Alin Achim | PubMed Health

PubMed2006

SAR image filtering based on the heavy-tailed Rayleigh model.

Synthetic aperture radar (SAR) images are inherently affected by a signal dependent noise known as speckle, which is due to the radar wave coherence. In this paper, we propose a novel adaptive despeckling filter and derive a maximum a posteriori (MAP) estimator for the radar cross section (RCS). We first employ a logarithmic transformation to change the multiplicative speckle into additive noise. We model the RCS using the recently introduced heavy-tailed Rayleigh density function, which was derived based on the assumption that the real and imaginary parts of the received complex signal are best described using the alpha-stable family of distribution. We estimate model parameters from noisy observations by means of second-kind statistics theory, which relies on the Mellin transform. Finally, we compare the proposed algorithm with several classical speckle filters applied on actual SAR images. Experimental results show that the homomorphic MAP filter based on the heavy-tailed Rayleigh prior for the RCS is among the best for speckle removal.

Algorithms↗

PubMedDate not supplied

QT & RR variability spots the earliest autonomic deregulation in diabetes. Fading of vagal sino-atrial drive but not of sympathetic ventricular responsiveness to life challenges.

27 consecutive insulin-dependent diabetic patients (pts), under 50 years, with blood glucose controlled within normal limits and no significant or multiple cardiovascular/neurological complications in the lights of clinical tests, went through a protocol as follows: laiddown at relaxed rest for 10 min, then stood-up quietly for 7 min, and finally experienced a stress-interview for 10 min while supine. A thoracic ECG lead was digitized at I ms (Codas, Dataq Instr.), RR and QT intervals were software-detected, resampled at 500 ms, and Fourier-transformed over 3 min epochs to get auto-or cross-spectra. RR-by-QT mean square coherence detached the RR-independent fraction of QT low fequency (LF) spectral power, called idioventricular QT-LF. We detected autonomic impairment of three types (discriminant score = 92.31%), presumably differentiated upon the locus of lesion, using RR's basal variance and mean RR shortening when standing as follows: (I) RR shortening > 200 ms in 10 pts; (II) normal RR shortening but no RR variance in 4 pts; (III) stiff RR around 600 ms and no RR variance in 2 pts. The above pts have been excluded from further analysis. The remaining 11 pts with no such impairments (5M and 6F, 36.4 y +/- 4.4 SD, history of 6.0 y +/- 5.2) have been compared with 11 normal subjects in an age and gender-paired control group in two steps. Step 1: Preliminary MANOVA/ANOVA showed significant effects on the ensemble of spectral variables of every single factor (status: normal or patient group; intervention; gender) with no significant factor interactions. Significant effects of intervention or status on main RR spectral variables and on a few QT spectral variables were also documented. Step 2: Non-parametric tests showed that diabetics had (mildly to moderately) shorter mean RR, while their RR-LF was always significantly lower than those found in normals--a difference propagated to QT-LF but not to idioventricular QT-LF. In the intra-group there were similar responses to interventions except stress with respect to mean RR. Consistent reduction in RR-LF under moderate or no change in mean RR suggests vagal down- regulation that, judging by idioventricular QT-LF showing, goes perhaps before a similar process with sympathetic control of ventricles. This phase delay may introduce an early arrhythmic risk worth dealing with in secondary prevention.

Adult↗