8 September 2006 Multi-object feature detection and error correction for NIF automatic optical alignment
Author Affiliations +
Abstract
Fiducials imprinted on laser beams are used to perform video image based alignment of the beams in the National Ignition Facility (NIF) of Lawrence Livermore National Laboratory. In any laser beam alignment operation, a beam needs to be aligned to a reference location. Generally, the beam and reference fiducials are composed of separate beams, as a result only a single feature of each beam needs to be identified for determining the position of the beam or reference. However, it is possible to have the same beam image contain both the beam and reference fiducials. In such instances, it is essential to separately identify these features. In the absence of wavefront correction or when image quality is poor, the features of such beams may get distorted making it difficult to distinguish between different fiducials. Error checking and correction mechanism must be implemented to avoid misidentification of one type of feature as the other. This work presents the algorithm for multi-object detection and error correction implemented for such a beam line image in the NIF facility. Additionally, we show how when the original algorithm fails a secondary algorithm takes over and provides required location outputs.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Abdul A. S. Awwal, "Multi-object feature detection and error correction for NIF automatic optical alignment", Proc. SPIE 6310, Photonic Devices and Algorithms for Computing VIII, 63100Q (8 September 2006); doi: 10.1117/12.682240; https://doi.org/10.1117/12.682240
PROCEEDINGS
10 PAGES


SHARE
Back to Top