Maximum likelihood estimation as a general method of combining sub-aperture data for interferometric testing

Peng Su, Jim Burge, Robert A. Sprowl, Jose Sasian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

39 Scopus citations


Interferometers accurately measure the difference between two wavefronts, one from a reference surface and the other from an unknown surface. If the reference surface is near-perfect or is accurately known from some other test, then the shape of the unknown surface can be determined. We investigate the case where neither the reference surface nor the surface under test is known. By making multiple modulated measurements where both surfaces are translated and rotated, we obtain sufficient information to reconstruct the figure of both surfaces. We have developed software that provides a maximum likelihood estimation of both surfaces, as well as an assessment of the quality of the reconstruction. This was demonstrated for the measurement of a large flat mirror, using a smaller reference mirror that has significant shape errors.

Original languageEnglish (US)
Title of host publicationInternational Optical Design Conference 2006
StatePublished - 2006
EventInternational Optical Design Conference 2006 - Vancouver, BC, Canada
Duration: Jun 4 2006Jun 8 2006

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume6342 II
ISSN (Print)0277-786X


OtherInternational Optical Design Conference 2006
CityVancouver, BC


  • Absolute surface shape metrology
  • Interferometry
  • Maximum likelihood estimation
  • Modulated sub-aperture interferometric testing
  • Optical testing
  • Sub-aperture testing

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering


Dive into the research topics of 'Maximum likelihood estimation as a general method of combining sub-aperture data for interferometric testing'. Together they form a unique fingerprint.

Cite this