Abstract: Methods for segmenting an ocean bottom area into multiple homogenous, rectangular sub-mission areas and generating new, composite sensor performance functions for each sub-mission area are provided. The method discretizes each voxel into a plurality of equally sized, square range bins and obtains a mean P-of-Y curve for each row and column based on a bottom characterization for each of the voxels. Zone parameters along each direction are iteratively calculated, with adjacent zones merged when their parameters are within predetermined values. Pooled variances are calculated for each direction and a preferred mission direction is chosen based on the direction with the smallest pooled variance.

Abstract: A technique is provided for determining expected maximum probability of detection of targets moving through a sensor grid to maximize performance of the sensor grid by changing the sensor grid variables. Non-spatial variables from a plurality of sensors of the sensor grid are removed to provide that sensor performance of the plurality of sensors is a function of spatial variables. A selection is made for the desired description of flow of targets such as pedestrians through the sensor grid between restricted stochastic flow and unrestricted stochastic flow. The specific technique and related equations for determining expected maximum probability of detection for each sensor depends on whether restricted stochastic flow or unrestricted stochastic flow is selected.

Abstract: A mine line detection method to identify mine-like contacts (MLCs) that fit into mine line patterns dictated by standard mine laying protocol. In general, MLCs define a minefield where the minefield has both mines and false alarms. The method classifies every MLC into one of three classes: fixed—a mine line pattern with fixed mine spacing; random—a mine line pattern with random mine spacing; or false alarm—the MLC does not fit into a mine line pattern and is not considered a mine. The method requires no knowledge of either the sensor probability of detection (Pd) or probability of false alarm (Pfa); nonetheless, it significantly improves both. A nonparametric anomaly detector identifies regions with unusually high MLC density for more detailed analysis. Suspected regions are analyzed for unusually regular patterns of MLCs. Probabilities that the observed target densities and target patterns could occur by chance are estimated.

Abstract: A method for detecting a target partially obscured in the image. Sensor data from an area of concern in converted to image data indexed to represent the area. Test signature data is provided representative of the image of a portion of the type of target of interest. The image data is normalized and filtered using the test signature data to increase the signal-to-noise ratio of portions of the normalized data that approximate the test signature data. The so-filtered data is digitized to a logic level "1" or "0" based upon threshold criteria. The resulting digitized data is convolved with a unity weighting function window to generate correspondingly indexed density data. The density data values whose first conditional probability is at least p times more likely than the second conditional probability are then converted to a logic level "1" while the density data values whose first conditional probability is less than p times more likely than the second conditional probability are converted to a logic level "0".