Detection of buried targets via active selection of labeled data: application to sensing subsurface UXO

When sensing subsurface targets, such as landmines and unexploded ordnance (UXO), the target signatures are typically a strong function of environmental and historical circumstances. Consequently, it is difficult to constitute a universal training set for design of detection or classification algorithms. In this paper, we develop an efficient procedure by which information-theoretic concepts are used to design the basis functions and training set, directly from the site-specific measured data. Specifically, assume that measured data (e.g., induction and/or magnetometer) are available from a given site, unlabeled in the sense that it is not known a priori whether a given signature is associated with a target or clutter. For N signatures, the data may be expressed as {x_i,y_i}_i=1,N, where x_i is the measured data for buried object i, and y_i is the associated unknown binary label (target/nontarget). Let the N x_i define the set X. The algorithm works in four steps: 1) the Fisher information matrix is used to select a set of basis functions for the kernel-based algorithm, this step defining a set of n signatures B_n⊆X that are most informative in characterizing the signature distribution of the site; 2) the Fisher information matrix is used again to define a small subset X_s⊆X, composed of those x_i for which knowledge of the associated labels y_i would be most informative in defining the weights for the basis functions in B_n; 3) the buried objects associated with the signatures in X_s are excavated, yielding the associated labels y_i, represented by the set Y_s; and 4) using B_n,X_s, and Y_s, a kernel-based classifier is designed for use in classifying all remaining buried objects. This framework is discussed in detail, with example results presented for an actual buried-UXO site.