ACT/SGER: Object Identification and Classification in Aerial Images

点击翻译按钮获取中文摘要

AST-0442037BertozziThis project supports research in mathematical algorithms and implementations to represent, manipulate, and analyze data, specifically including feature detection, and registration and characterization of natural images. By building on a class of active contour models known as image snakes, the investigators will develop a fast method for object identification, to be used on large volumes of aerial photographs. Combined with higher order nonlinear PDE-based methods for resolving piecewise linear signals, and with shape and size constraints, the final system should be particularly efficient at locking onto predefined shapes. This interdisciplinary work draws on mathematical ideas from fluid dynamics and incorporates datasets from the intelligence community.This award is supported jointly by the NSF and the Intelligence Community. The Approaches to Combat Terrorism Program in the Directorate for Mathematical and Physical Sciences supports new concepts in basic research and workforce development with the potential to contribute to national security.