Identification of image and blur parameters in frequency domain using the EM algorithm.
Emin AnarimHakan UcarYorgo IstefanopulosPublished in: IEEE Trans. Image Process. (1996)
Keyphrases
- frequency domain
- em algorithm
- expectation maximization
- space variant
- fourier domain
- gaussian mixture
- maximum likelihood
- cross correlation
- maximum likelihood estimation
- parameter estimation
- fourier transform
- likelihood function
- maximum likelihood estimates
- power spectrum
- expectation maximisation
- mixture model
- mixture distribution
- probability density function
- spatial domain
- discrete fourier transform
- image segmentation
- map estimation
- gaussian mixture model
- image data
- power spectra
- template matching
- image features
- feature extraction
- input image
- density estimation
- multiscale
- probabilistic model
- maximum a posteriori
- denoising
- spectral signatures
- subband
- bayesian framework
- edge detection
- blind deconvolution
- k means
- closed form
- density function
- standard deviation
- feature points
- multiresolution