Error Bounds for the Method of Alternating Projections.
Abstract
Given a collection of closed subspaces of a Hilbert space, the method of alternating projections produces a sequence which converges to the orthogonal projections onto the intersection of the subspaces. A large class of problems in medical and geophysical image reconstruction can be solved using this method. A sharp error bound will enable the user to accurately estimate the number of iterations necessary to achieve a desired relative error. The sharpest possible upper bound is obtained for the case of two subspaces, and the sharpest known upper bound for more than two subspaces. Keywords: Alternating projections; Error bounds; Image reconstruction.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 27, 1987
- Accession Number
- ADA176770
Entities
People
- Howard L. Weinert
- Selahattin Kayalar
Organizations
- Johns Hopkins University