• backscattering phenomenology
• feature extraction
• target classification

Before designing any target classification system which uses radar information as a basis for classification it is important to understand the link between the physical attributes of the target and the corresponding features in the observed target imagery. An understanding of this nature enables the identification of the image features which provide target discrimination [1]. In this thesis multi-perspective data acquired from an ISAR system will be analysed to understand the relationships between the observed data and the real data. Those data represent the history of range profiles of four rotating target placed on a turntable.