Covariance-based completion of partial Mueller matrices with one missing polarization row or column using the indices of polarimetric purity
A method is presented for completing Mueller matrices with one unmeasured row or column, excluding the first row and first column. Every covariance-compatible set of twelve measured entries admits a completion of covariance rank at most two, reducing hierarchical maximization of the indices of polarimetric purity to an algebraic selection. A convex preliminary stage corrects incompatible measurements when necessary. The completed representation enables the use of descriptors and decompositions developed for complete Mueller matrices, including in polarimetric imaging. These quantities characterize the selected representative, not necessarily the unknown sample. Python and MATLAB implementations are provided.
Authors
- Ignacio San José
- Jose Gil
Publication Details
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1364/opticaopen.35004595.v1
- Primary Topic
- Optical Polarization and Ellipsometry
- Type
- preprint