DCT-FY08 Target Detection Using Multiple Modality Airborne and Ground Based Sensors.

摘要

Automated 3D modeling of building interiors is useful in applications such as virtual reality. We have developed architecture and associated algorithms for fast, automatic, photo-realistic 3D models of building interiors. We have developed an ambulatory human operated backpack system made of a suite of sensors such as laser scanners, cameras, orientation measurement units (OMU)s which are used to both localize the backpack, and build the 3D geometry and texture of building interiors. We have developed a number of localization algorithms based on merging laser, camera and OMU sensor information, and compared their performance using a high end IMU sensor which serves as the ground truth. We have shown our average position error for a kilometer path to be 50 cm or 0.1%. Once the backpack is localized, a 3D point cloud can be generated and 3D meshing or plane fitting algorithms are applied to generate texture mapped 3D models. We have generated 3D models for multiple floors of 8 buildings over the past few years.