Multitarget Multiple-Instance Learning for Hyperspectral Target Detection

In remote sensing, it is often challenging to acquire or collect a large data set that accurately labeled. This difficulty usually due several issues, including but not limited the study site’s spatial area and accessibility, errors in global positioning system (GPS), mixed pixels caused by an image’s resolution. We propose approach, with two variations, estimates multiple-target signatures from training samples imprecise labels: multitarget multiple-instance adaptive cosine estimator (MTMI-ACE) spectral match filter (MTMI-SMF). The proposed methods address abovementioned problems directly considering multiple-instance, imprecisely labeled set. They learn dictionary of target optimizes detection against background using (ACE) (SMF). Experiments were conducted test algorithms simulated hyperspectral set, MUUFL Gulfport collected over University Southern Mississippi–Gulfpark Campus, Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) Santa Barbara County, CA, USA. Both real experiments show are effective at learning performing detection.