Multi-platform target detection using multi-channel coherence analysis and robustness to the effects of disparity

The use of multiple disparate platforms in many remote sensing and surveillance applications allows one to exploit the coherent information shared among all sensory systems thereby potentially reducing the risk of making single-sensory biased detection and classification decisions. This paper introduces a target detection method based upon multi-channel coherence analysis (MCA) framework which optimally decomposes the multi-channel data to analyze their linear dependence or coherence. This decomposition then allows one to extract MCA features that can be used to implement a coherence-based detector. This detector is applied to a data set of simulated disparate sonar imagery provided by the Naval Surface Warfare Center (NSWC) - Panama City. This database contains images of both targets and non-targets with various variabilities with respect to resolution, signal-to-noise ratio (SNR), target and non-target types, etc. Sensitivity analyses are then carried out in order to gauge the performance under such variablities that may be encountered in disparate multi-platform detection problems. Performance of the detection method will be given in terms of probability of detection (P_d), probability of false alarm (P_{f a}), and the receiver operating characteristic (ROC) curves.