Sensor and Detection Algorithm Based Clutter Metrics

Clutter metrics we important image measures for evaluating the expected perfomance of sensors and detection algorithms. Typically, clutter metrics attempt to measure the degree in which background objects resemble targets. That is, the more Target-like objects in the background the higher the clutter level. However, it is critically important that the effects that the sensor system and the operating characteristics of the detection algorithm be included in the measure of clutter. For example, clutter to a coarse resolution sensor coupled with a pulse thresholding detection algorithm is not clutter to a second generation FLIR with a man-in-the-loop. We propose to build on the present state-of-the-art, using existing first and second order clutter metrics and their respective perfomance studies and evaluations to derive a new class of clutter metrics which explicitly use characteristics of the sensor and detection algorithms. We believe that this approach is somewhat unique and will lead to robust and accurate classes of clutter metries suitable for sensor and detection algorithm evaluation. The technology and methodology derived from this program of work will be used to develop sensor based texture and clutter metrics for robotic and remote sensing applications. Identification of clutter is important for robotic and machine vision applications where the rejection of clutter or the quantification of clutter levels is important.