A current trend in information fusion involves distributed methods of combining both conventional "hard" sensor data and human-based "soft" information in a manner that exploits the most useful and accurate capabilities of each modality. In addition, new and evolving technologies such as Flash LIDAR have greatly enhanced the ability of a single device to rapidly sense attributes of a scene in ways that were not previously possible. At the Pennsylvania State University we are participating in a multi-disciplinary university research initiative (MURI) program funded by the U.S. Army Research Office to investigate issues related to fusing hard and soft data in counterinsurgency (COIN) situations. We are developing level 0 and level 1 methods (using the Joint Directors of Laboratories (JDL) data fusion process model) for fusion of physical ("hard") sensor data. Techniques include methods for data alignment, tracking, recognition, and identification for a sensor suite that includes LIDAR, multi-camera systems, and acoustic sensors. The goal is to develop methods that dovetail on-going research in soft sensor processing. This paper describes various hard sensor processing algorithms and their evolving roles and implementations within a distributed hard and soft information fusion system.