See Better by Moving: Robust In-Situ Specular Surface Roughness Estimation Using Active Camera Motion

Shiho Inaba, Toshiaki Tsuji

Keywords: Multi-view Vision, Robotic Polishing, Surface Roughness Estimation, In-Situ Measurement

Abstract:

This study proposes an in-situ multiviewpoint surface measurement approach for a polishing robot using fringe-pattern illumination.
By exploiting the angular dependence of specular reflection, we utilize a weighted average, $C_{w\mathrm{\_avg}}$, which quantifies viewpoint reliability using a mask image.
This enables robust surface roughness estimation against ambient light fluctuations, even with strictly limited calibration data.
Moreover, it demonstrated a separability ($J = 62.61$) approximately 2.1 times higher than the best fixed viewpoint, quantitatively confirming its robustness against lighting variations. This study aims to realize a fully autonomous mirror-polishing system.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 21