Consider the problem of controlling a network of surveillance sensors that are capable of selecting which areas to observe and which modes to observe these areas. In this paper, we study the problem of controlling the observations of these sensors adaptively in order to classify accurately a collection of objects using information on their observed features. Our proposed approach is modeling objects as templates of 3-D features, and modeling sensors as observing features of individual objects, subject to degradation by noise, obscuration, missed detections and background clutter. We exploit a statistical framework based on random sets similar to those used in multit-arget tracking to model the statistical relationship between observed features and object types to compute information-theoretic estimates of the probability of error in classification. We present a novel approach for computation of these distances between distributions of random sets using k-best assignment algorithms. These estimates are combined with real time information to generate predictions of the information value of individual measurements for sensor management. Using these predictions, we develop assignment algorithms to compute sensor management strategies to minimize this bound. The resulting sensor management algorithms are capable of solving problems involving a large numbers of objects in real-time. We show simulations of the resulting algorithms for classifying 3-dimensional objects from 2-dimensional noisy projections that illustrate how the algorithms select complementary views to overcome obscuration and provide accurate classification. Our results show that our information-based sensor management algorithms achieve comparable classification accuracy to adaptive simulation-based approaches that evaluate the value of information, while requiring nearly five orders of magnitude less computation.