Challenges on the Applicability of Adaptive Relevance Vector Machine for Image Reconstruction in Soft-Field Tomography
| Autor: | Fernando L. Teixeira, Daniel Ospina Acero, Qussai Marashdeh |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computer science
Posterior probability ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 02 engineering and technology Iterative reconstruction Bayesian inference 01 natural sciences Field (computer science) Relevance vector machine Statistics::Machine Learning Entropy (classical thermodynamics) 0202 electrical engineering electronic engineering information engineering Entropy (information theory) Entropy (energy dispersal) Entropy (arrow of time) business.industry Entropy (statistical thermodynamics) Covariance matrix 020208 electrical & electronic engineering 010401 analytical chemistry Pattern recognition 0104 chemical sciences Artificial intelligence Tomography business Entropy (order and disorder) |
| Zdroj: | 2020 IEEE SENSORS. |
| DOI: | 10.1109/sensors47125.2020.9278898 |
| Popis: | Relevance Vector Machine (RVM) is a machine learning technique relying on Bayesian inference that can be used to solve tomography image reconstruction problems under a probabilistic framework. By highlighting discrepancies between entropy estimates and inaccuracies of the posterior distribution covariance matrix estimates, we demonstrate how the adaptive RVM framework does not offer reliable and consistent operation for certain soft-field tomography problems, herein exemplified by electrical capacitance volume tomography. This has important consequences on the practical applicability of RVM for image reconstruction problems involving such sensor modalities. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|