Lossless compression of SAR imagery using a multiple-pass gradient adaptive lattice filter

Synthetic aperture radar (SAR) is useful in many applications, including oil slick monitoring, terrain mapping and navigation, and automatic target recognition. Previous technological advances have resulted in data collections over larger areas at higher resolutions, generating massive amounts of data. Many effective compression techniques exist that can reduce the volume of data for storage and transmission with moderate loss of information. However, some SAR applications do not or cannot tolerate any compression loss (interferometry for example). This paper addresses a lossless SAR compression scheme that uses multiple passes of an adaptive filter to decorrelate the data prior to entropy coding.