Warning: pg_query(): Query failed: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 in /dati/webiit-old/includes/database.pgsql.inc on line 138 Warning: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 query: SELECT data, created, headers, expire, serialized FROM cache_page WHERE cid = 'https://www-old.iit.cnr.it/node/1160' in /dati/webiit-old/includes/database.pgsql.inc on line 159 Warning: pg_query(): Query failed: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 in /dati/webiit-old/includes/database.pgsql.inc on line 138 Warning: ERROR: missing chunk number 0 for toast value 29512337 in pg_toast_2619 query: SELECT data, created, headers, expire, serialized FROM cache_page WHERE cid = 'https://www-old.iit.cnr.it/node/1160' in /dati/webiit-old/includes/database.pgsql.inc on line 159 Iterative image restoration with non negativity constraints | IIT - CNR - Istituto di Informatica e Telematica
IIT Home Page CNR Home Page

Iterative image restoration with non negativity constraints

Abstract: In many image restoration applications the nonnegativity of the computed solution is required. General regularizationmethods, such as iterative semiconvergent methods, seldom compute nonnegative solutions even when the data are nonnegative. Some methods can be modified in order to enforce the nonnegativity constraint. Other methods, which can be derived from the Kuhn-Tucker conditions of a constrained maximization problem, naturally embed nonnegativity. In this note we aim to compare the performances of different iterative regularization methods which produce nonnegative images from various point of view, i.e. the computational cost, the efficiency and consistency with the discrepancy principle as standard technique for choosing the best regularization parameter and the sensitivity to this choice. An extensive experimentation on both 1D and 2D images has shown that the most noteworthy methods are truncated CGLS from the point of view of the computational cost and EM for the reconstruction efficiency. Both methods appear to be consistent with the discrepancy principle and not too sensitive to the choice of the number of iterations suggested by this principle.


2006

Autori: Favati P., Menchi O., Lotti G., Romani F.
Autori IIT:

Tipo: Rapporti tecnici, manuali, carte geologiche e tematiche e prodotti multimediali
Area di disciplina: Information Technology and Communication Systems
Rapporti tecnici IIT - 2006-TR-010