It is widely acknowledged that tree roots and other forms of buried biomass can have an adverse effect on the performance of ground-penetrating radars (GPRs). In this paper, we present analyses that examine that effect for ground-contacting GPR systems. A test site containing extensive root infiltration at Eglin Air Force Base, Florida, was excavated, and the root structure and soil were thoroughly characterized. A numerical simulator based on the discrete dipole approximation, which is an integral-equation-based method, was developed, validated, and subsequently used to compute scattering from root structures modeled by an ensemble of buried cylinders. An examination of the results is presented that quantifies the potential for false alarms and increased clutter due to buried roots.

A numerical simulation is carried out to study the effect of the thin metal outer case of an antitank mine and the top air gap of an antipersonnel mine on the passive infrared imaging signature. In addition, an antipersonnel surface mine is also analyzed in the present investigation to show its effect on the soil thermal content. The effect of short- and long-wavelength radiation as well as the convective heat transfer is incorporated in this analysis. The temporal development of the temperature distribution over a diurnal cycle is presented for both buried mines. The results show that the thin metal outer case of a buried antitank mine and the top air space of a buried antipersonnel mine have a pronounced effect on the depthwise temperature through the soil. Also, the results show that both buried mines have a noticeable effect on the intensity of the landmine signature on the soil-top surface over a diurnal cycle. A nonexisting mine signature on the soil-top surface is established for an antitank mine with a thin metal outer case. An almost nonexistent signature is also in evidence for the antipersonnel mine with or without an air gap. The results of the present investigation show that the thermal signature of a surface mine produces much larger temperature extremes than the thermal signature of a buried mine. These results play an important role in producing more effective techniques for mine imaging detection.

We present and compare methods for feature-level (predetection) and decision-level (postdetection) fusion of multisensor data. This study emphasizes fusion techniques that are suitable for noncommensurate data sampled at noncoincident points. Decision-level fusion is most convenient for such data, but it is suboptimal in principle, since targets not detected by all sensors will not obtain the full benefits of fusion. A novel algorithm for feature-level fusion of noncommensurate, noncoincidently sampled data is described, in which a model is fitted to the sensor data and the model parameters are used as features. Formulations for both feature-level and decision-level fusion are described, along with some practical simplifications. A closed-form expression is available for feature-level fusion of normally distributed data and this expression is used with simulated data to study requirements for sample position accuracy in multisensor data. The performance of feature-level and decision-level fusion algorithms are compared for experimental data acquired by a metal detector, a ground-penetrating radar, and an infrared camera at a challenging test site containing surrogate mines. It is found that fusion of binary decisions does not perform significantly better than the best available sensor. The performance of feature-level fusion is significantly better than the individual sensors, as is decision-level fusion when detection confidence information is also available ("soft-decision" fusion).

Ka-band (27 GHz to 36 GHz) radar images of a 2.5-m red maple tree, measured in the Ohio State University ElectroScience Laboratory (ESL), Columbus, compact range, are presented. Radar images of the tree with and without leaves were obtained, as well as images of a 6 in diameter metal sphere obscured by the tree for investigation of attenuation effects. The large bandwidth (9 GHz) of these measurements provides a cm scale range resolution, so that backscatter from individual components of the tree canopy can be separated. In addition, measurements made as a function of azimuth angle in 0.1 degree steps allow inverse synthetic aperture radar (ISAR) images to be obtained with cm scale resolutions in down and cross range. A single scattering model, in which tree branches are represented as a deterministic collection of cylinders and leaves as a distributed collection of curved dielectric sheets, is combined with a detailed measurement of the tree geometry to provide theoretical ISAR images for comparison with the measured data. Although incomplete knowledge of tree geometry limits a direct quantitative comparison of the measured and modeled returns, a qualitative comparison of images shows that single scattering models may be sufficient for obtaining many features of foliage objects in Ka-band radar images.