Dust particles contamination monitoring in the backscattering light experiment for the LISA mission

Introduction In the context of space-based optics, contamination of the optics due to particle deposition is inevitable and constitutes a critical issue. This gets more challenging for the sensitive heterodyne measurements of... [ view full abstract ]

Introduction

In the context of space-based optics, contamination of the optics due to particle deposition is inevitable and constitutes a critical issue. This gets more challenging for the sensitive heterodyne measurements of the Laser Interferometer Space Antenna (LISA), the space-based gravitational wave observatory to be launched in 2034. Therefore, table-top experiments need to be developed for a better understanding of how µm- to mm-size dust particles present on optical surfaces affect LISA measurements.

Methods

We realised an experiment for constant monitoring of dust deposition on a mirror surface via a reflex camera and developed image processing methods to analyse the pictures. The set-up consists of a 12.5mm diameter mirror under test (MUT) placed on a micrometric translation stage and shined from the side (fig.1). We use a remotely controlled reflex camera with a macro objective constantly looking at the MUT and taking pictures. This set-up will be integrated to a homodyne Michelson interferometer designed to measure the coherent backscattering (CBS) from an optical surface with small roughness or contamination. Once integrated, the two experiments will allow to correlate the CBS signal with the physical properties of the dust particles identified on the MUT. 

Dust particles are analysed via image processing techniques in MATLAB [1]. At a first instance, a sparse image composed of compact structures is estimated from the original image (fig.2) using convex optimisation theory and sparse representations framework. The constructed image is then analysed via MATLAB built-in functions where segmentation is performed, followed by the classification of the detected spots (size, position, flux, peak intensity). The smallest spot-diameter identified has a value of 5µm (camera sensor pixel-size is 3.9µm) (fig.3). Spot-sizes range from 5µm to 390µm.

Conclusion

This analysis applied on consecutive images helps to detect the apparition of dust. Improvements to reduce computational time and comparison with other image processing techniques are planned. The first results of the simultaneous measurements of the CBS signal and image processing on the pictures will be presented.

Acknowledgments

This work is co-funded by: RegionPACA, Thales Alenia Space, Observatoire de la Cote d'Azur, CNES, CNRS

Bibliography

1 A Dabbech PhD Thesis https://tel.archives-ouvertes.fr/tel-01191496/document