Bivariate Gamma Distribution for Wavelength-Resolution SAR Change Detection

• Published in: IEEE transactions on geoscience and remote sensing Vol. 57; no. 1; pp. 473 - 481
• Main Authors: Vu, Viet Thuy, Gomes, Natanael Rodrigues, Pettersson, Mats I., Dammert, Patrik, Hellsten, Hans
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Bessel functions; Bivariate; Bivariate analysis; Bivariate gamma; Bivariate gamma distribution; CARABAS; Change detection; Clutter; Detection; Distribution; False alarms; Hypotheses; Parameter estimation; Parameter modification; Probability density function; Probability density function (pdf); Probability density functions; Probability distribution functions; Probability distributions; Probability theory; Radar imaging; Random variables; Resolution; SAR (radar); Scale and shape parameters; Shape; Statistical analysis; Statistical hypothesis test; Statistical tests; Statistics; Surveillance; Synthetic aperture radar; synthetic aperture radar (SAR); Synthetic aperture radar (SAR) images; Tracking radar; Wavelength
• ISSN: 0196-2892; 1558-0644; 1558-0644
• DOI: 10.1109/TGRS.2018.2856926

A gamma probability density function (pdf) is shown to be an alternative to model the distribution of the magnitudes of high-resolution, i.e., wavelength-resolution, synthetic aperture radar (SAR) images. As investigated in this paper, it is more appropriate and more realistic statistical in comparison with, e.g., Rayleigh. A bivariate gamma pdf is considered for developing a statistical hypothesis test for wavelength-resolution incoherent SAR change detection. The practical issues in implementation of statistical hypothesis test, such as assumptions on target magnitudes, estimations for scale and shape parameters, and implementation of modified Bessel function, are addressed. This paper also proposes a simple processing scheme for incoherent change detection to validate the proposed statistical hypothesis test. The proposal was experimented with 24 CARABAS data sets. With an average detection probability of 96%, the false alarm rate is only 0.47 per square kilometer.