Repeat station imaging: An approach to highly-accurate and time-sensitive change detection

Authors: Douglas Stow*, San Diego State University
Topics: Remote Sensing, Hazards, Risks, and Disasters, Land Use and Land Cover Change
Keywords: remote sensing, aerial imaging, image registration, change detection
Session Type: Paper
Day: 4/12/2018
Start / End Time: 5:20 PM / 7:00 PM
Room: Bayside C, Sheraton, 4th Floor
Presentation File: No File Uploaded

Repeat station imaging (RSI) is an approach to multi-temporal image change analyses that minimizes misregistration effects and enables real-time (i.e., on-board) change detection. RSI involves returning the same sensor to the same airborne imaging stations (defined by specific horizontal and vertical positions) over time, and geometrically processing images on a frame-by-frame basis to maintain the benefits of replicated view geometry. This means that baseline (time-1) imagery and flight line and image capture station locations must be available prior to capturing imagery at later points in time. Platform navigation and location-based camera triggering is enabled by global navigation satellite systems. Replicated view geometries mean that automatically generated match points and fairly simple geometric transformation are adequate for achieving image registration near the pixel level, and in near real-time. Anomaly detection algorithms can then be applied to distinguish changes of interest from other changes and radiometric noise.

The goal of this presentation is to provide theoretical and technical background of RSI, demonstrate its implementation for a range of geographic applications, and conclude with a vision of the follow-on research that is required to move RSI to operational status for most applications. Theoretical and technical elements of flight planning, image capture, geometric and radiometric pre-processing, information extraction, and data and information delivery will be covered. Geographic applications that will be covered include post-hazard damage assessment, law enforcement surveillance, monitoring vegetation growth/decline, and measuring wildfire spread. Many of the follow-on research requirements pertain to integration of RSI with small unmanned aerial systems.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login