No abstract

We present a sequential framework for change detection. This framework allows us to use multiple images from reference and mission passes of a scene of interest in order to improve detection performance. It includes a change statistic that is easily updated when additional data becomes available. Detection performance using this statistic is predictable when the reference and image data are drawn from known distributions. We verify our performance prediction by simulation. Additionally, we show that detection performance improves with additional measurements on a set of synthetic aperture radar images and a set of visible images with unknown probability distributions.

No abstract

No abstract

This paper presents an algorithm for detecting handguns in terahertz images.Terahertz radiation is capable of penetrating certain materials which are opaque at optical wavelengths, such as clothing, without the harmful effects of ionizing radiation.The approach taken is to segment objects of interest and classify them based on shape.We use a modified version of an active contour algorithm found in the open literature.Modifications include: a pre-processing step that includes clutter filtering and seeding of an initial contour; and a post-processing step that removes clutter pixels from the segmentation. The features used for classification are moment-based and Fourier shape descriptors. Classification as handgun or non-handgun from these features is done via Fisher's linear discriminant.iii