about 1e-6 of it penetrates in the dichroic plate and is reflected by the back face of the dichroic, creating a defocused second light path which produces on the VIS detector a ghost image of the primary mirror obscured by the secondary mirror and its three spider arms.

The output product of the Optical Ghosts Calibration pipeline is a JSON file containing a dictionary with a ‘header’ section, with some metadata like the production date and reference frames used, and a ‘data’ section which contains the actual ghost position model.

The ghost position model consists in two 4x4 matrices, A and B, which are used to produce the X and Y ghost centroid offset relatively to its originating star position.

The star position is first converted from sky coordinates to detector pixels using the detector WCS, then from detector pixels to FPA positions in millimeter using the ‘as required’ FPA geometry stored in the Mission DataBase. These coordinates are then multiplied by the two A and B matrices to produce the X and Y offset in millimeter from the star centroid to the ghost centroid. Finally, these X and Y offsets in millimeter are converted back to pixels in detector to find the actual ghost centroid position in the image.

Details on how to compute and use the A and B matrices can be found in Samuel Ronayette’s note: https://euclid.roe.ac.uk/dmsf/files/4270/view.

The code implementing these operations is located in https://gitlab.euclid-sgs.uk/PF-VIS/VIS_Tasks/-/blob/develop/VIS_Ghosts/python/VIS_Ghosts/VIS_ghost_model.py?ref_type=heads.

Detailed Tech Note can be found: https://euclid.roe.ac.uk/dmsf/files/17192/view.

The VIS ghost position model doesn’t take into account the stars’ spectral energy distribution.

The VIS ghost position model doesn’t implement second order reflections because a star creating a visible 2nd order ghost would be so bright that it would make the whole science image unusable (tested in CSL Thermal Vacuum Cycling Chamber).

A new model is built every month during the visit of the self calibration field (F-001), for monitoring and validation.

Name of file:

EUC_VIS_MDL-GST__[YYYYMMDDThhmmss.s]Z.json

where the date is the production date.