In This Paper, We Propose A Multiframe Marked Point Process Model Of Line Segments And Point Groups For Automatic Target Structure Extraction And Tracking In Inverse Synthetic Aperture Radar (ISAR) Image Sequences. To Deal With Scatterer Scintillations And High Speckle Noise In The ISAR Frames, We Obtain The Resulting Target Sequence By An Iterative Optimization Process, Which Simultaneously Considers The Observed Image Data And Various Prior Geometric Interaction Constraints Between The Target Appearances In The Consecutive Frames. A Detailed Quantitative Evaluation Is Performed On Eight Real ISAR Image Sequences Of Different Carrier Ships And Airplane Targets, Using A Test Database Containing 545 Manually Annotated Frames.
