TY - JOUR

T1 - Square root filtering in time-sequential estimation of random fields

AU - Chint, Toshio M.

AU - Karl, W. Clem

AU - Mariano, Arthur J.

AU - Willsky, Alan S.

N1 - Funding Information:
parts by Office ofNaval Research Grant N00014-91-J-1004, National Science Foundation Grant MIP-9015281, and Air Force Office of Scientific Research Grant AFOSR-92-J-0002,
Funding Information:
Drs. Chin and Mariano were supported by Office of Naval Research Grant N00014-91-J-1120 from the

PY - 1993/4/8

Y1 - 1993/4/8

N2 - As a time-sequential and Bayesian front-end for image sequence processing, we consider the square root information (SRI) realization of Kalman filter. The computational complexity of the filter due to the dimension of the problem - the size of the state vector is on the order of the number of pixels in the image frame - is decreased drastically using a reduced-order approximation exploiting the natural spatial locality in the random field specifications. The actual computation for the reduced-order SRI filter is performed by an iterative and distributed algorithm for the unitary transformation steps, providing a potentially faster alternative to the common QR factorization-based methods. For the space-time estimation problems, near-optimal solutions can be obtained in a small number of iterations (e.g. less than 10), and each iteration can be performed in a finely parallel manner over the image frame, an attractive feature for a dedicated hardware implementation.

AB - As a time-sequential and Bayesian front-end for image sequence processing, we consider the square root information (SRI) realization of Kalman filter. The computational complexity of the filter due to the dimension of the problem - the size of the state vector is on the order of the number of pixels in the image frame - is decreased drastically using a reduced-order approximation exploiting the natural spatial locality in the random field specifications. The actual computation for the reduced-order SRI filter is performed by an iterative and distributed algorithm for the unitary transformation steps, providing a potentially faster alternative to the common QR factorization-based methods. For the space-time estimation problems, near-optimal solutions can be obtained in a small number of iterations (e.g. less than 10), and each iteration can be performed in a finely parallel manner over the image frame, an attractive feature for a dedicated hardware implementation.

UR - http://www.scopus.com/inward/record.url?scp=0347055956&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0347055956&partnerID=8YFLogxK

U2 - 10.1117/12.143140

DO - 10.1117/12.143140

M3 - Conference article

AN - SCOPUS:0347055956

VL - 1903

SP - 51

EP - 58

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

T2 - Image and Video Processing 1993

Y2 - 31 January 1993 through 5 February 1993

ER -