Neural networks for linear inverse problems with incomplete data especially in applications to signal and image reconstruction.
Andrzej CichockiRolf UnbehauenMarkus LendlKlaus WeinzierlPublished in: Neurocomputing (1995)
Keyphrases
- image reconstruction
- incomplete data
- limited angle
- neural network
- signal reconstruction
- reconstructed image
- missing data
- bayesian networks
- missing values
- super resolution
- em algorithm
- pattern recognition
- signal processing
- maximum a posteriori
- tomographic reconstruction
- total least squares
- sampled data
- frequency domain
- radon transform
- nearest neighbor
- expectation maximization
- maximum likelihood
- random projections
- high resolution
- training data
- image sequences
- image processing
- computer vision