Multiple Scatterer Detection Over Artificial Media Using Sar Tomography and High-Resolution Spectral Estmation Techniques

This paper addresses multiple scatterer detection over man-made areas using SAR tomography. Diverse high-resolution tomographic estimators are compared, showing that stochastic maximum likelihood and signal subspace fitting techniques are more efficient and accurate than deterministic maximum likelihood technique when estimating coherent scatterers. Based on these two techniques, detection schemes are proposed and their effectiveness for scatterer detection is demonstrated by using multi-baseline L-band SAR data over a test site containing man-made objects .