The authors propose that hidden Markov models (HMMs) and multiaperture synthetic aperture radar (MASAR) imaging can be used to construct automatic target detection algorithms. Their preliminary studies show that HMM MASAR detection: effectively exploits anisotropic scattering differences between target and multiple clutter classes; is computationally efficient; can be used with either single polarisation or multipolarisation SAR imagery; and can be used with either coherent or noncoherent subapertures. Further, the results indicate that the accuracy of HMM MASAR detection is comparable to other techniques while requiring orders of magnitude less computation. These results suggest that the HMM MASAR detection technique could be effectively deployed in fielded automatic target recognition systems.

Detecting anisotropic scattering with hidden Markov models

In this thesis, a new algorithm to improve automatic target recognition techniques on High Range Resolution (HRR) Profiles is presented and also a number of ways are investigated for target detection using Synthetic Aperture Radar (SAR) images. A new 1-D hybrid Automatic Target Recognition (ATR) algorithm is developed for sequential High Range Resolution (HRR) radar signatures. The proposed hybrid algorithm combines Eigen-Template based Matched Filtering (ETMF) and Hidden Markov modeling (HMM) techniques to achieve superior HRR-ATR performance. In the proposed hybrid approach, each HRR test profile is first scored by ETMF that is then followed by independent HMM scoring. The first ETMF scoring step produces a limited number of " most likely " models that are target and aspect dependent. These reduced numbers of models are then used for improved HMM scoring in the second step. Finally, the individual scores of ETMF and HMM are combined using Maximal Ratio Combining to render a classification decision. Classification results are presented for the MSTAR data set via ROC curves. An ultra-wideband (UWB) synthetic aperture radar (SAR) simulation technique that employs physical and statistical models is developed and presented. This joint i v physics/statistics based technique generates images that have many of the " blob-like " and " spiky " clutter characteristics of UWB radar data in forested regions while avoiding the intensive computations required for the implementation of low-frequency numerical electromagnetic simulation techniques. Comparative results from three SVD-based subspace filtering approaches on target detection algorithms are reported. These approaches are denoted as " Energy-Normalized SVD " , " Condition-Statistics SVD " , and " Terrain-Filtered SVD ". Approaches towards developing " self-training " algorithms for UWB radar target detection are investigated using the results of this simulation process.

Improved target recognition and target detection algorithms using hrr profiles and sar images

Multi-aperture SAR (MASAR) is an extension of conventional SAR imagery that allows anisotropic returns from man-made objects to be exploited for detection. In this paper, we propose a MASAR ATD algorithm based on hidden Markov models (HMMs). We show that this algorithm economically locates anisotropic target returns. Using simulated L-band MASAR imagery containing M35 trucks, generated by the Xpatch-es software written by Loral, we derive HMM structures that efficiently model the sub-aperture radar return trajectories for target, grass, and tree pixels. We compare HMM detection results for the simulated MASAR imagery over a 105 degree angle of integration with two alternative methods: MASAR split-aperture (SA) change detection and conventional SAR two-parameter CFAR detection. To obtain our results we group detected pixels into target-sized clusters using a clustering algorithm. The results show that HMM detection far outperforms CFAR detection while requiring 1/10th as many FLOPS per pixel. Further, HMM ATD is nearly as accurate as SA change detection while requiring less than 1/500th as many FLOPS per pixel. Finally, for a more practical 45 degree angle of integration, we show that HMM detection and SA chnage detection have equivalent performance, while HMM detection requires less than 1/185th as many FLOPS per pixel.© (1995) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Layne R. Flake

Papers

Multiaperture SAR target detection using hidden Markov models