Robustness of optimal binary filters : analysis and design /
all state-specific optimal alters.
| Main Author: | |
|---|---|
| Format: | Thesis eBook |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1999.
|
| Subjects: | |
| Online Access: | Link to OAKTrust copy |
| Summary: | all state-specific optimal alters. applied to image processes that deviate from design are derived. Sufficient conditions are given under computed for various text models: independent sparse conditional probabilities of template transitions criterion, which is mean absolute error for the binary derives bounds for robustness, provides analytic each state and the state possessing maximal robustness estimating model parameters. A global filter is expressing the optimal alter analytically in terms of expressions for measuring filter robustness by first for processes statistically close to the one for which from image data and approximation the robustness by image from an observed image. Optimality is with respect to some error images. Both the ideal and observed images are random introduced that is applied across all states, introduced. Filter design is examined from the is determined. To investigate the problem of designing it has been designed. This problem is crucial for model in which analytical expressions for robustness model is proposed for determination of robust states noise, sparse edge noise, noise under rotation, and of robust binary alters in the Bayesian framework, Optimal binary alters estimates an unobserved ideal aphics. parameter states. Based on the prior distribution of parameterizing the Ideal and Observation random sets particularizes the entire analysis to a sparse noise perspective of robust states of nature. A simplified plays an important role in robustness. By practical application since filters will always be probability laws. The optimal filter is found relative processes. The present work treats the general concept restoration of rotation. It is shown that even when robust when its performance degradation is acceptable robustness (as opposed to robustness by state) is sets and these are governed by parameterized the most robust filter, concepts of the mean the noise is independent of the signal, the signal the robustness of alter design is examined relative to the states, a measure of robustness is defined for to the observed random set. Robustness surfaces are to these laws. Qualitatively, a filter is said to be under the degradation operator transforming the ideal which the global filter is uniformly more robust than |
|---|---|
| Item Description: | "Major subject: Eletrical Engineering". Vita. |
| Physical Description: | xii, 128 leaves : illustrations ; 28 cm. Also available online. |
| Bibliography: | Includes bibliographical references (leaves 126-127). |