We present three-dimensional (3-D) target tracking based on fused radar and infrared (IR) sensor data with the inclusion of target orientation in the measurement vector. We provide the noise statistic of IR-sensor measurements, including target orientation measured from the IR image. The track-to-track fusion with extended Kalman filter is used to combine radar with IR sensor data. In conventional tracking approaches, there is a fundamental limitation in that it is difficult to accurately estimate the current acceleration of the target, even with nearly perfect measurements of range and angle relative to the target. The correlation between target orientation and velocity can be used to overcome this limitation. We evaluate tracking performance to show how much improvement is obtainable through the inclusion of the target orientation in the measurement data for a realistic 3-D scenario.
Target range estimation is traditionally based on radar and active sonar systems in modern combat systems. However, jamming signals tremendously degrade the performance of such active sensor devices. We introduce a simple target range estimation method and the fundamental limits of the proposed method based on the atmosphere propagation model. Since passive infrared (IR) sensors measure IR signals radiating from objects in different wavelengths, this method has robustness against electromagnetic jamming. The measured target radiance of each wavelength at the IR sensor depends on the emissive properties of target material and various attenuation factors (i.e., the distance between sensor and target and atmosphere environment parameters). MODTRAN is a tool that models atmospheric propagation of electromagnetic radiation. Based on the results from MODTRAN and atmosphere propagation-based modeling, the target range can be estimated. To analyze the proposed method’s performance statistically, we use maximum likelihood estimation (MLE) and evaluate the Cramer-Rao lower bound (CRLB) via the probability density function of measured radiance. We also compare CRLB and the variance of MLE using Monte-Carlo simulation.
Target range estimation is traditionally based on radar and active sonar systems in modern combat system.
However, the performance of such active sensor devices is degraded tremendously by jamming signal from the
enemy. This paper proposes a simple range estimation method between the target and the sensor. Passive IR
sensors measures infrared (IR) light radiance radiating from objects in dierent wavelength and this method
shows robustness against electromagnetic jamming. The measured target radiance of each wavelength at the IR
sensor depends on the emissive properties of target material and is attenuated by various factors, in particular
the distance between the sensor and the target and atmosphere environment. MODTRAN is a tool that models
atmospheric propagation of electromagnetic radiation. Based on the result from MODTRAN and measured
radiance, the target range is estimated. To statistically analyze the performance of proposed method, we use
maximum likelihood estimation (MLE) and evaluate the Cramer-Rao Lower Bound (CRLB) via the probability
density function of measured radiance. And we also compare CRLB and the variance of and ML estimation
Moving target tracking in an infrared (IR) image sequence under high clutter and noise power has been recently
under intensive investigation, and the track-before-detect (TBD) technique based-on dynamic programming (DP)
is known to be especially attractive in very low SNR environments (3dB). In this paper we present a novel 3-
dimensioanl(3D) TBD-DP technique using multiple IR image sensors. Our approach, which does not require
a separate image registration step, uses the pixel intensity values read off jointly from multiple image frames,
to compute the merit function value required in the DP process. To overcome the computation burden related
with the 3D TBD-DP process, we also propose a novel technique that progressively changes the resolution or
the level-of-detail (LOD) of the image. And we analysis the detection performance of these algorithm, detection
probability P<sub>d</sub> and false alarm probability P<sub>FA</sub>. Our simulation results demonstrate that the proposed algorithm
has good track detection performance with the computation load of less than an order of magnitude compared
with the straight-forward 3D TBD-DP, not employing the LOD technique.