1 July 1991 Decomposition and inversion of von Neumann-like convolution operators
Author Affiliations +
Methods are presented for the decomposition of two-dimensional von Neumann-like convolution operators into sums and products of smaller von Neumann-like operators. Consequences of the techniques include the face that ever second-order operator is the sum and product of five or fewer first-order operators. A totally symmetric second-order operator can be written as the sum and product of three or fewer first-order operators. In general, an nth-order von Neumann-like operator can always be written as the sum and product of three lower order operators. The following inversion result is also discussed. If the (circulant) von Neumann mean filter operator is defined on a square coordinate set with n rows and n columns, then it fails to be invertible only if either the integer 5 divides n or the integer 6 divides n. This result provides a partial solution to a question posed by P. Gader.
© (1991) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zohra Z. Manseur, Zohra Z. Manseur, David C. Wilson, David C. Wilson, } "Decomposition and inversion of von Neumann-like convolution operators", Proc. SPIE 1568, Image Algebra and Morphological Image Processing II, (1 July 1991); doi: 10.1117/12.46113; https://doi.org/10.1117/12.46113


Max-polynomials and template decomposition
Proceedings of SPIE (June 30 1991)
Decomposition methods for convolution operators II
Proceedings of SPIE (October 31 1990)
A Model For Parallel Image Processing
Proceedings of SPIE (January 08 1984)
New variants of the POCS method using affine subspaces of...
Proceedings of SPIE (October 31 1992)

Back to Top